Intel's Itanium Will Get x86 Emulation

pissoncutler writes "Intel has announced that they will be releasing a software emulation product to allow 32-bit x86 apps to run on Itanium Processors. According to these stories (story 1, story 2), the emulator is capable of the x86 performance of a 1.5Ghz Xeon (when run on a similar speed Itanium.) Who said that no one cared about x86 anymore?"

Comments

i object !

[ibbie]

I haven't seen enough info on the new IBM PowerPC 970 CPU expected shorty.

watch who you're calling shorty, farm boy.

Re:i object !

[mblase]

watch who you're calling shorty, farm boy.

There, there. Most women will say it's not the size, but what you do with it that counts, anyhow.

Re:i object !

[FroMan]

This is an apple.slashdot.org post, that should be iObject, not i object.

Re:i object !

[Per Wigren]

Most women will say it's not the size, but what you do with it that counts, anyhow.

...and most men will say it's the inside that matters.

Don't believe everything people say!

Re:i object !

[Vegan Pagan]

This is an apple.slashdot.org post, that should be iObject, not i object.

I object!

Stolen, but insightful.

[Anonymous Coward]

http://www.macslash.org/comments.pl?sid=03/04/23/1 82250&cid=3

First off, the disclaimer: this is my pet theory, i.e., a total, wild, pulled-of-out-my-*ss speculation, okay. I have no inside info or contact with people who might know this, but here is my speculation of why this AMD thing keeps coming up despite the fact that the use of the IBM 970 is almost a certainty.

Put this AMD thing into a bigger context of recent events.

We've heard that 10.3 will include a more integrated Classic environment where Classic Mac OS apps will be given many of the benefits of Aqua.

Apple quietly releases their implementation of XWindows system, X11. Despite the fact that this news set Slashdot buzzing for days on end and probably should have had some mention from Steve Jobs in the keynote he'd given a few days prior, it was released very quietly. Interesting.

Next, the somewhat unexpected news that Microsoft was buying Virtual PC. What on earth could Microsoft want with VPC? We can speculate that they want greater control over emulation of Windows on the Mac, but that sounds weak. They still control the operating system that gets installed on VPC so from that perspective they've gained nothing by buying out VPC.

And then these weird, peristent, inexplicable rumors that Apple is in talks with AMD about something or other. Who knows what. It's very doubtful that it's about a chip that would replace the PPC since we've read many, many well-informed examinations of such a move and the technical hurdles would likely ruin Apple.

So what could all this possibly point to? Apple has given us a system that can basically run software from three different operating systems: the classic Mac OS, Mas OS X (the Next OS), and Unix. They recently brought the Unix world closer with the release of X11. Wouldn't it be amazing if hardware in the near-future included an "add-on" chip (something like Altivec that works in conjuction with the PPC processor) that emulated the x86 hardware? Maybe it would give Mac users the ability to run Windows and PC software, not via software emulation, but with hardware assistance. Imagine the interest Apple could draw if they presented the world with a machine that runs the Classic, OS X, Unix and Windows applications... all in one environment and almost seamlessly.

Now does Microsoft buying VPC make sense? Maybe? Maybe not. Maybe MS Mac Business Unit caught wind of this and wants to one-up Apple somehow. Any thoughts?

AMD would be a likely partner is such a move since one could imagine the problems with Intel assisting Apple with this. If it was popular, Intel would be killing their own business. AMD, on the other hand, wouldn't, if I understand the situation correctly.

Anyway... like I said... wild speculation, but that's what all this says to me.

Re:Stolen, but insightful.

[Anonymous Coward]

These same thoughts wandered through my mind and then out again, simply because apple have Been There Done That over and over before. there were PC compatibility cards for Pluses, for the Mac II, for the Quadras, and for PCI PPC macs... none of which were particularly succesful

Then again the fact they've done it so many times before could mean they're likely to bash their head against this particular wall one more time

Re:Stolen, but insightful.

[inertia187]

I like it, I want it, where can I buy it? And can I get it as a PCMCIA card for my TiBook? Would that work??

Wild speculation or not, it's fun to think about.

Re:Stolen, but insightful.

[ruiner13]

"Imagine the interest Apple could draw if they presented the world with a machine that runs the Classic, OS X, Unix and Windows applications... all in one environment and almost seamlessly."

They used to have this... you could buy some of the Performa lines waaaay back in the day with a card that was basically a mini-PC, and had a 586 and later a pentium on it. OrangeMicro used to have a line of em, could even cut and paste between environements, IIRC.

Re:Stolen, but insightful.

[Surak]

So what could all this possibly point to? Apple has given us a system that can basically run software from three different operating systems: the classic Mac OS, Mas OS X (the Next OS), and Unix. They recently brought the Unix world closer with the release of X11. Wouldn't it be amazing if hardware in the near-future included an "add-on" chip (something like Altivec that works in conjuction with the PPC processor) that emulated the x86 hardware? Maybe it would give Mac users the ability to run Windows and PC software, not via software emulation, but with hardware assistance. Imagine the interest Apple could draw if they presented the world with a machine that runs the Classic, OS X, Unix and Windows applications... all in one environment and almost seamlessly.

Ummm...I'm pretty sure Apple already tried this once. They sold some PowerMacs with cards that had 486 processors on them so you could write Windows on it. Wasn't that thing a dismal failure?

Re:Stolen, but insightful.

[lameland]

Sun still makes such a beastie: here. I've used an older model. It looks very similar to running VMware, you see the whole POST and boot screens in a window. Performance was pretty good, the only problem is that the HD is shared, which brought the machine down somewhat.

Re:Stolen, but insightful.

[liquidsin]

Sounds like fun! What I personally would like to see out of this is Apple releasing their own BSD distro. They did a lot of work to get Darwin to where it is. They could be working with AMD to optimize it further for x86 (64 bit?) and gussying up X with some Aqua-alike eye candy. Pretty much OSX without the ability to run Mac apps. I'd buy!

Re:Stolen, but insightful.

[shotfeel]

I'd think more along the line of the PPC processor IBM was rumored to have in the works back in the 601 days that included an X86 compatible core on die (was that the 610?).

That would be cool.

Re:Stolen, but insightful.

[Brandon Sharitt]

I think I remember reading that thread the other day and I have to agree with the poster that brought up the OS/2 comparison. Nobody wrote native OS/2 apps since it could run windows 3.1 apps so seemlessly, so it died(yeah I know it was a bit more complicated but that was a major contributing factor). It's already hard enough to get native ports to the Mac, and it will be even harder if Macs can run Windows apps seemlessly.

Re:Stolen, but insightful.

[truenoir]

PC compatabilty cards were basically PCs on an expansion card, with extra drivers to allow them to share some of the Mac resources (floppy, HD, etc). Orange Micro made plenty for adding on after Apple stopped trying to use them.

You can see specs here http://www.orangemicro.com/productarchive.html

The main problem with these was really the expense. Despite a few tradeoffs, they worked pretty well. Added bonuses like easy file sharing between the "PC" and Mac made life easier if all you were doing was say, testing a web page in Windows. When PCs started becoming cheaper for a whole machine than and OrangePC card, it was merely the space saving that kept them around. Slot based CPUs didn't help either.
Nowdays there's no real point. Get yourself a $400 Dell and a VGA switcher.

Re:Stolen, but insightful.

[tmasssey]

It was the 615, and it never saw the light of day.

The early days of PPC were wild. Apple and IBM working together on hardware and software (Taligent and Pink, some of which got rolled into OS/2's System Object Model). The possibility of running OS/2, Windows *and* MacOS all on the same computer all at the same time via Microkernel... Cool stuff.

A lot of things were attempted but never worked. The 615 is an example: a PPC with a 486 core (IBM has rights to Intel CPU's second only to Intel themselves). The 620 was another: an Itanium-like (without the VLIW) CPU with tons of pipelines and multiprocessor capabilities that never made it into production. Then there's PREP, CHRP, OS/2 for the PowerPC...

1994 was a wild time for vaporware...

Re:Stolen, but insightful.

[Anonymous Coward]

I know everyone always suspects strategic and mysterious ulterior motives for everything Apple does, but the X11 stuff was added with little fanfare because it was an afterthought.

What I've heard, in a nutshell: Steve asked Pixar to try to make a switch to OSX for a portion of their desktops. Porting all their tools to Cocoa wasn't going to be as feasible as they thought, so they needed solid X11 support. An small effort for Apple, a smoother trasition for Pixar, and a freebie win with the *NIX geek community. No secret moonbases.

Re:Stolen, but insightful.

[Glock27]

It's very doubtful that it's about a chip that would replace the PPC since we've read many, many well-informed examinations of such a move and the technical hurdles would likely ruin Apple.

Those "well-informed examinations" were obviously hogwash.

NeXT supported several, quite disparate, computer architectures - x86, HP PA-RISC, SPARC, and 68xxx. Not only did they use different CPUs, but these systems all had different busses and peripherals! Yet, in general, NEXTSTEP/OPENSTEP ran quite well across all those platforms.

Apple could make a very nice transition to Opteron/Athlon64. The should do a G4 emulator, allowing acceptable (if not barnburning) performance for PPC executables, just as they did to facilitate the 68k->PPC transition. The native apps would of course scream - and Apple could immediately offer 4-CPU systems, which it has never had. Those systems could carry big margins.

I also feel Apple should stick with PPC on the notebook side. PPC is well suited to portable applications, and this gives a clear upgrade path for people who really want to keep their current applications. It also allows Apple to hedge it's bets by keeping current on two architectures. If PPC 980 (or whatever) turns out to be a big win over Opteron2, it's not that big of a deal to switch back. I find that scenario unlikely, however.

Note that Apple could still easily stop people from selling "white box" Opterons running MacOS X, if they choose to keep their (poor in my mind) current business model. It could simply produce Apple Opteron motherboards, incorporating a (probably encrypted) Apple ROM, required for Opteron MacOS X to boot. The DMCA takes care of the rest... :-/

If Apple chooses to take that approach, it won't solve some of the bigger problems regarding Apple hardware. For instance, Apple being a single-source supplier, and also limited hardware availability at times. Regardless, Opteron looks like a very good option.

Re:Stolen, but insightful.

[squiggleslash]

I suspect Apple would rather have people take Darwin, add GNUstep or something similar, PPC emulation, and something like Executor's OS, and give people the ability to run Mac apps... without the Mac GUI. (ie with NeXT widgets, or even Windows XP ones.)

Apple's selling point is the design, from the box containing the computer to the scroll bars on the screen. The lack of software support and cross platform compatability is a minus point for them, not a plus.

Re:Stolen, but insightful.

[Razzak]

If switching to an AMD processor would be the death of apple, allowing windows programs to run transparently on OSX would be chopping apple into one-inch bits and sprinkling them throughout the world.

1) Run windows binaries transparently.
2) No more programs designed for OS X
3) Slower, more expensive computers that run the exact same programs as windows
4) No one buys macs anymore
5) PROFIT!

Re:Stolen, but insightful.

[pmz]

Ummm...I'm pretty sure Apple already tried this once. They sold some PowerMacs with cards that had 486 processors on them so you could write Windows on it. Wasn't that thing a dismal failure?

I'm curious why it would fail for Apple but not Sun. Sun, for many years, has sold x86-based motherboards on a PCI or SBus card for running Windows. They integrate nicely with CDE, put Windows where it belongs (in an CDE window and in a flat-file emulated hard drive), allow mounting Solaris UFS directories as network drives, among other useful things.

Quite honestly, I would love for Sun to make a PowerPC card, too, so I could play with Mac OS X. Imagine Solaris 9, Windows 2000, and Mac OS X all in one box. This isn't too far from what Apple would accomplish (classic, OS X, Windows). I've also seen instructions for putting Linux on Sun's PC cards, so it could even be Solaris, Windows, Mac OS X, and Linux all in one enclosure running simultaneously (multiple SunPCi cards can be installed...what fun!).

Re:Stolen, but insightful.

[podperson]

Actually just sticking an x86 processor in there and letting it run VPC and Cocoa threads could be viable, BUT...

I think the AMD HyperTransport chipset + IBM PPC 970 is far more likely.

Apple has already released several Macs with built-in PC hardware to vaguely enthusiastic responses.

Re:Stolen, but insightful.

[mingot]

I disagree. People who ran OS/2 tried to stay away from the WinOS2 box as much as possible. Native applications sold well (considering the userbase). The problem was there were never enough people using the OS to justify large scale development.

Re:Stolen, but insightful.

[Okonomiyaki]

We've heard that 10.3 will include a more integrated Classic environment where Classic Mac OS apps will be given many of the benefits of Aqua. Where did we hear this? I'd be a little surprised and disappointed to find out that Apple has spent their resources enhancing classic. I guess it's fair to expect it to get some modest enhancements with each major system release but is the Classic Environment really that important? Apple has done a better job than I think anyone expected in getting developers to embrace Carbon and Cocoa and as a result Classic has no value to most OSX users. Anyway, I don't think I'm alone in hoping Classic is more of an afterthought in 10.3.

Re:Stolen, but insightful.

[Anonymous Coward]

I remember my high school used to have a few computers with these cards installed. Very few adults, not to mention students, knew about them or how to use them. I remember you just had to press a couple of keys and it would switch right over to the dos prompt and you could start windows. Then you could switch back and forth by pressing the keys, it was fairly impressive(to me at the time). The best was when someone was using one of the computers and you switched it to dos on them, they would freak out.

Re:Stolen, but insightful.

[Chazmyrr]

If you've already spent 15k on a workstation, what's 700 more for a PCI card?

Someone who has spent 3k on an Apple is more likely to spend that 700 on a Dell and stick it on the network.

Re:Stolen, but insightful.

[yuvtob]

Wouldn't it be amazing if hardware in the near-future included an "add-on" chip

The only people that would say that it's amazing are Slashdot users. Most of the people who buy PCs (especially those who buy macs) are not interested in this - all they want is to run Office/iLife/Internet software and they don't care what it is. In fact, I believe that a major part of Apple's success in the last few years is the fact that people don't need to run a special app they bought at the office, but rather create/read documents and surf the web - which can be done on all platforms.
While Apple might use AMD chips (for CPUs, or maybe for something else?) is not farfetched, but it WON'T be to emulate...

Re:Stolen, but insightful.

[0x0d0a]

Wouldn't it be amazing if hardware in the near-future included an "add-on" chip (something like Altivec that works in conjuction with the PPC processor) that emulated the x86 hardware? Maybe it would give Mac users the ability to run Windows and PC software, not via software emulation, but with hardware assistance. Imagine the interest Apple could draw if they presented the world with a machine that runs the Classic, OS X, Unix and Windows applications... all in one environment and almost seamlessly.

Not going to happen.

(1) IP issues.

(2) Reliability issues. It can't work 85% of the time.

(3) Apple wants to keep control over their (quite lucrative) market. Using Windows software takes that away.

(4) Cost. Macs are expensive, and this would make it worse. It's cheap to get an old PC.

Re:Stolen, but insightful.

[Surak]

I'm curious why it would fail for Apple but not Sun. Sun, for many years, has sold x86-based motherboards on a PCI or SBus card for running Windows. They integrate nicely with CDE, put Windows where it belongs (in an CDE window and in a flat-file emulated hard drive), allow mounting Solaris UFS directories as network drives, among other useful things.

The CAD and scientific workstation markets. It used to be that you had to spend $25K on a Sun workstation in order to do high-end CAD or scientific applications. The idea was space savings and to some extent cost savings. You could have your workstation apps and your Microsoft Office running on the same box. Now with desktop PCs rivaling the performance of high-end RISC-based workstations, this is no longer necessary -- all those things can now run on the same processor even. Unfortunately, more often than not, the OS ends up being Windows NT/2000/XP. :(

Re:Stolen, but insightful.

[ToPAz3in6]

This is not any insider info... this is a dammed good educated guess. And if it happens... WOOHOO! Chipzilla and Redmond get a run for their money.

Simply put: Next X-Servers on a CHOICE of x86-64 or 970... while desk/laptops get 970.

All the Unix/Linux/OSS programs now work with OSX when re-compiled or with minimal tweaking... and Apple Instantly gets a huge software base for its servers. (x86-64 for the desktop would not be a big benifit as of yet... they dont need to push it, if a want develops- ok they'll get it later, if not- no worries)

Consequence: Wintel loses ground in the server market.
Period.

Re:Stolen, but insightful.

[MoNkaholic]

The lack of software support and cross platform compatability is a minus point for them, not a plus.

Apple is-say it with me now-a hardware company.

Re:Stolen, but insightful.

[sribe]

Next, the somewhat unexpected news that Microsoft was buying Virtual PC. What on earth could Microsoft want with VPC? We can speculate that they want greater control over emulation of Windows on the Mac, but that sounds weak. They still control the operating system that gets installed on VPC so from that perspective they've gained nothing by buying out VPC.

Oh good grief. The simple fact is that virtual machine technology has gotten hot on Windows servers, and Microsoft was caught without a product, and Connectix had an end-user level product shipping with the server-level version in beta. Something about all those security bugs and DLL and registry conflicts makes virtual servers on Windows even more compelling than on Linux or Solaris. So there's your easy answer, Microsoft wanted the Virtual PC for Windows product, and the (original) Mac version was just part of the package.

Now a little rant, the persistent rumors that Apple will switch to x86 Real Soon Now are the biggest, stupidest, crock of shit I've ever heard. Apple makes the vast majority of their revenue from hardware sales. Now assuming they came out with an x86-based Mac OS that would only run on their own x86 boxes, exactly how long do you think it would take for the patches to appear to get Max OS 86 Edition to run on generic x86 PCs? Poof, there goes 80% of Apple's revenue and the company would be out of business within a year. Failure to notice that problem demonstrates the sheer stupidity of all the pundits who in pushing Apple to do this are essentially saying "I really like the Mac, but I don't want to actually pay for its advantages".

Having gotten that off my chest, I must say that the idea of Apple working with AMD to add something to the 970 to speed up emulation is very intriguing. Running Windows at decent speed on a PPC-based Mac might offer Apple an actual competitive advantage, rather than killing them off. What are the odds? I certainly wouldn't bet on it. But yes, it does at least seem within the realm of possibility.

Or, instead a bunch of wild speculation, why not realize that Apple and AMD are both a part of the HyperTransport consortium and are (presumably) both very interested in 64-bit computing on the desktop...

Yep, trying to leave fantasy aside and live in the real world it does seem like this more boring explanation is most likely...

Actually I believe that the complete opposite is true. Apples market share could go up 10x overnight if they released Mac OSX for x86. Hardware is a tough place to make money, the hardware COSTS money. Lost of it, profit margins are slim.

Yes, but have you actually tried to determine whether or not 10x improved market share for the OS would replace the revenue lost from the hardware? No, of course you didn't because you're a bonehead! Apple's profit margins are not slim; while the rest of the industry struggles to get to or maintain 5%, Apple manages around 25%. And of course "10x overnight" is wildly optimistic to the point of complete absurdity. More likely would be something 25%-50% the first year, 3x within 5 years--MAYBE! I've seen analysis of this option from people who actually understand something about Apple's sources of revenue, and it's pretty scary stuff. (Oh, and if Apple used the x86 then of course their laptops would use the crippled "mobile" versions just like everybody else, instead of being able to use the regular full-speed CPU.)

Re:Stolen, but insightful.

[jkabbe]

One thing you're forgetting is the NeXT basically failed. They were nearly irrelevant.

Remember that the feasibility of something is quite apart from the practicality and advisability of following that course.

Re:Stolen, but insightful.

[pmz]

Someone who has spent 3k on an Apple is more likely to spend that 700 on a Dell and stick it on the network.

Having a PCI card offers several advantages for a workstation: better desktop integration, fewer moving parts, fewer physical enclosures to manage, and fewer monitors and keyboards. For an office/home office desktop, the PCI cards really would be a perfect fit for many people.

Also, the cards don't have to cost $700. Used ones with 400MHz CPUs go for about $100 to $200. From Sun, ones with 733MHz Celerons are $500, and ones with 1600+ Athlons are $700. There are also unofficial ways to upgrade the CPUs (at least on the SunPCi I cards), so they aren't necessarily fixed configurations.

Re:Stolen, but insightful.

[kalidasa]

1. X11 didn't get a lot of press because only nerds care about X11.

2. The new hardware will not be able to boot OS 9 and earlier.

3. Most speculation says that MS bought VPC for the emulation engine for use with multiple virtual machines on one server.

4. Why would I want to cripple my IBM 970 machine (if one were available) by adding an x86 emulation card, when I can do the same thing in software with VPC (since I have to pay the MS tax for the OS anyway, I might as well pay for VPC, too) and not mod the box?

Re:Stolen, but insightful.

[Glock27]

One thing you're forgetting is the NeXT basically failed. They were nearly irrelevant.

Except, that it was bought by Apple and essentially took it over. MacOS X is OPENSTEP with a different GUI.

The challenge for Apple, today, is twofold: how to increase marketshare, and how to offer competitive CPU power.

I'm quite sure Opteron systems (especially if Apple did allow white box competition) would satisfy both goals.

And yes, I mean without sacrificing margins.

Re:Stolen, but insightful.

[jkabbe]

Except, that it was bought by Apple and essentially took it over. MacOS X is OPENSTEP with a different GUI.

I know. The point I was making was that just because you can it doesn't mean you should.

Where would Apple place Opterons on their marketing? Would they market to the high-end desktop crowd? If they did, would the makers of high-end software actually create two versions of their code, one optimized for each platform (Opteron might be fast but it won't have Altivec)? Even assuming that development for the two environments would simply be a checkbox away ("Compile for PPC or Opteron?") Apple would basically be either killing PPC or dooming Opteron to failure. Software companies would probably pick one system to optimize for and ignore the other (heck, many don't even optimize for Altivec now).

And with all this risk, what exactly would the reward be? What is the prize? Avoiding some future that may never come where IBM doesn't produce a fast enough PPC chip? Not worth it. Why kill the company over a pipe-nightmare?

Re:Stolen, but insightful.

[Alan Partridge]

spot on.

I've been a Mac user for over 10 years and therefore have a great many classic apps on my machine. I hardly ever touch them now.

Classic performance is irrelevant to me, I want OSX and HARDWARE performance boosts for my NEXT machine.

Re:Stolen, but insightful.

[Anonymous Coward]

It was the 615, and it never saw the light of day.

The early days of PPC were wild. Apple and IBM working together on hardware and software (Taligent and Pink, some of which got rolled into OS/2's System Object Model). The possibility of running OS/2, Windows *and* MacOS all on the same computer all at the same time via Microkernel... Cool stuff.

A number of prototypes were put together, there were serious performance issues. Making Mac and OS/2 run on the same PPC had two problems, PM used so much shared memory that it was incredibly difficult to make it run on a Mach like microkernel with any speed. Then the other was that switching between little and big endian requires cache flushes that made it difficult to practically run both of them. It'd be nice if a book about that stuff was put together, there are some wickedly important software engineering and business lessons to be learned but most of it was secret.

A lot of things were attempted but never worked. The 615 is an example: a PPC with a 486 core (IBM has rights to Intel CPU's second only to Intel themselves). The 620 was another: an Itanium-like (without the VLIW) CPU with tons of pipelines and multiprocessor capabilities that never made it into production. Then there's PREP, CHRP, OS/2 for the PowerPC...

I don't know of a 615 ever got passed the rumor state. The 620 was created though, it was the first 64bit PowerPC. Not a lot of pipes in it, just a fast 64bit chip, hard to say if it was a failure or an experiment; the people who pour their hearts and souls in to that stuff never speak clearly about those kinds of issues. They made thme, I think Bull may have been the only customer and it was quickly replaced with the 2 "630" processors that were built by the AS/400 group. "Boxcar" and "Northstar" were the code names. Those chips became the first version of AS/400 running no PowerPC like processors. If I'm not mistaken those 2 chips also formed the backbone for the first sysplex processors that are in the zServer family now. It was a hazy time, RS/6000's have different needs than AS/400's that have different needs than the people Intel sells to. There were pushes to make fast integer chips to go head to head with Pentium; there were also real business demands to make high bandwidth chips to satisfy IBM's database and high end market. The whole chip industry of that era kind of split the market that way, the sales types and traderags never understood that. The 620 could have legitimately been a 64bit experiment (mind you alpha was being fabbed with, maybe, 30% of them coming off the line and working.. very low yield) but there was talk and expectations.

Vapor or not, you won't see Apple on AMD any time soon. AMD might fab chips for them but they will be PowerPCs.

Re:Stolen, but insightful.

[Anonymous Coward]

Classic has no value to most OSX users

Yeah, except those OS X users who use QuarkXPress or need to connect to an Exchange server for mail and calendaring (so no, turning on IMAP or POP/SMTP and using Mail.app wouldn't cut the mustard). Almost all of my clients are now in this category. I could not in good conscience let my clients pay the ripoff prices Apple is charging for their G4s that still boot into OS 9, so they got X-only G4s and I thought Classic would work well enough for them to tide them over-- it's only two apps, right? Wrong! They not only crash with amazing frequency, they often can bring down the whole Classic environment in the background, without the Mac even telling me-- I don't notice until I want to switch back to do something in one of them, and notice the "active app" triangle under the dock icon is gone.

I always thought Quark were a bunch of customer-hostile dicks, but lately the amount of grief I've had to endure because their fucking app still isn't OS X-native has changed my opinion: Now I think they're bunch of customer-hostile dicks who need to die, as slowly and painfully as possible. And Microsoft, while the state of their Mac apps has improved dramatically in recent years, deserve to eat shit over not starting an X-native Exchange client for so long. People were clamoring for it since 10.0 was released, for Christ's sake. It was probably a strategic decision, to keep OS X from gaining a foothold in businesses for a while.

Re:Stolen, but insightful.

[DrXym]

While OpenStep lovers might go on about the elegance of the system (and it is elegant), it is certainly not intuitive nor particularly easy to program, especially if you're a conventional programmer used to event based APIs rather than a wacky Objective C message model.

Perhaps with decent help it would be easier, but Apple's Cocoa documentation stinks and the help system seems about on par with the IBM Redbook system I used to program OS/2 ten years ago. In other words it stinks. In the end I bought a book, to learn it; something I've rarely had to do with any programming system.

I think Cocoa programmers must be masochists, or perhaps Apple are sadists.

Re:Stolen, but insightful.

[Alan Partridge]

has it not occurred to you that Apple, in optimising OSX to run on the PPC 970 will also be optimising OSX to run on Power 4 and Power 5? Sure, there's no Altivec support in Power, but a server needs Altivec like I need a bag on my hip.

Re:Stolen, but insightful.

[Chazmyrr]

I'll give you fewer physical enclosures. Not necessarily fewer monitors and keyboards. KVM switch or KVM over IP takes care of that.

I'm on the fence about desktop integration.

Someone who buys an Apple for video editing, graphic design, or publishing should have all the tools they need available without resorting to PC software. If there is a need for some specialized software only available for the PC, it's not clear to me that it would run as well on a card as opposed to an actual PC. If there isn't such a need, why buy the card in the first place?

If you absolutely positively have to be able to run both Mac and PC programs and space is at a premium, why not an XServe and a Dell PowerEdge? Stick them in a closet. KVM over IP to a flat panel, keyboard, and mouse. Don't even need to run Cat5 if you set up a wireless bridge.

Re:Stolen, but insightful.

[Jeremy Erwin]

The major problem with supporting multiple hardware platforms is the fragmentation of the market. Although it is possible to create Fat binaries, containing executables for AMD, PowerPC and SPARC, some publishing houses won't bother to support every architecture.

Re:Stolen, but insightful.

[GlassHeart]

[Apple's] (poor in my mind) current business model.

Niche marketing can be counter-intuitive, but it's also the classic question of whether you want to be a big fish in a little pond, or a little fish in a big pond. Apple chose the former, exerts great influence in its chosen market, and makes a profit in probably some of the worst years ever in the industry.

There are far worse business models, and technically it's nearly impossible to provide the kind of hardware-software integration that Apple currently sells without controlling the hardware. It's not so much they like being a hardware and software company as that they can only distinguish themselves in the market by also selling hardware.

Let's examine your business plan, where:

Apple could make a very nice transition to Opteron/Athlon64. [...] I also feel Apple should stick with PPC on the notebook side.

which in effect nearly triples the development effort for a Mac software vendor. First, you need to build and test an Athlon version (which is not going to be compatible with the Windows version), build and test a PPC version, and then test the PPC emulation version. Thereby making Apple's already small marketshare even more fragmented, when the obvious sensible thing to do is to get a new high end PPC CPU, drop the G3, and improve G4 compiler optimizations.

If PPC 980 (or whatever) turns out to be a big win over Opteron2, it's not that big of a deal to switch back.

That would be plainly insane. Apple's third party software vendors tend to be smaller, and would have a very hard time hopping from platform to platform. Even some big ones have not completed the OS X transition, and you're talking about going to x86 and back?

Re:Stolen, but insightful.

[eMilkshake]

Well, yeah, but that was also the model w/o a CD-ROM eject button and placed the power button directly under the CD-ROM tray.

And what do you think the Windows users did? And why do you think they hated the Mac?

Re:Stolen, but insightful.

[jcr]

What I personally would like to see out of this is Apple releasing their own BSD distro.

And what, exactly, would Apple gain by doing that?

Pretty much OSX without the ability to run Mac apps. I'd buy!

Ok, so there would be one customer. Unless you're willing to spend about $50M for your copy, it's really not a good prospect for Apple to release a BSD with X windows on intel.

-jcr

Re:Stolen, but insightful.

[arlow]

Wouldn't it be amazing if hardware in the near-future included an "add-on" chip (something like Altivec that works in conjuction with the PPC processor) that emulated the x86 hardware? Maybe it would give Mac users the ability to run Windows and PC software, not via software emulation, but with hardware assistance. Imagine the interest Apple could draw if they presented the world with a machine that runs the Classic, OS X, Unix and Windows applications... all in one environment and almost seamlessly.

A nice idea, but there are huge implementation hurdles to acomplishing this:

you either have to

1) run a whole windows os as a process in mac os x a la virtual pc, but this would seem very inelegant to most mac users -- the fact that classic apps still have different looking widgets pisses a lot of UI snobs off; could you imagine how people would react if the entire classic environment had to run in a window??

2) emulate the (enormous) win32 api -- perhaps make a fork up wine project and beef it up, (like theyve done time and again with other open source software,) and then run windows apps through x11/wine. if they could pull this off itd be damn cool. however wine is pretty unreliable as is, and mac users just want their stuff to work out of the box. i really doubt they could pull this off in fewer than 5 years. by then, microsoft will have screwed with the api enough to shaft apple and the wine project.

as much as i wish that apple would take a crack at #2, i am a realist, so i have to face thet fact that its probably not going to happen. oh well.

Re:Stolen, but insightful.

[Graymalkin]

How do you propose anyone make a G4 emulator "scream" on a processor architecture with only 16 GPRs?

Re:Stolen, but insightful.

[slimak]

not likely, remember that apple is a hardware company. Releasing an OS Xish distro for non-apple architecture would only hurt them.

Re:Stolen, but insightful.

[jcr]

One thing you're forgetting is the NeXT basically failed.

Umm... Sold to Apple for $400M? I hope my next venture fails like that..

-jcr

Re:Stolen, but insightful.

[jcr]

I think Cocoa programmers must be masochists, or perhaps Apple are sadists.

Spoken like someone who has absolutely no idea what he's talking about.

If you can't make the shift to OO concepts in programming, then yes: Cocoa's not going to be any fun for you. For those of us who don't like doing the same work over and over again for each app we write, Cocoa is a godsend.

-jcr

Re:Stolen, but insightful.

[Llywelyn]

" The should do a G4 emulator,"

You are losing credability fast here.

Do you know anything about emmulation and the difficulties involved with this or are you just running off on this one?

" and Apple could immediately offer 4-CPU systems, which it has never had"

IIRC they could offer that now with the G4 and they can *certainly* offer it with the PowerPC 970.

"I also feel Apple should stick with PPC on the notebook side. "

Credability is *gone*.

This would mean that they would need an Opteron Emmulator on the mac, or else this would horribly schism the market.

Nevermind, it would horribly schism the market either way--you do realize that emmulation between these is not going to be a walk in the park? Further, getting people to support fat binaries, such as what happened in the 68k switchover, or to recompile and/or *test* for two separate platforms is not going to happen in most development companies?

"f PPC 980 (or whatever) turns out to be a big win over Opteron2, it's not that big of a deal to switch back."

You've hit rock-bottom in your credability and started digging.

" if they choose to keep their (poor in my mind) current business model"

You do realize that they are turning a profit in a failing economy and that they are horribly undervalued at the moment?

"If Apple chooses to take that approach, it won't solve some of the bigger problems regarding Apple hardware. For instance, Apple being a single-source supplie"

You *do* realize that this is intentional on their part, right?

"and also limited hardware availability at times."

I generally don't notice this. I plugged in a wireless--two-button--mouse into an iBook the other day that the PC crowd had just been using for their presentation and it just worked.

" Regardless, Opteron looks like a very good option."

Only if you completely don't understand the way Apple works as a business.

Re:Stolen, but insightful.

[tomdarch]

M$ probably bought VirtualPC in order to sell Windows upgrades. You have to remember that VPC also runs on Windows:

company: No, Ms. M$ salesperson, we have a bunch of apps that will break if we buy your Windows upgrade! We might as well port to Linux at this point.

M$ sales: Ah ha, with Microsoft VirtualPC you can spend more money to run your old version of Windows inside a virtual machine inside the new version that has features you don't need!

company: Oh! In that case we'll take 10,000 licenses!

As usual, the Mac market is tiny compared to the Windows market.

Re:Stolen, but insightful.

[borschski]

Nice logic.

The only place that I'd beg-to-differ with you is on the reasons behind buying Connectix. It makes A LOT more sense that they purchased them due to Windows Server 2003 and the ability to sell complete, stable, turn-key environments (Linux; NT, etc.) that would run seamlessly on WS2003. Virtualization is growing stronger (albeit slowly) and this makes more sense than a Mac play.

BTW, I'm a Mac OS X lover and run VPC with WinXP daily. It's slow but quite usable on a dual 1.42ghz PowerMac with 1.5gb of RAM.

Re:Stolen, but insightful.

[nedron]

Well, OS/2 for the PowerPC wasn't vaporware, as I ran it. And so did several other people at CompuServe, who were all working on the "CompuServe Information Manager for OS/2 for the PowerPC" (how's that for a name?). I also was in Boca Raton on loan from CompuServe working with IBM to fix a couple of IOCTL problems in the kernel.

That's not to say that it ever got released, but it definitely wasn't vapor.

-David

Re:Stolen, but insightful.

[DrXym]

Thanks but I program Objective C and Cocoa so I know exactly what I'm talking about. The documentation stinks, the help system stinks, the dev tools stink. Where is the context sensitive help that the likes of MS DevStudio has had for years. Where is the debugger than can handle (horrors!) multiple threads properly? Where is the stable Interface Builder that doesn't arbitrarily crash?

As I said, I think OpenStep / Cocoa is elegant but it is done no favours by the crappy tools or the crappy documentation.

Re:Stolen, but insightful.

[djward]

Imagine the interest Apple could draw if they presented the world with a machine that runs the Classic, OS X, Unix and Windows applications... all in one environment and almost seamlessly.

No. Here's why. All developers would immediately stop coding native apps for the Mac-PPC. One version release, runs on Wintel and Mac-x86. But no matter how well Apple integrated it, you'd still be running Windows applications, with associated interface and problems inherited from being Windows programs.

Well-written programs, even share/freeware, are a major part of the Mac experience. Look at how much nicer Office for OSX is vs. Office for Windows. Take that away, you end up killing the Mac.

With things like Virtual PC or PC-emulator cards, you pay a price in terms of performance and you are still runnning Windows, so there is a distinction between when you are using your Mac and when you are not. Lose this distinction and the performance hit, developers write one version, the Mac turns into Windows on less-cheap hardware, and it does a painful death.

It would be friggin cool to be able to run my windows progs direct on the Mac, I agree, but it would kill Apple.

Re:Stolen, but insightful.

[tmasssey]

No, that's pretty much the definition of vapor: a product that never shipps. Not that doesn't exist at some level, but never ships.

I've seen OS/2 for PowerPC, though I was not fortunate to have run it myself. However, it never existed as a product; hence, vaporware.

Interestingly, there are a *ton* of products that IBM has built all the way to the point where the are ready to ship, but never shipped them. There's a good reason. When IBM ships something, they then must support it. Nearly forever. The cost of a product is not so much the cost of building it, but the cost of supporting it. That's one reason why OS/2 PPC never shipped.

Of course, there's another reason: by the time it was ready, it was obvious that there wasn't a hardware platform to run it on, and there wasn't going to be a platform. IBM wasn't interested in running OS/2 PPC on RS/6000 hardware: AIX was already there. They wanted it to run on PREP/CHRP PC's: what IBM saw as the next consumer/standard business PC. But by the time OS/2 PPC was ready, PREP and CHRP were dead.

It's too bad. I would love to be typing this on a PPC running OS/2... Instead it's a Thinkpad running Windows 2000... Sigh.

Re:Stolen, but insightful.

[Pharmboy]

Now does Microsoft buying VPC make sense? Maybe? Maybe not. Maybe MS Mac Business Unit caught wind of this and wants to one-up Apple somehow. Any thoughts?

Very interesting. I had not thought of that angle. Now I have to wonder if Apple is going to throw some resources over to the WINE project, making Virtual PC less relevent. I could even see IBM throwing some muscle that direction as well, since IBM has no love for MS. Purely speculation, but remember, IBM and AMD still have a working relationship as well, and obviously IBM and Apple have for years.

If you had all these OS's, with Apple taking the lead on the desktop, and BSD/Linux/Unix on the server end, and the ability to run legacy applications from MS..... The more you think about it, the more curiouser it gets :)

Re:Stolen, but insightful.

[Shuh]

Apple could make a very nice transition to Opteron/Athlon64. The should do a G4 emulator, allowing acceptable (if not barnburning) performance for PPC executables, just as they did to facilitate the 68k->PPC transition. The native apps would of course scream - and Apple could immediately offer 4-CPU systems, which it has never had.

It wouldn't work. The 68k->PPC transition had PPCs emulating 68k faster than 68k could run within a year! This is not possible with x86, especially with the arrival of the PPC 970.

Re:Stolen, but insightful.

[Shuh]

One thing you're forgetting is the NeXT basically failed.

Umm... Sold to Apple for $400M? I hope my next venture fails like that...

Steve Jobs was lucky he still had pull with his old company (Apple Computer) and was able to convince them to pay to bring NeXT (and himself) back on board. NeXT floundered because it got into the x86 space... let's hope Apple knows better.

Re:Stolen, but insightful.

[Shuh]

That would be plainly insane. Apple's third party software vendors tend to be smaller, and would have a very hard time hopping from platform to platform. Even some big ones have not completed the OS X transition, and you're talking about going to x86 and back?

Apple made the 68k->PPC transition look so easy in '94 that people with only a passing interest in Apple think PPC->x86->PPC is a slam dunk! ;)

Re:Stolen, but insightful.

[ruiner13]

If I remember correctly, Cyrix made the "586". The Pentium was Intel's version (although still technically a 586). They were marketed seperately, with Cyrix's part getting the 586 moniker. But i've done lots of drinking since then, I could be wrong (it has happened before).

Re:Stolen, but insightful.

[Glock27]

Niche marketing can be counter-intuitive, but it's also the classic question of whether you want to be a big fish in a little pond, or a little fish in a big pond.

Apple could sell it's OS for x86 boxes at say $300 a pop. If they manage to capture 3 million installs a year (about 6% of the PC market), that would be $900 million in essentially "free money" (compared with selling hardware anyhow). CDs are quite cheap to duplicate.

Apple chose the former, exerts great influence in its chosen market, and makes a profit in probably some of the worst years ever in the industry.

Along with Dell, which has the opposite business model. Don't be so sure something in the middle wouldn't work...

which in effect nearly triples the development effort for a Mac software vendor. First, you need to build and test an Athlon version (which is not going to be compatible with the Windows version), build and test a PPC version, and then test the PPC emulation version.

You should talk to some ex-NeXT developers. Unlike some other environments, things worked very well across architectures. Further, in this situation the various busses and peripherals are identical, simplifying things further.

Thereby making Apple's already small marketshare even more fragmented, when the obvious sensible thing to do is to get a new high end PPC CPU, drop the G3, and improve G4 compiler optimizations.

Sure, if the new high-end PPC CPU is actually competitive with Opteron, both in price and performance. Also it needs to be available in large volume, though the same could be said for Opteron.

That would be plainly insane. Apple's third party software vendors tend to be smaller, and would have a very hard time hopping from platform to platform. Even some big ones have not completed the OS X transition, and you're talking about going to x86 and back?

If circumstances warrant, that is a nice option to have. As I stated, though, I doubt it would happen that way.

At any rate, I think you are vastly overestimating the complexity involved. Cocoa code is very portable.

Re:Stolen, but insightful.

[Glock27]

How do you propose anyone make a G4 emulator "scream" on a processor architecture with only 16 GPRs?

I don't care if it "screams", only that it works at some tolerable pace. Software upgrades to native binaries would be necessary for great performance.

However, with something like a 30-40% CPU power advantage over the 1.25 GHz. G4s, the Opterons should do all right.

Re:Stolen, but insightful.

[York the Mysterious]

This brings back memories of CHRP. I was so pumped up when that was all about to come out. Geez middle school was great. Hehe. Then they had to kill them right when Power Computing had that sweet G3 CHRP design. -Tim

Re:Stolen, but insightful.

[Glock27]

The should do a G4 emulator,

Do you know anything about emmulation and the difficulties involved with this or are you just running off on this one?

What difficulties do you forsee, exactly?

and Apple could immediately offer 4-CPU systems, which it has never had

IIRC they could offer that now with the G4

Then why haven't they? Traditional processors like the G4 need special chipsets to hit 4-way. Opteron doesn't.

and they can *certainly* offer it with the PowerPC 970.

If and when it ships, I suppose you mean. Further, what do you think the cost of the 970 will be versus the Opteron? My guess is that it will be quite a bit more expensive.

I also feel Apple should stick with PPC on the notebook side.

Credability is *gone*.

In your humble opinion. ;-)

This would mean that they would need an Opteron Emmulator on the mac, or else this would horribly schism the market.

Nevermind, it would horribly schism the market either way--you do realize that emmulation between these is not going to be a walk in the park? Further, getting people to support fat binaries, such as what happened in the 68k switchover, or to recompile and/or *test* for two separate platforms is not going to happen in most development companies?

No, it means that software vendors should be strongly encouraged (or forced) to supply fat binaries for the two architectures. This is the same scheme that worked fine for NeXT (a much smaller company!) supporting four architectures at the same time.

If Apple chooses to take that approach, it won't solve some of the bigger problems regarding Apple hardware. For instance, Apple being a single-source supplier

You *do* realize that this is intentional on their part, right?

Intentionally stupid, perhaps. This limits their market tremendously, since many large government and private organizations won't buy single-source equipment, period.

and also limited hardware availability at times.

I generally don't notice this. I plugged in a wireless--two-button--mouse into an iBook the other day that the PC crowd had just been using for their presentation and it just worked.

I was referring to limited availability of Mac hardware either because a) Apple can't get enough of certain components or b) Apple can't manufacture enough to meet demand. White box MacOS X boxes would solve these problems nicely. :-)

Regardless, Opteron looks like a very good option.

Only if you completely don't understand the way Apple works as a business.

Think Different.

Re:Stolen, but insightful.

[Glock27]

Apple could make a very nice transition to Opteron/Athlon64. The should do a G4 emulator, allowing acceptable (if not barnburning) performance for PPC executables, just as they did to facilitate the 68k->PPC transition. The native apps would of course scream - and Apple could immediately offer 4-CPU systems, which it has never had.

It wouldn't work. The 68k->PPC transition had PPCs emulating 68k faster than 68k could run within a year!

It isn't necessary that the emulators run faster, just "nearly as fast". The native binaries would kick.

This is not possible with x86, especially with the arrival of the PPC 970.

You people keep talking like this thing is shipping now. News flash: it isn't.

Even when it does, its price/performance vs. Opteron may be pretty poor.

Re:Stolen, but insightful.

[MsGeek]

AMD made the 5x86 chip, which was basically extended 486 architecture. Made it up to 166MHz. Not a bad little chip, actually. I think it's still being used in the embedded market.

Re:Stolen, but insightful.

[Glock27]

Except, that it was bought by Apple and essentially took it over. MacOS X is OPENSTEP with a different GUI.

I know. The point I was making was that just because you can it doesn't mean you should.

Where would Apple place Opterons on their marketing? Would they market to the high-end desktop crowd?

Their entire desktop and server lineup would switch to Opteron (blade servers, OTOH, might be a good place to offer a PPC option, for lower power consumption).

If they did, would the makers of high-end software actually create two versions of their code, one optimized for each platform (Opteron might be fast but it won't have Altivec)?

Opteron has SSE2, and plenty of software shops are already supporting that. ;-) I'm sure Adobe would be quite happy to ditch Altivec, for instance.

Even assuming that development for the two environments would simply be a checkbox away ("Compile for PPC or Opteron?") Apple would basically be either killing PPC or dooming Opteron to failure.

How so? Doesn't each CPU have it's strengths? If not, why support both? It has to be one way or the other...

Software companies would probably pick one system to optimize for and ignore the other (heck, many don't even optimize for Altivec now).

Perhaps, but given decent tools it doesn't have to be that way.

And with all this risk, what exactly would the reward be? What is the prize?

World class performance, which Apple is far from today. Check the article on Adobe's site benching Photoshop on P4s vs. Macs.

Avoiding some future that may never come where IBM doesn't produce a fast enough PPC chip?

Just out of curiousity, has Apple announced it is using an IBM chip? I hadn't heard that, myself...

Not worth it. Why kill the company over a pipe-nightmare?

Why is this scenario a nightmare? Lose your prejudices (against x86 for instance) for a second, and imagine being able to get a Mac that is as fast as any desktop machine on the planet, faster than any Intel box (for real this time), and competitive on price.

All this would come at the "cost" of vastly increased Apple market/mind share. Just think if software starting coming out for MacOS X before Windows... ;-)

Re:Stolen, but insightful.

[Graymalkin]

The reason I ask is there are absolutely no decent PowerPC emulators for the x86 architecture and figured since you thought it was so easy you might have one lying around. The primary reason for this is the absolute lack of GPRs on x86 and even x86-64 processors. The PowerPC ISA defines 32 32-bit GPRs, the x86 ISA defines 8, the x86-64 defines 16. In order to make up for the x86's lack of GPRs your emulator has to do a LOT of caching just to execute a single instruction. The x86 always ends up doing far more overhead work than actual execution work when emulating PowerPC code.

When going the other way around, emulating x86 or 68k code on the PowerPC the situation is very different. Because of the abundant GPRs, interpereted instructions aren't going to be writing to or reading from registers that have been cached out to the processor cache or main memory. The 68k emulator introduced with the PowerPC based PowerMacs was pretty much a direct map of instructions. For nearly every instruction on the 68k there was an equivilent PowerPC instruction and since the 68k only had 8 GPRs mapping them to registers on the PowerPC was very easy. The instruction ratio was something close to 2:1, for every clock it would take a 68k to execute an instruction a PPC would take 2. At 66MHz the 601 was already beating the pants off the fastest 68040s used in previous Macs.

Even the fastest Athlons and P4s out right now would have a hell of a time emulating so much as a 300MHz G3, not even bringing to question G4s and their vector units. An Opteron is not going to do a whole lot better despite its performance advantage over current x86 chips. Wow Apple could release x86 Macs that run their current software like a 300MHz G3 (ie. like crap) and then use a processor no native software would run on. That means a short while after they get their third party developers moving away from Classic and Carbon development they expect them to release their software compiled, bug tested, and supported for an x86 platform. Their third party developers would drop them like a bad habit. With no third party developers Apple would be little more than NeXT with a cooler logo. MacOS X would die from neglect just like OS/2 and BeOS did.

Re:Stolen, but insightful.

[Glock27]

The reason I ask is there are absolutely no decent PowerPC emulators for the x86 architecture and figured since you thought it was so easy you might have one lying around. The primary reason for this is the absolute lack of GPRs on x86 and even x86-64 processors. The PowerPC ISA defines 32 32-bit GPRs, the x86 ISA defines 8, the x86-64 defines 16. In order to make up for the x86's lack of GPRs your emulator has to do a LOT of caching just to execute a single instruction. The x86 always ends up doing far more overhead work than actual execution work when emulating PowerPC code.

I'm not sure what you mean by "a LOT of caching". I'd simply implement using the eight x86 GPRs for the first eight PPC GPRs. This would likely be enough for a surprising fraction of routines. For those using more than eight registers, I'd simply use 24 more variables (r9-r32). If these variables are referenced frequently (the appropriate use for a register) they will tend to stick in L1 cache and not suffer too badly in terms of performance. If they are not referenced frequently, the use of a register wouldn't have helped the routine's performance much regardless.

[snip historical 68k emulation discussion]

Even the fastest Athlons and P4s out right now would have a hell of a time emulating so much as a 300MHz G3, not even bringing to question G4s and their vector units.

An Opteron is not going to do a whole lot better despite its performance advantage over current x86 chips.

Despite having four more general purpose registers and such low memory latency? I disagree.

I'm pretty sure the performance would be more like an ~1 GHz. G4, sans Altivec. I doubt emulating Altivec would be worthwhile (though perhaps some routines could be mapped to SSE2). My approach, though, would be to report the emulator as a G3. All Mac software supports the G3.

Wow Apple could release x86 Macs that run their current software like a 300MHz G3 (ie. like crap)

I see you've taken your 300 MHz. theory and run with it... ;-)

and then use a processor no native software would run on.

Except any recompiled Cocoa or Carbon app, or any out of the box Java app. I expect software developers would have little difficulty recompiling, testing and distributing. I'm pretty sure they'd be happy with that given the potential for 10x or more growth in the Mac software market.

That means a short while after they get their third party developers moving away from Classic and Carbon development they expect them to release their software compiled, bug tested, and supported for an x86 platform.

Yes, by checking a second architecture check box.

If NeXT/Apple can't pull this off flawlessly by now, they might as well go out of business. NeXT did a great job with far fewer resources over fifteen years ago.

Their third party developers would drop them like a bad habit.

That's certainly what I'd do if Apple came out with best-of-breed hardware. Not.

In fact, if Apple actually had the world's fastest Unix workstations, I suspect they'd gain very substantial marketshare. Marketshare attracts developers, it doesn't alienate them.

With no third party developers Apple would be little more than NeXT with a cooler logo. MacOS X would die from neglect just like OS/2 and BeOS did.

If your previous prediction actually happened then this would follow...however I'm quite sure that moving to Opteron (especially if third party clones were allowed) would boost Apple's fortunes, not the opposite.

As I told another poster, Think Different. ;-)

Re:Stolen, but insightful.

[ablair]

"What I personally would like to see out of this is Apple releasing their own BSD distro."

And what, exactly, would Apple gain by doing that?

Answer to first: They make a disro already, it's called Darwin and yes, as a *BSD it runs most *BSD code with little or no tweaking (both on PPC or x86 systems).

Answer to second: I guess the same gains as any company basing their core software on open source projects. Companies such as Red Hat and now Apple that do this benefit from masses of enthusiasts continually improving their operating system cores, letting them devote their limited resources to value-added and unique features on top that customers actually pay for. Read Martin Fink's The Business & Economics of Open Source to find out more.

Re:Stolen, but insightful.

[byolinux]

How is that hip-bag working out for you? :)

Re:Stolen, but insightful.

[Graymalkin]

I present the fact there are currently no fast PowerPC emulators available on the x86 architecture as evidense the task is a hopeless one.

Simply mapping the eight GPRs of an x86 chip to the first eight GPRs of a PowerPC chip does not do you any good at all. Most PowerPC instructions accept at LEAST three registers for input, two input registers and an output variable. Typically x86 instructions (like add) take only two registers. In order to complete a single PowerPC add you'd need at least one x86 mov instruction to get a value into a GPR, the add instruction and then another mov to get that value into a GPR you've mapped to the first eight on the PowerPC.

In short you're fooling yourself thinking you can get any sort of decent performance out of an x86 emulating a PowerPC. The overhead for simple instructions is fairly hefty, more complex operations require the host processor to do even more work. An extremely well done emulator MIGHT get a tenth of the performance out of emulated code over native code. For a 3GHz P4 that is equivient to about say a 300MHz G3. Hey that looks familiar. I'm sorry but a 300MHz G3 isn't going to do anyone a lot of good unless they're content to run Safari and Mail exclusively. Even a fifth of the performance of native code is still only a 600MHz G3 equivilent. You're wagering Apple could easily pull off an ISA transition based on technologies that have no current implementation.

You're obviously misinformed about the Carbon API if you think it is portable to the x86 ISA. Carbon exists to allow developers to more easily transition their applications to OSX from ToolBox from classic MacOS. Carbon is the most used 80% or so of all the classic MacOS managers and libraries. It is written specifically to let PowerPC binaries run on both classic MacOS and OSX.

Even with the Opteron Apple would still have to sell their platform to people. They would still need to convince PHBs that their platform would be a good move for the company, they'd have to convince schools their products were worthwhile investments, they would have to still convince home users they would be happier using a Macintosh to browse the web and get their e-mail. No matter what processor Macs were powered by they would have to sell their platform. Their platform is what does or does not sell, not the processors.

What you really don't seem to pick up about Apple is they are a big fish in a little pond. In the world of Macintoshes they are the king and reigning champion. They have no competition with other people selling Macintoshes and can thus do more with their products. As soon as they enter the x86 realm they become a leeeeettle tiny fish in a great big pond. Apple would end up like VA Lin^H^H^H Software. They would be selling a non-Windows operating system on commodity hardware. That is not a business plan for success.

I would suspect your solution to that problem is license the OS out to people, that of course would increase adoption. Apple would have to sell ten times more copies of MacOS than they currently sell Macintoshes to make up for the difference in the margins. With sales numbering around a million machines a year they would need to increase that number by ten just to make as much money as they do now. In order to make the OS attractive to people they would need to support all or at least a signifigant portion of the hardware and hardware configurations that Windows does. As we know, hardware vendors are none too friendly in developing non-Windows drivers for their products. In the Linux world this slack is made up by developers in their free time. Apple to be competitive with Microsoft would need to do this work themselves. This would increase overhead and lower profits and require even MORE people to buy their OS to break even.

You might want to drop the Different part and just do more of the Think.

Re:Stolen, but insightful.

[nitehorse]

Point 1: Any systems that Apple ships with a PPC970 in them will get "World class performance" on a par with a P4, if not handily spanking it.

Point 2: IBM makes all of the G3s that Apple currently ships. You buy an iBook? You get an IBM G3 inside. You buy an older iMac? IBM G3. So it's not like Apple would be moving to a new untested company who hadn't proven themselves to be capable CPU producers.

Point 3: The PowerPC 970 was designed with the same HyperTransport support as the Opteron. The same exact system. IBM is planning on making 4-way and 8-way PPC970 boxen to sell. Apple could easily do the same, and we're talking about performance that wouldn't just "compete" but blow away anything else except possibly the Opteron. Also, keep in mind that Altivec really does turn the tables in a LOT of cases... SSE2 is nice, but it hasn't shown itself to be nearly as easy/nice to code to, nor has it shown the startling difference with regards to performance that Altivec has.

All in all, my prediction is that a move to PPC970 is all but inevitable. Opteron would be interesting, but Apple is in the middle of a software changeover right now... there are still a few million MacOS Classic users to convert to Jaguar/Panther, you know. You tell all of your current users that their brand-new PPC hardware is actually obsolete, because new hardware will be x86-64 and all future applications will be x86-64, and you've just successfully cannibalized your own market and signed your death warrant.

Re:Stolen, but insightful.

[Surak]

Hopefully, if PCs are to rule the CAD market, the vendors will have the sense to begin using Linux-based or BSD-based systems (I've read Pro/E will be available for Red Hat sometime, which is encouraging).

I hope the same thing. I really think it's a no-brainer: if you have the app compiled on Sun already (Unigraphics in fact has a version for Solaris x86), and it works with OpenGL, it should be zero problem to port the thing to Linux.

The thing is that although UG has a Solaris x86 version, people aren't using it. They figure if they're going with x86 hardware, they're going to run Windows. Why? Why? Why? I don't know. I hate it.

Re:Stolen, but insightful.

[Surak]

And what do you think the Windows users did? And why do you think they hated the Mac?

Three words: one mouse button. :-P

Re:Stolen, but insightful.

[The Lynxpro]

How is Microsoft keeping Apple from running Linux? Why would Apple want Linux anyway? The OS is built on BSD...that makes no sense...

Re:Stolen, but insightful.

[Shuh]

You people keep talking like this thing is shipping now. News flash: it isn't.

News flash: It's maybe 2-3 months (tops) behind the Opteron as far as commercial fab production is concerned.

Even when it does, its price/performance vs. Opteron may be pretty poor.

The estimated SPEC marks indicate they are similar. But SPEC doesn't include Altivec/3D-Now, however, and Altivec is a much nicer SIMD unit. Additionally, the cost (but maybe not the price) will be lower for the 970 because it is a smaller piece of silicon.

Re:Stolen, but insightful.

[Glock27]

Point 1: Any systems that Apple ships with a PPC970 in them will get "World class performance" on a par with a P4, if not handily spanking it.

All right, you've finally spurred me into looking into this chip. The performance figures from IBM, at 1.8 GHz. (fastest initial speed grade, same as Opteron) are:

937/1051 SPECINT2000/SPECFP2000

That is somewhat lower integer performance and about the same FP performance as the 1.8 GHz. Opteron (fastest speed grade currently shipping).

So yes, competitive performance...but at what price? If IBM ships this at a reasonable price point (or even lower than Opteron), great.

Point 2: IBM makes all of the G3s that Apple currently ships. You buy an iBook? You get an IBM G3 inside. You buy an older iMac? IBM G3. So it's not like Apple would be moving to a new untested company who hadn't proven themselves to be capable CPU producers.

I knew IBM had produced chips for Apple in the past, but wasn't aware they were currently. Thanks for the info.

Point 3: The PowerPC 970 was designed with the same HyperTransport support as the Opteron. The same exact system. IBM is planning on making 4-way and 8-way PPC970 boxen to sell. Apple could easily do the same, and we're talking about performance that wouldn't just "compete" but blow away anything else except possibly the Opteron. Also, keep in mind that Altivec really does turn the tables in a LOT of cases... SSE2 is nice, but it hasn't shown itself to be nearly as easy/nice to code to, nor has it shown the startling difference with regards to performance that Altivec has.

All good points, and Altivec is noteworthy. It'll be very interesting to see head-to-head benchmarks between this and Opteron.

All in all, my prediction is that a move to PPC970 is all but inevitable. Opteron would be interesting, but Apple is in the middle of a software changeover right now... there are still a few million MacOS Classic users to convert to Jaguar/Panther, you know. You tell all of your current users that their brand-new PPC hardware is actually obsolete, because new hardware will be x86-64 and all future applications will be x86-64, and you've just successfully cannibalized your own market and signed your death warrant.

If you read my original post, you'll find I advocated keeping PPC for portable applications. PPC hardware wouldn't be obsolete, it'd be used in the hottest area. Once again, Apple could easily support two processor architectures in MacOS X.

However, presuming Apple and IBM can move things along in a timely manner, the 970 does look like a very reasonable path. I was impressed to see it's power consumption at only 42 W. for the fastest part - that is about 1/2 of Opteron. There is also a 1.2 GHz. 1.1 V. part that only consumes 19 W. That is impressive.

The less impressive part is Apple continuing along as single-source niche player. Its sad that Apple takes that tack, when it could be so much more. MacOS X has every potential to take on Windows across the board. Whether Apple is willing to go beyond the "Switch" ads is the question...

All that said, if PPC 970 Macs come in at around the same price as a comparable Opteron system, my first 64-bit computer might well be a Mac... ;-)

Re:Stolen, but insightful.

[Glock27]

I present the fact there are currently no fast PowerPC emulators available on the x86 architecture as evidense the task is a hopeless one.

You don't think that it's just that no one cares, particularly?

I'm not going to continue back and forth on the emulation question. IMO it is possible to do decent PPC emulation on x86, given enough time and effort. 'Emulation' could even be accomplished by doing a one-time conversion from the PPC binary to x86 ahead of time - sort of a cross-compile. At least that would eliminate runtime overhead.

You're obviously misinformed about the Carbon API if you think it is portable to the x86 ISA. Carbon exists to allow developers to more easily transition their applications to OSX from ToolBox from classic MacOS. Carbon is the most used 80% or so of all the classic MacOS managers and libraries. It is written specifically to let PowerPC binaries run on both classic MacOS and OSX.

Carbon (IIRC) consists of the '32-bit clean' set of APIs that could be supported under MacOS X without causing performance problems. There is no reason this toolkit couldn't be ported to x86.

Even with the Opteron Apple would still have to sell their platform to people. They would still need to convince PHBs that their platform would be a good move for the company, they'd have to convince schools their products were worthwhile investments, they would have to still convince home users they would be happier using a Macintosh to browse the web and get their e-mail. No matter what processor Macs were powered by they would have to sell their platform. Their platform is what does or does not sell, not the processors.

The platform sells based on many criteria including performance, price/performance, availability and public perception of the company's stability. Apple suffers badly right now in several of these comparisons.

What you really don't seem to pick up about Apple is they are a big fish in a little pond. In the world of Macintoshes they are the king and reigning champion. They have no competition with other people selling Macintoshes and can thus do more with their products.

However (as the switch ad makes clear) Apple does have competition from people selling Windows boxes. If you don't think so, you're living in denial.

As soon as they enter the x86 realm they become a leeeeettle tiny fish in a great big pond. Apple would end up like VA Lin^H^H^H Software. They would be selling a non-Windows operating system on commodity hardware.

They'd be selling a very polished, Unix based OS against Windows. Windows is annoying a lot of people these days. If even 10% of Windows users switched to MacOS X, that would be a tremendous win for Apple.

That is not a business plan for success.

Not surprisingly, I disagree. ;-)

I would suspect your solution to that problem is license the OS out to people, that of course would increase adoption. Apple would have to sell ten times more copies of MacOS than they currently sell Macintoshes to make up for the difference in the margins.

What do you think Apple's margin is on a $2799 Powerbook? I seriously doubt it's over 10%. One copy of MacOS X at $300 (97% margin) comes very close to making the same profit. Don't forget that Apple has to produce the OS one way or the other.

With sales numbering around a million machines a year they would need to increase that number by ten just to make as much money as they do now. In order to make the OS attractive to people they would need to support all or at least a signifigant portion of the hardware and hardware configurations that Windows does. As we know, hardware vendors are none too friendly in developing non-Windows drivers for their products.

Apple already has support for a tremendous (and IMO sufficient) PC hardware base - you know, the same hardware that's currently used with Macs. AGP cards, PCI cards and USB peripherals. Nothing needs to chang

Re:Stolen, but insightful.

[GlassHeart]

Apple could sell it's OS for x86 boxes at say $300 a pop.

Considering loyal Apple users are unhappy with the $130 Jaguar, what makes you think that price is feasible? Remember that somebody who switches to x86 OS X has to replace all their software, so a "switcher" can easily be looking at $1,000 or more in software (that they already have Windows versions of).

Secondly, Apple will have to support a thousand drivers for various obscure peripherals. That will dramatically increase engineering costs, and most likely decrease the stability of the kernel. Apple will be in the same boat as Linux (or even worse) when it comes to drivers.

Thirdly, Microsoft just won the anti-trust suit, and is unlikely to be bothered by the US government again soon. You think they'll just sit there and let OS X reach 6%? At the very least, I would expect no native Microsoft Office or IE for the x86 OS X, if not for the PPC altogether. Then what? OpenOffice.org? Then why not run Linux?

Along with Dell, which has the opposite business model. Don't be so sure something in the middle wouldn't work...

But I'm not saying that it won't work. I'm pointing out that Apple's business plan does work, despite your not liking it.

Unlike some other environments, things worked very well across architectures. Further, in this situation the various busses and peripherals are identical, simplifying things further.

It doesn't matter. Commercial software need to go through full testing cycles for each platform they support. The quality of the API or the software does not decrease the testing requirements, only the number of bug fixes. It's an expensive activity, likely to be a significant burden on the relatively smaller Mac software vendors.

When you're done testing, you then have a fragmented tech support team. Assembling such a team to support the myriad configurations of x86 hardware (as your 6% $300 OS requires) is also much more expensive.

[...] if the new high-end PPC CPU is actually competitive with Opteron, both in price and performance.

So why are you assuming (by virtue of your Opteron advocacy) that it can't?

At any rate, I think you are vastly overestimating the complexity involved. Cocoa code is very portable.

I'm a professional software developer, which means that I'm risk averse. While I love Cocoa, Objective C programs are just as easily made non-portable as C programs. Moving platforms is always dangerous, and taken only when the alternative is worse.

Today, the 970 is clearly a far better option, even if it doesn't surpass the x86. Altivec optimizations will pay off except on the G3s which are on their way out. No additional expenses that I point to in this post, and no need to compete directly with Microsoft with a $300 OS.

Re:Stolen, but insightful.

[Graymalkin]

This discussion has become absolutely absurd. Your entire argument is based on your personal opinions.

There's a lot of interest in PPC emulators but the proposition is not feasible! Go read up on PowerPC assembly for a little while, then take a look at some x86 assembly. It is not a simple task to emulate one on the other. Several groups have been trying for years to build a PPC emulator on the x86 architecture, it just isn't in the cards.

You need to stop pulling things out of your ass and research them. Apple's AVERAGE margin on hardware is about 25%, read their quarterly statements for that. They make beaucoup cach off their hardware because they're the only ones that sell it and have tight control over their production process. Their high end equipment like PowerBooks have really nice margins reaching towards 30%.

Apple's third party hardware support is larger than most PC fanatics assume but is not nearly enough to license the OS to third parties. OSX would need to support everything Windows does to be a viable alternative to Windows. No 10% of Windows users are going to switch without more functionality than Windows provides. The SAME thing happened to BeOS and OS/2. Both were better operating systems than Windows but they lacked serious hardware support. Without it neither OS was a viable alternative to anyone. If Apple has to invest the time and money to increase hardware support their margins selling the OS disappears.

Did you eat paint chips when you were young or do you not understand the pond concept? Apple makes very good money selling products to a niche market. Success does not always necessarily equate to gigantic market share. Gigantic market share for some companies is a very bad thing. You can't seem to get your head around this concept. We'll contrast Apple's business plan to Dell's in hopes you can understand.

Dell competes directly with all of the other PC manufacturers because they make a product that are exactly like all of their competitor's products. Some of their products actually come from the same manufacturer that their competitors get their products from. In order to survive in that sort of market (big pond) Dell has to keep their prices competitive with other companies in the business. This most often means running razor thin margins in order to sell their hardware. Without selling services which they make good margins on Dell would go down the tubes.

Apple's business plan on the otherhand allows for making a real profit margin on their hardware sales. They forcus their sales at niche demographics and niche markets. Their business plan REQUIRES they sell to a niche audience (little pond). Because they sell to a niche market and are the only company providing that particular product, they can sell their products for a price that doesn't completely destroy their profit margin.

Dell buys their hardware from an ODM, the people who actually design and manufacturer most of their equipment. They give a piece of hardware to a marketeer and tell them to figure out who that product they've got is good to market to. If they have a cheapo laptop they can say its great for students, if they have a expensive PC they can say it is a workstation and market it to businesses.

Apple starts at the other end and designs a piece of hardware based on the target demographic. The iMac was designed for people who did not want the hassle of setting up a PC which in 1998 was a PITA for most people. It also catered to the needs of schools who needed easily maintained computers for labs and classrooms. The iBook was designed almost specifically for students with its rubber padding, ballistic plastic case, and handle.

Apple's is a big fish swiming around in a little pond. They cater to the people other computer companies either miss entirely with their marketing or ignore because they desire selling to a lower common denominator. Dell is a little fish in a very big pond. They have to fight all the other little fish, even the fish building PCs in their garages out of white box parts, to survive. If you dumped Apple into the big pond they would cease being a big fish and would also find they weren't equipped to survive in the big pond.

Re:Stolen, but insightful.

[jcr]

Where is the context sensitive help that the likes of MS DevStudio has had for years. Where is the debugger than can handle (horrors!) multiple threads properly? Where is the stable Interface Builder that doesn't arbitrarily crash?

I found them all at developer.apple.com.

-jcr

Re:Stolen, but insightful.

[Ben Hutchings]

I don't know of a 615 ever got passed the rumor state.

I believe there was serious work done on it, and some of the people involved now work for Transmeta.

Re:Stolen, but insightful.

[Glock27]

This discussion has become absolutely absurd. Your entire argument is based on your personal opinions.

And yours, of course, is based on the Truth.

There's a lot of interest in PPC emulators but the proposition is not feasible! Go read up on PowerPC assembly for a little while, then take a look at some x86 assembly. It is not a simple task to emulate one on the other. Several groups have been trying for years to build a PPC emulator on the x86 architecture, it just isn't in the cards.

OK, after a brief Google search, I found SoftMac From the site:

"All of our emulators run Mac software very quickly, typically delivering 50% to 70% of the clock speed in emulation. In other words, a 500 MHz PC running Windows 98 can emulate the equivalent of a 300 MHz Macintosh Quadra running Mac OS 8. Your results will vary depending on the exact type of PC you are using, the amount of RAM present, the version of Windows being used, video drivers, etc. For a good explanation on exactly what type of PC to use and how fast your results can be, see our Hardware Requirements and Benchmarks page."

This happens to be much more in line with the performance I would expect than your (unsubstantiated) claims would suggest. It is also a concrete example of a 100% complete PPC emulator running on x86.

You need to stop pulling things out of your ass and research them. Apple's AVERAGE margin on hardware is about 25%, read their quarterly statements for that. They make beaucoup cach off their hardware because they're the only ones that sell it and have tight control over their production process. Their high end equipment like PowerBooks have really nice margins reaching towards 30%.

Fine, I'll grant you the point regarding Apple's margins. With the new figures it will require 3-4 copies of MacOS X to equal one Powerbook.

I'm going to cherry pick the rest of you're article since we've already argued many of these points at least once.

OSX would need to support everything Windows does to be a viable alternative to Windows.

Not at all, as long as there is a logo program and an easy location to find the hardware compatibility list. The important thing is that high quality hardware is supported, not junk. And guess what, as Apple marketshare grows, more and more OEMs will support Apple.

Did you eat paint chips when you were young or do you not understand the pond concept? Apple makes very good money selling products to a niche market.

Surely you see that it does not follow that Apple wouldn't make even more money in a larger market. Especially if Apple creates additional market segments (say, the white box Mac crowd). Those who buy white boxes would likely never buy a Mac, and vice versa.

Apple's is a big fish swiming around in a little pond. They cater to the people other computer companies either miss entirely with their marketing or ignore because they desire selling to a lower common denominator.

No, in many markets (like education) Apple is head-to-head with the likes of Dell, Gateway, and Sony.

Dell is a little fish in a very big pond.

Dell is the biggest whale in the ocean.

They have to fight all the other little fish, even the fish building PCs in their garages out of white box parts, to survive.

So does Apple. If you think there aren't plenty of (for instance) students who look at iBook then decide a Compaq laptop is a better value for a variety of reasons, you're delusional.

If you dumped Apple into the big pond they would cease being a big fish and would also find they weren't equipped to survive in the big pond.

Apple could keep all of it's current branding, style and so on. The group that finds that appealing (about 3% of new computer buyers, IIRC) will continue to buy those boxes at Apple's rather obscene margins.

For the other 97%, something different is needed.

Thanks for an interesting discussion, and have a great day!

Re:Stolen, but insightful.

[Glock27]

One last link for you to check out. This person basically agrees with my position, and marshals a lot of evidence to support his views:

Mac OS on x86... or PowerPC 970?

Re:Stolen, but insightful.

[Graymalkin]

Wow you found someone with the same stupid argument as you, do you want some kind of award? A biscuit perhaps?

Here's two situations

1. Apple switches to an entirely different processor architecture which forces them to leave behind all their old PowerPC software. The hope is some really fast processors will make people spend several thousands of dollars on new Macintoshes as well as the software to run on them.

2. Apple switches to a faster chip using the same instruction set as their current processors use. All of their old software runs without incident. Customers can buy new hardware as they see fit and upgrade the software when 64-bit clean versions are released. Existing software still works fine and runs better than it did on their old equipment.

Of those two situations which do you think is more likely? I'll give you a big hint, it isn't the first option. Drop this ridiculous Opteron crap. Apple is a co-developer of HyperTransport with AMD. An Opteron would make really fast Macintoshes that had no software to run really fast and would thus be useless until third party developers ported their software. It is much more likely Apple wants to use a faster PowerPC chip (like the 970) on top of a HyperTransport bus than switch to an entirely new processor. Porting Carbon and Cocoa to the new processor would be a hell of a lot less work than porting these frameworks to x86. You've made so many stupid conclusions it is downright painful to keep responding to them.

All Opteron switch proponents base their claims on the hope someone is going to perform a miracle and spit out a super fast PowerPC emulator. They don't seem to understand emulating PowerPC on an x86 is nothing at all like emulating a 68k. Hoping this is going to be magically possibly is futile. Thus basing hopes of Apple's with Opterons is just as futile. Faster processors is not the only aspect of a computer system that draws in customers no matter what slashdot weenies would like you to believe.

The article you point to is a guy making a bunch of assumptions based on VERY LITTLE real evidense. Half of his "evidense" is second or third hand rumors while the other half is entirely circumstantial. The conclusions he draws are based off of circumstantial situations relating to third hand rumors. I'm sorry but the Register and MOSR are not viable nor reliable sources of information.

I've read all of this crap before many times. You're not the first nor last person to think Apple will move to Opteron processors. MacOSX running on an x86 processor will not make people flock to the platform. Linux running on x86 likewise does not make people flock to the platform. The platform is not Windows it is not the "standard" and thus people are frightened of it. Whiney slashgeeks who proclaim "when OSX runs on Intel I'll buy Apple" are full of it. They will not buy a Macintosh, they never intend to.

An Opteron in Macs would not increase their sales enough to make up for the people ditching the platform because their mission critical software was not available any longer. It likewise wouldn't make up for the people abandoning the platform because any new hardware purchase would mean several thousands more in software purchases. OSX would have never made a mark on the world had it not been for the legacy compatibility it has maintained since day one. Carbon and Classic applications were all OSX had for a very long time. A new version without the ability to run old software would kill Macs and kill Apple. That is not a good business strategy.

Re:Stolen, but insightful.

[Glock27]

Wow you found someone with the same stupid argument as you, do you want some kind of award? A biscuit perhaps?

All you do is spout the same tired, circular arguments that show a complete lack of imagination. I'm sure Apple really wants 3% marketshare rather than 30%. After all, why would it want to make all that filthy money?!? Shareholders? Who cares...

Let's look at one of the silliest arguments in your latest polemic:

Linux running on x86 likewise does not make people flock to the platform.

1) People are flocking to Linux. It is the fastest growing operating system overall, and is taking marketshare much faster than is MacOS X, for instance.

2) Linux lacks commercial applications. MacOS X would solve this particular problem swimmingly. Apple has plenty of commercial software support.

3) Linux lacks UI polish. Guess what, this is MacOS X's strong point. Linux' user unfriendliness and rough edges are often cited as a reason its "not ready for the desktop". If I'm not mistaken, you'll agree that MacOS X is "ready for the desktop".

Therefore, your argument is what is called a "non sequitor". The conclusion doesn't follow from your premises.

We'll just have to agree to disagree. I hope Apple management has a more open mind than you do, but sadly they are probably still content being a niche player rather than an industry force. If the "Mac experience" is really better than everyone should want it for the most people possible.

Think Different.

Re:Stolen, but insightful.

[DrXym]

And those docs are on the dev tools cd as well and they still stink. In MS DevStudio, you can view 700Mb of help by index, by topic, hierarchically, search by keywords in seconds, context sensitive help and more. P

In OS X you get a link to the top page and the fun of hunting fruitlessly for what you're trying to do. The search function is so painfully slow (minutes) it may as well not exist. At least you can bookmark pages but it's still completely inadequate. Some functions such as the Posix / Darwin layer don't even appear in the same help system and appear as man pages. How crap is that?

Re:Stolen, but insightful.

[Glock27]

NeXT floundered because it got into the x86 space...

Yah, it had nothing to do with $800+ OS licensing fees or $5000 development environments...

Sure.

Re:Stolen, but insightful.

[Llywelyn]

>What difficulties do you forsee, exactly?

1) Compare and contrast the x86-64 assembly and the PPC 970 Assembly. This is *nontrivial* and we are talking about a tremendous slowdown.

2) The PPC line has more registers. This means caching, which is *slow*. Even assuming full use of the registers, a good scheduling algorithm, and lots of L1 cache this is going to *hurt* performance. Even 50% speed emmulation is too slow if you want people to have a reasonable chance of switching.

3) AltiVec support. SSE/MMX is a completely different beast from AltiVec and Apple has put far too much emphasis into marketing AltiVec at this point. Also, having AltiVec, it is a hell of a lot easier to work with than SSE.

4) Testing and quality control. 'nuff said.

>Then why haven't they? Traditional processors like the G4
>need special chipsets to hit 4-way. Opteron doesn't.

Lack of a market to support development costs.

You can parallel a series of systems easily enough that I think they just concluded individuals looking for a four processor were more likely to get two dual rackmounts than to actually bother with the additional cost.

The G4 supports multiple cores as well. It all comes down to development cost versus how much they expect to make.

>In your humble opinion. ;-)

You are the one advocating that Apple support two platforms.

>No, it means that software vendors should be strongly
>encouraged (or forced) to supply fat binaries for the two '
>architectures ...and in the commercial software world, this means that they have to be thoroughly tested and reviewed on both platforms, and optimized separately for both platforms (particularly if you mean for your software to be used on both), maintained on both platforms, synched with the hardware/drivers for both platforms (heh), and then supported on both platforms.

Brilliant.

Do you even realize why the 68k-PPC transition worked or any of the difficulties that were involved?

>Intentionally stupid, perhaps.

They are making money in a failing economy, have good economic outlooks, and you call them stupid?

>This limits their market tremendously, since many large
>government and private organizations won't buy single-
>source equipment, period.

Big fish, little pond.

Little fish, big pond.

Take your pick: either-or.

>or b) Apple can't manufacture enough to meet demand.
>White box MacOS X boxes would solve these problems
>nicely ...and increase their development costs by an order of magnitude trying to support so many additional pieces of third party hardware, it would also through hardware-software integration out of the window.

>Think Different.

First you have to Think before you can Think Different: consider the implications of what you are proposing, what it would mean from a marketing standpoint, and the difficulties involved.

Re:Stolen, but insightful.

[DrXym]

Gosh, how brave, an anonymous coward. Yeah, you're right. I should take the hint from Apple and pricks like you and stay silent. OS X development is so perfect, how could I possibly quibble?

Re:Stolen, but insightful.

[dnahelix]

...not to mention the folks who don't want to pay hundreds of dollars for Photoshop upgrades that don't actually give you anything new.

All About the Hardware

[Darth Hubris]

If Apple can control the hardware on which it will run, an Opteron doesn't seem too out of the question. Running an opteron != Apple on PC board/hardware.

If...

[dreamchaser]

If Apple starts producing AMD based systems, which I doubt will happen, don't expect an open architecture. You can bet that there will be proprietary elements to the platform and OS/X won't run on commodity x86 hardware.

Re:If...

[MuckSavage]

Of course not. That would destory apple's hardware sales.

Re:If...

[Anonymous Coward]

The best thing they could do is partner with AMD and NVIDIA to make a box that will run Windows, but also run the Mac OS with special additions to the CPU and video card. IOW digital keys. Apple could charge NVIDIA and AMD a smallish licensing ($20-$30) fee for this and then users would be able to dual boot a mac OS. And Apple would still esentially get their hardware revenues.

Remember, MS is breaking with NVIDIA for xbox2. They want a provider dedicated to MS. NVIDIA has other aspirations. AMD is still the up and comer against Intel. If they could get and help to expand the Mac market, they could become #1.

I'd gladly buy a box (or components for a box) that cost $100-$150 more if it ran the MacOS. As long as it runs Linux and Windows, too. I've always wanted to run OSX but the hardware is just too much and too limiting.

Re:If...

[Some Dumbass...]

If Apple starts producing AMD based systems, which I doubt will happen, don't expect an open architecture. You can bet that there will be proprietary elements to the platform and OS/X won't run on commodity x86 hardware.

It would be interesting to see what those "proprietary elements" are. There isn't much that's proprietary about current Macs, right? (I could be wrong about this). Basically they're stock memory, hard drives, etc. with a Motorola CPU (Okay, the DVI output is unique, but that's about it, I think). Anyway, If you change the CPU, then what's so proprietary about them that it could prevent an OS running on that hardware from running on other hardware? Most differences (like that DVI port) could be fixed by a properly coded driver, couldn't they?

Re:If...

[Mr Pippin]

That won't neccesarily be true. Remember that the core of OS X is Darwin (an open source kernel that is already running on x86). Someone could simply install the latest Darwin at the time and move the OS X (x86 version) binaries across to run on a non Apple branded x86 system. Then again, it would not be a SUPPORTED distribution of OS X. My 2 cents.

Re:If...

[dadragon]

It would be interesting to see what those "proprietary elements" are. There isn't much that's proprietary about current Macs, right? (I could be wrong about this). Basically they're stock memory, hard drives, etc. with a Motorola CPU (Okay, the DVI output is unique, but that's about it, I think). Anyway, If you change the CPU, then what's so proprietary about them that it could prevent an OS running on that hardware from running on other hardware? Most differences (like that DVI port) could be fixed by a properly coded driver, couldn't they?

DVI ports are standard. They are used on high end PC video cards as well. I think you're talking about ADC, but that too is a standard, it just isn't used by anybody else. It's documented, and can be implemented by anyone.

The ADC port is power, USB, and DVI video signal on one cable. There is a product called "DVI Excractor" which takes ADC and gives you only DVI. There is also a DVI to ADC adaptor which plugs into a power outlet, USB port, and DVI port and spits out an ADC signal to the monitor.

As for proprietary parts on a Mac, I don't think there are any anymore. The PMU is non-standard I think, but that's it. OpenFirmware: Open, PCI: Open, AGP: Open, USB: Open, well you get the idea.

Re:If...

[dadragon]

What's more likely is that Apple is using AMD to get their HyperTransport chipsets. Apple, IBM, and AMD are all on the HyperTransport consortium, so that makes sense. Apple motherboard with AMD chipset and IBM processor.

Another, less likely, but more likely than Operon CPUs is AMD manufacturing PowerPC processors for Apple.

Maybe for a while...

[ruiner13]

"Will we see Mac OS X running on two different platforms/CPUs?"

If we do, I think it would probably be only for a brief transition period, like when they switched from the 68K line of processors to PPC. But who knows. I really hope they don't switch to AMD, that would make people less inclined to write software that is still compatible with the PPC architecture I own (assuming they don't make binaries compatible with both... i don't think they can, can they?).

Re:Maybe for a while...

[RealErmine]

that would make people less inclined to write software that is still compatible with the PPC architecture I own

Too late.

Re:Maybe for a while...

[lcracker]

The type of bundles that Apple uses in OS X (from its NeXT roots) allow for support for however many architectures you want in the same bundle. Bundles are not binaries, they are collections of binaries and resources in a set of folders with a particular layout to them. For example, MyApplication.app/Contents/MacOS is where MacOS X PPC binaries live. If the app and relevant frameworks were ported to Win32 for example, the exe file(s) could live in MyApplication.app/Contents/Win32. I believe if the bundle has support for multiple architectures, it's called "fat". Also note that using these kind of bundles requires support from the launch services of the OS, and probably the linker as well (Although launch services could get around that on some platforms I guess).

Re:Maybe for a while...

[ivan256]

If we do, I think it would probably be only for a brief transition period, like when they switched from the 68K line of processors to PPC. But who knows. I really hope they don't switch to AMD, that would make people less inclined to write software that is still compatible with the PPC architecture I own (assuming they don't make binaries compatible with both... i don't think they can, can they?).

This is the biggest reason I've doubted the Apple/Opteron rumors from the start. When Apple switched from 68k to PPC they chose a processor that was capable of emulating the old platform at full speed to ensure a seamless transition from the user perspective. I doubt Apple would be interested in anything but a seamless transition this time as well. Opteron, however, doesn't have enough registers (among other problems) to do a good job at emulation the PPC architecture. I would guess that there would have to be AMD chips that are 10x faster than PPC chips (they're getting there, but PPC isn't that far behind yet) or Apple would not be satisfied with the PPC emulation experience. I would believe the use of Itanium more that the use of Opteron, just because Itanium is much better suited to PPC emulation. Unfortunatly a single Itanium CPU costs more than most complete Apple systems right now, so that's probably unrealistic as well.

As for all the people that say the 970 is vaporware because of the lack of hype, well there's always been much less hype from IBM and Motorola about their new CPUs than from Intel, AMD, and (formerly) Digital (remember the old Digital Alpha CPU ads back in the late 80s/early 90s? "We're on our third generation 64bit architecture. Our compitition hasn't even started designing their first." It was the first CPU specific TV ad I remember seeing. Classic). IBM markets to manufacturers, not to end users, so unless you're a developer you don't see the hype. IBM and Apple are well suited for each other because IBM has a history of licensing portions of their CPU cores and using them to put together custom processors for the customers. Apple would love to have that kind of control, and they won't get it anywhere else.

Re:Maybe for a while...

[Grant_Watson]

Well, I could see Apple allowing the use of a "fat" binary that could run on either machine; they did it with 68K/PPC, right?

The catch is that the programmer would still need to port the program; you need machine code for both platforms. There's just no way to get around porting. (I suppose you could ease the process by using a cross-platform toolkit, ala wxWindows, but Apple would probably have you hunted down and shot for violating their interface guidelines. Maybe they would provide such a toolkit?)

Anyway, the whole Apple/AMD thing sounds highly unlikely.

Re:Maybe for a while...

[zorander]

Short answer yes...

When you control the object file format and the loader and the executable format, that kind of stuff is trivial, at the expense of storage space...Though probably not even that much space, since only code segments would be doubled, maybe increasing binary size by 50%. Not a huge deal at all with today's 200Gig drives...Even 300% would be quite tolerable, though it would probably effect load times without a good precache or intelligent loader (Again, no problem).

Brian

Re:Maybe for a while...

[Brandon Sharitt]

If Apple did anything with x86 it would probably be in something like the Xserve or something else that wouldn't rely on third party software as much so arcitecture wouldn't matter as much.

Re:Maybe for a while...

[jazuki]

You're a little off. The mach-o binary format that NeXT used and that Mac OS X still uses (though not exclusively) supports multi-architecture binaries in the same file. In the NEXTSTEP days, it was not uncommon to have binaries that supported NEXTSTEP/OpenStep on PA-RISC, SPARC, x86, and 680x0 in the same file. These were commonly referred to as objese binaries. Someone can corrent me if I'm wrong, but I believe the PEF binary format (used for Classic PPC and Carbon applications) can support similar multi-architecture binaries in the same file.

Apple doesn't use the bundle architecture so much to support different processor architectures as to support different operating systems. For example, currently, some application bundles contain both binaries for both MacOS (X) and MacOSClassic. This allows the application to run natively on both operating systems (Mac OS X and Mac OS 9) while allowing it to take advantage of features of each that are unavailable in the other, which can be a problem for Carbon binaries intedend for both OSes.

MacOS on x86 talk yet again?

[glam0006]

Am I missing something? After extensive discussions/explanations on Slashdot and all of the Mac sites, why do some people still think MacOS will ever be released for the x86 platform?

More info from IBM

[daveschroeder]

The story missed a major source of information about the 970 directly from IBM:

PowerPC 970 2002 Microprocessor Forum presentation

This contains a link to IBM Senior PowerPC Architect Peter Sandon's detailed presentation in PDF format.

Eh?

[The Bungi]

but feel Linux is still not ready for the desktop yet

Please mod story as (-1, Flamebait).

Re:Eh?

[Elwood P Dowd]

Not to mention:

"Could we be that lucky?"

Approx. 8 million (+5, Insightful) comments in stories past have pointed out that:

NO, MORON, WE WILL NOT BE THAT LUCKY.

Re:Eh?

[The_K4]

There are also alot of people who are NOT "linux people" that I have dealt with that will agree it's not ready. For "linux people" it's great....for Joe Enduser it's still intimidating in alot of ways.

Re:Eh?

[Anonymous Coward]

The phrase "not ready for the desktop yet" is a meaningless mantra mumbled by people who obviously haven't used linux and its BSD bretheren lately. I use Doze(TM) in the office and BSD otherwise, and lately it seems that Doze(TM) is the OS that's not ready for the desktop. Tried OS/x too. Yeah, it's almost ready, but too emphatic about being proprietary.

Re:Eh?

[The Bungi]

You might find this fantastically weird, but I completely agree with you.

Re:Eh?

[MaxQuordlepleen]

err.. I don't know what you're smoking but pass some this way ;)

I use Linux on my work computer, and OS X on my home computer. The difference between the two, once you get beyond the consistent interface of OS X (no small thing!), is the integration.

Now, I use Xandros on my Thinkpad and I consider it to be the best desktop Linux by far (Corel's file manager + apt-get = w00t), but the fact that the applications come from the open source community at large means that Apple-style integration is nearly impossible.

The Xandros people do a great job but they can't possibly integrate hundreds of third-party open source apps into a coherent whole. In consequence I get one app in which Search is Ctrl-F , and another app in which Search is Ctrl-S. The great thing about OS X is that the application interfaces are so consistent as to make using them an almost mindless activity. And yes, I think that's a good thing! I need to use my mind for my work, not remembering which key combination activates the search function.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Missed an option:

[FortKnox]

For those of us wanting to get away from Windows, but feel Linux is still not ready for the desktop yet, this might make Apple a more viable alternative.

You forgot "and have the extra cash". Lets face it. The only reason I haven't gotten myself an iMac, is because I don't have the extra grand or so to buy one.

Now if AMD jumps into the mix, things may get interesting...

Re:Missed an option:

[MuckSavage]

Now if AMD jumps into the mix, things may get interesting... So you'd be willing to pay for a mac with an amd processor, but not a mac with a PPC chip?

Re:Missed an option:

[__aaaaxm1522]

Yup, that's right. Keep perpetuating the myth. Have you looked at the Apple Store lately? iBooks are <$1K ... compare to a Dell or similar notebook and you'll find that Apple matches or beats their pricing.

True - the desktops are still somewhat pricey. $1000 more? No. Not if we're talking iMacs, and if you're comparing them to a machine purchased from a major manufacturer like Dell or Compaq - If you're talking beige-boxes, well then yes. Apple computers are $1000 more than a beige box ... as are the Dell, Compaq, HP, IBM and Gateway machines.

But keep in mind, Apple is really focusing on the portable market segment this year, so that's where most of the value is going to be.

Re:Missed an option:

[Anonymous Coward]

Then get an eMac, they're only $999. You know, a grand over a comporable pc.

Re:Missed an option:

[FortKnox]

competition drives down prices

Re:Missed an option:

[Anonymous Coward]

I see dell laptops for around $650. A little creative deal hunting and you can get the 15'', 2.0 Ghz p4 for $50 more.

iBooks aren't close to that.

Re:Missed an option:

[Anonymous Coward]

That Dell has no firewire, no built-in wireless support, and weighs about twice as much with half the battery life.

You're right. iBooks aren't even close to that.

Re:Missed an option:

[0x0d0a]

No...Apple *is* fairly competitive in laptops, but they still don't match their competitors on price.

A bottom-of-the-line iBook is $999, and a bottom-of-the-line Dell (Inspiron 1100) is $749 (after $150 rebate). (The Dell has 16 megs less of VRAM, which I didn't consider significant, since hardly anyone is going to be doing 3d gaming on a bottom-of-the-line laptop anyway).

As for the desktops, Apple's computers are a percentage more than PCs, not $N thousand more. So, sure, at the bottom end there isn't a $1000 difference.

Apple doesn't have any interest in competing on price. Once you have a commodity product, you make *far* less money.

Re:Missed an option:

[Dr. Manhattan]

I just this month got a dual 2GHz Athlon 2600, 1GB RAM, 2x120GB drives. I scavenged the DVD drive, CD-RW drive, SCSI card, Firewire card, sound card, keyboard, and mouse from my old system. Got a GF4 MX 440 as well. Total cost: ~$1600 (overestimate). Add in $200 for WinXP Pro (yes, I run Linux too, but I game and my wife wants her greeting cards) and we're up to $1800.

Just hit the Apple webpage, and went for the Ultimate Powermac G4 (dual PPC 1.4GHz), and customized it as close as possible to that. (1GB RAM instead of 2, GF4 MX instead of TI, etc.) Total: $3,270.00. That includes a DVD/CDRW drive, and 180GB disks instead of 120GB, though. Let's give them $250 credit for the drive, and $150 for the difference in disk size. I think that's pretty generous. Estimated comparable total: $2870.

Difference: $1070. That's enough to buy another kick-ass PC, or the lowest-end eMac plus tax.

I seriously looked at Macs; I'd really like to run OSX. But I'm not going to be able to upgrade for three to five years, so I need top-end stuff now. I just couldn't justify the Mac's price tag.

Re:Missed an option:

[f3r0c7ty]

It is not just the cost of hardware. Once I go out and spend the extra $900 on an iMac then I need to buy new software (what there is of it).

Personally, I love the new Mac's and OS-X but I am not in a financial situation where I can run out and replace 1 or even all 3 of my personal computers with a MAC. It just costs too much to do so. Even if I did what would it get me? Great hardware running a great OS with a limited amount of software. I just cannot do it...and this I believe is the issue Apple has to resolve in order to grab market share.

Re:Missed an option:

[chasingporsches]

with a limited amount of software

like the other guy said, that is helping to further and spread the myth. the only reason why there are MORE software programs for windows, is because blind codemonkeys code for windows because it's easy. but if you'll notice the quality of most windows software compared to those for MacOS... you can find the same or better software to do just what you want it to. and if not, install virtualpc.

Re:Missed an option:

[labratuk]

Apple computers are $1000 more than a beige box ... as are the Dell, Compaq, HP, IBM and Gateway machines.

Exactly. At least with the Dells et al. you have an alternative where you can be in the same league. People keep going on about how much you get for your money with a mac. I don't dispute that. I also don't want half that crap.

If I just want an x86, to run linux, I have to spend about $300.

If I want anything that will run OSX, it's going to cost me at least $1000. I don't have that kind of money. And if I did, I could sure as hell think of something better to spend it on than computer hardware.

Yes you can make the same Ferrari argument as everyone always does, but I wish people would stop preaching at me to use OSX.

Do you go around saying Ferraris aren't expensive?

Do you go up to an econobox driver and start preaching that they should get a Ferrari?

...so that's where most of the value is going to be.

I don't care about value, I care about price.

Re:Missed an option:

[f3r0c7ty]

is because blind codemonkeys code for windows because it's easy Come on now...I am a software developer who has developed for both platforms and Cocoa applications are much easier(and elegant) to develop than Win 32. Carbon on the otherhand sucks but so does OS 8/9. The .NET platform has definitely levelled the field but Win 32 based programming is anything but easy(or elegant). What it is though is a guarantee that your app will *should* run on 90% of the personal computers out there and if you a writing an app to make a living this might be a driving force. I am not pro MS/Intel but I work at a shop that does work on both and there are only a handful of people who run MAC's. All for the same reasons I mentioned.

Re:Missed an option:

[f3r0c7ty]

Now that I think about it further...even if it was easier to develop an app for Windoze would that be a bad thing?

Re:Missed an option:

[thatguywhoiam]

Difference: $1070. That's enough to buy another kick-ass PC, or the lowest-end eMac plus tax.

Just so I'm clear, you
scavenged the DVD drive, CD-RW drive, SCSI card, Firewire card, sound card, keyboard, and mouse from my old system

... which obviously saved a bit of money.

Then, on your AMD, you're likely missing FireWire, Gigabit Ethernet, probably wireless support (i.e. antennas built into the case)... and yes, the kick-ass enclosure which is definitely worth something in terms of design and functionality (drawbridge access). Oh, and you paid for 1 GB of Apple RAM.

I'm too lazy to calculate that out, but I'm guessing the real difference, for the same functionality, is less than $200.

Now, if you didn't need those extra things, then you have the right machine I suppose. Apple desktops are only competitive if you will really use all the niceties. Myself, I need FireWire for the iPod, and the GigE is great when doing Mac-to-Mac transfers (another nicety - you don't need crossover cables for Macs, they autonegotiate) for instance.

Re:Missed an option:

[chasingporsches]

i was not referring to real programmers... i said "blind codemonkeys", blind meaning they can't tell whats going on around them. proof in point, look at software on download.com or tucows.com. except for the commercial software, its not high quality at all. i wasn't meaning anything personal, i respect true developers for windows.

Re:Missed an option:

[Dr. Manhattan]

Just so I'm clear, you scavenged the DVD drive, CD-RW drive, SCSI card, Firewire card, sound card, keyboard, and mouse from my old system

... which obviously saved a bit of money.

If you'd read what I wrote, I accounted for that. I gave Apple a $250 credit for their DVD/CDRW drive, right? Plus a $200 credit 'cause I have dual 120GB drives instead of dual 180GB. Everything else is comparable (including the case... An Antec Plus 1080AMG, opens easily, extra fans, nice layout, etc.) Keyboard and mouse is lost in the noise. If you want, we'll call it $70 and now the difference is only $1000 even.

Then, on your AMD, you're likely missing FireWire...

But you clearly didn't read what I wrote, because (in the part you quoted!) I said I scavenged my Firewire card. Sure, it's 400Mbps instead of 800, but that's all my camcorder will do anyway.

And so far my 1GB of ECC RAM has performed flawlessly through several tests (overnight kernel compiles, memtest86, etc.) I fail to see what "Apple RAM" has on that.

Re:Missed an option:

[Krusty Da Klown]

That's just it. Don't pay for Apple RAM. Buy it from the same place at the same price as the RAM you bought for your PC.

Re:Missed an option:

[Scudsucker]

You're perpetuating the falacy by comparing OEM prices to build-it-yourself prices, so of course your box is going to be cheaper.

Re:Missed an option:

[Rinikusu]

You could pick up an older Powermac (beige).. We have a G3 350 here at work that runs OS X reasonably well (certainly not any worse than our P166 OS2 machine).

Re:Missed an option:

[Jucius Maximus]

" I see dell laptops for around $650. A little creative deal hunting and you can get the 15'', 2.0 Ghz p4 for $50 more."

And piss-poor build quality. Sure, the specs are high, but if you press on the back of the display, you can see the shape of your finger on the front. Good luck getting the sleep function to work. I doubt the dell has firewire. Try the keyboards on the dell and then the iBook. You will find that the Dell one is very cheap. You won't get a suite of software with the dell.

The numbers definitely favour the dell, but if you actually use each one for a period of time, the iBook would be overall the better choice.

Re:Missed an option:

[TheRaven64]

The problem, for me, with moving to a Mac is that it's a large cost, rather than a lot of small costs. I own a PC, which I built. Every so often I identify a bottleneck in the system, and replace that part of it, spending a small amount each time. If I were to buy a new Mac, then it would be a large capital investment. If I were to buy a PC it would be the same, but I have no intention of buying a new PC. If I already owned a mac, then of course the situation would be different, and I would continue buying upgrades for it. If apple released a system with just motherboard and CPU, that I could plug my existig disks, graphics card, ram monitors etc. then I might be tempted, but they don't, and they probably couldn't for a reasonable price without undercutting themselves.

Re:Missed an option:

[bnenning]

Just hit the Apple webpage, and went for the Ultimate Powermac G4 (dual PPC 1.4GHz), and customized it as close as possible to that.

Apple's prices for RAM and HD upgrades are insane. You'd save quite a bit getting those from third parties, although it wouldn't completely erase the difference.

Re:Missed an option:

[thatguywhoiam]

Okay, I have to apologize, I was hasty.

I didn't really get that you had incorporated the costs of the older parts, but they were minor as you said.

Also, I actually noticed the FireWire option after the fact, and tried to add a correction to my post - only to be stymied by the Lameness Filter, so I gave up. Sorry about that.

The mention of Apple RAM had nothing to do with capability btw, I was referring to the fact that Apple (like many other comapnies) charges a fortune for RAM compared to individual vendors, who don't have to pre-buy in huge batches.

I still maintain that the price disparity is not that great all things considered. Versus a beige box a Mac will never win of course on price, but I really don't think Apple is trying to compete there - those companies make no money. The consumer benefits if he/she's wise (as you are) but other than that it is not a sound business proposition.

Again, sorry for the confusion.

Re:Missed an option:

[jo_ham]

If you want to game, then the Mac isn't really for you anyway, even if you want it.

There are some good games out for it, but nothing like you'll get on Windows.

If the only reason you have that copy of MS Crayola is to play games, stick with your PC - as you say, you can get it cheaper for more power.

If you want to do other things then a Mac becomes a good option. The Powermac is expensive right now, but the iMac, Powerbook and iBook ranges are very good value for money. They all lack the expansion of a Powermac system (although they all have at least 2 USB ports and a firewire port - iMac has 3x usb and 2x firewire).

If you're looking at a laptop at any point in the future, serioulsy consider either a 12" or 14" iBook. Go to an Apple store or another retailer and actually use it for a little while. Open and close it, pick it up, type on it, mouse around. You'll be impressed, trust me.

Same with using an iMac and a Powerbook.

The Powermac is too expensive for what it is right now, but the rest of the Apple lineup is well worth your time and money.

Re:Missed an option:

[jo_ham]

Go and look for someone selling an old Sawtooth or Mystic G4 box.

The Dual G4 450 Mystic in particular would be excellent - they have a pair of 450MHz or 500Mhz G4s and can support up to 2 gigs of Ram in OS X.

They can be had for cheap, and will run OS X more than adequately - it will fly along.

You can take most of your PC stuff and plug it in - hard drives, Ram (pc100/133 dimms), optical drives, graphics card (or just attach your monitor(s) to the Rage 128 that came with it in the agp slot if you don't need fancy graphics).

We're using a Dual 450 as a Final Cut Pro 3 workstation, producing broadcast videos, and there's nothing special about it - it's a stock dual 450, but with a large amount of firewire storage added.

Second hand, but very good stuff.

An obvious explaination....

[Chmarr]

Me and a few friends have long held the belief that Apple releasing OS/X for the x86 platform would KILL Apple. Unlike Microsoft, Apple's 'coin' is their hardware platform, rather than software. The software is just there to sell the hardware. If they released OS/X for x86, then their hardware sales would plummet.

Yes, they could make some money off selling OS/X. However, they would then have to become MUCH more interested in ensuring their software is not being pirated, and that means some kind of DRM. A lot of folk love Apple because of their anti-DRM stance, and a lot of that love would disappeaer if Apple went down this route. As it is, Apple don't seem THAT concerned about piracy of their software, instead relying on those that want to 'do the Right Thing' with Apple, which is a fair percentage of their user base.

Instead, this is my theory on the Apple/AMD relationship, if there is one.

- It would be STUPID of Apple to rely on a single-source for their new processors, so, who better than to ask as a 'second source' than AMD? Yes, I'm sure Apple/IBM will get a leetle percentage out of all the chips that AMD make, but I'd bet my dollars that's what's going on.

Of course, the other possiblity is that AMD HAD talks with Apple, and they consisted of "Hey, lets go do lunch." "No." :) But... that's still 'talks', isn't it? :)

Re:An obvious explaination....

[ckaminski]

I wonder if the real answer is that AMD will recode the Opteron with a code engine that can execute big-endian PPC ops? (Is the Mac still big-endian?)

AKA Transmeta, with all the performance of the AMD64 engine. It'd be a good move for Apple, since they'll get a top-of-the-line 64bit CPU that actually has some vendor turnover. And if it's simply a matter of altering the instruction decoder... With the performance gap between current Mac's and this new CPU, even a small performance lost doing instruction translation will mean a much faster Mac platform.

And if the comment about Apple and Hypertransport is true, that makes it all the more POSSIBLE (however highly improbable?).

Re:An obvious explaination....

[TheAwfulTruth]

Actually I believe that the complete opposite is true. Apples market share could go up 10x overnight if they released Mac OSX for x86. Hardware is a tough place to make money, the hardware COSTS money. Lost of it, profit margins are slim.

And even if Apple made the OS free-standing for any x86 machine, that would not stop Apple from being able to build ultra-stylish, high-priced, boutique x86 machines and put their own OS on it now would it? They'd probably sell exactly the same # of machines that way as the die hard Apple aficinados would probably still prefer to buy apple branded machines, AND it would grow their OS market share tremendously, maybe even driving their branded machine sales higher in the future.

Frankly I think not porting OSX to x86 was a huge blunder that will hold the company back another 5 years till the next actual achetecture change is forced upon them.

Re:An obvious explaination....

[SqueakyFerret]

I don't really put much faith in Apple really having an "anti-DRM" stance, actually. Unless there's some statement they have made that I missed, the only evidence we have of Apple's anti-DRM stance is their "Rip, Mix, Burn" ad campaign. But I saw that as more a question of them just capitalizing on what everyone was already doing in order to sell a few iMacs

I seriously doubt there's a true anti-DRM sentiment in Steve's head. Apple might stay mostly silent about DRM, and might not copy-protect their OS, but only so long as there's not real profit motive for them to do so. I think it's just convenient right now that, because of the way their market works, they don't need DRM very much. This allows them to sit back and let everyone think they are anti-DRM, and are therefore on their side. Don't believe it.

Re:An obvious explaination....

[Chmarr]

Well, given that Apple were making more systems than Gateway, and there's a LONG held belief that the Apple systems are overpriced, profit margins can't be THAT slim :)

Again, if Apple release OS/X for the x86 platform, they're going to have to get a lot more interested in DRM. This isn't just a policy change; it's a complete about-face.

Re:An obvious explaination....

[smcavoy]

I don't think x86 makes sense quite yet.
Their still dragging developers and users kicking and screaming into OS X. Having to support x86 and PowerPC would be BAD.
You'd need Photoshop, [insert other mac app here], etc. ported to x86, and PowerPC....

Re:An obvious explaination....

[TheAwfulTruth]

I believe that they have already incorporated full DRM in their music products and an upcoming huge music announcement for some on-line music service will probably be taking full advantage of it. The about face is already happening.

The DRM question of "Will Apple have DRM" no longer exsists, but is now just "How will it be used?" is left to see.

Re:An obvious explaination....

[akuma(x86)]

Apple will never port their software to PC-based x86 systems. What they *may* do is build a closed apple motherboard/system with an x86 CPU. Let's face it - Motorola and IBM have fallen way behind what AMD and Intel have to offer in terms of raw CPU performance. AMD and Intel are pretty reliable suppliers.

Re:An obvious explaination....

[TheAwfulTruth]

At this point I completely agree. Personally I think that making the switch when the moved to OSX the first time would have been the best. Since it is as you say a "Dragging kicking and screaming" thing already, changing the processor at the same time would not have been much worse.
In fact, re-compiling an app for a new processor is HUGELY easier than porting it to a new OS which is what they are doing now moving from OS9 to OSX.

But now I think they ARE stuck until everyone is completely over to OSX and everything is humming again before they can even consider it. So... another 5 years?

Re:An obvious explaination....

[jimmcq]

If they released OS/X for x86, then their hardware sales would plummet.

If Apple Did start using AMD chips, OS/X could still be a closed/proprietary platform that wouldn't run on non-Apple hardware.

Re:An obvious explaination....

[scrotch]

Actually I believe that the complete opposite is true. Apples market share could go up 10x overnight if they released Mac OSX for x86. Hardware is a tough place to make money, the hardware COSTS money. Lost of it, profit margins are slim.

Market share != Profit;
10 * $129 < $1500; // 10* OSX cost < average Mac price

Hardware is a tough place for Windows PC makers to make money. Apple has been doing pretty well there. Dell and Gateway have problems and losses because they're in competition with each other and with your cousin who makes PCs in his garage. Your suggestion that Apple would make more money by competing with Dell and your cousin is strange.

Re:An obvious explaination....

[PCBman!]

You'd be surprised.

In the x86 market, you're lucky to get 15% margin on a machine (10% if you're trying to undercut somebody)--keep in mind I'm talking about LOCAL computer shops selling hardware.

Apple makes A LOT more then this. I wouldn't be surprised if they sold much of their hardware at 50% markup--maybe as high as 75% depending on WHAT it is. Why and how they can do this is easily answered. People may grumble, but they still buy.

Yes software can have 10x the markup (by that I mean 1000%), however, there's a lot more of a fight in that field vs hardware.

It'd be nice to get OS X for x86, and the margins on software are huge, but Apple really does make its money on its systems and Microsoft would find a way to eviscerate them in the software market.

Re:An obvious explaination....

[Chmarr]

You can't say "I believe that...", offer no proof, and then say that the question doesn't exist :)

The question DOES exist, and here it is: Does apple's technology have DRM in it? And, prove it!

Re:An obvious explaination....

[jbolden]

PC hardware is a tough place to make money. Apple has profit margins of 25-30% on most systems. 25% of a $2000 system = $500 = 60% of $800. You think OSX would sell well for $800 on PCs?

Re:An obvious explaination....

[jbolden]

Steve has taken Pixar out of the DRM consortium publically even when Disney threw a fit over the issue. Apple is about the only major hardware vendor not a member of the TCPA.

Re:An obvious explaination....

[kwerle]

Apples market share could go up 10x overnight if they released Mac OSX for x86.

Reality check. Assuming that their market share is 4 or 5 percent, you're saying that they could take 40%+ of the market overnight.

I don't think so. (though I type this from OSX)

Re:An obvious explaination....

[Drakonian]

Revenue != Profit

Re:An obvious explaination....

[TheAwfulTruth]

First of all their current market share is closer to 2%, and frankly yes, maybe not "overnight" (Sorry about the hyperbole) but I think they actualy would have a huge chance for success considering that the mac already has a huge cache of mature and popular commercial applications for it. But considering the price difference and the availability difference for their hardware, the choice most logically falls wo wintel. If they evened the hardware playing field I believe they would be MUCH more successful.

Re:An obvious explaination....

[TheAwfulTruth]

If their sales of OSX went up 10x at $120 (with 95% profit margin on materisls) then yes they would be making more money. And as I said before if they got a 20% market share using osx that could only lead to an INCREASE on properly marketed apple branded hardware as well.

Re:An obvious explaination....

[SewersOfRivendell]

NeXT already tried going the x86 software route with OpenStep, and it got them nowhere. I don't think Steve Jobs is likely to make that mistake again. Whatever else you say about him, he does learn from experience.

Re:An obvious explaination....

[Llywelyn]

While I appreciate your point, your math is way way off.

The original poster:

"Apples market share could go up 10x overnight"

Not by 10 people, but an order of magnitude--the original poster is way off in this assertion, but let's run with it for the moment. Let's also work with raw numbers because I don't know the number of people who currently own a computer off the top of my head (it also doesn't affect calculations, so long as we are consistent).

Thus, lets say that 1 million people currently are using MacOS X (yes, I know that this is low--but it makes math easy). This means that, for their marketshare to go up "10x overnight" 9 million people have to purchase MacOS X.

9 million * $129 = $1.161e+09 > $1 billion dollars.

Now, I still disagree with the original poster's assessment and think that your point is valid, but I had am impulsive need to correct the math :-)

Re:An obvious explaination....

[GlassHeart]

Apples market share could go up 10x overnight if they released Mac OSX for x86.

I think you mean MacOS X's market share, not Apple-branded computers. In any case, why would it?

Is it because of the iApplications? Well, a Windows user needs to throw away all his old apps and pay $130 to get them. I think most switchers see them as a nice bonus, but not a primary factor.

Is it because of Mac-only applications? I think we can safely rule that out.

Is it because of a Unix core? Since Linux hasn't exactly ruled the world yet, I think we can rule that out, too.

Is it because of Aqua? Possibly, but think about losing all your apps.

Is it because OS X can comfortably dual boot with Windows? Oops, it can't. First Apple will need to support VFAT and NTFS, then write HFS drivers for Windows.

So who exactly will buy these things? People who always wanted to run Unix on the desktop, can't afford a (used) Mac, and can't figure Linux out? They also have to somehow afford OS X and repurchase all their old apps. Ten times, overnight, you said?

Personally, I doubt even the "I'd switch if only" subset of Slashdot would, but that's because I'm cynical. I'd like to see where you got your projections, though.

Re:An obvious explaination....

[jbolden]

Lets play the game. Lets say OSX sells for $120 retail that means it wholesales for about $75. I'll grant $5 for materials. With 10x the sales what about additional customer service and support costs? Also there is shipping. Lets say this makes it $60. Even with 10x their current OS sales apple is still losing money fast now assuming no hardware sales.

Now if you think the x86 version of OSX leads to an increase in PPC sales I'd like to know why. What does OSX x86 do to increase someone's willingness to pay Apple's hardware margins?

Re:An obvious explaination....

[Rinikusu]

Be ported to the x86 and experienced the kind of boom in marketshare the other poster mentioned.
Where's Be again?

Re:An obvious explaination....

[scrotch]

Thank you for concern, and by all means meet those impulsive neads.

My math perhaps doesn't represent my point well at all :)

Part of the thought back there was that if only a tenth of the people that want OS X enough to pay for it for a PC bought a Mac instead, Apple comes out ahead. Such that if instead of 10 copies of OSX they sell a single Computer, their revenue is higher.

Revenue does not equal profit, as the other poster posted. But selling a computer also has longer term benefits, like the 3 or 4 major OS upgrades that owner will buy for that box.

Re:An obvious explaination....

[Chmarr]

Okay, I have a developer account... lead me to the documentation.

Re:An obvious explaination....

[akuma(x86)]

The XBox hack is something a little different. People hacked it to run unsigned/foreign code on proprietary closed hardware. What people would do in the Apple case is to try to run Apple software on non-closed open x86 systems.

IBM was the first to do this with the original IBM-PC and then some people hacked the BIOS and created the 'clones'.

I'm sure that others could clone an Apple-based system, but they would do so on pretty weak legal grounds these days of the DCMA.

Re:An obvious explaination....

[n3m6]

or they could release an OSX version for x86 which will run only on AMD Opteron. That would drive AMD, and Apple's software sales.

Re:An obvious explaination....

[TheAwfulTruth]

I did not mean to indicate that x86 OSX would lead to PPC OXS sales. I figured it was clear that by switching to x86, the PPC version of OSX would never have happened. I meant that there would still be at least as many die-hards buying Apple branded x86 boxes with their "WOW" factor designs as there are now, maybe even more as more and more people got real exposure to OSX.

As far as accounting goes, how about we just look at the industry instead of making up numbers?

Apple is just about the only company left on the lower end where the OS and the hardware are tied togeather. And they are only barely alive. Everyone else figured it out long ago or died. Even Sun has Solaris for x86 and they are running Linux on Sparcs. Even IBM is wising up!Why can't apple?

Actually if you read todays WSJ there is an article that pretty much gives insight to why. Steve just isn't all there. That's all there is to it. He's just a /little/ too much of a dreamer and a /little/ to little of a realist. In the end I thing the slightly upset scale is doing nothing but hurting them and their customers.

Re:An obvious explaination....

[jbolden]

Everyone else in the OS market is dead (or very small) as well. Its not like selling OSes for the x86 has proven a winning strategy. As for Solaris x86 that is a big money loser which Sun wanted to discontinue but couldn't for fear of alienating their developer community.

As for IBM they make their money doing what Apple does with zSeries, iSeries, pSeries. Further they plan within months to go head to head with Intel.

Anyway your original point was that selling OSes they could make the money they do today. The "made up" numbers show that isn't the case even if they do great selling OSes. The point is that they sell hardware so that they can charge a lot for the OS.

GNU-Darwin, x86 compatibility

[generationxyu]

It's not exactly Mac OS X, but it's the Darwin core -- http://www.gnu-darwin.org.

Apple did make an "x86-compatible" Mac a few years back, I think it had a 486 chip alongside the PPC (or even 040?) I don't remember too much about this, I think it worked by pressing Cmd-return, at which point it would switch to the 486, while maintaining state on the PPC. Essentially like the Orange Micro PC compatibility cards they used to make. (NuBus what!)

I'd love to see an Apple/AMD collaboration, either a licensed port of the whole Mac OS X to x86 architecture, or a dual-processor machine. It'd be pretty badass.

Re:GNU-Darwin, x86 compatibility

[Anonymous Coward]

Perhaps you are confusing GNU-Darwin with the actual Darwin codebase that is availible at opensource.apple.com ? There are other projects, including OpenDarwin (opendarwin.org), not only GNU-Darwin. Apple provides ISOs for Darwin for x86 from that site above, as does OpenDarwin.

Re:PPC 970 == Vaporware

[Capt'n Hector]

Vaporware? If I'm not mistaken, the PPC 970 is ahead of schedule. In fact, it's hitting the market a good deal faster than many other chips out there, so i wonder why you're calling it vaporware. IBM is not dragging their feet. On the contrary, they're moving extremely fast.

What's more, the PPC 970 is not shrouded in secret, (at least from an apple hardware point of view) If you think the 970 is shrouded in secret and is vaporware, I wonder what you think of the Moto G5.

PowerPC 970 and Power4

[clbyjack81]

I am a little ignorant on this, so please enlighten me. Can instructions for the PowerPC 970 be migrated to the Power4 chips without too much trouble? The point being, is there the prospect that Apple will put the Power4s in some new, really high end Xserves?

If so, that could let Apple break out of just the 1U market and compete with 2U and 4U servers with more than just two processors.

Re:PowerPC 970 and Power4

[Znonymous Coward]

Can instructions for the PowerPC 970 be migrated to the Power4...

If you asking will a Power4 32-bit application work on the PPC970 the answer is yes. If you would have read this, you would have answered your own question.

Re:PowerPC 970 and Power4

[jbolden]

Apple is not going to move to a Power4. OTOH IBM's development dollars on the Power4 worked towards the 970 and the 980 will be based on the Power5....

Enough already!

[psyconaut]

Why do we have to have a story about "MacOS on x86" every few months on Slashdot?!

-psy

Re:Enough already!

[angst_ridden_hipster]

Same reason you only see beautiful people on TV.

Manifestations of Wishful Thinking.

(Still... NeXT ran on multiple hardware platforms ... and there are a hell of a lot of beautiful people in Los Angeles, etc).

Re:Enough already!

[feldsteins]

Why do we have to have a story about "MacOS on x86" every few months on Slashdot?!

Because so many x86 users want Mac OS X on their cheap-ass boxes, and so many Mac users want Mac OS X on a cheap-ass box. Put simply: wishful thinking.

Re:Enough already!

[Atomizer]

They compliment the "Amiga is still Alive" stories quite nicely, don't you think?

In fact, I hear the next Amiga is going to be coming out on the PPC 970. ;)

Re:Enough already!

[pudge]

FWIW, I didn't even think of this as a "Mac OS on x86" story. We have two new big processors coming out, and we have Apple talking to AMD and we have someone talking about the IBM chip, and we have wild speculation on what will happen, only some of which involves "Mac OS on x86".

this is NOT offtopic

[Anonymous Coward]

...because Apple would be like Microsoft if they had the marketshare. In fact, worse: try doing anything to Apple hardware or software (excluding the open kernel) and see count the hours until the C&D nastygram.

Re:this is NOT offtopic

[GlassHeart]

Apple would be like Microsoft if they had the marketshare.

That's probably right, and maybe Apple will be even worse.

However, that's lazy thinking, because it implies that we should support an ethical company into monopoly marketshare, and count on it to do the right thing all the time. That's wrongheaded, because what we should do is to introduce and maintain competition instead. An Apple with 95% of the PC market might be disaster, but an Apple with 40% of the PC market might be nothing but good news for everybody except Microsoft. For one, many more open standards would probably be in effect, and not only because Microsoft neglected to butt in at the birth of the Internet.

PowerPC 7400 was "vaporware" too

[daveschroeder]

Remember, Motorola didn't even release the PowerPC 7400 (G4), much less any information about it, until apple introduced the Power Mac G4. Does that mean it didn't exist?

Apple probably can't play those same kind of "keep it secret until we announce our product" games with IBM, but keep in mind the only thing IBM has really done was introduce the concept for the processor at MPF. Judging from how Apple has rolled out new processors in the past, it wouldn't be surprising to find if further information about the 970 is being withheld at Apple's request (Apple being a potentially huge customer for this chip).

Also, remember that before CeBIT, IBM posted press releases on its German site talking about 970, the fact that it featured AltiVec, and how IBM was going to be demoing several 970-based blade servers at CeBIT. The press release suddenly got pulled and there was no further information about the 970 from IBM.

One way this could be interpreted is that the 970 is vaporware.

The more likely scenario, however, based on how apple has done releases of new processors in the past, including several iterations of the 7400 family, is that more information is being withheld until Apple releases a system featuring this processor. Then the floodgates will open.

The only reason we may know anything of it at all is that IBM felt it fundamentally important enough to present at MPF - we haven't heard a peep since.

Re:PowerPC 7400 was "vaporware" too

[Thumpnugget]

The more likely scenario, however, based on how apple has done releases of new processors in the past, including several iterations of the 7400 family, is that more information is being withheld until Apple releases a system featuring this processor. Then the floodgates will open.

Or alternatively, that information will be withheld not until Apple releases such system, but announces they'll be making such systems.

IBM is an 1800-lb gorilla that makes its money these days by tantalizing potential OEMs with the specs of new products to get them to sign on to manufacturing contracts months in advance of the actual release, and having an 80-lb chimp like Apple telling it not to talk about the 970 probably sounded like a real annoyance to IBM execs who only see the bottom-line income generated by these multiple partnerships. And when one partner gets out of line, they tend to whack back. Especially when its already doing that one partner (Apple) a huge favor.

Anyone consider that maybe that's the reason why Apple moved WWDC to June and to Moscone Center in SF? That IBM can't stand sitting on the specs of these new chips and is forcing Apple to go public with the details so they can start selling them to other people?

AMD and OSX

[Gleep The Dragon]

My guess is that if Apple starts using AMD chips for anything it would be for a server product where you don't need high end graphics, just fast data schlepping.

Stop it. AMD + Apple == Hypertransport

[wazzzup]

So Apple says this is the year of the laptop, right? If I'm not mistaken, AMD chips run hotter than just about anything out there. So who wants a laptop with 15 minutes of laptop life and the capability of burning your wang to a small, blackened stump of carbon (or for the ladies a sizzling fajita)?

Besides, are they or any of the Mac software vendors going to support two versions of their Mac products? No.

OK, What's new?

[Biff Stu]

The most recent article linked in the post is Feb 27. While I am looking forward to the enhanced Mac performance that the 970 promises, this doesn't bring any new information to the table.

Apple & AMD... rumors... again

[Lord Satri]

This was posted yesterday: again rumors for Apple//AMD.
Linked to The Register article.

Hypertransport

[vanced]

Does nobody remember that both AMD and Apple sit on the consortium for Hypertransport? If you look at Apples current lagging hardware specs you'll see a need for two things. A faster Bus and a faster CPU.

AMD == Hypertransport && IBM == P970

Re:Hypertransport

[deadfishhotmail.com]

Thank god someone else realized this...I was scrolling down waiting to see soething on hypertransport but it looks like about everyone forgot. It's not like CPU's are the only thing AMD and IBM make.

I think everyone is suffering from a severe case of tunnel vision.

Re:Hypertransport

[astrox]

Yes, I remember. If you take that same fact and think about it for a couple milliseconds longer you'd realize that Apple doesn't need AMD to get Hypertransport. And to expand on that, IBM's PPC 970 already has support for a 6.4 GB/s usable bandwidth bus, Hypertransport or not.

Re:Hypertransport

[shotfeel]

Somebody has to make the hardware to support Hypertransport, don't they?

Wouldn't it make as much sense for Apple to source this out to AMD as to do it themselves?

With regard to the 6.4 GB/s bus on the PPC 970, you're going to need hardware support to bridge this bus with hypertransport (or whatever else you use). Wouldn't that also be a place where AMD could help out?

Re:Hypertransport

[TinkersDamn]

Mod Parent up! Hypertransport is *why* they're working together! If they implement Hypertransport on the P970 like the Opteron does, ie. 3 channels, then you can go to 8 processors really easily. One of the challenges of Altivec is that it swamps the bus, with Hypertransport they can solve that issue, and got multiprocessor in a big way. Expect this in server versions first. As to the x86 issue, is there anything stopping them from building a multiprocessor with an Opteron instead of one of the P970s? With a Hypertransport connection and local RAM you don't have to worry about welding two disparate busses together. This could be *way* cheaper than the older solution which required emulating a Wintel box...

Application Split

[Oculus Habent]

While it's extremely unlikely that Apple would pursue two completely different platforms at the same time, I think we would be most likely to see different processors in different markets, i.e. the 970 in the consumer line, the Opteron in the server line, or some division like that.

Re:Application Split

[Anonymous Coward]

Ummm... considering that the 240 opteron would be cheaper than the 970, that is highly unlikely.

Re:Application Split

[shotfeel]

A bit out of my league in this area, but if they wanted a better server CPU, why not go for the much more compatible Power4?

Re:Application Split

[lowmagnet]

Power4 chips have 2 processors on die, consume much more power than G4 or 970, and cost an order of magnitude more per chip. Believe me, I work for a company that uses AIX (where most Power4 chips go).

Lucky?

[Logic Bomb]

"Could we be that lucky?"

Define "lucky". You mean, not only having to find Mac software, but now having to find software for your particular Mac platform? appleppc.slashdot.org along with appleamd.slashdot.org? Developers throwing up their hands in disgust and walking away when confronted with a platform redesign two years after the last one? Sounds lucky to me.

Seriously, give whatever Jobs has up his sleeve a chance. If he wants a decent PowerPC chip, he'll get one.

Re:Lucky?

[Tide]

That simply isnt true. Those NeXT boys had the multiple platform thing down-pat. Simply check a box in project builder and your .app contains code for both platforms in one neat file, at only a 10% increase in size. NeXT ran on Intel, PA-RISC, and NeXT's own creations.

Re:Move away from Linux?

[tmark]

Not everyone wants to roll up their sleeves and start coding just to use "desktop" software. There *are* people out there who just need to write documents/work on spreadsheets/balance their checkbook, and not all of them share the Open Source agenda: do you really think they all ought to participate in Open Source, instead of just switching to some OS they feel suits them better ?

Something It Seems Everyone's Forgetting

[Farley Mullet]

Apple has heavily optimized OS X and the so-called iLife apps (iTunes, iPhoto, etc.) for AltiVec, the special vector instruction set that the G4 has. That's why OS X runs much more nicely on G4's (which have AltiVec) than on G3's (which don't). The reason all the buzz started about Apple migrating new Macs to the 64-bit IBM chips in the first place was that IBM introduced AltiVec workalike instructions for their new chips, so Apple could move up without sacrificing the AltiVec optimizations. Moving to x86-ish hardware would mean that they'd lose all the AltiVec optimizations they've made, so it seems unlikely to me.

Re:Something It Seems Everyone's Forgetting

[ipjohnson]

Could be done but the whole point of altivec was for optimization (i.e. speed increases) ... Do you think you'll still have the speed increase when you start transalting it?

Re:Something It Seems Everyone's Forgetting

[GreenHell]

I don't know much about MMX/SSE, or Altivec either. However, I do know something (granted not much) about the CPUs they run on, and that doesn't make it very likely for Altivec emulation if it follows the same trend.

Ever wonder why you can emulate an x86 on a Mac but why you can't emulate a Mac platform beyond the old M68k systems on an x86?
Simple, it's much easier to emulate CISC processors on a RISC machines than it is to emulate RISC processors on CISC machines.
Take a look in one of the previous posts somewhere, and someone's done a slightly better job of explaining why than I could ever hope to do.

I don't know for sure, but I'd assume this RISC emulation problem would provide a great deal of headaches (and yes, the translation would essentially be emulation) to the people developing the code to do the translations and would more than likely result in a signifcant performance loss that switching to a faster processor would be unlikely to offset.

(As I said, I'm really just making an inference here from what I know about other things. Feel free to prove me wrong.)

Re:Something It Seems Everyone's Forgetting

[RzUpAnmsCwrds]

"Ever wonder why you can emulate an x86 on a Mac but why you can't emulate a Mac platform beyond the old M68k systems on an x86?"

There are PowerPC emulators for Windows. They don't run Mac OS, but they do run Linux.

Re:Something It Seems Everyone's Forgetting

[GreenHell]

Much appreciated. Never heard of them to tell the truth.

Any ideas on performance?

not x86

[tjw]

x86-64.

There's a difference. You won't be able to run OSX on Intel chips, nly high end AMD chips (Opteron).

Someone had to say it....

[word munger]

This would give new meaning to the term "PC Card"!

<emote>groan</emote>

desktop

[g4dget]

For those of us wanting to get away from Windows, but feel Linux is still not ready for the desktop yet, this might make Apple a more viable alternative.

Come on, don't hide behind "not ready yet". Just spit it out: "I don't like the Linux desktops". Now, that wasn't too hard, was it?

That's fine, I don't like the OS X or Windows desktops either. That's why they make so many different kinds. But let's not pretend that there is a single desktop that is oh-so-much-better for everybody than any of the others.

Your statement makes about as much sense as saying that "vanilla ice cream isn't ready yet for the kids of America, but strawberry, which is clearly so much better, is too expensive".

Re:desktop

[leifm]

That's not a fair assertion. I don't feel Linux is ready for the desktop either, but I also miss some of the flexibility it gave me now(I replaced my Dell with an iBook). I liked the way I had my Gnome desktop set up, I liked KDE, I liked a lot of the applications. But then you had the dependency chasing to deal with, fucking with fonts to get thinks looking decent, the inability to copy and paste between some apps, little things like that. I believe linux is very close to being a great desktop, but it's not quite there. OS X gives me much of what linux did, but also a GUI I don't have to fuck with, and that's why I am on OS X.

Re:desktop

[g4dget]

I liked the way I had my Gnome desktop set up, I liked KDE, I liked a lot of the applications. But then you had the dependency chasing to deal with, fucking with fonts to get thinks looking decent,

If you are having dependency problems on Linux, maybe you are just using the wrong distribution. Gnome and KDE just installed and update automatically for me. At some point (once they were ready), FreeType and antialiased fonts just appeared as an option after doing a software update on my Linux box.

And, of course, while installing Gnome or KDE may have been cutting edge on some old version of RedHat, I suspect that the latest versions just ship with everything ready and enabled (I think RH9 has Gnome2 and antialiased fonts out of the box).

So, if "dependency problems" and "lack of anti-aliased fonts" are the major issues why you think Linux isn't ready for the desktop, those seem to have been overcome by at least some major distributions.

the inability to copy and paste between some apps, little things like that.

I dunno--in that regard OS X doesn't seem much better. Applications like IE, Mozilla, the i* stuff, and the OS 9 stuff all works differently. For example, a lot of apps ignore my OS X keyboard mappings (presumably because they go through older APIs) and Mozilla gets confused by some some OS X pathnames because it's still using some OS 9 APIs. A lot of stuff I would expect to be able to drag and drop I can't drag and drop on MacOS (Safari is particularly limited in that regard, but so are other apps).

Every desktop has its own historical baggage. On Linux/X11, the three different cut-and-paste mechanisms are a minor nuisance, on OS X, it's the three different APIs (emulated, Cocoa, Carbon). On Windows, well, let's not even go there. Pick your poison.

Re:desktop

[leifm]

You make a good point, and I shouldn't have bitched about those things since I haven't really played in Linux for close to a year. Our household has two iBooks and two white box PCs. My girlfreinds kids have the PCs, which run XP. I would love to get them Macs, but I don't have the cash, and I would love to get them to stop using Windows, but one attempt at that was very ugly. Maybe that's where Linux developers need to look in terms of desktop usability. Go find a 12-16 year old girl, put Redhat on her box and then listen to the never ending bitching. Write down the issues and you have what Linux needs to own the world:)

Re:desktop

[g4dget]

Go find a 12-16 year old girl, put Redhat on her box and then listen to the never ending bitching.

That's what 12-16 year old girls have 12-16 year old boyfriends for, who have ulterior motives for putting up with the bitching.

Write down the issues and you have what Linux needs to own the world:)

Well, more mature nerds do that for wives and parents. From my own personal experience, I have to say that the differences in bitching about Windows, OS X, and Linux from those sources are small compared to the absolute levels of complaints--all of those systems have serious usability problems, and it doesn't look like anybody knows how to fix them.

Re:desktop

[Chucker23N]

"Come on, don't hide behind "not ready yet". Just spit it out: "I don't like the Linux desktops". Now, that wasn't too hard, was it? "

That's not what he was trying to say.

An operating system whose premier window server doesn't even have one standardized widget toolkit is simply inferior regarding its GUI when compared to Mac OS X, Windows or even BeOS.

Dont forget

[Erect Horsecock]

The Huge Future Apple CPU Thread. A very informative read focusing on the PPC 970, 980, and Moto 7457.

Motorola "G5" already shipping

[daveschroeder]

Motorola's 5th generation of PowerPC host processors - the "G5" - have been shipping for quite some time.

They're just not at all interested in the desktop market anymore.

Re:Motorola "G5" already shipping

[tmasssey]

That's in fact why Apple is in the hurt they're in.

They signed a deal with Motorola to only buy from Motorola any chips that Motorola builds. IBM has faster G3 and G4 processors than Motorola, but Apple can't buy them. Why? Their deal.

Why doesn't Motorola ramp up the speed of the G4? AltiVec. AltiVec is *critical* for things such as DSP operations. It's AltiVec that makes PPC a powerful option for switch and router designers and other embedded marketplaces. These applications don't need 2GHz CPU's, they need efficient 400MHz CPU's. Motorola focuses on that marketplace. And *that* is why Apple can't get a CPU faster than 1.2GHz.

IBM, though, builds lots of PPC computers: AS/400's and RS/6000's (excuse me: eServer i-series and eServer p-series). They need fast CPU's, and they don't care about the DSP garbage. Though it seems that with the 970 they've included both.

Re:Motorola "G5" already shipping

[dadragon]

They signed a deal with Motorola to only buy from Motorola any chips that Motorola builds. IBM has faster G3 and G4 processors than Motorola, but Apple can't buy them. Why? Their deal

Actually, nearly all G3 processors come from IBM these days. My iBook has an IBM G3 600mhz in it. All newer iBooks do too.

Re:Motorola "G5" already shipping

[Shuh]

Take a closer look. The G5 is the 85xx series. Moto is not shipping, and has not shipped, an 85xx series PPC.

Take and even closer look... the 8540 was one of that guy's links. I really have to wonder about this... because a DDR-RAM controller is missing on the 7455, but is there on the 8540.

This bears some looking into!

Re:Motorola "G5" already shipping

[Guy Harris]

They're just not at all interested [motorola.com] in the desktop market anymore.

They may not be interested, but they still mention it on the parent of the page you cite:

Motorola Host Processors, Integrated Host Processors and Microcontrollers based on the PowerPC RISC architecture can power your notebook or desktop PC, workstation, or server, as well as high-end networking and telecommunications embedded applications.

Speculation du jour

[angst_ridden_hipster]

I think everyone(*) is anxious for Apple to jump ahead in the GHz game. Considering how fast the Intel/AMD folks are cranking up the chips, it feels like we're being left behind.

We can talk until the cows come home about how CISC/Hybrid MHz are not RISC MHz, but the fact is we all want our machines to be faster. Even if they're already really, really fast.

But I can't see Apple making a transition to a platform that's not binary-compatible with PPC. It was painful enough when they went from 68xxx to PPC, and then to force everyone to buy all their applications again with the transition from OS 8/9 to OS X.

To do it again, within a year or two of the last major transition, would be disastrous. While I'm sure the software companies wouldn't much mind forcing everyone to buy a new version of all of their applications, how many users would put up with this? How long would people wait for Photoshop 8?

(* at least all the Apple users, and maybe a fair number of Unix/BSD users)

Re:Speculation du jour

[tgd]

I think very few people are anxious for Apple to jump ahead of the game. Contrary to what the /. crowd makes the tech marketplace look like, very few people want plexiglass windows, neon green alien decals, hard drive cables that glow under blacklights, the latest 7ghz processor and a $400 video card.

I think everyone(*) wants their computer to be able to take care of what they want it to, and everyone(**) is probably pretty happy with where Apple's hardware is right now, because everyone(***) knows how much more efficient they are on those computers.

(* everyone except the vocal minority of computer users represented on /.)
(** everyone who isn't in college living off Mom and Dad's money so they've got cash to burn on the latest and greatest hardware)
(*** everyone who's actually used a modern Mac day-to-day, and just smiles knowingly when they see stories talking about how Macs are overpriced, underpowered, or destined to fail when Linux wins the desktop on Slashdot)

Re:Speculation du jour

[jbolden]

Last year IBM demod a buggy 6ghz 970. The 970 is also much faster per clock cycle than the G4 (something like 2-3x). The 1.8 ghz 970 will crush everything except the 3ghz PIV with HT. If the 2.5ghz 970 hits soon....

Mac OS X on PPC and X86?

[iJed]

It is certainly possible that we will see Mac OS X on x86 at some point in the future. It is another question, however, if Mac OS X x86 will be able to run on any x86 hardware and not just proprietary hardware from Apple.

It is rumored that Apple do currently have Mac OS X running on x86 in the form of project Marklar and that it is kept up-to-date with the PPC version. It is also true that NeXTStep ran on 68K, x86, Sparc and PA-RISC so this shows that the Mac OS X team is likely to capability to easily port this software.

I suppose all we can do is wait and see...

Apple will not use two platforms

[stilwebm]

For Apple to run OS X (or it's descendants) on Apple branded computers, they would have to create serious confusion and frustration among Apple users. Apple users don't want to think about "what processor version installer should I run." Sure there are so-called fat binaries that include binary code for multiple platforms, just as Apple used when transitioning between Motorola 680x0 (aka 68K) processors and PowerPC processors. However, that was a one way transition. People knew that PPC was the future or all Apple as well as an upgrade. PowerPC processors could run 680x0 code through emulation quite well with no user intervention. With a transition to x86, however, Apple would have a huge problem with backwards compatability for existing applications. PowerPC emulators are in the works for x86 (actually, at least one will work on most modern architectures), but believe me, they are not an acceptable solution for production use - especially among most Mac users.

Using two simultaneous platforms is a big problem for sales and developer relations. Which is better? Why even bother with the other platform then? Or, why is the new platform so much better yet it has little available software? Why bother porting to the second platform when sales are sluggish on that platform? Then existing customers get angry. Why is my platform being abandonned? New customers feel the same if the gamble doesn't pay off and gets killed. The only partial exception is if one platform does not substitute for the other, say appliances vs. desktops and servers. Think Sun's purchase of Cobalt.

Re:Apple will not use two platforms

[stilwebm]

They aren't complete emulators, but more processor simulators.

They are:
* PSIM http://sources.redhat.com/psim/
* MicroLib http://www.microlib.org/

Re:Apple will not use two platforms

[afantee]

Why do you think Apple can't support 2 platforms, while IBM and HP have been selling multiple architectures for decades.

Darwin supports X86 right from the beginning, and there are strong evidence that OS X does indeed run very well on X86, as well as G4 and G4.

Apple could instantly double its OS market share by releasing OS X for X86, and it's not impossible for them to make Macs using both PPC and x86.

Re:Reading OSnews lately?

[stratjakt]

linux desktops have had "a lot of promise" for the last decade.

It's always been a case of "just wait, the next release will solve everything!". Zealots chant it as their mantra.

It's not going to happen. FOSS, by it's very nature, will never produce anything more than a patchwork clone of other desktops.

There's no technical reason that a desktop as slick as OS/X couldn't be built on top of linux the way OS/X is on top of BSD, but that kind of effort requires management and discipline. Only a corporate effort can pull that off.

In the OSS world, if you dont like the way a projects going, you go ahead in your own direction. And that's fine, after all, its unpaid hobbyists doing the work.

But in a corporate environment all the coders have to be thickskinned when their nifty super-duper new subsystem proposal gets nixed, and buckle down and get the job done. If linux desktop was a corporate project, there would be no KDE vs Gnome vs Enlightenment vs blah vs blah discussions. There would be one project.

Short of some for-profit coming in and getting it done (which I think may eventually be the case), I just cant see it happening.

Re:Stop it. AMD + Apple == Hypertransport

[realdpk]

Hypertransport - good point. They could be using that. Why you didn't mention it in the post is beyond me though.

AMD doesn't run incredibly hot any more - check out the Barton and Opteron chips some time.

Re:Stop it. AMD + Apple == Hypertransport

[10Ghz]

If I'm not mistaken, AMD chips run hotter than just about anything out there.

A common claim. Unfortunately it's wrong. Athlon XP doesn't really run any hotter than Pentium 4 does for example. In fact, you coulöd say that XP runs cooler than P4 does.

For facts on this issue, go here:

http://www.aceshardware.com/read.jsp?id=50000365

all wrong...

[liloconf]

sorry but apple is a hardware company not a software one. If you check there income you'll see they make very little on there OS and ilife products. If Apple came out with a new computer using an AMD chip they would be hurt drastically by those building there own apple computers instead of spending the premium in the apple store. The ibm 970 will happen, AMD might be involved but only with helping Apple on hyperthreading i think...

Apple could do it with Mono

[ShatteredDream]

If apple were to officially port Mono to OSX and help build it into a robust development tool and environment then it would be possible for them to build .NET bindings for OSX that would run on both the PPC 970 and Opteron.

Re:Apple could do it with Mono

[mingot]

Haven't you heard? Microsoft is bad. Mono is based on a Microsoft product. Therefore Mono must be a waiting devil that will eat your children and impregnate your pets. Yeah yeah, mod it down baby.

Concept....

[Anonymous Coward]

even if Apple doesn't move to x86(which would, IMO, make them either the next M$ or dead), why can't the OSS community make a comparable variant? I mean, the whole Cocoa app structure is built to the OPENSTEP standard, with some whizbang additions. Why we start with GNUstep and make a source-compatible version using the Linux kernel? I mean, it would be fairly easy, much easier than cloning Windoze 'cause the groundwork is already done(again, GNUstep & WindowMaker).

Re:Concept....

[shotfeel]

Instead of using the Linux kernal, wouldn't it make more sense to use Darwin? Its the core of OS X, runs on X86 hardware and is open-source.

AMD thing in bigger context

[daveschroeder]

Or, instead a bunch of wild speculation, why not realize that Apple and AMD are both a part of the HyperTransport consortium and are (presumably) both very interested in 64-bit computing on the desktop, and that:

1. One of HyperTransport's most commonly supported speeds is 6.4GB/sec;

2. Apple is desperately in need of a revamp of the entire desktop architecture, especially memory and system bus (aside from processor itself);

3. The IBM PowerPC 970 cooincidentally supports a system bus speed of 6.4GB/sec.

Doesn't the HyperTransport relationship seem a bit more logical than all this off-the-wall stuff about Marklar, Apple switching/adding processors, etc.?

Re:AMD thing in bigger context

[Brandon Sharitt]

Exactly. Even though AMD's most visible products are their CPUs, they do a lot more than just that.

Re:AMD thing in bigger context

[bnenning]

Doesn't the HyperTransport relationship seem a bit more logical

Exactly. A year ago it might have made sense for Apple to switch to x86, but with the impending 970 release it would be silly. It would substantially reduce the currently huge demand for the 970, as buyers would fear the machines being obsoleted if Apple abandoned the PowerPC entirely. But HyperTransport is win-win for everybody (well, not MS and Intel, darn).

Re:AMD thing in bigger context

[Angelwrath]

Yes, this makes far more sense - the AMD exec may consider Opteron and HyperTransport to be integral technologies designed with each other in mind. He may have thought the reporter was asking specifically about the Opteron with regards to the Tier 1 comment, and the exec responded with an admission of work on the HT project, with nothing to do with Opteron.

As for the speculation - I doubt Apple is going to switch to Opteron. The PowerPC 970 will be the easiest transition to current performance hardware that could hope for, and is "good enough" to Opteron performance that the minimal gain would be received at a tremendous cost and hassle and potential risk ot the company. Highly unlikely. We already know that Apple has plans for the 970 by reading the Asian newspapers and tech websites

On the other hand, how about emulation? Microsoft bought Connectix for a reason, but what isn't widely discussed is talk of new emulation technology that may potentially find itself into Connectix products that would have revealed a dramatic rise in Windows emulation performance. Who knows what might have been, but certainly Connectix is the kind of company that would have investigated this, out of business interest.

Did Microsoft squash a technology that would have made Windows on the Mac "good enough"? That would be an interesting question to try and solve. We've all heard by now they're paying id waste money to prevent id from releasing Doom 3 into our hands, does it surprise us that they specifically act to squash new products? I've heard another company has access to the same technology.

Whatever the case, the remainder of 2003 should be very interesting to Apple.

Re:AMD thing in bigger context

[Angelwrath]

They would squash Windows performance on the Mac because if WinOnMac became that popular, people could do things like switch to Mac, but use VPC for individual programs like Simply Accounting, AutoCAD, and other PC-specific programs. That would significantly reduce the chances of those people using MS software, because there are a ton of PC users who stick with the PC over just a few apps. With Office, Internet Explorer and Entourage for Mac, plus an integrated, simple environment, and a platform that doesn't crash as much as PCs do anymore, the Mac is interesting to a lot of PC customers.

So there's the answer - it steals all the potential related software sales from Microsoft, and hands them to Apple.

Re:Chimera Cons

[Morky]

this means that apple must do something drastic and something soon. but what are the alternatives?

What are you talking about? Didn't you at least read the post? Aren't you a little curious about the PPC970 and what kind of performance to expect? Why would you even write a three-paragraph post on a subject you have no clue about? I hope you're just trolling.

if MS can port windows to handhelds, why can't Apple do it?

Apple did it before anyone. Ever hear of the Newton?

I doubt it

[p3d0]

Look at the "success" IBM had making OS/2 run Win32 software. Result---developers only target Win32 because they get OS/2 for free, and so nobody is developing for OS/2 any more.

Though I suppose if it were an "add-on chip" that didn't ship with Mac boxes by default, that might prevent the OS/2 syndrome. People would be less likely to rely on the Windows software running on Macs if most Macs couldn't run it.

Re:I doubt it

[jbolden]

IBM dropped the ball in a hundred ways on OS/2. I think this idea that Win32 compatability killed OS/2 is nonsense. There have been tons of Win32 incompatabile operating systems and few did as well as OS/2. "A better Dos than Dos a better Windows than Windows" was a major selling point. People btw did write OS/2 native apps: Microsoft Lan Manager being a classic example.

Re:I doubt it

[mingot]

OS/2 Did NOT run Win32 software. At least not by IBM's hand. And the offering that IS availible via a 3rd party (Odin) is not all that hot. So there was never a "write win32, get OS/2 for free" mentality in place.

The mentality was "Write win32, OS/Who? Isn't that thing dead?"

Re:I doubt it

[bnenning]

I think this idea that Win32 compatability killed OS/2 is nonsense.

Granted IBM's marketing was terrible, but I believe this was a major problem as well. I used OS/2 Warp in college and the lack of native apps was annoying. I recall a number of software companies whose response was "our Windows version works fine in OS/2".

Re:I doubt it

[jbolden]

I suggest you take a look at some of the OS/2 threads on slashdot for stories. I was a huge fan of OS/2 from version 1.3-3.0. But for example I got 1.3 for $99 from an 800# you needed to learn about via. word of mouth. IBM's OS/2 division spent millions running commercials telling people to demand OS/2 on their next desktop while their hardware division wouldn't ship OS/2 with the systems.

There were some good OS/2 native apps at different times. You really need to be specific about years here. Lan manager was a perfect example Dos / Windows systems couldn't handle modem + 1 app very well yet OS/2 286 systems did modem + ethernet + multiple apps fine. That created all kinds of neat lan server based apps (like POS systems).

By the 2.0/2.1 days Lotus 1-2-3 for OS/2 could support something like 512 megs of ram while Windows spread sheets could support 16 megs. OS/2 word processors handled huge documents much better than Windows ones. OTOH there wasn't really much demand for this functionality. The key home / small business advantage of OS/2 was it gave people in 1993 the ability to multitask windows and dos apps in a way that wouldn't be possible with windows until NT 4.0.

Take that away and what would OS/2 have brought to the table? It wasn't as good a server / network OS as the 386 Unixes like SCO (which was a very good product 10 years ago) and not nearly as app rich as windows.

Re:Move away from Linux?

[Llywelyn]

"Open Source requires participation; coding and community."

Does it? I don't think I have ever joined in the R-Project's mailing lists, but I use their software regularly. I know a lot of Linux users who lack the time or the skill to be kernel hackers.

You /do/ realize that most of us choose a platform that /does what we want it to/ and /works for us/ and not for religious reasons? For my purposes, Linux isn't ready for the desktop. If I joined the team, I could help it get ready, and in a matter of years, it might be with my help (it also might be without my help).

OTOH, I could just continue using my Mac and actually meet my deadlines and get my work done.

My spare time, incidentally, does go into an Open Source projects, either: The Swarm Project or Equation Service. You want me to start working on Linux now as well?

Speaking of education

[Anonymous Coward]

iBooks start at must run a Free OS, then by all means stick to them.

It looks like you never got any.

Re: That would mean the end of Mac apps

[DavidinAla]

If Macs could suddenly run Windows applications (without something like VPC), why would anyone write anything except Windows apps? The big companies that now target both platforms could just drop their Mac software and tell Mac users to buy the Windows version. Companies that now specialize in the Mac market could start making Windows apps and sell to both platforms. Apple would totally lose control of the integration that has made the Mac experience what it is today. I just can't see any other reasonable result of what the poster suggests.

Let's pull it all together...

[stienman]

Will we see Mac OS X running on two different platforms/CPUs? Could we be that lucky?

In short, NO.

Firstly, as everyone knows, Apple makes money off one thing, and one thing only - HARDWARE. They make great software only to sell their hardware.

The benefits of controlling the hardware are

1. A better user experience
2. Lower tech support costs
3. Better quality control
4. Specialized/customized designs with an eye toward aesthetics

They CANNOT allow others to create hardware upon which their software will run. This means that they have to use a special BIOS, and manufacture their own boards. IF they switch to an OS that can be run on an x86 processor (and custom mothboard/bios/etc), you will find, the very next day, a crack for the software which will allow it to run on any generic motherboard, and further down the line a BIOS image which will allow an unmodified software to run on a non-custom motherboard.

Right now they can control it because a 'commodity' PPC motherboard costs more than the same apple motherboard. It would surprise me if Apple wasn't applying some pressure to various suppliers to prevent the widespread availability of commodity PPC equipment which is very similar to Apple's own. This is common in the industry. Furthermore, they may even have a slightly altered/customized version of the various PPC chips they use.

The only way for Apple to play against WINTEL is to not compete - not competing means selling essentially different products. Apple would die if they had to sell their OS and try to make a profit at it - the company is simply not designed to compete against MS. (Although if they did Windows would improve dramatically)

Put another way, Apple is a whole user experience company. They don't want the user to go to a generic theatre, sit in seats made by some strange company, eat food purchased from GFS, and watch a movie made by three different movie studios. They want you in their theater, their seats eating their food, and watching their entirely controlled movie.

This is good for those who only want to deal with one company, and are willing to pay for it. They know their market. They may be trying to expand it a little towards the geek segment that play with software but don't care about hardware (we run unix!). It is unlikely that they will ever capture the imagination of the hardware geek, they know it, and they aren't courting us.

So stop posting freaking stories about OS X on any commodity hardware, ok?

-Adam

Re:Let's pull it all together...

[laertes]

They want you in their theater, their seats eating their food, and watching their entirely controlled movie.

No wonder Macs are such a niche market; the seats are alive!

Re:Let's pull it all together...

[ilsie]

In short, NO.

I am forced to post an Onion article from olden days.

Despite Claims, Long Story Not Made Short
SCHENECTADY, NY--Contrary to her pre-account vow, area resident Barb Schuyler's long story of how a series of cashier foul-ups at the grocery store Monday made her 25 minutes late for a dental appointment was not made short. "So then, it turns out the stupid woman forgot to ring in my Savers Club discount," Schuyler told friend Gloria Conlon nine minutes into the non-abbreviated tale. The story is the 1,643rd Schuyler has failed to make short since 1994.

Why do we need x86?

[jceaser]

Could someone please explain to me why there is such a "need" to have Mac OS X on an x86 processor? Why is it a good thing to run on a processer with 4 registers (8 if you use the address registers for non address calculations) and an outdated asm languages when 32 registers and risc is just so much fun? Their are a lot of different processors out there and I really don't think x86 is the best in the world. Why would anyone what to run any code on anything made by intel. I'm not trying to start a flame here, I just want to know why so many people want x86 over anything else (mips, sparc, hp-risc, power-pc).

Re:Why do we need x86?

[ocelotbob]

Because X86-64 doesn't have just 4+4 registers. They've added 8 more general purpose registers, plus 8 more registers for working on SIMD code like SSE and SSE2, bringing the total of general purpose and special registers to 16 64 bit registers and 16 128 bit registers. While 8-32bit x86 assembly is ugly, x86-64 has provided a good number of features that make it more like a good RISC processor. Same goes for Itanium, where technically it has 128 registers, with 32 of them being visible through "traditional" means, and the others being visible through a register stack mechanism.

Comment removed

[account_deleted]

Comment removed based on user account deletion

What ever happened to CHRP?

[nycroft]

I think it was back in '95 or '96. IBM and Motorola were in development of dual-platform supporting processor called CHRP or Common Hardware Reference Platform.

The Common Hardware Reference Platform (CHRP) Specification describes a family of machines based on the PowerPC(tm) processor that are capable of booting multiple operating systems including Mac OS, Windows NT, AIX® and Solaris(tm).

Wouldn't that have been cool? What ever happened to that idea? Here's the old documentation.

It appears that IBM has some information on their site that is still recent, dated Sept. 2002. Weird. I'd love to have one of those machines. PowerPC 970? Forget about it.

AMD fabbing 970 Chips?

[Alpha_Geek]

Why isn't this a possibility? Didn't AMD's fab in Dresden take over some slack for Motorola at some point in the past?

I think its quite possible that Apple wants AMD to fab some chips with some spare production time they might have. IBM might not be able to fab the amount of chips that Apple wants to have ready by September.

Seems like a reasonable non-conspiracy related theory to me.

Re:AMD fabbing 970 Chips?

[tmasssey]

I doubt it.

IBM is another of those companies that fabless chipmakers (such as Cyrix, when they were building chips) came to when they needed extra capacity. IBM makes an unbelievable number of chips, from PPC processors to x86 processors (there are still a *lot* of embedded designs that use 80186's, for example) to memory controllers, to you name it.

In fact, AMD doesn't have a lot of capacity for their own stuff. Their biggest problem is on the high end: .13 micron fabs. They have lots of lower-end fab capability, but it's unlikely that Apple needs that kind of capacity...

Re:AMD fabbing 970 Chips?

[stilwebm]

In fact, AMD doesn't have a lot of capacity for their own stuff. Their biggest problem is on the high end: .13 micron fabs. They have lots of lower-end fab capability, but it's unlikely that Apple needs that kind of capacity..

AMD is also spinning off their flash memory division into a joint venture with Fujitsu, of which AMD will own 60%. The flash memory division represents a significant amount of their fabrication capacity. Fab 25 in Austin is going to be dedicated to this venture, and the 50-50 fab venture in Japan between Fujitsu and AMD will also to be included.

http://biz.yahoo.com/bw/030331/315443_1.html

Re:success (or lack thereof) of PC cards in Macs

[King_TJ]

The biggest reason those cards weren't "wildly successful" was their price, if I recall correctly.

In the heyday of these offerings, it was about the same price to buy a complete, seperate PC system. Many folks said "Where's the logic in adding PC support to my Mac when I can own a full PC system for the same money?"

The only market they really captured was the niche of people wanting to run both PC and Mac applications, but not willing to give up any more space in their home or workplace for another computer.

Also, these devices were still add-on cards, which always lack some of the integration of having the compatibility truly "built in" to the system. The beauty of a PC, in many ways, is the "box of slots" nature of the thing. You have thousands of possibilities in the way of PCI, AGP (or in the past, ISA or EISA) cards. Want a special purpose graphics card? Just buy it and drop it in! Special high-speed serial ports for a multi-line BBS system, perhaps? Just buy a "Digiboard" and get 8 or more ports. With a PC on a card, you're limited to what's actually on the card itself, or what it's able to use on the Mac's own board.

While I'm not so sure Apple has any interest in going the "PC compatibility" route again - I do think it would be a much different story if the compatibility was truly on the motherboard.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Re:Move away from Linux?

[christurkel]

Yes, participation is key. I cannot code. I cannot write even a shell script, but what I do do is provide feedback to developers of open source software I use. Developers appreciate constructive feedback and suggestions for changes or improvement. I have never once encountered a developer who said, "Don't like it, code something else". I've gotten many "thank you for your feedback"s or "I'll keep that in mind for a future version". In this way, the whole development process improves.

You do not have enough info on the Chip.....

[williamyf]

ArsTechnica to the Rescue:

* Inside the IBM PowerPC 970 Part I: Design Philosophy and Front End
http://arstechnica.com/cpu/02q2/ppc970/ppc970 -1.ht ml

* Ars Technica Newsdesk A Brief Look at the PowerPC 970
http://arstechnica.com/archive/news/103475624 5.htm l

* Ars Technica - CPU and Chipset Guide
http://arstechnica.com/cpu/

Hope it helps fill that Gap.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Multiprocessor 970 is a benefit for apple

[Master Of Ninja]

If you read the IBM link, you can see that the 970 is multiprocessor enabled. Once apple gets their hands on it they can easily create 2 way systems, and probably 4 way systems and up. I'm not sure about benchmarking, but linking processors in this way will help offset the x86 processor speed advanatge. And with IBM technology behind them, I'm sure it's easily possible.

BTW, I think AMD are trying to pull off a similar trick with the multiproc. Opterons, and eventually Athlons.

Re:Multiprocessor 970 is a benefit for apple

[Jeremy Erwin]

The PowerMacs are already dual processor machines.

Re:why not a SMP system with AMD and G4/IBM970 ?!?

[stilwebm]

Because cost of parts, manufacture, and engineering would be ridiculous.

two suppliers

[gtmac]

Surely the answer to the AMD rumors is obvious. Apple can not be dependent on a single processor supplier. Motorola are rapidly removing themselves from the game. When the IBM 970 comes out the G3 and G4 will be dead within a year. Motorola have no processors to complete and are heading deep into embeded land.

Apple need another supplier so they limit their risk. They maybe getting AMD to fab a PowerPC type chip.

Alternatively....

Maybe they are just going to use AMD64 chips to build 8 and 4 way XServes?

NeXT used to have fat binaries compatibility across NeXT Black hardware, Intel, Sun, HP and Alpha.

Anthony

Re:two suppliers

[noewun]

Except that all G3s used by Apple are made by IBM.

Re:two suppliers

[gtmac]

except?

IBM - G3 DEAD

MOTO - G4 DEAD

IBM 970 - ALIVE

equals one supplier IBM.

Re:two suppliers

[mixmasta]

Nope, the G4 is still great for notebooks, etc

Re:two suppliers

[Frodo2002]

I don't think you are entirely correct. First of all, IBM manufactures the G3 as far as I know. Secondly rumour has it that the G3 can go much much faster than it currently does. Apple does not buy faster G3's from IBM because it would look rather silly if your G3 had a higher clock-rate than your G4. (This is ignoring of course that the G4 has that altivec unit which means that it would still beat a faster G3 on altivec optimized apps. But your average consumer probably does not understand these things.) Advantage of getting everything from IBM? You keep your G3 line going, but ramp up the speed considerably. The G3 goes into low end laptops. You drop Motorola completely and put a powered down version of the 970 into your high end laptops. (Rumour again is that at 1.4 GHz the 970 consumes energy at the same rate as the current G4's). The downside? As you correctly observed, Apple then has all its eggs in one (IBM) basket. But the situation does not seem to be as bleak as you make out.

Re:two suppliers

[aphor]

Except that the instruction decode unit is a modular part of AMD's chip offerings... As in, the AMD chips execute their own native sub-instruction internal code, and they have a programmable front-end to translate the x86 stuff. If you wanted a PPC chip from AMD, Apple would only have to give them the specs, and EXISTING AMD chip cores could be quickly (software) adapted to execute PPC machine code.

All that's left is the pinouts and power/heat-dissipation requirements to sort out. The AltiVec is another side issue, to which I'm sure AMD would gladly offer a next-gen solution...

Re:two suppliers

[Pharmboy]

Maybe they are just going to use AMD64 chips to build 8 and 4 way XServes?

This is EXACTLY what I was thinking. Of course, with opendarwin, someone will have some fun trying to port osX over to the AMD anyway, if for no other reason, than just to do it. (like Xbox, etc.)

I would love to be able to have more choice, and IF Apple did port osX over to the new AMD64, I would buy one in a SNAP, just to see if we can finally replace our windows boxes.

Re:two suppliers

[noewun]

The "except" being your suggestion that the G3 will somehow magically die when the 970 comes out. I wouldn't be surprised if it lasts for a while, as the IBM G3s can scale to 2 GHz.

Re:two suppliers

[Datafage]

I was under the impression that fat binaries worked by having specific code for all supported architectures, not by being very high level, and thus could be as optimized as the programmers cared to make it.

Re:two suppliers

[Pharmboy]

It's a very nasty situation when an OS has two different processors (and two different sets of binaries).

Linux, the cousin of BSD (the parent of OS X) runs on many processors, as do many of the other *nix varients. It would be much easier to run a *nix based OS on multiple platforms than Windows, because it already runs on other platforms, and was designed to from the very beginning. Berkeley designed 4BSD to do exactly this. Windows' support of other CPUs, is a less sucessful story.

Even though the one article focuses on 4 to 8 way desktops, the server market would be the likely first target: You can get more per unit out of the gate, and less application support is need to get them in the field. Put a 4 to 8 way box, with 16 to 64 gigs of ram, and you have a great web server. It might take longer to get all the multimedia and other desktop niceties up to snuff, but I would bet that support for apache, bind, sql, and other OSS services would come fast, since OSS can potentially develop faster when properly motivated.

Since the 970s are designed specifically for SMP and to be reasonably priced, and the server market is not sold purely by the gigahertz rating, but rather by real world performance, AND it being produced by IBM who is very likely to support Linux and *nix in general on this CPU, these could get popular fast.

I wouldn't want to own any SUN stock when it comes out.

Re:two suppliers

[evilviper]

Linux, the cousin of BSD (the parent of OS X) runs on many processors, as do many of the other *nix varients

There are several big differences though.

1. Most people using Unix on non-i386 platforms (or just Linux at all), are far more advanced than your average Windows/Mac users.
2. Most applications used on Unix are open source... That means the CPU hardly matters at all.

For Windows/Mac OS X, most software is binary-only, and companies are going to decide that it's not worth the effort of supporting processor X, when it only has a fraction of the users. So, which ever gains popularity will be the defacto only system to use, and users of the other will be out of luck.

Re:two suppliers

[gtmac]

you continue to miss the whole point. If you read what I said again you will find that my point is that Apple will want to have two suppliers of its chip. If the G3 by IBM scales to 2 Mgz so what.... they still will have one supplier IBM.

Re:two suppliers

[Pharmboy]

For Windows/Mac OS X, most software is binary-only, and companies are going to decide that it's not worth the effort of supporting processor X, when it only has a fraction of the users. So, which ever gains popularity will be the defacto only system to use, and users of the other will be out of luck.

Which is exactly why I said Apple on AMD, if it ever occured, would be on the server side, where the source is available because most is open source. Apple ships Apache with OS X, for example. Porting over the other server applications would likely come from the open source community. Apple would have to port over OS X, which is not trivial but since it is BSD it would be more portable than windows by a long shot. Porting the applications afterward would be much easier.

Keep in mind, Apple has gotten pretty serious about the server market lately. Their offerings NOW are decent for many applications. This would give apple a 64 bit server, with backward compatability for existing 32bit applications. This is one way to get into serving huge databases. Of course the 970 will offer similar capability, but is actually designed to be a desktop cpu, not server. The Power4 cpu, the big brother of the 970, is designed for server use. This would make the AMD cpus capable of mulitboot Linux/OS X/Windows more easily. It may also give better performance for Windows apps running on top of OS X.

I am not necessarily EXPECTING this to happen, but Apple has a better chance of developing a specialized OS for servers on a different CPU than Microsoft. If memory serves me correct, MS gave up support for other CPUs a version to two back. Its all wild speculation, but interesting possibilities exist. Remember, the OS X kernel has been opened up in the OpenDarwin project, so OSS programmers may be helping with the dirty work.

As to the users being more advanced: If you port the kernel on top of any cpu, and put aqua on top of it, or X11, which Apple is also involved in, then how geeky you are is not longer a factor.

Re:two suppliers

[benzapp]

Yes, the original design spec of the NexGen Nx586 upon which The K6/Athlon/Operton is based allowed for the process to switch from 386 mode to its native RISC instruction set, which ultimately was to be PowerPC compatible.

Some of this I believe may have had something to do with the processor being manufactured by IBM around the same time as OS/2 PowerPC edition was being finalized... That all fell apart however.

Re:two suppliers

[sidb]

First of all, IBM manufactures the G3...

The 350mHz G3 that used to be in my PowerMac before I upgraded the processor was made by Motorola. I just dug it out of my closet and looked. Mine is rather old, though. IBM may make them all now days.

Enough Already

[glwtta]

Ok, we know, we all know - Apple is a hardware company, they make their money selling hardware. You are not any cooler for saying that yet again. It's common knowledge, you are not smarter than the rest.

Two suppliers...

[gtmac]

Surely the answer to the AMD rumors is obvious. Apple can not be dependent on a single processor supplier. Motorola are rapidly removing themselves from the game. When the IBM 970 comes out the G3 and G4 will be dead within a year. Motorola have no processors to complete and are heading deep into embeded land.

Apple need another supplier so they limit their risk. They maybe getting AMD to fab a PowerPC type chip.

Alternatively....

Maybe they are just going to use AMD64 chips to build 8 and 4 way XServes?

NeXT used to have fat binaries compatibility across NeXT Black hardware, Intel, Sun, HP and Alpha.

Anthony

Microsoft and VPC

[jbolden]

Actually I think there is a very good reason for Microsoft to buy VPC that has nothing to do with Apple. Intel has indicated they are switching focus over to the Itanium line, and over the next 5+ years the limits of the x86 platform are going to become more troublesome (things like 64 gig limit of addressable memory...).

The Itanium's x86 emulation is only so-so. VPC makes a product which allows an entirely alien architecture to run x86 apps almost perfectly providing you have an x86 OS. It would be possible for the VPC guys to take their PPC code and recreate it for Itanium to create the same level of compatability for Itanium architectures. That would be functionality that Microsoft would want to offer their customer base.

Re:Microsoft and VPC

[joeykiller]

Microsofts wants the Virtual PC technology as a means for customers to run older operating systems such as Windows NT 4.0 on newer versions of Windows.

As strange as this may seem at first, it makes sense: Microsoft is now in the process of stopping support for Windows 95 and Windows NT 4.0. But some customers still have to run applications that requires these operating systems, and VPC will allow them to do just that:

Quote from an article in Server Watch:

Part of Microsoft's attraction to Connectix's technology may be because it adds depth to its forthcoming Windows Server 2003 family by allowing existing NT 4 customers to keep their NT 4 applications running as virtual machines. This makes the technology a ready-made ramp to migrate customers from NT 4 to the new Windows platform.

Re:Microsoft and VPC

[homer_ca]

VirtualPC does nothing to address the problem of running IA32 software on IA64 hardware. It lets you host IA32 virtual machines on IA32 or PPC hardware through native execution on IA32 h/w and emulation on PPC h/w. Itanium can already run IA32 software correctly but slowly through its own emulation.

Most pundits think MS bought VPC so they could add the virtual machine feature to Windows 2003 server and they're likely to neglect the Mac version.

Re:Microsoft and VPC

[jbolden]

Itanium emulation is more buggy than VPC's. Most Windows software won't run at all on Itanium x86 emulation.

Re:Microsoft and VPC

[jbolden]

I've heard that as well. But in that case it would have made more sense to license the technology than buy the company.

Re:Microsoft and VPC

[kill-1]

Why?

Microsoft has always had enough of cash to just buy whole companies to get the technology.

Re:Microsoft and VPC

[klez23]

Microsoft doesn't license technology, silly. They digest it.

Re:Microsoft and VPC

[WhiteDragon]

except that this is Microsoft, they always buy the company :-)

Re:Microsoft and VPC

[jbolden]

Generally you would buy a company to be able to control / redirect its efforts. You would license a technology when you want it in essentially its existing form.

Re:Microsoft and VPC

[mixmasta]

maybe, maybe not.

Arent companies super cheap now? Arent stocks at a 6 year low?

Re:Microsoft and VPC

[jbolden]

I know in the case of the Alpha it was way more complicated than this. It involved on chip emulation ccapacity + very complex libraries used for compiling and recompiling + strange addons to the file system to allow for "versioning" that is invisable to everything but these libraries.

My guess is that the above is way to complex and specific to Alpha to port.

Re:Microsoft and VPC

[jbolden]

I think Connectix was a private company but I may be mistaken. As for super cheap it depends on your relative time frame, in general I'd say no -- they've gone from insanely expensive to just very pricey.

Re:Microsoft and VPC

[jkovach]

I think you're confusing two things here. One of my friends bought a Multia many years back when they were still current, supported systems and was running Windows NT on it. NT for Alpha came with an x86 emulator built in. In NT 3.51 it was a 286 emulator and in NT 4 it was a 486 emulator. I remember running Norton SysInfo on the x86 emulation in NT 3.51, and for the BIOS info or something like that it said "Insignia Solutions".

DEC later came out with a product called FX!32 which is probably what you are thinking of here. FX!32, which was a free download from DEC and not included with Windows, basically attempted to translate the x86 binary code to native Alpha code using various techniques. It could do the translation on the fly or do it offline before the program was started. FX!32 had higher performance than the Microsoft/Insignia emulation, but still paled in comparison with running native code.

Re:Microsoft and VPC

[jbolden]

I think you are right about my confusion. I was thinking the FX!32 was the emulation. Thanks for the correction.

There is NO conspiracy theory, AMD isn't just x86

[MarcQuadra]

I have a better reason: AMD can fab those CPUS easily and IBM has better things to do than fab chips for Macs. Apple needs to make sure it doesn't have to stop the assembly line for IBM to fab more CPUs. My guess is that Apple will have AMD produce IBM 970 chips alongside IBM. IBM probably doesn't want to be the first in line for Macintosh CPUs, there's not enough money in it for a multi-faceted operation like IBM. AMD can produce ample chips and they might be able to make a profit doing it.

This has nothing to do with Macs and x86, AMD produces a LOT more than just athlon chips, they'll be pumping out AMD-970s with their extra capacity.

Re:From the blatantly-obvious-dept.

[jceaser]

I was afraid that would be the answer.

970 info at Ars Technica

[cygnus]

if you're looking for 970 info, Hannibal has a decent article over at Ars Technica, and a followup is on the way. also there's a +1 thread of deth in Ars' forums.

Clickable offering

[bstadil]

Why not spend a few more minutes making the link clickable?

Inside the IBM PowerPC 970 Part I: Design Philosophy and Front End

Ars Technica Newsdesk A Brief Look at the PowerPC 970

Ars Technica - CPU and Chipset Guide

Added a MS benchmark I found interesting vis a vis Opteron

Opteron the curent King

Existing reasons

[gidds]

I think you're reading too much into these things.

For example, the quiet release of X11 is perfectly logical without any of this. It's clearly a Good Thing(tm) for to be able to run just about any bit of Unix software on Mac OS X. But X11 just isn't part of Apple's way of doing things. For years they've succeeded in making their system predictable and easy to understand and use by keeping everything consistent: the same menu entries in the same places doing the same things, the same keyboard shortcuts, the same icons and GUI elements that work in the same way, the same mindset and way of working, &c &c.

Unix programmers, OTOH, often seem to hate this sort of restriction, and rejoice in the freedom to make their apps look and work competely differently from everyone else's. Unix users may have no problem with this and are used to it, but non-technical Mac users might well find it confusing and awkward, especially when sharing a screen with Aqua apps. (I probably would, and I'm far from non-technical!)

So it already makes perfect sense for Apple not to promote the release in their usual way. Unix enthusiasts will get to hear about it anyway, and will realise its significance. But it will make clear the separation between Unix GUI apps and the core platform they've worked so hard on.

Re:Existing reasons

[shotfeel]

For example, the quiet release of X11 is perfectly logical without any of this.

Not to mention the fact that its still a beta (currently at public beta 3).

Maybe the time to announce it is when you're ready to go with a final release? -Sorry, forgot what decade I was in.

Re:Existing reasons

[iomud]

I consider myself a power user. I've run various unix-like operating systems and Windows of course before switching over to a mac. I have to say that now I notice UI inconstancies more often when I need to use windows for something. Often times I think "Now what do they really mean by the dialogue box?", maybe using the Mac has dumbed me down or maybe it really is more consistent.

Re:Would you install MacOS in your Intel/AMD machi

[oscast]

$5000?! Which Mac are you buying? The average Mac is about $1500

Re:apple hardware is dead

[MacDaffy]

The truth is, Apple as a proprietary processor company is dead.

The PowerPC isn't a proprietary processor. If you'd like to design a motherboard that uses it, go ahead. No one's stopping you (unless it infringes on an Apple design, that is). The hard part would be selling it...

Apple will not destroy its PPC customer- and developer-bases by tossing them aside after all the time, money, and effort expended on Mac OS X. Apple will adopt the PPC 970, take Motorola out of the CPU development loop, and provide Mac OS X for a tightly proscribed x86 configuration (including its own branded boxes--almost everything but the CPU in a Mac is now commodity parts, so that perceived barrier is long gone).

Steve Jobs is a patient man when it comes to the world-at-large. He knows that Apple probably won't ever replace Microsoft as the dominant player in the x86 market, but he also knows that this is probably the perfect time to give them some competition. Microsoft faces a number of challenges to its dominance: its attitude toward DRM, its "trusted computing" initiative, the quiet debacle it's weathering vis-a-vis virtual weekly security updates to XP and other critical software, the growing popularity of open source software, its enterprise licensing scheme, and the increased scrutiny it's under after losing the anti-trust case (like IBM before it, the loss itself will prove more damaging than the punishment).

Apple will continue to produce Mac OS X for PPC. The x86 version would be--in the beginning--a loss-leader. It would get noses into the tent from every market segment. That interest would fuel developer interest (notice how quickly the "there's no software for the Mac" discussion abated in the flood of Open Source offerings it now enjoys).

Once that interest is cultivated, Apple has a whole slew of products/ideas "on the shelf" that would benefit from this renewed interest. There's an advantage to being ahead of your time if you survive long enough.

Re: bundles

[ubiquitin]

Mac OS X apps are "bundles" which means that they are a directory with a structured content inside. Grab a terminal on OSX, cd into your /Applications directory and then cd inside of an app and look around. This paradigm fits perfectly with multi-chipset binaries inside of the bundle. So imagine a Photoshop.app that has both PPC and 970 and 64bit and x86 code inside of it. Sure, the app size is larger, but all of the platform-independent resources are conserved. I can't really imagine a cleaner way to handle multi-platform compatibility.

Conversions & Consumers

[buckhead_buddy]

Sticking with the PPC 970 gives Apple an easy way to maintain backward compatability for its current market of folk unable to break their addiction to legacy, non-native Mac software.

Moving to an AMD or Intel chip gives Apple an easy way to attract the business users looking for cutting edge innovation like 64-bit compatability and a Unix based OS.

My guess is that because they'll have two well defined audiences, that won't overlap it would be right up Apple's alley to do both. Here's our line of computers for the "Classic" mac crowd unable to leave Quark 4 and PixelPaint. But here's our line of cutting-edge, Classic-free boxes that you can order with your choice of AMD or Intel chip inside.

The wilder rumors of a "Classic" like environment on AMD or Intel processors may have a bit of truth, but they're anxious to ditch the Mac legacy crap so if anything it'd be a Classic environment designed to boot your copy of Windows, OS/2, or other legacy 80x86 OS. I still don't believe them, but once you dump the Mac legacy stuff, moving to a different processor is much simpler.

Re:Conversions & Consumers

[gerardrj]

The problem with your plans is market share. Software makers already claim they can't find the financial profitability in porting their software to Mac becuase of such a small audiience.
If Apple splits its market in to two incompatible processor technologies, it would be even LESS likely that new software would be ported, and it would have to be ported twice. That means twice as many SKUs, twice the inventory and shipping problems, twice the testing issues, all for what? Perhaps 20% grater market share?

Reading too much into this...

[Goldfinger7400]

"And we'd caution against reading too much into the AMD official's comments."

I nearly cracked up when i read this sentence, keeping in mind that the whole rumor-mongering session was started by an assumption that "talking with all first and second tier clients" must have included apple.

What about "Not ready to pay for Apple hardware"?

[Wee]

If someone submitted a story with a statement like that subject, they'd hear about it. But (Mac) people can bag on anything non-Mac and get away with it easy enough. Anything that isn't Apple is beneath consideration, probably out of habit or aesthetic reasons or past history, whatever. I personally like Mac OS X very much, but I'm more pragmatic in my choices than many Mac users seem to be. I just can't use OS X because Mac hardware "isn't ready for me".

The last completely new PC I bought was 4 years ago, and it was built from components. I've since upgraded quite a few of the parts in it (video card, CPU/RAM, motherboard), but quite a bit of it is still original (hard drives, SCSI cards, CD/CD-R, sound card, case, monitor, etc). I'll probably keep buying a new motherboard/CPU/RAM combo for the next 5-6 years or so. My previous PC lasted 9 years, and if I could find modern AT motherboards I'd still be putting new parts in it.

The point is that every two or three years I can throw $300 at my initial $2000 investment and basically get a very workable system (with more horsepower than current Macs, I might add). Until I can do that with Apple hardware, Mac OS X will never be "ready for me" -- not because I don't like it, but because it's just not the best value. Besides, Linux on my desktop works just fine (and it's also free, as is 98% of the software I use, so that's a nice bonus). Sure it ain't perfect, but what is?

-B

Technical Note

[neurostar]

Frankly I think not porting OSX to x86...

Apple does maintain an x86 port of OSX. In fact portions of it are opensource (tm).

neurostar

Re:Technical Note

[TheAwfulTruth]

Obviously I meant the entire thing.

Re:Technical Note

[neurostar]

The entire thing is ported to x86.

neurostar

Wild Speculation

[shotfeel]

I was thinking if Linux can be successful as a server OS, why not a Darwin Server running on am Apple made PC?

I can think of a hundred arguments against this, but it makes as much sense (to me) as half the stuff I've read so far.

The recipe

[flappinbooger]

Let's see, we've got IBM, Windows, Linux, Apple, OSX, and AMD all in one topic. POST that sucker!

Darwin is Open Source AND i386

[Zo0ok]

Darwin is Open Source and runs on both PPC and i386. You mean that IF Apple releases Mac OS X for i386, it will not be based on Darwin? Or they will close source for i386 darwin? Or they will build something into AQUA/the proprietory parts of OS X that wont run on standard i386 hardware, while Darwin still will?

I too doubt Apple will produce AMD based systems. But if they do, I doubt they will seriously try to lock people out from it. (The can see how successful MS was trying to stop people from running Linux on XBox, OS X on Dell is probably even more interesting for hackers).

Re:Darwin is Open Source AND i386

[LordNimon]

No, what he's saying is that if Apple builds a system based on Opteron or any other non-PPC chip, it will not be based on commodity PC hardware. For instance, the BIOS will be completely different than what PCs have today. And since the BIOS won't be PC-compatible, it won't be able to boot DOS or any other OS that uses PC BIOSes in any way (i.e. all of them).

Re:Darwin is Open Source AND i386

[lowmagnet]

If they were to port Open Firmware to AMD64, I think they'd be set. It would be open, proprietary and different in one move.

Re:From the blatantly-obvious-dept.

[jceaser]

You can't by the rom at your local store.

Re:Stop it. AMD + Apple == Hypertransport

[dadragon]

A common claim. Unfortunately it's wrong. Athlon XP doesn't really run any hotter than Pentium 4 does for example. In fact, you coulöd say that XP runs cooler than P4 does.

It is still considerably hotter than PowerPC. Apples didn't have fans until the latest PowerMac G4.

Re:There is NO conspiracy theory, AMD isn't just x

[Drakonian]

AMD!= x86 indeed. Outside of the Slashdot world this is fairly well known. Just speculation but I'd guess x86 is their least profitable area. They make tonnes of ICs for various purposes. Their flash memory is probably the most common. Lots of A2Ds, D2As and stuff too.

Re: bundles

[angst_ridden_hipster]

Yeah, that's a good solution -- but it *does* still require that developers build a version for each chipset.

And you can bet that they'll pass the expense of doing so on to the consumer, probably in the guise of a version update.

That's the sticking point, not the technical problems involved.

Re:Stop it. AMD + Apple == Hypertransport

[10Ghz]

True, but I was comparing it to Intels CPU's since they are Athlons primary competitors. Right now PowerPC and x86 aren't really competitors. There is a wall between PowerPC (Macs), and x86 (Windows). Well, Linux can run on either, but for mainstrean, the two are separate.

Re: bundles

[Raffaello]

" So imagine a Photoshop.app that has both PPC and 970 and 64bit and x86 code inside of it."

Unfortunately, Adobe would have to do more than just imagine it - they'd have to rewrite platform specific portions, recompile, test, and debug it.

Unless they've already done this, along with MS (for Office), and a dozen other major application vendors for their Mac products, why would anyone buy a high end workstation Mac, that can't run *any* of their existing software?

Control at MOBO/BIOS level

[Mr. Underbridge]

You can bet that there will be proprietary elements to the platform and OS/X won't run on commodity x86 hardware.

Depends what their market is. If they can make it so that they control the MOBO and BIOS's like they do now for PPC architecture, then it doesn't matter. They could build systems based on the same Intel or AMD chips I can buy off the shelf, but it doesn't get me any closer to having an actual system because I can't get a MOBO except from Apple. And it's not like generics can legally reverse engineer the BIOS's like back in the day (thanks DMCA!). Even if someone did, all they need is a check in the BIOS that checks who made the MOBO and maybe another in the OS that refuses to run on a non-MacOS BIOS. So control of their platform is not a problem.

At that point, the only reason they'd have to even bother mussing around with the chip is to control the upgrade market...but do they care that much? I'm not saying they *should* go Intel/AMD, just that they *could* and it wouldn't matter that much.

Re:Control at MOBO/BIOS level

[lowmagnet]

All they have to do is put in PRAM and run OpenFirmware. Instant proprietary. But, as mentioned above, they'd still be PCI/AGP/FW/USB/ATA/etc.

Re:There is NO conspiracy theory, AMD isn't just x

[Gogo Dodo]

IBM is pushing their fab services quite a bit, so losing the Apple order would make a dent. See the Manufacturing section on Digitimes.

Additionally, I'm sure that IBM will use the 970 somewhere in it's own product line instead of just for Macs.

Re:Stop it. AMD + Apple == Hypertransport

[Surlyboi]

A common claim. Unfortunately it's wrong. Athlon XP doesn't really run any hotter than Pentium 4 does for example. In fact, you coulöd say that XP runs cooler than P4 does.

Well, considering the average P4 makes desktop
fusion a reality, that's not saying much, is it? =)

How is this a troll?

[Tighe_L]

This is not a Troll, geez I was making my comments! That's not fair! I think of Trollers as people what are posting "first post" etc. Or is that "Flame Bait"?

I was honestly upset that some people don't think Linux is mature as a desktop. I think it is. I my have been a little ferial, but geez!/P

Re:How is this a troll?

[Tighe_L]

OK That is more like it! thank you to those who moderated my post! I am very happy!

While we're discussing chips

[aliens]

I know smaller die fabs allow for cooler chips etc, until you start filling everything in with more transistors etc.

If you produced a Pentium 1 core or even a 2 using a .13micron process could you make a very cool running chip as compared to the P4's?

Just something I never figured out. Thanks for any replies.

Re:While we're discussing chips

[Anonymous Coward]

Chip temperature doesn't really depend on power consumption, but rather on power density (watts/sq. mm) and the thermal resistance between the chip and the heat sink. So, smaller die doesn't necessarily mean cooler. Here's an illustration: a small flashlight bulb (non-LED) uses a fraction of a watt, yet its tungsten filament gets up to around 2000 deg. F or so. Why? because the filament has a very small surface area (= high power density) and it's thermally insulated from the ambient by either vacuum or low thermal conductivity gases.

From what I've seen about power consumption and die sizes, Intel's silicon is increasing in power density as the dies shrink, which means they're gonna run hotter still. PowerPC is lower than Intel in both power dissipation and power density, due to SOI technology and better design. Intel's policy towards processor cooling is, "screw'em - it's up to the OEM to figure out how to cool our chips!".

Look at this again

[chasingporsches]

from a non-geek perspective. I personally agree with this guy. When i can install linux on my mom's machine, and have her be able to figure out how to use it, then, and only then, will i consider linux ready for the desktop. And i'm pretty sure the guy who posted this story agrees with me. I'll admit, linux is ready for the desktops of geeks, but for the others? Keep trying, red hat. You're almost there. IMHO, until then, OS X is a much better desktop OS.

Re:Look at this again

[dadragon]

When i can install linux on my mom's machine, and have her be able to figure out how to use it, then, and only then, will i consider linux ready for the desktop.

By that criterion, Windows isn't ready for the desktop. Neither is Mac OS X (though it's better than Windows), or even OS 9. Linux is the worst of the four, and OS/2 and DOS are out of the question.

Re:Dear Apple,

[markomarko]

And people complain about eminem...

Re:Dear Father O'Day:

[markomarko]

Dear Anonymous Coward, If you spent your time working instead of trolling slashdot, you might be able to afford a new trenchcoat and move out of your parent's basement.

Re:Move away from Linux?

[labratuk]

Yeah, whenever I hear someone say the phrase '...ready for the desktop...' I completely tune out now.

It really has become a complete cliche, and I don't think people know what they're talking about anymore. It's turning into 'BSD is dying'. It's very easy to say.

All I know is that Linux is on my desktop, and it sure as hell is not going anywhere fast.

Re:Lock in.

[markomarko]

In my case, I prefer my hardware (a powerbook) to the pc offerings, so you could say I am happily "locked into" my hardware. Plus I much prefer the operating system I have chosen to "lock myself" into. In fact, as far as being "locked into" the software goes, you could call me a happy prisoner. Hmm, does this metaphor suck or what?

Re:Chimera Cons

[u19925]

-->Aren't you a little curious about the PPC970 and what kind of performance to expect?
So far, nothing impressive. if you have any impressive thing about it, would you mind sharing it? -->Why would you even write a three-paragraph post on a subject you have no clue about?
When we talk about things to come, it is easy to do mud-slinging since nothing can be proved. but that is not my style, so i will give you benefit of doubt. may be i am totally clueless.

-->Apple did it before anyone. Ever hear of the Newton?
I am old enough to have used GM refrigerator. That doesn't mean I would tell everyone GM makes refrigerator. Apple Newton is a thing of past. It survived when there wasn't any competition. Once competition sprang, it was dead. i am talking about today. this is one market segment, where style and differentiation still matters, for which apple is famous and they can make some inroad. Apple doesn't sell any PDA today, but I won't be surprised if they come out with one soon.

Re:There is NO conspiracy theory, AMD isn't just x

[olePigeon (Wik)]

I have a better reason: AMD can fab those CPUS easily and IBM has better things to do than fab chips for Macs. Apple needs to make sure it doesn't have to stop the assembly line for IBM to fab more CPUs. My guess is that Apple will have AMD produce IBM 970 chips alongside IBM. IBM probably doesn't want to be the first in line for Macintosh CPUs, there's not enough money in it for a multi-faceted operation like IBM. AMD can produce ample chips and they might be able to make a profit doing it

I think you have this backwards. AMD just recently signed up to use IBM's new manufacturing plant to increase production yield on chips and allow for higher process manufacturing (.09 micron.) IBM wouldn't be disrupting anything to "just" manufacture chips for Apple. Since AMD will be booming in the embedded business when/if HyperTransport takes off, they'll need the extra manufacturing space to produce their chipsets.

You're also overlooking a very obvious clue to the PowerPC 970 being the chip of choice for Apple: the fact that IBM has included an AltiVec engine (and by that name, too.) IBM has stated before and stated again that they will not be using AltiVec, that it's simply there for 2nd and 3rd party vendors to take advantage of.

Can you name one practical vendor that utilizes AltiVec other than Apple? I highly doubt IBM is catering to Amiga.

The whole thing about Apple being in talks with AMD is more plausible if it's put in terms of HyperTransport chipsets and software compatability, and not switching their entire platform over to AMD64. As noted before, IBM and Apple are both on the HyperTransport consortium, it's only reasonable that they need to talk to each other now and again regarding HyperTransport issues. If you see on The Register or some other place about Apple being a purchaser for chips from AMD, please keep in mind that it's most likely HyperTransport chipsets and not Opterons.

Re:IBM PPC 970? huh?

[pressman]

And upon what do you base this? Any links to documentation you'd care to provide?

How Does a Cheaper CPU = Less Profit for Apple?

[neildiamond]

Okay, regardless of whether or not Apple ever designs X86 computers, if they did it wouldn't lower their profits. Why? They can still charge more their computers if the OS only runs on their version of x86 hardware! If it runs faster, the Mac faithful will be pleased. Sure someone is likely to hack it to work on a white box PC, but as far as average end users are concerned, it is not a big issue and any piracy issues would be easily offset by the number of new people buying slightly cheaper Windows compatible Macs. (Heck, I might even consider it.)

I also suspect that OSX (if written properly for a small set of sound/video cards) would be faster than Windows on the same machine. Even if it isn't, people crave the Mac experience. Mac users have never minded paying more. They don't even care that Macs are the slowest on the block right now. It's about the user experience folks. Plain and simple.

codePost

[ennerseed]

if (/.everyUser.post.codePost == (mod.point > 4)){ /.ennerseed.post.mod.point = 5 (nerdable);
}else{ /.everyUser.post.codePost.stop (redundant);
}

Sure Apple makes money from software

[benwaggoner]

While I doubt iLife makes a lot of money, the professional apps like Final Cut Pro absolutely do. $1000 per boxed copy - with a higher gross margin than most of Apple's computers I'd warrant.

Although with Apple's pricing strategy for Shake ($5K for Mac w/ free network rendering, $10K for Linux, plus ~3K per rendering node), they're definitely trying to sell more Xserves there right now.

Re:IBM PPC 970? huh?

[kalidasa]

Yes, it IS for desktop computers. To read the spec:

IBM PowerPC -- In the Hand, On the Desk,
and Everywhere else
Lisa Su
Director, PowerPC and Emerging Products

PowerPC 970 is the first in a family of new 9xx 64-bit Microprocessors
Key Features:
Based on award-winning Power4 technology
Up to 1.8 GHz
Proven 64-bit microprocessor architecture with native
32-bit application compatibility
Up to 6.4 GB/s system interface
Implements SIMD coprocessor
Full Symmetric Multiprocessor (SMP) support

Target Applications

64-bit Linux Applications
Desktop/Workstation
Entry Server

See where it says "Target applications: Desktop?"

http://www.ibm.com/jp/chips/forum0/pdf/05.pdf

Re:success (or lack thereof) of PC cards in Macs

[Jeremy Erwin]

My father bought one of those cards for his 6100. Games comprised the vast majority of his PC software. Now he acomplishes the same thing with VirtualPC.

missing the point

[Scudsucker]

Its not technical issues that are keeping Apple from releasing a port for x86. If they did that, two issues would bring them down: supporting the wide array of hardware on x86 and piracy of the operating system. Right now I might be pirating a copy of Jaguar to run on my Mac, but I have to have bought a Mac in the first place, so Apple has at least gotten some money out of me. If I take a copy of OS X for x86 and install it on a Dell, Apple sees nothing.

Re:missing the point

[Glock27]

Its not technical issues that are keeping Apple from releasing a port for x86.

That is a central part of my thesis, yes. ;-)

If they did that, two issues would bring them down: supporting the wide array of hardware on x86

They already support a wide enough array of hardware - and guess what it is all the same as the PC hardware. AGP, PCI and USB...sound familiar? Heck, NVIDIA is even making Fireware chipsets these days... ;-)

and piracy of the operating system. Right now I might be pirating a copy of Jaguar to run on my Mac, but I have to have bought a Mac in the first place, so Apple has at least gotten some money out of me. If I take a copy of OS X for x86 and install it on a Dell, Apple sees nothing.

First of all, I addressed this exact issue in my post.

Even if Apple wanted to allow white-box MacOS X boxes (with no special Apple ROM), all they need is at least as good of a licensing/authentication scheme as most of the computer games out there now. Not a big deal, really.

Apple has amazingly good relationship with Intel

[Glasswire]

You said: AMD would be a likely partner is such a move since one could imagine the problems with Intel assisting Apple with this.
I believe this is not at all true... Note that

1) Jobs and Andy Grove are old buddies
2) Jobs addressed the Intel Sales Convention this year and got a standing ovation
3) Pixar just bought a ton of Intel Xeon-based servers for rendering (not Apple or IBM or AMD)
Watch for MORE Intel/Apple projects not less... coming from here...

Re: That would mean the end of Mac apps

[Duck_Taffy]

Why? Well maybe because it's easier to program in Cocoa and use Project Builder and Interface Builder than Visual Studio, besides which, they're free. The only thing I miss in switching from Visual C++ 6 to Project Builder is auto-completion.

you ignored his point

[Scudsucker]

If I want anything that will run OSX, it's going to cost me at least $1000.

If I just want an x86, to run linux, I have to spend about $300.

Not from Dell, Gateway or HP you aren't. Origional poster was comparing OEM machines to OEM machines, not OEM's to something cobbled together from cheap parts with a power supply borrowed from a toaster.

Re:you ignored his point

[labratuk]

That's exactly my point.

With x86s, you are able to throw something together from cheap parts with a power supply borrowed from a toaster, and run linux just fine, for $300.

This is simply not possible with Apple machines. $1000 is the starting price.

Re: bundles

[jcr]

Grab a terminal on OSX, cd into your /Applications directory and then cd inside of an app and look around.

Or, just control-click on an app, and select "Show Package Contents" from the popup menu...

-jcr

Re:There is NO conspiracy theory, AMD isn't just x

[MarcQuadra]

I stand corrected. I think you're right-on about the chipsets, too.

Apple will need someone to glue all the parts together, and AMD can bring HyperTransport to the table (IBM is PCI-X, right?). That makes a lot of sense, they'll be IBM CPUs running on AMD chipsets.

Re: That would mean the end of Mac apps

[DavidinAla]

I agree that it's MUCH easier to use Project Builder (and IB) than the tools to build Windows apps. I concede that some small shops and solo developers would still use it for that reason, but the ones who are in business to make a company really profitable would go the other route. David

Dual architecture: been done before by Apple

[willmc]

Will we see Mac OS X running on two different platforms/CPUs?

It sure looks like it. Heck, they did it before with Mac OS 7 and 8, it ran on both Motorola 680x0 and PowerPC architecture. There was a bit of growing pain then, given that PowerPC binaries wouldn't run on 68k machines, and then there were "fat" binaries (which would run on both), but it wasn't terrible and people got through it in one piece in the end. Now, obviously that wasn't the same as a shift from 32-bit to 64-bit computing, but it looks as if there will be similar binary compatability issues again, so it's probably worth looking backwards in time to the introduction of the Power Macs with their new-fangled PowerPC processors.

Re:Chimera Cons

[gerardrj]

It wasn't so much competition that killed hte Newton, as poor management at Apple at the time of its release.
Those where the dark days of Apple when Steve was out of the driver's seat and the company was releasing a new desktop or laptop model every week. The company was killing itself from the inside out.
Steve came back and killed the Newton (along with 90% of the product line), because he needed to get the company back to its foundations in order to frow stong agin.
I personally ahve little doubt that had Steve never been ousted, and the Newton released with him at the helm, that it would have been the top dog in the PDA market.

PowerPC 615

[John Bayko]

Here's what the "Great Microprocessors" list has to say (in the Transmeta section, not the PowerPC section):

In the early 1990s, Apple decided that the Motorola 680x0 series was not keeping up with the Intel 80x86 series, largely because PCs were Intel's primary market, while Motorola CPUs were used more in embedded systems. RISC designs were simpler and could be improved with less effort, so Apple switched to the PowerPC CPU in 1994 (after prototypes in 1991 using the 88K), but to maintain compatibility, needed to emulate the 680x0. The initial emulator interpreted 68LC040 (without FPU) code, and a later version stored translated blocks of code, and ran faster than Apples previous high end Macintoshes.

This impressed IBM engineers enough that a project was started to emulate the 80386+ architecture on a PowerPC (known as the PowerPC 615), but the project was cancelled (apparently after successful versions were completed - possibly because of performance, problems with efficiency using the PowerPC architecture (the 80x86 much more awkward and complicated than the 680x0), marketing decisions, or strategic/management decisions - I don't know, but the computer industry was very volatile at the time, and the path of the future was not at all clear). However development on the conncept continued with the DAISY project (Dynamically Architected Instruction Set from Yorktown), which translated to a hypothetical VLIW CPU instead of the PowerPC. Both the DAISY system, and a later project called Dynamo from Hewlett-Packard (which ran PA-RISC on PA-RISC), could optimise code as it ran (Dynamo could improve PA-RISC performance by up to 20% over non-emulated code).

Several engineers (many from Sun, such as David Ditzel, designer for Sun's UltraSparc CPU, and Bob Cmelik who wrote instruction profiling tools for SPARC programs) helped found Transmeta, which created the missing VLIW processor, and created a new dynamic translator (called a "Code Morpher" by Transmeta) to emulate the 80x86. [...]

This site has awsom specs

[eadint]

After loking at this site apple would be crazy not to use the 970.

http://www-3.ibm.com/chips/techlib/techlib.nsf/t ec hdocs/A1387A29AC1C2AE087256C5200611780

It's all good...

[qon]

Let them go Opteron or 970... either way, I'm happy. The G4 is really showing some age and I'd like to see Apple move on. The irritating perception that Apple is slower than a similar x86 box will probably go away once they get to 64 bit computing.

q

question I have never seen asked

[zogger]

--I'm sure it HAS been asked, just I musta missed it. As long as everyone else is speculating and guessing, I'll give it a shot:

Why again doesn't apple just build their own dang chips? Is it really the simplest explanation, cost? What does it take to build a fab, a billion and a half now? Seems like for the past buncha years, it's always this problem "apple has to negotiate with ibm, moto, no they are going x86 no intel, no amd, no..." I mean REALLY, wouldn't it make more sense for them in the long run to just do it, make their own, and then they could have what they want, a really good chip and OS combo, and not have to always go begging to the various chip guys? Sort of a sun deal, but geared towards desktops and low end servers?

How this relates to their "talks" with IBM and amd I don't know. Perhaps purchasing a fab from one, purchasing a license from another,using some concepts and designs they develop inhouse, running those ideas past the chip guys for a fee, to make their own products? Or maybe someplace somewhere there's a radically NEW design waiting to be released, it just needs to be built? And isn't part of the cost difference with apple always been that they bought more expensive chips? if they made them themselves, well...it would be theirs, right? profit is profit! they'd be paying themselves eventually, could use that to drop prices, become much more attractive, still keep total control, sell more boxes and OS copies and etc.

Re:question I have never seen asked

[gerardrj]

Apple doesn't phyisically manufacture their own chips because it costs too much to start such a factory. Then you have to re-tool the factory every few months every time you want to use some new technology or technique (like smaller trace sized, SOI, copper interconnects, etc). This is also a similar question to "Why buy milk when you could just own a cow?" (or produce any other commodity you normally purchase). It's a matter of scale. For you to own and care for one cow is not cost effective. It's cheaper for you to pay someone else to deal with that hassle and let them make a profit from it, than for you to do it directly. This is the same reason that Apple doesn't own an LCD display fab or an injecetion molding fab, etc. Outsourcing is easier and keeps Apple focused on what they do best: design neat hardware, write killer software and make some deliciously addictive Kool-aid.

Apple DOES however use a chip they designed. Apple was one third of the consortium that designed the PPC platform. You may recall the AIM(Apple, IBM, Motorola) partnership? The group sat down and designed a common CPU reference platform that could be scaled to many uses, be power efficient, and be economical to produce. As such Apple does not need to license anything from anyone to have someone build the PPC for them.

The thing is though, that the PPC specs are more like black boxes than blueprints. In other words, the specs tell you what you put in, and what you get out, not what happens in the middle. The spec doesn't tell you what the chip should look like or how to construct it, just how it should work. Hence Apple negotiates with different manufacturers to impliment the PPC specs in the best technology available.

If you recall the whole reason Apple stuck with Motorola for so long is because IBM refused to impliment the "Altivec" enhancements in silicon. Apple was torn because IBM's processes are FAR better than Motorola's (faster clock speeds, lower power draw, lower heat output, higher yields). Now IBM is willing to do Altivec, and Apple gains all the other IBM nicities like SOI, copper, and smaller process size.

If Apple wanted to they could negotiate/contract with Intel, AMD or and of the myriad of chip fabs in the world to make the PPC. One has to ask themselves though... why would Intel or AMD be interested in building the PPC? Who other than Apple would be a customer? And what extra costs would Apple incur because the new company had to start with a design from scratch?

Rumours

[Jucius Maximus]

According to certain rumours, the IBM 970 15" Powerbooks are already in production and ahead of schedule:

http://www.envestco2.com/macwhispers/archives/0000 57.php

This is supposedly from a 'reliable source' in the engineering dept of an equipment supplier.

WE HAVE A WINNER!

[sfgoth]

ding ding ding ding! Give this man a prize.

It's amazing how many people don't see this obvious consequence.

It's the same thing with Classic apps. If Apple makes them "too" well integrated, why should developers move forward?

Very few applications adapt with changing APIs. OS Venders need to maintain backwards compatibility so there are products for their new OSes, but write new APIs so there will be new products to replace them.

Putting too much energy into supporting your "old" APIs is a recipe for stagnation.

Re:apple hardware is dead

[erwass]

The reason Apple will make an x86 configuration is simple. It will save Jobs the embarassment of having to stand up at every MacWorld and show the Mac vs. Wintel benchmark comparisons..." and if you squint just right at the bars for the Photoshop Lorentzian Blur you can see the 1.63 GHz G4 just blows away the Intel 9GHz machine". Apple hasn't done it until now not because they fear clones. Their lawyers can shut down x86 cloners just as fast as PPC cloners. Rather, until OS X, Apple had no means of migrating to the x86. The only scenario I can see against this eventuality is if the successors to the 970 just mop the floor with the x86's progeny. I'm not holding my breath. As long as Apple can't measure up in the MHz war they are hamstrung. The mass market just can't see past clockrate.

Re:To run OSX comfortably all you need is a G3

[labratuk]

Is it really fast enough to run OSX? Well?

Will it run OSX as fast as my dual PII 400 runs Linux?

(It cost under $200, incidentally)

Apple should support both PPC and X86

[afantee]

With Safari, Keynote and quite probably another sectret Apple office application that could kill MS Office on the Mac, fewer and fewer Mac users are going to pay MS $500 for its bloated and buggy software, so Apple has really nothing to worry about upsetting the Beast anymore.

Apple is in a very strong position to grow and compete with MS and the Wintel box makers by releasing OS X on X86 and making computers based on both PPC and AMD / Intel.

Apple is the best in industrial design and capable making powerful and stylish computers better than those clueless box makers like Dell.

Amazingly, despite being 60 times smaller than MS, Apple actually produces more and better software than the Redmond bully: WebObject, Final Cut Pro, DVD Studio Pro, Shake, FileMaker, AppleWorks, Safari, Keynote, Darwin Streaming Server, iLife, ...

PPC 970 is for Apple and IBN

[afantee]

IBM will use the PPC 970 in their blade servers, well before Apple. So they have strong incentive to develop the chip.

Re:Chimera Cons

[Morky]

There are links in the post that should help you out, if you're wondering why it's so impressive. It should basically erase the Intel/Apple speed gap (based on initial estimates) and provide an easy segway into 64-bit desktop systems. Huge memory bandwith, too. It will probably be a little behind the Opteron initially, though. Still, there shouldn't be any significant complaints about speed for awhile after it comes out.

Re:IBM PPC 970? huh?

[Morky]

Yeah, except for the part where IBM says it's a chip for workstations. I'm sure Altivec was thrown in there because IBM thought is was neat.

Windows Emulation, x86 is a Bad Idea

[General_Tso]

First of all, OSX is a feather in Apple's cap in terms of marketing. It's really increased Apple's mindshare lately: Desktop Unix is a big deal. Why challenge that with Windows emulation? Virtual PC is good for legacy Applications, but I don't think Apple needs any help from the Windows community. There are plenty of awesome applications already, and there will be more on the way, particularly with help from the Open Source community. Why make Safari and Keynote if they wanted to cozy up to Microsoft with a Windows emulator. Even if OSX is ported to x86, they won't emulate for Windows applications. And they won't switch to x86. Intel's too caught up in stuff like Palladium.

970 all the way

[Enrique1218]

People hear are asking for an open hardware system that functions as well as a Mac. But, that request is problematic because the functionality is directly related to proprietary nature of the platform. Apple hands picks it hardware, it knows how it will behave and thus programs it OS and application around it. It then releases, those spec to third party developer so they can do the same. X86 doesn't have this advantage. You have all these variations out there that it is impossible to guarantee perfect functionality. Windows still has issues, Linux-x86 still has issues. No matter how much faster they are they still won't match the Macintosh on functionality. Thus 970 processor looks to be a better candidate than athlon-64

Re:970 all the way

[MacDaffy]

Apple hands picks it hardware, it knows how it will behave and thus programs it OS and application around it. It then releases, those spec to third party developer so they can do the same.

This works both ways. If Apple can get in on the ground floor of an advanced x86 chip, there's no reason they can't design a machine around it.

The most daunting aspect of the x86 problem is legacy support. The sanest way to handle it is to not provide it. x86 machines designed from the ground up for Mac OS X could be exploited by Apple and licensed to third parties --especially if those boxes could still run Windows and Linux and were reasonably priced. Rather than closing off the current hardware/software revenue stream, Apple would be opening new (and prospectively more lucrative) streams for its hardware, software, and ancillary products. The move would be fraught with problems, but they would be problems Apple would love to have.

Lastly, Apple has been driving hardware design industry-wide since the iMac was introduced. Innovations introduced on the Mac have routinely found their way into PC's (I can tell the vintage of most PC laptops by identifying the PowerBook they most resemble and counting forward six-to-nine months).

IIRC, Apple is shooting for around five per cent market share as a first step. If they can hit the right price-performance-stability sweet spot with an x86 box (and combine it with a PPC 970 box that puts its teeth directly into the Pentium's ass), I have no doubt whatsoever that they'll achieve that goal...unless, of course, they mess up the marketing.

Re:970 all the way

[gerardrj]

Huh?
What do you mean "...functionality is directly related to proprietary nature of the platform..."?

What exactly is proprietary about the Mac?

The CPU?
Nope, PPC is arguably on of the most open CPU formats on the market. Almost anyone willing to can join the PPC consortium and produce processors based on the platform.

The RAM?
Nope. Macs use standard RAM modules that are commodoty parts.

Peripheral bus?
I think most people in the world can get their hands on a PCI or AGP card pretty quickly.

Disk storage?
Hmmm. IDE and SCSI interfaces are the most popular and widely produced drive interface formats out there.

Network?
Ethernet 1000bT, at least 100bT standard on most models. Just plug in and let DHCP do its thing.

Okay, I think we're starting to narrow it down.

CDROM?
Again, IDE and SCSI cover it. Most any off-the-shelf drive pops in and "just works"

Display ports: VGA monitors work with all Mac hardware. You may need an adapter for Mac/VGA though. I know, I have three CRT displays on my Mac at the moment, one via the built-in Apple large port, one via a VGA port and one via a digital video port with a digital/crt adapter on PCI display cards.

Mouse and keboard?
USB devices are supported completely. In fact, before the iMac adopted it, USB was pretty much just "that funny connector" on most PCs. Yea, Intel invented USB, but Apple brought it to the mainstream.

Okay, so the hardware isn't proprietary, well it must be the software then.

Networking: TCP/IP, DNS, HTTP, SMTP, ARP nope, nothing proprietary there. But oh yea, you still have AppleTalk available for all those small networks that just don't want to bother with TCP/IP.

Graphics: OpenGL and PDF. Industry standards. In fact, some of the most open standards available.

Data interchange: OS X is probably the leader in adopting and supporting XML.

Multimedia: Quicktime supports most all industry standard codecs: MPEG2, MPEG4,AAC,MP3,sorenson, AIFF,etc.

Hmmm.... so tell me: what exactly is so proprietary about the Mac again? What you must mean by proprietary is that Windows' x86 binaries don't run natively on the machine under Mac OS.

To get anything proprietary hardware wise on a Mac you have to go back quite a few years to when they were all ADB, mini-din serial ports, NuBus based systems, and that was quite a while ago.

Re:Would you install MacOS in your Intel/AMD machi

[Dr Reducto]

Obviously they aren't very upgradeable. That's what happens when you put laptop guts in a small, cute, and ever-so huggable computer. That's why the iMac has a GeForce 2go gpu.

eMac: unsafe at any price.

[MsGeek]

The eMac, sad to say, is a complete and utter P.O.S.

They made a lot of mistakes when designing it, similar to mistakes made when Michael Spindler wanted ultra-cheap 603-based Macs and had his engineers take an LC-class motherboard and graft a PPC chip to it.

The mistakes on the eMac are thus: the 17" monitor was basically grafted onto an unchanged CRT iMac motherboard and power supply. The iMac had a 15" CRT from day one. Anyone who's worked around computer hardware for any time knows that a 17" CRT monitor sucks more juice and gets hotter than a 15". When you consider that the third-gen CRT iMac lost the fan and was cooled by convection alone (Steve HATES fans in computers) any change in heat production would render such a machine unable to exhaust enough heat to sufficiently cool itself. Also add to this the power draw problem, and you have a recipe for a completely unstable machine that breaks down at the drop of a hat.

A friend of mine fixes Macs for a living. He has to deal with dying eMacs practically every day (he has a few contracts with schools) and detests them.

The eMac is going to wind up in the Road Apple category at Low End Mac eventually. Just watch. Also just watch Apple quietly retire the machine and build another all-in-one Mac for scholastic use.

Apple had better...

[ppc970]

...use the PPC 970. Because otherwise my slashdot nick won't be so cool!!

AMD and Intel chips are too hot for Apple's taste

[afantee]

A 1.8 GHz Opteron or a 3 GHz P4 consumes about 80 W, compared to 40 W of a 1.8 GHz PPC 970.

More importantly, a 1.2 GHz PPC 970 burns only 19 W, which makes it possible for Apple to design cool and sexy fortables without huge heat sinks or noisy fans.

The low energy consumption is also critical for 24/7 servers, it reduces electricity bills and hardware failures. So I can't really see why Apple or anyone else should be too excited about the hot chip.

Re:Stop it. AMD + Apple == Hypertransport

[Matty_]

Your comment about fans is untrue.

My Sawtooth PowerMac G4 which I purchased in March of 2000 has one large case fan in the middle, which blows air over the CPU heat sink and cools the entire box inside.

Ooops!

[Glock27]

Before you jump all over me, I just realized that SoftMac only emulates 68xxx processors.

Sorry I got that wrong - I'll concede the entire emulator point for the time being.

Take it easy.

Fun

[inertia187]

Fun. So now they realize after they create the chip that they want 20 years of backwards compatibility. The PowerPC knew they wanted this, according to this slashdot article.

Mirrors:
story 1
story 2

Re:Fun

[sharph]

CNET and Information Week aren't going to get slashdotted, don't worry.

Re:Fun

[jbs0902]

Actually, when we created Merced (1st Itanic) it was designed to be able to be FULLY backwards compatible (i.e. boot MS-Dos 1.0). 25%-33% of the chip was actually a HARDWARE ia32 to ia64 translation engine.
You could put the chip is EPIC (ia64) mode and everything would run though the normal pipeline or ia32 mode and things 1st ran through the ia32 translator then most of the normal pipline. Yeah, you took a performance hit in ia32 mode, but it was the price you paid for "100%" backwards compatibility.

So, I am not sure why the change to a software emulator, unless:
1) they ditched the hardware emulator to get back some real estate of the die, or
2) they didn't like the switching the chip between ia32 & ia64 bit modes.

Also, you can tell I've been out of the Itanic design loop for 5 years now. So, some information is out-of-date or lost in the fog of memory. And, I'd like to say that Merced was such a horribly managed project I left engineering.

Re:Fun

[TopShelf]

And so did the Opteron as well - for such a large technological leap, backwards compatability is a must (read: lessened risk for the corporate consumer)...

Re:Fun

[karlm]

Maybe they just wanted a second option. Hardware emulation probably runs some apps faster and software emulation probably runs others faster.

IMHO, the software emulator is a better long-term solution. A hardware emulator uses some power even if you're not using it and drives up the cost of the chips by taking up realestate and increasing the defect rate. Your design-test cycle is also much faster for a software solution. There's also the marketing point of "we're doing this well so far, and will give you an even better version when it comes out, for free". They can't easily upgrade hardware for free at a later date. The software emulator probably has a lot of overlap with the compiler group, so you might get compiler research almost for free.

Also, I assume most of the guys writing the software emulator aren't experts in hardware design and vice-versa. The two projects are completely independent and likely don't steal personell from eachother.

Re:Fun

[khuber]

DEC Alpha tried something similiar with their x86 emulation.

I think that's a different situation. For starters, Itanium already does IA32 in hardware (it's just really crappy apparently).

DEC wasn't in the x86 market to start with so FX!32 extended their market by making NT/Alpha more attractive. With the 21164, Alpha introduced data handling functionality in hardware that was intended to accelerate x86 emulation. It was probably too late by then.

There must still be major management/direction problems with the Itanium project for them to resort to this kind of hack. It's embarassing that they can outperform their hardware implementation in software.

The only software emulation I can think of that was successful was Apple's 68k emulation for PPC, but their approach was brilliant and well thought out IMO (smooth transition, fat binaries including code for both chips). At the time, PPC was compelling. I don't think Itanium performance is as compelling even though Itanium 2 is pretty decent from what I've seen. I think for a straight 64 bit Linux system, Itanium 2 is a much better chip.

I suspect Intel and friends (oops almost typed fiends!) will be back with improved hardware support for IA32 because people won't be satisfied with the emulation performance. AMD has to feel pretty good about having Intel/HP in this position.

-Kevin

Re:Fun

[drinkypoo]

A hardware emulator uses some power even if you're not using it and drives up the cost of the chips by taking up realestate and increasing the defect rate.

If you make the emulator separate enough from the core of your new architecture, you can switch the power off when you're not using it. A number of big pieces of silicon in our lives do this, including mobile video solutions (I think the latest mobile radeons, and maybe some of the desktop chips as well?) and some CPUs.]

The only really good thing about a software solution is that you could have a microcode update, as you say, and of course it takes up space, that's always a bummer.

Re:Fun

[Webmonger]

The x86 emulation circuitry on Itanium is really slow. So slow that they think software emulation would be faster. . .

Re:Fun

[Anonymous Coward]

...if they're working with the old DEC stuff (via Compaq via HP), how come they don't dust off the code that they used on the RISC boxes and then Alphas that would do JIT emulation of Vax binaries, and end up saving the resulting RISC binaries back as it hit the code?

Then, they could do the x86->Itanic code conversion rather invisibly and dynamically as the need arose.

Re:Fun

[atam]

From I read in the article, I think it is not an emulator per se. It is more like a just-in-time compiler/translator. Probably it is something similar to what the Transmeta Crusoe or Alpha FX!32 does. Both of these products already proved that you could do it in software implementation pretty efficiently.

Re:Fun

[nixil]

You do know that Intel owns FX!32 with one of their deals with HP/Compaq. So this could just be a revised version of the code for the Itanium.

Re:Fun

[Hoser McMoose]

It will be almost exactly like what FX!32 does, because that software is no longer "Alpha FX!32", but rather "Intel FX!32". Intel bought pretty much all of the old Alpha technology, software and design teams, as well as the plant that Digital used to build the chips in. This occurred over quite a number of years (starting just before Compaq bought Digital), but this is the first really obvious sign of Intel technology to come out of it.

Re:Fun

[pesc]

DEC wasn't in the x86 market to start with so FX!32 extended their market by making NT/Alpha more attractive.

The only software emulation I can think of that was successful was Apple's 68k emulation for PPC

Actually, one of the top priorities for the Alpha design was to replace the VAX computer. Since much of VMS is written in VAX assembly, an Alpha compiler was developed for VAX assembly code. (Just treat VAX assembly as another high level language, compile it and optimize it.)

To make the transition for customers easier, another tool was developed (VEST) that translated linked binary programs from VAX to Alpha. (Basically, the program disassembled the user application to assembly source, and then recompiled that source to a new Alpha program.)

The FX!32 tool was developed later from this technology. (FX!32 does not emulate; it recompiles like VEST, on the fly, as needed)

So I would say that the VAX to Alpha transition is an example of a successful architecture transition that predates 68K->PPC.

By the way, VMS is now being ported to Itanium. Of course there will be a binary translator that translates binary applications from Alpha to Itanium.

Re:Fun

[leandrod]

> DEC wasn't in the x86 market

Yes, it was. It had some pretty sweet machines as far as PCs go. BTW, did you never notice post-merger Compaq desktops and notebooks had more DEC heritage than Compaq's own?

I would like to run...

[vDave420]

...a C64 emulator on an x86 emulator on 1.5Ghz...

Or something like that... =)

-dave-
Get BearShare! for your p2p needs!

Re:I would like to run...

[Alien Being]

"...a C64 emulator..."

Just one? Or a, oh never mind.

Re:I would like to run...

[calags]

I guess that would make it a 64-bit Commodore 64!

Re:I would like to run...

[GnarlyNome]

how about a KIM-1 full 1Mhz Emulation (wow lunar lander in 1.5K of static ram)[oh the feeling of power}

Re:I would like to run...

[samhalliday]

or a VIC-20 on a C64 emulator on an amiga emulator on x86 emulator (bochs) within an x86 emulator on 1.5Ghz... :-D

Re:I would like to run...

[usotsuki]

Bochs runs Personal C64 adequately on my Celeron/700. I bet it would run PC64 *well* on your 1.5GHz machine, comparable perhaps to a 486/50, so about the same as a 6510/1 in emulation.

-uso.

Opteron

[whig]

Sounds like a defensive reaction to the release of the Opteron. If AMD is offering a 64-bit chip with support for full-speed 32-bit x86 software, then Intel has to have a competitive answer *before* industry adopts the AMD64 over IA-64 for future migration.

Re:Opteron

[LBArrettAnderson]

no - Intel has been planning for emulation the whole time. AMD still has the advantage with full compatability at full speed. But you're right; it sure does sound like it.

And industry won't really adopt a certain chip - I'm sure it'll be just like the x86's today; you can go back and forth between Intel and AMD pretty easily with each new computer you buy - unless you're anti-Intel because they have that agreement with microsoft.

Re:Opteron

[LBArrettAnderson]

well, i don't have any primary resources, but i do remember reading on the /. comment sections one time that intel has an agreement with microsoft, and obviously i believed it because all of the computers that i see in stores have both pentium 4's and windows on them.

Re:Opteron

[Jace of Fuse!]

What agreement with Microsoft?

Don't kid yourself. Microsoft and Intel are in bed together and have been for a very long time. Once again, I don't have primary sources much like the parent (I'm sure someone will post some) but I know that Microsoft works very closely with Intel, moreso than they do with AMD if they do at all.

Re:Opteron

[ocelotbob]

And industry won't really adopt a certain chip - I'm sure it'll be just like the x86's today; you can go back and forth between Intel and AMD pretty easily with each new computer you buy

Actually, this is a pretty major fork between AMD and Intel. Unless there's a new processor made by one of them, the two competing 64 bit "x86" systems are mutually incompatible with each other. People are going to have to commit to one or the other, because the instruction set, hell the coding style, is markedly different in the two architectures. AMD's offering, x86-64, is very much a cleanup of the x86 instruction set, with a few features that should have gone into the architecture long ago. IA-64, on the other hand, is essentially a complete abandonment of x86, which, as others mentioned, is something that really hasn't happened with intel since they made the 8080 decades ago.

While I feel that eventually there's probably going to be in-processor emulation of the competetor's code, that's not the case now. This is perhaps where the AMD-Intel war gets truly ugly. Since the days of the 286, the rivalry has been essentially tit for tat, a few added features by one side gets picked up by the other. This is a lot different -- there is no easy migration back and forth.

Re:Opteron

[aminorex]

x64 and ia64 are entirely distinct and incompatible
instruction set architectures. you're not going to
be able to run your x64 kernel on an ia64 chip.
it's not in the least similar or analogous to the
ia32 situation.

Re:Opteron

[AvitarX]

AMD does not have full compatability.

According to their site if you want to run x86-64 code you can not use 16-bit legacy apps.

Yes, technicaly the chip can run all those apps, but then it is just the next athalon, and not 64-bit chip with extra registers.

Re:Opteron

[Erect Horsecock]

Microsoft has been very public about supporting Alpha, AMD64, PocketPC and pretty much anything non-Intel they can find.

Most (if not all of them currently available) PocketPC PDAs use the Intel XScale processor in them though

I know I know, IHBT yadda yadda

Re:Opteron

[Vaughn Anderson]

unless you're anti-Intel because they have that agreement with microsoft.

*ahem* a slight reminder about a very interesting court appearance by the CEO (or somebody) from AMD on the behalf of MS... let me refresh your memory it ended something like this...

States Attorney: So what has motivated you to come forward with this information?

AMD Executive: That really nice rich man, Mr. Gates, told me to say it, and said he'd let me play ball in his sandbox if I did.

so, what were you saying about "agreements" with MS? Please fill me in, as I am a little out of the loop...

Re:Opteron

[Hoser McMoose]

You might want to read over those tech-docs again, or more specifically the "AMD64 Architecture Programmer's Manual Volume 1:", in section 1.2: Modes of Operation.

<quoting sub-section 1.2.3>
Compatibility mode--the second submode of long mode--allows 64-bit operating systems to run existing 16-bit and 32-bit x86 applications.
<end quote>

However, I think what you're looking for is a little earlier in the section.

<quoting sub-section 1.2.1>
Long mode does not support legacy real mode or legacy virtual-8086 mode, and it does not support hardware task switching.
<end quote>

Now, this may seem like a bit of a loss, since DOS was run in real mode, and Linux 1.xx made use of the hardware task switching, but neither of these operating systems are ever going to run in Long mode on an x86 chip since they're both long since being EOLed. Even running DOS programs under Win2003 won't require real mode (unless I'm really off as to how the DOS window works).

In short, this is just cruft that would never be used in x86-64 long mode anyway.

Re:Opteron

[jasonditz]

That's not too clear at all, maybe AMD better clarify that.

How on earth can you claim legacy support for, say, a 486, and not claim legacy support for 8086? The 486 supports legacy 8086, so neccesarily anything that supports it would have to support 8086.

Thinking about that from an assembler language programmer, how would that even work? I mean, obviously you've got all the registers of an 8086, right? I'd really be interested in knowing what exactly happens if you try to bootstrap the machine off a non-protected mode kernel.

I'm not sure how you can trim off something as intimate to the x86 as Real Mode and still call it compatibility. And why on earth did they bother to support 16-bit protected mode if they're not supporting real mode? Are there still a lot of people using Phar Lap?

1.5ghz Xeon?

[JanusFury]

Is it really emulation or does it convert x86 assembly so it can run on the Itanium? If you can get 1.5ghz worth of performance out of EMULATION on the Itanium, then I need a new processor.

Re:1.5ghz Xeon?

[afidel]

It's probably a bit of both, it is probably very similar to FX!32 for the Dec Alpha version of NT4. What this did was emulated many x86 functions, but if something was getting called a lot it was dynamically recompiled to native Alpha code. Worked pretty well overall.

Re:1.5ghz Xeon?

[frostgiant]

Depends what you mean by emulation. There are many varities of emulation, I would expect that it is using some sort of dynamic recompilation on the x86 code.

Re:1.5ghz Xeon?

[cehardin]

This is exactly what AMD and Transmeta chips do. They convert x86 CISC instructions into blocks of VLIW instructions. We have been running VLIW machines for many years, actually.

Re:1.5ghz Xeon?

[aminorex]

Smells like bytecode JIT to me.

Re:1.5ghz Xeon?

[afidel]

Yeah except years earlier and with real machine code not idealized bytecode =)

Better C|Net story

[Webmonger]

Here's a more detailed C|Net story.

(Yes, it's linked from the posted C|Net story).

Duh..

[OmniVector]

The Itanium had a lot of good ideas, but no matter how much you want to drop an old architecture and start over from scratch like the goal of that project was, you've got to provide a transition period. Athlon's doing this with the Opteron, Apple is doing this with OS X using the Carbon Toolkit, etc etc. The *key* to getting a user base to switch from an older architecture to a newer one has to be a compatability layer.
Perhaps that is what doomed Itanium 1 to failure form the start. (Well that combined with the horrible heat output and power consuption of the Itanium 1).

Re:Duh..

[Okonomiyaki]

For the record, Carbon is not a transitional API. People will be writing apps with Carbon long after compatibility with the classic MacOS has ceased to be a concern. OSX's Classic Environment is a better example of a transitional compatability layer. An even better example would be the 68k emulator built into the MacOS during the transition to the PowerPC architecture because it's actually an emulator, Classic is not.

Re:Duh..

[xneilj]

"Any problem in computer science can be solved with another layer of indirection."
--- David Wheeler, chief programmer for the EDSAC project in the early 1950s.

Scarily, it's still just as true today...

Lets see how well it runs java or .net code

[Billly Gates]

The epic architecture is flawed in the sense that it can not run anything not in parrallel without having optimizations lost.

Code in .net and java can not be compiled in parrallel so all the features of the itanium will not be enabled.

Itanium seems to be a nitch chip in which it can do certain things like scientific apps which can be compilied in parrallel really fast but it sucks at anything else.

Re:Lets see how well it runs java or .net code

[lkaos]

The epic architecture is flawed in the sense that it can not run anything not in parrallel without having optimizations lost.

I don't think you really mean parrallel :-) I think what you're referring to is interleaving.

Compilers that support interleaving can achieve parrallelism up to the number of stages on the pipeline (something ridiculus ia64 like 13 or something).

Now of course, if a compiler can optimize for interleaving without programmer intervention, a JIT can optimize for interleaving.

Re:Lets see how well it runs java or .net code

[msgmonkey]

Actually Java and .NET bytecode style applications are likely to beable to be better optimized than their C and C++ equivelants with a decent virtual machine when running on Itanium.

A VM bytecode program contains alot more structural information of how the original program looked than C or C++ programs. On the Itanium the compiler has to take a "best guess" or some profile data to compile for the most common program-flow, this is one of the largest factors that limit Itanium peformance since alot of the run-time hardware optimisations are n't and cant be there.

A VM could analyse program-flow and compile different versions of the same function, dynamically changing which is used for example.

Of course this does n't help the vast majority of C/C++ code out there, but your assertion is hardly correct.

Re:Lets see how well it runs java or .net code

[Erich]

I don't think you understand translation, or maybe you don't understand EPIC.

EPIC (or VLIW, which is pretty much the same thing) atchitectures define instructions to be executed (or, at least, that can be executed) in parallel in the encoding of the instructions. Most superscalar machines evolved from single-issue architectures (PPC, Alpha, x86) only have sequential instructions.

That being said, there are almost always instructions that can be executed in parallel. The only difference between EPIC/VLIW and sequential dynamically-parallel machines is whether the hardware or the assembler/compiler/optimizer determines whether or not instructions can be executed in parallel.

Now, software that interprets (for instance) java can utilize parallelism, because it can do things like lookup the place to jump for the next bytecode while it's executing the code to handle the current bytecode.

Software that translates (for instance) java to native code can utilize parallism, both in the translation phase, and in the native code output. Of course, scheduling code is a complex thing, but it doesn't mean it's impossible. And given that you have the scheduler for a good compiler for your architecture (which Intel does) you may have many of the problems already solved.

Translating x86, or PPC, or any other architecture is very similar... you have interpretation (usually somewhat slow), and varying levels of translate-to-native (which ends up with varying quality of output). Some architectures are easier to interpret than others (Java is nice, because it has no flags), but in the end the problem is pretty much the same between architectures.

Itanium's flaws are not in it's EPIC/VLIW nature. VLIW chips can be very small and efficient. Most high-performance embedded chips are migrating towards VLIW (see: TI DSPs, StarCore DSP, TransMeta) because you can get parallelism without having to add hardware complexity for out-of-order execution, hazard detection, etc. At the very least, you can get a VLIW that performs the same as a single-issue computer.

The problem with the Itanium is that it has lots of features. It's very complex. It has lots of predicates, lots of cache prefetch instructions, etc. that make it very cycle-efficient for either a hand programmer or a compiler that knows everything.

Unfortunately, the complexity makes design rather challanging, and many of the features aren't being used by production compilers, because they're hard to implement. Normal instructions are easy.

Sigh. It would be really great to see a good, simple VLIW with a good compiler on the market. In the end, however, it would only be a little more cycle-efficient and a little less costly than a superscalar chip with similar features. And the superscalar has the code base and the existing, proven compiler. Guess which one wins?

Your architecture prof for the day,

Re:Lets see how well it runs java or .net code

[khuber]

Good post Mr. 151!

I agree about the validity of VLIW. We know the parallelism is there or superscalar wouldn't work.

To me, VLIW is the logical extension of superscalar RISC, but (big but) superscalar can deal with the evolution of the functional units and architecture of a chip and VLIW can not.

Executables statically scheduled for Itanium2 may not be optimal for ItaniumN. How do you deal with that? Add OOO to ItaniumN? Oh boy... Itanium2 already has hardware to deal with "split issue" - adjacent instructions that cannot execute in parallel.

-Kevin

Re:Lets see how well it runs java or .net code

[conway]

Compilers that support interleaving can achieve parrallelism up to the number of stages on the pipeline (something ridiculus ia64 like 13 or something).

Itanium 1 had 10, Itanium 2 has something closer to 15, don't remember the exact number.
Not only do you have to fill the whole pipeline, you should also fill all the execution units for best performance. (6 execution units on the Itanium 1 -- 2 int, 2 float and 2 branch).
However, the processor tries to give you a break, and can forward results from an instruction still in the pipeline as input to an intstruction at an earlier stage (for some instructions, for some stages).
In general, compilers are just not smart enough for these sorts of architectures yet. Give them time to mature.

Re:Lets see how well it runs java or .net code

[lkaos]

In general, compilers are just not smart enough for these sorts of architectures yet. Give them time to mature.

Interleaving is perhaps the most difficult thing I've had to do as a programmer. It requires really strong math skills and an extremely strong understanding of the architecture itself.

You have to more or less be both a mathematician, a chip designer, and a software developer to properly optimize. I'd be willing to wager that the most important CS problems to be solved in the near future are interleaving algorithms.

Re:Lets see how well it runs java or .net code

[rabidcow]

Actually Java and .NET bytecode style applications are likely to beable to be better optimized than their C and C++ equivelants with a decent virtual machine when running on Itanium.

They'll definitely compare favorably if you consider that Java and .NET are already translated on the x86 while it would be new overhead for C and C++ code. (Assuming that your VM is itself natively compiled.)

Java and .NET were also designed to be translated whereas x86 wasn't, which gets into what you're saying.

Re:Lets see how well it runs java or .net code

[msgmonkey]

Of course running under a x86 VM under an emulator is going to run like a snail. The gist of was that somehow Itanium was unsuited for Java or .NET applications compared to other processor which is not true.

Emulation

[whig]

Also, it's worth noting that Itanium has always supported running x86 software without emulation. It just turns out their hardware implementation is slower than emulating the same thing in 64-bit IA-64 mode.

The way I see it

[blitzoid]

This is great and all, but it's still EMULATION. x86 support in the Itanium seems very 'tacked on', unlike AMD's idea of simply extending the regular x86 instruction set to the realm of 64 bit. The way I see it, AMD chips will always be faster than Intel at x86 stuff. And when everyone is changing over, that's CRITICAL.

Re:The way I see it

[hendridm]

> And when everyone is changing over, that's CRITICAL.

Pffft. If you want to run 32-bit, get a P4 or Xeon. If you want to run 64-bit, you're most important application(s) is/are 64-bit anyway, right?

What uses would a company have to go 64-bit? Big ass database? High performance workstation perhaps? In the database scenario, you'd probably be running a 64-bit database anyway (or you'd be wasting your time and money). It is likely this would be your only, or at least most important, service running on the box.

How about a high performance workstation, like CAD or something. Well, that CAD engineer will probably have 64-bit CAD, which is what he/she will use most of the day. Who cares if MS Outlook or WordPerfect run at only the speed of a 1GHz processor (or whatever the actual emulation speed equivalent is)?

I don't see what the big deal is, but I know the average Slashdotter has a "AMD inside" bumper sticker on his modded chassis.

Re:The way I see it

[Junta]

It's not as simple as that. The people wanting to run 64-bit appas for big things don't have a problem. In essence, Itanium was made to solve a problem already solved. PA-RISC, Sparc, PPC, MIPS, and others already have 64-bit variants. Companies that need the 64-bit address space and such already have solutions, and don't care a bit about MS offerings on their servers. This is evidenced by MS withdrawing their unpopular ports of WnNT to non-x86 platforms years back.

Itanium may be a true server class chipd and capable of pulling off the same stuff PA-RISC and Sparc can. But if there is *any* performance advantage, it is so slight that it is overshadowed by pathetic industry and software support. Sure you will soon be able to run Windows, and have been able to run linux, but ultimately there isn't much to run on those systems.

AMD has struck a cord here. A lot of large environments (especially clusters) have been getting by on 32-bit architecture because of the great applications support and price/performance ratios. The Opteron falls into the same price/performance league as those 32-bit systems in use, can equal or best those processors in 32 bit tasks, and as the software matures and gets recompiled, smoothly migrate to 64-bit operation without a hiccup. When these huge clusters are running software packages that costs millions to develop, there is a vested interest in continueing to use them while simultaneously ironing out the kinks in their 64-bit versions.

There is a damn good reason why IBM and others are finally acknowledging AMD as worthy of building servers around. Itanium sales have been pathetic, and there has been much more customer interest in the possibility of upcoming Opteron products than the reality of existing Itanium systems.

Re:The way I see it

[debrain]

The way I see it, AMD chips will always be faster than Intel at x86 stuff. And when everyone is changing over, that's CRITICAL.

I think you are right, but not necessarily about what you think. It is "CRITICAL" because it is "The way [you] see it". I believe it is not the speed that is important here, but the perception.

Re:The way I see it

[sco08y]

On large multiuser servers, you certainly need people to be able to use x86 apps. If you're trying to centralize your Windows apps on a big server running Terminal Services, you'd want good 32 bit emulation.

High-performance workstations are still *personal* computers. There's always that one app someone wants. Or there's that one app that hasn't been ported... see Quark and Mac OS X.

And while software emulation is nice, the truth is that it's always going to be somewhat flaky, and as processors start to become more powerful than customers need, they'll demand better stability.

Re:The way I see it

[hendridm]

> High-performance workstations are still *personal* computers. There's always that one app someone wants. Or there's that one app that hasn't been ported... see Quark and Mac OS X.

As I suggested in my original message, you can still run your Microsoft Outlook and WordPerfect, but who cares if they don't run as fast as your 64-bit apps? How fast do you need your e-mail? I dare to say the average person would notice no lag at all.

Re:The way I see it

[leandrod]

> PA-RISC, Sparc, PPC, MIPS, and others already have 64-bit variants

Not quite. Each has not exactly variants, but have been fully 64 bits for years now. Not even sure you can still get new 32 bits.

And BTW, PPC is 32 bits only, that's why it's the PowerPC. The 64 bits is called POWER. Also, SPARC is 32 bits; the 64 bits is UltraSPARC.

Do you know what this means???????

[peculiarmethod]

Wow, I mean WOW.
NOOOOW I can watch my old dos demos from Unreal and The Humble Crew in less time than my brain can percieve them. Just what topped last years christmas list.

pm

Re:Do you know what this means???????

[MrLint]

let me know when someone figures out how to play all of those bootable only floppy games:)

Re:Do you know what this means???????

[Anonymous Coward]

vmware, bochs, virtualpc.

Re:Do you know what this means???????

[Michael's a Jerk!]

The Humble Crew

It's The Humble Guys...

Geez - Am I really this old?

Yes, this is offtopic.

Re:Do you know what this means???????

[damiam]

NOOOOW I can watch my old dos demos

A 3Ghz P4 or Athlon 3000 would be twice as fast, cost an order of magnitude less, and probably have better compatibility with DOS demos.

Re:Do you know what this means???????

[usotsuki]

Not good enough. MESS is closer.

-uso.
Still trying to make a full PC emulator.

And I thought it was just going to be ...

[craigeyb]

And I thought it was just going to be a space heater.

Emulator, converter?

[Trillan]

Ultimately, all an emulator does is convert instructions from one architecture to another. It's almost always more efficient to translate instructions in blocks

To come up with a really primitive, simple example, imagine a simple instruction set with a load, add, and branch if zero-set.

Code might look like this:

lda avar
add bvar
bre label

Now imagine we were translating to an instruction set that had mostly the same instructions, but needed a compare instruction to set our conditional flag

Instruction-by-instruction conversion might turn out like this:

lda avar tstz
add bvar
tstz
bre label

Now if the conversion was done on the entire block, we might end up with this:

lda avar
add bvar
tstz
bre label

Granted, this is a pretty simple example, but I hope it makes my point. Block conversions allow a great deal more optimization than instruction conversions.

This optimization might sound like a lot of work for the host processor, but if the block in question is a tight loop you more than make that up.

Re:Emulator, converter?

[AtATaddict]

Could the second example be achieved more easily(if not more effectively in all cases) by translating on an instruction-by-instruction basis, then optimizing using whatever normally optimizes the native instruction set?

Re:Emulator, converter?

[khuber]

Could the second example be achieved more easily(if not more effectively in all cases) by translating on an instruction-by-instruction basis, then optimizing using whatever normally optimizes the native instruction set?

What do you mean by "whatever normally optimizes the native instruction set"? What normally allows the optimization of assembly is data flow analysis of constructs in the higher-level language being compiled (like C). You can only do limited types of optimization with the raw assembly (peephole). Additionally, we know for a fact that the VLIW/EPIC codes in IA-64 require advanced compilers to produce efficient executable programs.

At the machine code level you have lost all that high level information so it needs to be inferred by profiling and block analysis. I agree with the parent poster on the importance of blocks.

-Kevin

Sounds familiar.

[Grenamier]

This actually reminds me of when Apple's emulation strategy back when they migrated from the old 680x0 series to PowerPC. It was well orchestrated and was actually something of a triumph for them. I hope that bodes well for Intel's attempt.

For Intel to have a long term future without the embarassment of junking the whole architecture, they need Itanium x to run IA32 credibly. Advances in x86 performance keep coming at such increasing development costs that I think they would have to be able to migrate the market to IA64 within 5-10 years from now.

I would like for both the IA64 and the Hammer architectures to flourish, but Intel's taken an extremely bold step with EPIC, and I don't want to see them get punished in the market for that alone. I like the spirit of aiming higher.

Re:Sounds familiar.

[Animats]

This actually reminds me of when Apple's emulation strategy back when they migrated from the old 680x0 series to PowerPC. It was well orchestrated and was actually something of a triumph for them.

Well, no.

Actually, it was a painful transition. Horrible hacks were required to make it work, and Apple lost considerable market share.

From the user perspective, all the applications that used the FPU stopped working. Worse, the PPC only had (has?) a 64-bit FPU, while the 68K and x86 have 80-bit FPUs. So a simple recompile often wasn't enough. Most of the engineering applications (CAD, EDA) were never ported to the PPC at all. There were unsupported 3rd party FPU emulators for the 68K FPU, but they were really slow, since they had to emulate a wider FPU.

Most of the OS ran in 68K emulation mode for years after the "transition". The PPC interrupt model was mainframe-like, assuming that you didn't do much in an interrupt routine except activate some process. The 68K interrupt model was minicomputer-like, with multiple interrupt levels used as the main locking primitive. Hammering those two together was painful. There were some things you just couldn't do in PPC mode; you had to drop into 68K emulation to prevent interrupts.

The old MacOS had what was euphemistically called "cooperative multiprogramming". That didn't mean you had threads without time-slicing, like a real-time OS. It meant you didn't have real context switching at all. You plugged your code into callbacks at different levels of processing, like "system tasks", "VBI tasks", "timer tasks", "interrupt tasks", etc., none of which could block. No mutexes. No locking. Only interrupt prevention. Trying to do anything in the background was very tough. (I know; I wrote a PPP protocol module for the 68K Mac. I had the only one that could dial the phone in the background without locking up the whole machine, and it wasn't easy.)

Worse, the 68K emulator depended on a jump table with 65536 entries, one for each of the first 16 bits an instruction could have. Early PPCs didn't have enough cache to keep that entire table in the cache all the time. But if it wasn't all in the cache, 68K emulation performance was terrible.

Amusingly, much of the perceived performance advantage of the early PPC machines came from the miserable 68K code generators used on the Mac. The Apple and Zortech compilers were clueless about 68K register allocation, preferring to do all arithmetic in register A0. The PPC code generators were much better. Some high-end apps used to be cross-compiled on Sun 68K machines because the Mac code generators were so bad.

Most of these problems were papered over using the Jobs Reality Distortion Field. But this was the period when Apple started losing market share big-time. Arguably, the PPC transition cost Apple its preeminence.

What Apple really needed was faster 68K CPUs, not a new architecture. Technically, that was quite possible. The Motorola 68060, (never used by Apple, but in the last 68K Amiga), was faster than the PPC of the same vintage. But Jobs had cut a deal with IBM under which IBM was supposed to make MacOS compatible machines (!), and that was the motivation for the PPC.

Re:Sounds familiar.

[ahchem]

A lot of what you said was very interesting, and probably accurate.

But you made one major error:
Jobs was not at Apple when they made the PPC transition. He was at NeXT.

I remember very clearly reading an interview given by Jobs where he ripped Apple's decision to switch to PPC.

Re:Sounds familiar.

[SewersOfRivendell]

Actually, it was a painful transition. Horrible hacks were required to make it work, and Apple lost considerable market share.

Well, no. Interestingly, you are technically correct on a couple of complex points, but you seem clueless on others. Perhaps your memory has faded. Think C 5's code generator was far better than MPW (Apple's) C or Symantec C++, but Metrowerks C was ultimately much, much better. MPW C tended to frequently do shit like (actual example from disassembling the 7.1-era Finder, IIRC):

mov.l a0, a5
mov.l a5, a0

Note lack of peepholing.

What you call "cooperative multiprogramming" is actually called "interrupt time." All documentation of which I'm aware refers to it as "interrupt time." No euphemism required.

Jobs had been fired for over seven years when John Sculley cut the PowerPC CPU deal, and It had nothing to do with PowerMac clones.

Most of these problems were papered over using the Jobs Reality Distortion Field. But this was the period when Apple started losing market share big-time. Arguably, the PPC transition cost Apple its preeminence.

No, dude. I was there. Apple never had "preeminence" or much market share. Apple was always struggling under the "Apple is dying" myth (and still does in some quarters today). In the mid-nineties, Apple had a series of crises caused by Sculley and his successor's ineptitude. Worse, Apple stopped playing to it's traditional strengths (industrial design and hardware/software) under Spindler, a problem that, combined with vigorous and useless penny-pinching in all the wrong places -- Apple's hardware & software quality hit the lowest point they'd ever reach at the end of Spindler's reign -- ultimately led to the ouster of Spindler. Amelio failed to recognize this (or much of anything else about Apple), which ultimately led him to buy his own doom in NeXT and the return of Jobs.

Re:Sounds familiar.

[Animats]

I'm down with a bad cold today; some of that was wrong.

I never used Think C, but I used MPW and Symantec/Zortech, and later Metrowerks. As you point out, MPW and Symantec/Zortech weren't very good. Metrowerks was a big improvement.

You're right about Jobs not being there at the PPC transition.

Apple had more market share than IBM in the Apple II days, and it was all downhill after that. When Gil Ameilo came in, Apple's market share was about 7%. Now, it's around 2.5%. (Apple likes to emphasize higher numbers for "home users" or "educational users"). One commentator points out that Apple market share moves like that of a religion, not a business - very slowly.

Re:Sounds familiar.

[Anonymous Coward]

That is the genius of it. It is most certainly factually correct in a general way, including the detail about unsupported 3rd party fixes. Of course by that definition linux is an unsupported 3rd party fix, but that is another story. The whole issue, if you can call it that, is Apple did not emulate the fpu which I think was integrated on the 040 but not the 030. Not everyone had an FPU then. Most software used Apple's math library, an abstraction layer which used the FPU if available or software if not. If you used the library you had no problem with PPC. Software that went straight to the FPU needed this "unsupported 3rd party" emulator. Despite the original posters 64 bit 80 bit mumbo jumbo FP was significantly faster on the PPC than the 040. The only apps that didn't make the transition were two shareware fractal generators and unfortunately the maintenance mode science and engineering apps you speak of. On the bright side, if any of those are still around, they are now likely just a compile away.

Re:Sounds familiar.

[Anonymous Coward]

Apple had more market share than IBM in the Apple II days, and it was all downhill after that.

I can only assume you mean Apple had more market share before IBM released the PC. Remember, the computer industry was very small back then compared to today. It took IBM less than a year to eclispe Apple's market share and less than five to establish the standard 95% to 5% ratio that has more or less held since then. The introduction of the Mac and then the LaserWriter blipped those numbers up but were countered later by other missteps. Also, by market share I mean installed base not the quarterly sales figure you are using. One of the problems of talking about Apple's installed base is that number is sometimes smaller than the margin of error of the statistical method used to measure it.

Re:Sounds familiar.

[IamTheRealMike]

What you call "cooperative multiprogramming" is actually called "interrupt time." All documentation of which I'm aware refers to it as "interrupt time." No euphemism required.

The confusion may come from the same on Win3.1 being called co-operative multitasking, due to the need for apps to relinquish control manually.

Re:Sounds familiar.

[Textbook Error]

It is most certainly factually correct in a general way... The whole issue, if you can call it that, is Apple did not emulate the fpu which I think was integrated on the 040 but not the 030

There was no FPU on the 020, 030, or 040LC - in fact the only Mac chip which had one built-in was the 040, so the possibility of using the math library you mention was well known at the time. People who did use it had no problem (from a math point of view) moving to PowerPC, and those that did were well aware that they were using instructions which were only supported by one type of 68K Mac.

The loss of 80 bit doubles is also a red herring - you now had native 64-bit doubles on all machines (which was a big deal at the time, since you could now assume there was a fast FPU present on all machines and not worry about trying to hack together implementations using fixed point math), and a (slower) 128-bit long double if you really really needed precision.

The original poster sounds like he has a massive chip on his shoulder about something (e.g., the fact that 90% of the Toolbox remained 68K code didn't really matter, since the critical 10% that was called most often - e.g., drawing bottlenecks - was native), but if you're interested in a more technical description of what the transition was like there's a useful article at BYTE.

Re:Sounds familiar.

[Guspaz]

You're leaving out another part of the painful transition; people who had just bought a 68K Mac suddenly found themselves being unable to run new software on their new Mac. Apple introduced the PowerPC, yet kept right on selling 68K models for the lower end, and every single person who bought one got screwed.

Nobody knew it at the time, but by paying a few hundred (And I really mean a small few) more dollars, you would get a machine that would be much more versatile in the future. But, at the time, the PowerPC machines provided no advantage, and cost more.

Re:Sounds familiar.

[jandrese]

Well, that's not full true. Although the people who bought the cheaper 68k machines were ultimatly out of luck, during the transition years everybody either built both a 68k and PPC version, or compiled it as a "fatbinary" which worked on both architectures. Metroworks had a compiler flag to choose which mode you wanted. It wasn't until the PPC603s came out and most of the market had switched that software houses started releasing only PPC versions of their apps. Usually it was for applications that wouldn't run acceptably on the old 68k machines anyway.

Re:Sounds familiar.

[Bert64]

Infact, the M68k architecture would have been a far more logical architecture to continue and extend than x86, and dumping m68k in favor of ppc was a major error on the part of motorola, and resulted in huge financial losses to motorola and countless other companies. Back in the days, m68k was the most commonly used architecture out there, sure x86 was used in ibm compatibles but not much else.... m68k was in macs, amigas, atari, the sega genesis, sun, sgi workstations etc, and countless embedded products and even cray supercomputers. There was a whole range, with the cutting edge chips being used in the high end servers, while the slightly older chips were mass produced cheaply for end user machines and games consoles. Had the m68k range continued to be developed, its likely the performance would be on a par or greater than x86, and many of these companies would still be using them.. whats more we would likely have userspace binary compatibility between most of the unix vendors, even if the kernel and driver sets were different. With this in mind, i can really see intel losing out to AMD bigtime in the near future.
Ofcourse in an ideal world, we could drop obsolete architectures and move immediately to new more modern and better thought out designs, or atleast have chips which are designed with future expansion in mind, rather than the "push the limit just out of reach, reach it next month" mentality of intel.

Re:Sounds familiar.

[Bert64]

But with the 68020 and 68030 there was the possibility of an external FPU, the 68881 or 68882. I believe many machines had the socket to accept one, but simply didnt ship with one by default because very little software took advantage of it, infact m68k FPU`s were always more common on the
Amiga platform, where 3d rendering was more popular.
Running a mac emulator on an amiga always seemed very fast, faster even than a real mac using the same cpu... and this while running the emulation environment as a process under a multitasking os

Re:Sounds familiar.

[leandrod]

> m68k was in macs, amigas, atari, the sega genesis, sun, sgi workstations

Hmm... let's see. Amigas had just died, or were about to; Atari had already went the way of the dodo. Sega Genesis never made much of a splash, Sun had already gone SPARC, SGI was already MIPS. And Apple wanted a future proof RISC system, which the M68K, while better than x86, certainly wasn't, especially in the absence of Intel's monopoly's wages.

Wow. Wrong on all counts.

Re:Sounds familiar.

[Bert64]

The genesis was VERY popular in the UK, ofcourse it was sold as the "megadrive" there, and was the dominant platform until the playstation took over.. Amiga was very much alive at the time too.. atari yeah, they were on their way out.. with their last ditch effort the jaguar, also based on an m68k chip..
Sun and DEC, not sure about sgi and hp tho... both dropped m68k because motorola told them it was nearing its end of life, they tried ppc, which was motorola`s official migration path, but chose to develop their own next generation processors instead.
At the time intel had no monopoly, i would wager that more m68k chips were being sold than x86 ones.

Re:Sounds familiar.

[leandrod]

> The genesis was VERY popular in the UK

One country alone doesn't keep an architecture afloat.

> Amiga was very much alive at the time too.. atari yeah, they were on their way out.. with their last ditch effort the jaguar, also based on an m68k chip.. Sun and DEC, not sure about sgi and hp tho... both dropped m68k because motorola told them it was nearing its end of life

Which timeframe are you thinking about? Seems to be you are uninformed about how old MIPS, SPARC, PA-RISC are, and for how long have Amiga and Atari been in trouble. About DEC it is clear, not only they already had their CISC processor, the VAX, so they were not M68K user, and they had a NIH syndrome not to mention the Alpha being so much better design.

> they tried ppc, which was motorola`s official migration path, but chose to develop their own next generation processors instead.

Ops! Not! The PPC is a derivation of POWER, sponsored by Apple, IBM and Motorola. Even Apple wanted to use the Alpha, but was turned down by DEC's VMS über alles attitude. When the PowerPC was conceived, POWER was already in full competition with the early generations of SPARC, PA-RISC, MIPS, Alpha.

> i would wager that more m68k chips were being sold than x86 ones.

Yes, just as there may be more PowerPCs being sold today as embedded chips, not for computers.

If you want to prove your point, you will have to give references, dates and numbers, not wild suspicions.

Re:Good work. sort of...

[Ryu2]

If this describes your CS program, then it would be a concern, but I don't think that your statement really applies for most.

Actually, most of the architecture classes at school I took used the MIPS architecture as an example -- an elegant, clean RISC design which makes it excellent for teaching. My work on other CS courses were done on platforms from Sun to Mac to SGI, but the X86 was conspicuously absent.

In fact, the only CS course involving any x86 programming whatsoever was a graphics class which used a cluster of Linux graphics workstations.

Details?

[CausticWindow]

Anybody got the technical details on this "emulation" versus the x86 compatibility in Opteron?

JIT compilation or instruction for instruction?

FX32! for Itanium

[msgmonkey]

I'd wager that this is FX32! (allowed you to do the same on Alpha) reworked for the Itanium. Considering Intel purchased all Alpha related technology I would n't be surprised. This is not really that bad a thing since FX32 was quite good at what it did (within its limits).

Re:FX32! for Itanium

[Dolphinzilla]

Lightbulb goes on - Your probably right, FX32 waa never that impressive, even on an Alpha. It could run solitair but not much more without bogging down severely.

Re:FX32! for Itanium

[gggandle]

Assuming my memory serves me correctly, FX!32 used to gradually convert IA32 code to Alpha code. After running the program a number of times the conversion would be complete and it could run 100% native mode.

Once that was done, the program should've run at a decent pace, assuming the binary conversion was reasonable.

Re:FX32! for Itanium

[Jeff DeMaagd]

Considering Intel purchased all Alpha related technology I would n't be surprised.

I don't know what they bought specifically, but I seem to only remember that they bought the fab for Alphas, as well as DEC's NIC and StrongARM technology. IIRC, DEC kept the Alpha technology, but having been bought by Compaq and then HP, I think there are enough cross-licencing deals in place that Intel might just have a lot of those rights available to them.

Re:FX32! for Itanium

[dracken]

Not - necessarily. I can speculate but. I dont know the exact details about the emulation but I can guess what is happening. Over the past few years, dynamic compilation, optimization and dynamic execution layer interface projects and papers have been doing the rounds in academic community. For example dynamo where a dynamic optimizer (which takes code and performs run time optimization on it - not emulation or translation) showed that the apps in fact ran *faster* even counting the overhead of optimization. This idea spawned DELI or dynamic execution layer interface which can dynamically translate instructions *as well as* perform optimizations on them using run time information. Researchers claim that execution is *faster* than running the same app on the native machine. All these are somewhat software equivalent of transmeta.

Now the interesting thing, both dynamo and DELI are from HP labs. So was HPL-PD an architecture that is the superset of itanium, invented and evaluated by HP waaaaaaay back (itanium is in fact based on HPL-PD). Now can the dynamic execution layer emulating x86 be based on DELI ? that is a speculation.

Re:FX32! for Itanium

[Shimbo]

I don't know what they bought specifically, but I seem to only remember that they bought the fab for Alphas, as well as DEC's NIC and StrongARM technology. IIRC, DEC kept the Alpha technology, but having been bought by Compaq and then HP

Intel bought the fabs off DEC. Compaq later sold the Alpha IP to Intel just before the HP merger, presumably because HP already had the Itanium as its migration path for HP-UX users.

Great! Now AMD should release a

[Utopia]

EPIC instuctions Emulator on Opteron.

PPC 970 would to emulate more than than IA32

[Grenamier]

To run existing applications on the 970, an IBM emulator would have to have a bit more to it than just emulation for the IA32 instruction set. Applications depend on an operating system and an actual API (as opposed to a "documented" API). If we want to run Windows applications to run on PPC, IA32 emulation is only a small part of it and most of the rest is already under development by other projects such as WINE.

It would help PPC for IBM to produce a software emulator for IA32, but it would also need to put some resources into helping Open Source projects fill in the gaps with the rest of the platform. I think Intel's IA32-on-IA64 emulation has a bit of an advantage here because the IA64 chips are supported by Windows, which hides the rest of the hardware platform from the applications.

Re:PPC 970 would to emulate more than than IA32

[N1XIM]

We call it Bochs.
http://Bochs.sf.net to be exact.

Re:PPC 970 would to emulate more than than IA32

[damiam]

If we want to run Windows applications to run on PPC, IA32 emulation is only a small part of it

Not really. If the emulation is good, you can just run IA-32 Windows on it and run apps from there, as is currently done with Virtual PC.

Re:PPC 970 would to emulate more than than IA32

[Grenamier]

I was a bit reckless with my use of the term "IA32".

I was only talking about the instruction set and the processor itself, because the post I was replying to seemed to assume emulating an x86 processor was enough to run x86 applications. Of course more than just the processor is necessary to support an app. Virtual PC does emulate the rest of the hardware required, and Virtual PC+Windows is a good solution for x86 apps on PPC.

Oh yeah, this'll work really well...

[TWX]

Having been the right school-age to had dealt with the first "PowerPC" Macintoshes, running System 7.5, this is a going to be a huge fiasco. The biggest problem that 7.5 had was that it was not running natively, the OS itself was being emulated. It sucked for performance. Yes, Apple did eventually get an all-PowerPC version out, with 8.0 or so, but at that point, it was geared toward the hardware of the time, which weren't 601's. School districts are still dealing with the effects of this screwup, and if they had simply built the OS in time to the hardware, this could have been averted.

And if you think that the commercial OS providers, all one of them that are mainstream, are going to have a version of their OS available to the general public for this machine, you're on something. They didn't even have support for more than 512 MB RAM in Windows Millennium, with a processor that can address 4GB.

Re:Oh yeah, this'll work really well...

[Rosyna]

Apple did not have an all PPC version of the Mac OS until Mac OS X. But even Mac OS X doesn't use all the features of the PPC processor (RTOC).

Mac OS 9.2.2 has 68k code in it. The window manager IIRC.

Re:Oh yeah, this'll work really well...

[norwoodites]

RTOC is just one of the general registers, r2, it is not a feature of the PPC, it is just a detail of the normal ABI.

Re:Oh yeah, this'll work really well...

[shotfeel]

And of course if you're running Classic on OS X you can still play all those old 68k apps.

One wonders how the older x86 apps will fare in comparitson 5 years from now.

Re:Good work. sort of...

[afidel]

Yep that's the difference between CS and the real world. The x86 ISA may not be the most elegant or clean but it is kicking the snot out of everything except maybe Power. Sure it can be seen as kludge on kludge but yet no one seems to be able to come out with something that beats it for perfomance without costing many times more.

Re:Good work. sort of...

[questamor]

My classes were based on x86, but also delved sideways into motorola 68k, perhaps with a 70/30 split there, just taking those two into consideration.

The rest was a good mix of 8bit embedded devices, and some build-your-own-cpu emulations. This was in the early 1990s however.

Like the OP, I've come across a few very new graduates who were taught x86 only. That wouldn't be so bad if it weren't for the attitude that they must have been taught it because it was the best and only real option.

Most popular perhaps, but not best.

This why open source will rock.

[Anonymous Coward]

You can easily recompile your apps to run at full speed, while all the nasty propietery stuff gets bogged. Now where did my gentoo for intanium install disc go?

Re:This why open source will rock.

[larry bagina]

recompiling actually involves a diffrent set of headaches...

• GCC hasn't (yet) been well tested with Itanium, and most of the developers work on non-CPU specific stuff, or i386 specific optimizations. You can bet big bucks that bad code is more likely to be generated for itanium, at least for a while.
• Itanium is 64 bits, whereas i386 is 32 bits. Most "Open Source" software is developed exclusively on i386 linux machines. The FSF programmers guideline states that you should assume a 32 bit machine (this was written when 16-bit computers where still common, and they didn't want to makem exceptions for them). There is a LOT of pointer arithmetic that assumes 32-bit pointers, 32-bit integers in open source software (this is a problem for lots of closed source software too). For early adopters, there will be a lot of code that won't run, or won't run correctly.

Re:This why open source will rock.

[afidel]

many, many linux programs have been 64bit clean for some time as linux runs on all of the major 64 bit platforms and has for quite a while. Sure there are a lot of small projects that are not 64bit clean but all of the core services are.

Re:This why open source will rock.

[nchip]

recompiling actually involves a diffrent set of headaches...

Recompiling is actually a bigger problem to propierty programs than to Open Source programs, since anyone can fix a OSS program to work on itanium, while propierty software needs to be fixed by the actual vendor.

GCC hasn't (yet) been well tested with Itanium

This is the biggest problem at the moment, but Intel/HP engineers are free to fix and optimize the ia64 specific stuff on gcc, which they have AFAIK already done. Again, something one could not do with propierty code.

Itanium is 64 bits, whereas i386 is 32 bits. Most "Open Source" software is developed exclusively on i386 linux machines

32bit/64bit big/small-endian problems have been fixed for ages. You can't get a software to debian stable unless it compiles and works on everything from m68k and arm to alpha and itanium.

Re:This why open source will rock.

[dvdeug]

There is a LOT of pointer arithmetic that assumes 32-bit pointers, 32-bit integers in open source software

Nonsense. Look at all the software in Debian, and realize that the vast majority of it is compiled for eight different architectures, including Alpha, with 32-bit integers and 64-bit pointers. Any even semi-major open source program has had those bugs fixed.

goodbye ia32 on-chip emulation?

[1nv4d3r]

Does this mean they can now take the ia32 hardware implementation out? I never liked that idea in the first place.

And, really, can't plenty of us just roll our eyes and go back to compiling our systems from source? I mean, once there's a linux kernel + glibc + gcc port, thousands of applications are instantly available to you.

<preachy>Every time you find yourself strapped to a single architecture, ask yourself why you have all this proprietary baggage holding you back. Whether it's that Word .doc format you used, or that built-on-contract accounting system you didn't obtain the source for, these days it's usually by your choice that you are in this predicament.</preachy>

Oops that should be FX!32

[msgmonkey]

Typo

One wonders why Intel didn't do this originally

[Thagg]

This is so clearly the right way to go, that one has to really wonder what Intel was thinking when they only released a unreasonably slow hardware emulator. I suppose the integration with the operating systems is a bit of a mess, and a moving target at that, but there would have to have been a number of engineers at Intel and HP that would have seen the tremendous performance difference from the beginning. It's not as if software emulation had never been done before.

This, tragically, does hurt AMD quite a bit. I had read multiple rumors about Intel having something up their bunny-suited sleeves, but most of these rumors had Intel supporting x86-64 -- that is -- copying AMD for the first time. This announcement takes away one of the unique advantages of the Opteron/Athlon64 without following AMD's lead.

There are a number of other advantages to the x86-64 architecture and implementations, and I do hope that AMD has a winner here, but it's not going to be the slam-dunk it looked like yesterday. The AMD prices, the Hypertransport links between the processors, the NUMA memory systems, the on-chip memory controller, the human-readable assembly language -- these are all good things. We'll see if they are enough.

thad

Re:One wonders why Intel didn't do this originally

[Meowing]

A wild guess would be that Intel and HP really didn't have the emulation and compiler expertise to do decent emulation right off the bat, but things changed when they got hold of all the DEC/Compaq experience with this kind of thing.

Re:One wonders why Intel didn't do this originally

[afidel]

Trust me Yamhill is still alive and well somewhere inside Intel, they would be stupid not to hedge their bets. At an intro speed of 1.4Ghz the Opteron will be faster at 32 bit code at a fraction of the cost (Newegg shows $315 for 1.4Ghz Opteron (Intel P4 equivilant of ~2.8Ghz) vs $2800 (at pricewatch) for a 900Mhz Itanium 2 (what will the 1.5Ghz part cost???)).

Re:One wonders why Intel didn't do this originally

[captaineo]

I would not be too hopeful about a software emulator. This sounds a lot like FX!32, the software binary translater that let you run x86 Windows programs on the DEC Alpha. While FX!32 was impressive for its time, and certainly a workable product, for most of its life it was not nearly performance-competitive with real x86 hardware. And the Alpha in its heyday was a MUCH faster chip than the Pentium. It's not clear to me that the Itanium CPU is inherently superior to x86 or x86-64 (if you optimize code specifically for it, sure, but that's a time-consuming, expensive proposition). IMHO Intel at this point is just trying to re-clothe the emperor before it's too late. Given their huge R&D investments it could certainly pay off, but I wouldn't put too much stock in that possibility.

Re:One wonders why Intel didn't do this originally

[Glock27]

This, tragically, does hurt AMD quite a bit.

I don't think so.

I had read multiple rumors about Intel having something up their bunny-suited sleeves, but most of these rumors had Intel supporting x86-64 -- that is -- copying AMD for the first time. This announcement takes away one of the unique advantages of the Opteron/Athlon64 without following AMD's lead.

If you think running 32-bit code half as fast (1.5 GHz. Xeon vs. 2.8 GHz. Xeon) on a processor that costs four times as much takes away any advantage you're confused.

To me it looks like Opteron is around 8x more cost effective at running 32-bit code.

Further, even on head-to-head TPC-C results, the least expensive Itanic system was only half as good. The Itanic running a non-Linux OS was only 1/7 as good.

Intel is in trouble.

Re:One wonders why Intel didn't do this originally

[bigbadwlf]

I'm pretty sure it's called Itanium. Is "Itanic" you way of trolling?

To me it looks like Opteron is around 8x more cost effective at running 32-bit code

I thought the point was to finally move away from "old" 32 bit code. This 32 bit capability is there for backward compatibility until such time as individuals and companies retire their old 32 bit apps.

Things change. The industry moves forward. This is why we don't all run 486's anymore.

Intel is in trouble.

Let's hope not. Competition is a good thing. Anyone who thinks AMD wouldn't gouge consumers with their pricing if they were the dominant player is naive.

Re:One wonders why Intel didn't do this originally

[John_Booty]

Things change. The industry moves forward. This is why we don't all run 486's anymore.

If the Pentium wasn't backwards-compatible, we might still be running on 486's after all!

Re:One wonders why Intel didn't do this originally

[conway]

Because Intel is a hardware company.
Hardware companies solve problems in hardware.
It just never occurred to them :)
HP on the other hand, who was co-designing the Itanium, wrote an emulator called Aries which emulated HP-PA by doing runtime translation, and that worked perfectly, and at about 80% of the normal app speed. Aries was shipped with HP-UX 11i on the orignal Itanium, and is greatly helping people transition from HP-PA machines to Itanium machines without recompiles, and no loss of performance (since Itaniums are usually faster than the PA's).
Apparently, after 10 years of hardware development of a crappy x86 emulator, Intel realized that its easier (and faster!) to do it in software.

Re:One wonders why Intel didn't do this originally

[PCBman!]

For a hardware company, they write some incredible compilers.

Re:One wonders why Intel didn't do this originally

[Texas_Refugee]

"Itanic" is the name that TheRegister.co.uk uses to refer to the Itanium.

They also call Intel "Chipzilla" and AMD "Chimpzilla" I believe.

Those cheeky Brits.

Re:One wonders why Intel didn't do this originally

[Glock27]

I thought the point was to finally move away from "old" 32 bit code. This 32 bit capability is there for backward compatibility until such time as individuals and companies retire their old 32 bit apps.

Ooops, I wasn't clear. The TPC-C results were 64-bit. Opteron has a pretty good edge on integer performance, 1100

Re:An IA-32 emulator for the PPC 970 would be fast

[thanasakis]

the implementation of the UltraSPARC III is rather poor

Could you explain this a little more? Really, I am not pretending to be a CPU expert here, but I was under the impression that the Ultrasparc-IIIi is a valid opponent to the intel CPU's because of its considerable throughtput.Ok maybe it does not have the horsepower of a 3 GHz pentium but in terms of overall performance (I/O AND computations) they are comparable. I do not see why a 10 year old mature 64 bit CPU architecture can be that worse compared to the IA64 which seems to be a it's very beggining. I would argue that surely an UltraSparc cannot emulate a Pentium very well (because it lacks the MHz's) but that does not constitute a proof that the implementation is poor.

Right

[DreadSpoon]

The Itanium2 is fricking fast. I mean, like, whoa kinda fast.

The Power5 might compete fovarably against the Itanium2, but not PPC 970.

PPC 970 is a consumer class chip tho, while Itanium2/Power5 are, well, not anything you'll find in a consumer's living room. ;-)

Re:Right

[AtATaddict]

Well, one PPC970 might not compete against an Itanium2, but how many 970s will you be able to buy(I'm assuming the pricing will be comparable to what Motorola currently gets for its high-end) for the price of a single Itanium2?

Re:Right

[Hoser McMoose]

The Itanium2 CAN be fricking fast, but it isn't always.

The performance of the IA-64, at this point in time at least, is really quite uneven. There are some situations where it's just blowing everyone else out of the water, being easily 2-3x (or more) faster then any other chip out there. Intel really likes these benchmarks and likes to show them off. However there are also a number of other applications where the Itanium2 seems to just slow to a crawl.

The trick is that IA-64 is HUGELY dependant on how well a compiler can optimize code for a particular application. As the compilers mature, the weak areas of the Itanium will shrink, though there will always be the odd case that just doesn't work well on IA-64.

As for right now, the Itanium2 and the Power4 are very close competitors in general. Some applications will run a lot better on the Power4, and some will run a lot better on the Itanium2, but most will perform similarly on both architectures. The IBM chip does have an advantage of being a bit more mature though, and it's performance tends to be much more consistant.

Now the only question is

[miketang16]

Will Intel concede to the AMD x86-64 architecture, or will they try to branch out on their own idea? (HINT: 2nd one is a bad idea... just look at the first "64 bit" itanium)

IBM PowerPC 970 clocks at 1.8 gigahertz.

[reporter]

Please read "Inside the IBM PowerPC 970" by "ARS Technica". The PowerPC 970 clocks at 1.8 gigahertz.

don't hold your breath

[rve]

IBM already offer this functionality for their iSeries, though not with a software emulator but with a hardware expansion. Windows, Linux, Unix and OS/400 can run in different subsystems on one machine at the same time.

The desktop market is not really a priority for IBM, so don't expect IBM to put immense amounts of effort into a piece of software that would almost certainly cost you more than buying an actual additional x86 PC to run your 32 bit apps on.

Software emulation?

[swtaarrs]

Since it's software emulation, this means that it's OS dependent. So if you want run x86 apps on an OS that Intel hasn't written their emulator for, you're out of luck unless somebody else writes one. Imho, AMD's Opteron native hardware x86 support is much better, and I think the AMD chips will be much cheaper, much like the Athlon XP is much less $$ than the Pentium 4.

Re:Software emulation?

[swtaarrs]

Ah, my bad. So if Intel released code to activate it then it's not that bad.

Re:Clean Design

[afidel]

Actually the RISC crowd was primarily right, they were just targeting the wrong area. All x86 cpus since the Pentium have been RISC internally with CISC externals. This works well because the larger words work well to minimize cache latencies (if you can fit more into each fetch then the impact waiting for it to arrive is minimized) and the RISC internals make it easier to ramp up the speed of the actual execution units. As you pointed out the PPC is seen as a "RISCish" cpu yet it shares many traits with the "CISCish" x86 cpu's. Pure RISC cpu's are a thing of the past, but it did have quite an impact on the overall design of CPU's.

Intel: Back to its roots

[satch89450]

Fun. So now they realize after they create the chip that they want 20 years of backwards compatibility.

Back before Bill Gates and IBM's Entry System Division thrust Intel microprocessors into every other home on the planet, electronic systems designers were actively courted by Intel by their claim to developing products that won't invalidate all existing design work in one swell foop. And, for the most part, they held up on their end of that promise, which is why the Pentium 4 still has a little bit of the 8080 in it.

Now, when the i432 came out, it was a completely different beast -- and the i432 died a justified death. The i860 didn't fare that well, either. The i960 has seen quite a number of design-ins, because the solution base the i960 was geared to was sufficiently different from the 80x86 that designers didn't try to replace 80x86 chips with the RISC-based i960.

Intel, that was a clue.

What Intel didn't foresee, but should have, is the great technological bust of 1999 put a number of companies under. Source code has flown to the four winds, in some cases the foreclosures also nailed every single backup. In short, the migration path via recompilation was no longer an option. (Not to mention that there were no dollars to make even the most trivial changes to the source to deal with 64-bit processors.)

So this announcement is surprising only in that it comes so late in the product development cycle, as Intel is coming out with its second generation of IA64 chips.

Competition. It's a good thing.

Re:Intel: Back to its roots

[ChadN]

Wasn't the "bust" a year 2000 phenomenon (late in the year, no less), not 1999? I remember 1999 as an enthusiatic year, where the sky was still the limit and everyone hopes were expected to be realized as the internet commerce age came to maturity. And the Y2K bug was still putting lots of people to work.

Re:Intel: Back to its roots

[snilloc]

During the US Prez campaign it became very obvious (mid to late 2000) that things were probably on a downturn. If it was obvious by that point, then the real damage happened way before that.

Go back to the first time Greenspan said "irrational exuberance" and start looking there for when the "real" economy started to tank. The stock market dive (again, mid Y2k-ish) was an aftershock of the real problems in the economy. Looking up stock market info would probably be trivial, yet I'm too lazy to do it. Some super-anal Alex P. Keaton could probably quote you all the relevant info, but I'm not that person.

why bother?

[g4dget]

As far as I can tell, there is no reason to switch to Itanium right now--it seems to be expensive, slow, and hard to compile for. This "compatibility layer" won't change that. When the Itanium is competitive in both performance and cost, then it's worth looking at again.

conversion?

[SHEENmaster]

If the IA-64 instruction set is similar to the IA-32 one, then it shouldn't be hard to wrote a program that converts binaries from one processor to another before running the code?

By murphy's law, dynamic linking or primitive datatype sizes will keep this from being practicle.

Re:conversion?

[Webmonger]

The IA-64 instruction set is not similar to IA-32. It's very, very, very, very, very different. Instead of being CISC (like x86) or RISC (like PowerPC) or VLIW, it's EPIC. The IA-32 compatibility is provided by special compatibility circuitry. If you're looking for a 64-bit instruction set that's similar to x86, you want AMD-64.

Re:conversion?

[julesh]

For those who don't know (I graduated in '97 and my computer architectures course had no mention of it, so it must be a fairly recent development...) EPIC standards for 'Explicitly Parallel Instruction Code' and basically means (as far as I can tell from a 5 min google) that stuff like instruction reordering for the parallel execution cores is handled by the compiler, rather than the processor (the theory being that the compiler should be better at it).

I think this makes it orthogonal to RISC/CISC/VLIW architecture and I believe Itanium owes the most to RISC, although this isn't explicitly acknowledged anywhere I've found and my theory might be a bit rusty...

but certainly, dynamic recompilation _has_ to beat emulation (except for startup speed). Apart from anything else, the new CPU has more registers to take advantage of so redundant store/load (or push/pop) cycles can be eliminated in the recompilation with a little data flow analysis...

Re:conversion?

[Webmonger]

Actually, EPIC is closest to VLIW. It's not orthagonal to the issue.

I din't think they have a plan

[Anonymous Coward]

The first Itanium was basically designed by Intel. Itanium 2 was HP's attempt to fix it (much better). Intel has lost it. They have great manufacturing and competent marketing, but they let all the best engineers leave (to AMD, IBM, Transmeta, etc.). And now they're starting to behave like Microsoft (threatening OEMs to try to stop them from using competing products). But their grip on the chip market isn't half as strong as Microsoft's on the OS market, so there's a serious risk of backfiring.

If you add AMD's engineering (Dirk "Alpha" Meyer & co.) to IBM's manufacturing (fabs nearly as good as Intel's and a lot more R&D), you have a pretty respectable force. Now all they need is decent marketing. And I suspect they'll get that from Microsoft (from Microsoft's point of view, anything that keeps Intel from growing too large is a good thing).

The fact that AMD is now insisting that "x86-64" be renamed "AMD64" might mean they know Intel is working on a "x86-64" CPU, and want to force them to use the new name. Wouldn't it be funny, an Intel CPU marked "AMD64 compatible"...? Now, does Intel care more about its money or it's honor? As Sir Francis Drake said, "you should fight for the one you have less of"...

Re:I din't think they have a plan

[Ben Hutchings]

I think AMD may be making a mistake in using the AMD64 name. I think they need a second vendor to give the architecture credibility and grow their own sales, and the only other x86 vendor interested in the high end is Intel (unless IBM is still making x86s). As you suggest, the new name will make it harder for Intel to get on board.

WHY...

[Uber Banker]

Is this modded as +4 100% interesting???

Why why why?

Pity there is no -1 100% wrong choice huh?

yeah, right

[Apreche]

and windows 95+ runs dos programs so well. I don't need a seperate box to play TIE Fighter or anything.

Re:yeah, right

[SuiteSisterMary]

Aye, but is that Win95 not emulating DOS, or Win95 not having all the TSRs for sound and the like?

In any event, go get the collector's editions of Tie and X-Wing; uses the TvX engine, which is DirectX 5 or so, gets you 3d accel, and all of the various addons.

If only they'd update it to the Alliance engine....

Crusoe

[Anonymous Coward]

So now Intel's Itanium is the exact same as Transmeta's Crusoe processors when running x86. Except Intel exposes the VLIW part of the chip.

And they call it EPIC instead of VLIW. This emulation can be pretty good. Transmeta was able to beat all native x86 chips when it first came out, and their manfacturing plants presumably weren't as nicely tuned as Intel and AMD's.

This should make it "trivial" for Itanium to also support x86-64/AMD64 if they wanted to "show how much more effecient Intel64 is".

Does this have anything *directly* in common with FX!32 I wonder?

Bad Reputation still here for AMD

[AvengerXP]

Whatever happens, even if the Opteron was 100% full backwards compatible and 2x faster than Itanium, nobody in the server segment or even the high end workstation segment will buy an Opteron because they think that AMD makes unstable cheap processors targetted at the nerdy overclocking enthusiast.

I personnally don't agree, but my opinion isn't worth jack inside the corporation and I already know the system's administrator has a "Intel Inside" sticker on his forehead, even if the chips cost 2x as much. They say they pay for "quality". Psssh, what a load of bull.

Re:Bad Reputation still here for AMD

[DeathPenguin]

Sad, but true.

Re:Bad Reputation still here for AMD

[glsunder]

There's a lot of people out there who aren't near as into CPUs as what you're thinking. There's a lot of small and medium businesses who have good relationships with non-tier 1 vendors, and if their vendor recommends the opteron, and it delivers what is promised, the customer will be happy. Throw in some tier 1 support from someone like dell, ibm or compaq/hp and I think you'll see AMD having a good chance to prove themselves. I think it'll come down to AMD delivering on their promises of stabilty, speed and value.

Re:Bad Reputation still here for AMD

[ProtonMotiveForce]

Yeah, those silly dolts. Paying attention to history. Pshaw.

Hell, I'm sure AMD will be around in 5 years, they're raking in so much money.

Re:Bad Reputation still here for AMD

[metamathica]

There's a reason why the people in the corporations feel that way about AMD processors. I was involved in desktop systems and internal servers purchasing at a previous job.

Our level of hardware stability complaint was more than double on AMD-powered systems. It is likely just as much the fault of the motherboard manufacturers as AMD directly, but this--along with the fact that P3 processors didn't require headache-inducing fans--led us to stop buying AMD altogether. The disparity in failure rates was even higher in hot rooms that had too many servers in too small a space. The boxes were all within specified operating temperature range, though they were indeed a bit hot.

Indeed the P4 runs hot just like the AMD chips do. However, the overheat protection still gives Intel an advantage. If a fan fails or some vents are blocked, it's a big advantage to know that your processor will still operate and will still be stable later.

I agree that most AMD processors are fine, at least if paired with a really good motherboard. However, the conventional wisdom about AMD stability problems has proven itself repeatedly to me and many other systems administrators.

Systems administrators at corporate level would understand that one stability problem costs more than the Intel premium on at least a dozen chips.

I'm sure that some disagree with me. I was personally skeptical of spending extra money on Intel for a while--which is why I could compare installations. Experience and my personal data convinced me that Intel was worth the premium--which is nowhere near 100%.

Re:Bad Reputation still here for AMD

[juhaz]

Both are, but of these two, Itanium is CLEARLY more "unproven" technology than Opteron is.

Re:Bad Reputation still here for AMD

[Bert64]

Which is exactly why itanium has failed sofar, and companies are still buying sparc/power/mips/parisc/alpha/x86 based machines instead, all of which are proven technologies.
I still think the idea to develop itanium atall was a stupid one, and they (intel/hp) would have been better off sticking with the alpha.
Ofcourse there are still plenty of people with less than half a brain.

Motives...

[mraymer]

I really have to wonder if Intel is doing this because of customer demand, or simply because they don't want AMD to have the upper hand.

From what I've seen, I would argue that their motive is the latter. Intel has show on several occasions that, these days, they simply don't give a damn about the end user. They care about market share, profits, and their precious stock price. Let's not forget the fact that "Pentium" was coined because Intel wasn't allowed to trademark the number 586.

Remember when they released an overclocked Pentium III to the public, and Tom's Hardware had that nice little article exposing it for the failure it was? It choked on GCC, among other things, while Intel steadfastly denied the problem. Then they actually recalled the processors. Competition at the expense of the end user... wonderful!

It is clear AMD is still going to come out on top in performance on this one, unless "software emulation" doesn't mean what I think it means. It is also clear to me that Intel has to do a lot more than throw some software emulation at an issue before I ever buy another Intel processor.

Intel's Missing the Point of the Opteron

[1stflight]

The point of the Opteron isn't the fact that it can do 32bit fast, but that it can do 64bits in a way that everyone understands and has been hammered out for decades.
The Itanium is a marvelous piece of work however, how's going to adopt something so unknown, vs something so familiar? That is the point Intel missed, 32bit is dead, 64 bit is here, which one will be chosen?

Will this be implemented a la 'code morphing'?

[0rbit4l]

I wonder if the emulation technique they'll be using will be similar to Transmeta's 'code-morphing'. I always wondered why intel didn't license that idea & use it on their Itanium. 'Code-morphing' achieved middle-of-the-road x86 performance on a VLIW (sound like a familiar goal?), but it was still far better than what Itanium gets with its current x86 support.

Re:Will this be implemented a la 'code morphing'?

[khuber]

Worked pretty well for Transmeta, didn't it? (laugh)

-Kevin

Re:Will this be implemented a la 'code morphing'?

[0rbit4l]

Transmeta hasn't succeeded for various reasons, all of which are arguable, but I don't think performance really was one. Most notably, their wild claims & marketing 'buzz' (lies) about power consumption didn't pan out. However, the point that x86-to-VLIW emulation in hardware actually works reasonably well (performance-wise) is usually lost to the masses that see a failed dot-bomb. This was an incredible feat. While they didn't get 700Mhz p3 performance out of a 1Ghz crusoe as they claimed, they got fairly close - and with a far simpler (and far different) processor than the p3. My point is, the technology for hardware JIT compiling & optimization to a VLIW from x86 is in place - intel should have at least looked at using it before now.

Re:Clean Design

[karlm]

The Alpha was the first CPU to 500 MHz (right?), and stayed competitive with x86 long after it got passed in clock speed. x86 is so good today because of the tremendous amount of research money its sales volume generates. Alpha provided much more bang for the research buck. They simply looked at what sorts of instructions are easy for compilers to utilize and designed a CPU around that. (It also helped that from the get-go they decided all Alphas were going to be purely 32+64-bit RISC chips. They had none of this stuff with IBM now supporting 3 different but closely related instruction sets on the POWER/PPC chips.) Are there any compilers that output x86 BCD arithmetic instructions anymore? AMD got rid of the 8087 mess in x86-64-bit mode. Maybe eventually they'll release a 64-bit-mode-only x86 CPU with a 8-16-32-bit mode emulation library.

In 64-bit mode, do they finally use constant-width instructions or at least limit themselves to 2-byte and 4-byte instructions? AMD has done some very smart things with x86, like 3DNow being much cleaner than MMX. I hope they continue to do nice clean things with the x86 instruction set (and depricate the kludge).

x86-64

[darthscsi]

An interesting thought is that the instruction format and register set of AMD's x86-64 is just an extention of x86, so if Intel has a good emulator for x86 running on IA64, then it should be (from a technical standpoint, not a licensing standpoint) fairly trivial to emulate x86-64 at speeds similar to the x86 emulation. THAT doesn't bode well for AMD.

And as for licensing, a clean room implementation should be very easy considering it is simply an extention of x86.

Re:x86-64

[bstadil]

THAT doesn't bode well for AMD

That would be the best thing that could happen to AMD.

It would validate AMD64 and give them a HUGE cost advantage. We are talking 5X or so at current pricing levels.

Re:Not quite - emulation of virtual machine unavai

[TrailerTrash]

Thanks for the pr0n link.

Shame on whoever modded this up.

on talking dogs and other wonders

[snarkh]

From the Inquirer (www.theinquirer.net):

In this 1996 Byte article, Selinda Chiquoine wrote that executing code written for one CPU on another "has always been like the talking dog" ? that is, the fact the dog could talk is a miracle in itself, even if all it could say was "sausages".

Re:on talking dogs and other wonders

[usotsuki]

Emulation of a CPU is nothing. There've been Apple ][ emulators for the PC for 15 years (1988!) now, and they generally had reasonably sane 65C02 CPU cores way back then. When I hacked together Dapple 0.1, I didn't consider it a miracle that the ][ software ran, but that *I* made the ][ software run.

-uso.

Will compiler tech ever get there?

[rabtech]

Itanium and the IA-64 instruction set depends very very very heavily on a good compiler to draw parallelism out of the code.

The reality is we may never get compilers that are that good, and we may never have many applications where much parallelism can be drawn out anyway... at least not enough to make it worthwhile.

EPIC is a huge gamble... one that may not pay off in the long run. I'm no fan of x86 per-se, but it seems that AMD has tried to bring it up to speed with x86-64... more registers (always the biggest complaint), dropping some of the cruft. I just have to wonder... with the major increase in die size and cost over the Opteron, the Itanium 2 sure doesn't bring that big of a performance lead to the table, and the Opteron still has lots of room to grow.

I think this latest effort is just Intel finally admitting that their x86 emulation strategy was a total failure (partial hardware emulation), so they are abandoning it and starting over with a software emulation layer. You cay pay through the nose for an IT2 system and get 1.5 Xeon MP performance. Well whoopity-do.

A0 math

[sacrilicious]

The Apple and Zortech compilers were clueless about 68K register allocation, preferring to do all arithmetic in register A0.

A0 was an address register... did you mean D0, or did they actually do math in an address register?

Re:A0 math

[Bert64]

Well he did say the code generator was pretty clueless, so it wouldnt surprise me atall :P

Misleading headline

[conway]

The headline is misleading.
Itanium has always had x86 emulation, just before it was done in hardware, and very very slowly. (The Itanium 1, at 800Mhz, ran x86 software at the speed of a 150Mhz pentium or so.)
A story at The Register, here explains that this new software will translate some of the x86 assembly to IA-64 assembly at runtime. (See picture)
This is the same way that HP's Aries works -- which translates HP-PA instructions into IA-64.
That works pretty well actually, delivering about 80% of the nominal speed most of the time. (We've used it a lot during development of HP-UX on Itanium, and actually ran a lot of the system binaries (ls, grep, etc.) on it until they were ported. Worked pretty well!).

What they still haven't done is implement something like this in hardware, but efficiently, like Transmeta does -- they translate x86 to a RISC core in hardware, and get really good performance.
But hey, this is Intel we're talking about :)

If Intel hadn't have done it

[Stonent1]

Someone else would have. Remeber NT 4.0 (and I think earlier versions) had an emulation layer for RISC processors to be able to run 386 binaries. I'm sure it wouldn't have taken long for MS or the open source people to come up with something similar. At least if Intel does it, they have the inside knowledge to be able to tune it for best performance.

Re:If Intel hadn't have done it

[sn0wcrash]

Maybe you are reffering to fx32. It was an emulator for NT 4.0 Alpha edition to run x86 binaries.

Re:Where to download this "FX32!" ?

[Dahan]

You can get it from here. At least I think that's it... I haven't had NT4 running on my alpha in years, and that file's an Installshield self-extracting EXE for NT Alpha, so I can't run it.

Re:Not quite - emulation of virtual machine unavai

[mu_wtfo]

Ok, not reading artiles before posting is common, and I guess by now I can deal with it - but not even reading a *comment* before moderating it!? People - you just modded up to +3 a link to www.teenmaidensonline.com! Good lord!!
I guess bashing .NET is just a sure way to get modded up...

Re:new? - NO -Why Mod him down?

[Glasswire]

Can't really understand why he was modded down -unless the rater was peeved that he hadn't heard the information himself?

I think it's an astute obsvervation that Slashdot sometimes reacts to online-press stories generated by official press releases about technology that everyone (or anybody who really cared) already knew about. If the reaction and discussion was to do with a company going public about old news, that would be fine, but folks that post about ancient history shouldn't delude themselves into thinking if it's on Slashdot, something technical just happened.

It is backwards compatable

[autopr0n]

But they say this emulator will run faster then using the built-in x86 decoder.

Please! (this is for all the people arguing this)

[pantherace]

Your second arguement doesn't make any sense. Show me any major OSS program that will not run on say Alphas because of pointer arithmatic, excluding Star/Open Office. (The reason for that is that they did do stupid things like that, and to get even a "64-bit" SPARC version, they had to do essentially the same thing Microsoft did on Windows NT for Alpha-clip it to 32-bits.)

The last program I had trouble of that sort was KDE 2.0-beta3 (may have been fixed anywhere from beta3-beta5) (and that may have been something else, as it was a wierd bug, because all the people I knew who had the problem (very limited set) all had alphas, and I didn't check or know anyone who checked on other archs)

Would all of you that make that arguement either 1-tell people their programming sucks, to the specific project, or more politely file a bug report. 2-stop with this linux/gcc/oss-in-general isn't 64-bit ready.

And if you want a time from when linux was 64-bit ready, look up a 533MHz Alpha's release date, and add about a year.(likely less, to determine how long I have been using a 64-bit ready linux)

Now, endianness may be a problem, or a cruddy instruction set, but don't confuse IA-64 or x86-64 as being the first 64-bit chips, and if you are going to argue that pointer arithmatic is problematic, then cite an example!

end rant on that subject

Why not dual CPUs ?

[billcopc]

I know nothing about systems design, but why couldn't they have a slave ia32 cpu just running in the sidelines, somehow tied into the architecture ?

Re:Why not dual CPUs ?

[iggymanz]

the funny thing is that the Itanium DOES have hardare x86 emulation, but it sucks so very badly they made this software emulator. How embarassing.

Re:Why not dual CPUs ?

[MSTCrow5429]

It wouldn't be cost-effective. The die size would be larger than it is already, the energy consumption would be over 200 watts, the heatsink would be the size of a small car, and you'd need a peltier cooler for the damn thing. And it would cost probably at least half again as before.

Wonderful example of spin doctoring

[XNormal]

The Itanic always had full 32 bit x86 compatibility and a significant percentage of its die real estate is spent on it. It just sucks so much that it's outperformed by software emulation. Needless to say, if you use the software emulation layer you would *still* be paying for the hardware emulation.

No they're trying to spin this story as if it's actually something good and not a patch for a white elephant.

See this story on The Register

Re:Clean Design

[renoX]

There is no such thing as being internally RISC: RISC is about a style of Instruction Set (Reduced Instruction Set Cpu).

Still, if you look at all the other CPU except the 80x86, they are all RISC CPUs, or VLIW.

The fact that 80x86 is CISC is irrelevant, 80x86 has won by being 1) cheap 2) being able to run the cheapest OS (Microsoft OSs), with lots of software.

Re:Clean Design

[renoX]

> In 64-bit mode, do they finally use constant-width instructions or at least limit themselves to 2-byte and 4-byte instructions?

No, they just added yet another prefix, this way they can mix regular x86 instructions and x86-64 instructions.
Yes it is another kludge, which will increase the code size of x86-64 code, but at least it gives you 16 registers.

3DNow! may be much cleaner than MMX (it isn't hard, MMX suck so much it isn't funny), but it is also dying a slow death: SSE and SSE2 will replace it eventually.

pretty major fork between AMD and Intel

[dpilot]

Actually, the fork is between Intel and Intel - at least until we see how good the emulation really is. With IA64 Intel has forked with its own X86 past. AMD hasn't forked.

Re:pretty major fork between AMD and Intel

[ocelotbob]

I was talking about compatibility between the two chipmakers. Up until now, they've been largely compatible. this is the first time when they have actually diverged in what is the best course for "mainstream" PCs

Re:pretty major fork between AMD and Intel

[dpilot]

I understand what you were saying. I was merely adding that it's Intel that broke the compatibility, not AMD. For that matter, AMD simply CAN'T clone IA64, even if they wanted to. From an IP perspective, IA64 is the most locked-down architecture on the planet. Can you say proprietary?

(I'm not a fan of IA64. Only Intel could get away with IA64 without getting laughed out of the conferences. Only Intel could get away with IA64 without EVERYONE retreating in fear of lock-in and loss of control. I guess this is the indication that Microsoft's power is fading, because Intel is stepping in.)

Intel fearful of Opteron

[m00dawg]

It looks like Intel is a bit concerned about Opteron after all; if they weren't, the probably wouldn't be putting in the x86 features into the Itanium shortly after the release of the Opteron. Either way, I think it shows that lately Intel is lacking some innovation, which is fine by me since I'm an avid AMD fan :) I'm not a huge fan of x86, but being able to easily migrate to 64-bit, or even just being able to have 8 more GPR and SSE registers is a nice idea that Intel should have done LONG ago.

Been there,done that (well, kind of)

[0123456]

Years back I tried an x86 Sinclair Spectrum emulator on SoftPC on a Tadpole Sparcbook: it was only running something like a 40-50MHz Sparc chip from what I remember, but the Spectrum games still managed to run pretty much at full speed. The main benefit of the Sparcbook for emulation, AFAIR, was that it had a real VGA chip so SoftPC didn't have to emulate that as well as the CPU and other components.

Re:Clean Design

[nickos]

That's silly. ARM's a clean and classic RISC architecture, and used extensively by embedded systems. Alpha is another good clean design, that was blazingly fast until Compaq killed it.

Recompiling isn't always enough

[0123456]

"And, really, can't plenty of us just roll our eyes and go back to compiling our systems from source?"

What do you do with OS-dependent code that's no longer supported and was written for an OS a couple of versions behind the one you're running on? Untar, make... oops, three bazillion build errors. Or even, say, what would have been perfectly portable Xview code in these days of Gnome and the like.

Sure, you have the source, you may be able to understand it, and you could put in days, weeks or longer to fix it to build on the new OS. But on a number of occasions in the past I've avoided doing so by running an old a.out binary on Linux while migrating to new software which did the same job and was still supported. There are very good reasons for wanting to be able to run older code without recompilation.

software 32 bit emu

[ratfynk]

Roll up your sleeves, pull them down, bend over and lookout here it comes 64 bit/with 32 bit emu home computing. So all you need is software emu, the national bird of New Zealand, that does not fly too well. Will hardware manufacturers be in any hurry to rewite their 32 drivers for the Win/Intel emu world, not if they can sell suckers new stuff. Get real you guys do you really think all the driver issues that are going to come out of the wood work will be that easy to address? The pci card issues alone will make your hair stand on end, so you will have to make sure you buy only Intel certified hardware. At higher prices, naturally. Forget your raid array buddy it will crap big ones. SCSI where are you? Will the emu do a good enough job of rerouting the instructions? Will the I64/32 be now portable or even possible, (given Microshaft software patents) in Linux? Hey they are talking about core Win32 code in the emu here, not just the stuff that Wine emulates. How easily one forgets Windows 3.0 and the first release of Windows 95, Win 98, not to mention the biggest piece of software crap ever unloaded on the public Windows ME/2000. We all had to rush out and buy new hardware, and hope that there were no driver and software issues.
Well its happenning all over again, the hardware manufacturers, (so called MS hardware partners) are getting hungy and they are about to feast on the beast. Intel and Microshaft have got more crap up their sleeves, and it is aimed directly at your wallet. Save money buy the Hammer! Or an Apple if you have lots of cash. Save yourself future head aches, learn to use real computers, with real software.

Ah - You're right ACs do go to 0 - not fair either

[Glasswire]

(As above)

Re:An IA-32 emulator for the PPC 970 would be fast

[jbx]

Ummm, no.

First off, you missed an extremely important technical detail of the PPC 970: they removed little-endian mode. Sure, you can still use instructions that store and fetch in little-endian mode, but you are forced to use indexed addressing mode when doing so. The upshot, for an emulator such as Virtual PC, is that a PUSH instruction on the x86 now takes 3 PPC 970 instructions instead of 1 G3/G4 instruction. Also there is no way on the 970 to load or store floating-point numbers in little-endian, so all those instructions take five or ten times longer.

Second off, emulators are NOT easy to write. Can't imagine where you got that. For one thing, you end up having to emulate hardware in addition to software, and the 20 years of IBM PC evolution is not pretty or elegant in any way. For another, x86 CPUs do automatic code flushing, which you have to emulate, meaning that every store instruction potentially is a code-modifying instruction that you have to beware of.

Finally, perhaps the biggest problem is still code size. 1K of x86 code still translates into 3-4K of PowerPC translation. Meaning a PowerPC would need 4 megs of instruction cache to do as well as an x86 with 1 meg of instruction cache. Lack of little-endian mode means it's even worse.

None of these problems get easier with the 970. Now, true, the 970 is a nice fast chip, and well-written PPC code will fly on it. But many of the ways in which the 970 is fast don't help emulation at all. And the lack of little-endian mode just blows chunks.

jbx (former VPC engineer)

Re:Clean Design

[karlm]

No, they just added yet another prefix, this way they can mix regular x86 instructions and x86-64 instructions. Yes it is another kludge, which will increase the code size of x86-64 code, but at least it gives you 16 registers.

Too bad. Constant width instructions would have meant they could start decoding an arbitrary number of instructions ahead without partially decoding any of the preceding instructions. As long as you're willing to to throw out 2/3 of your instruction decodings, you can also do the same with the current 3-length (1,2, or 3 bytes) system. Even on x86, which allows unaligned memory acess, aligned meory acess is faster.

The ARM "thumb" instruction set gets away with 16-bit instructions with 16 registers. Maybe getting down to 256 operations is too high a price to pay, so they'd go with 16-bit and 32-bit instructions. Sticking to 2-byte and 4-byte instructions means your loads are aligned and your instrcuctions never wrap across cache lines. The 8087 FP instructions don't work in 64-bit mode. Are all of the other x86 instructions functional in 64-bit mode?

I assume they wanted people to still be able to use their old 32-bit binary-only libraries linked with 64-bit code. How costly is a mode change to/from 64-bit mode? I trust AMD to make the design clean enough that a mode change only involves a change in the decoder unit.

Re:Good work. sort of...

[leandrod]

> The x86 ISA may not be the most elegant or clean but it is kicking the snot out of everything except maybe Power. Sure it can be seen as kludge on kludge but yet no one seems to be able to come out with something that beats it for perfomance without costing many times more.

You never heard of the wages of monopoly and economies of scale?

Give RISC the same engineering and manufacturing resources and it will be cheaper, faster, smaller, more energy-efficient, hands tied.

Insightful? Wow... the horror, the horror.

Re:Clean Design

[leandrod]

> There is no such thing as being internally RISC: RISC is about a style of Instruction Set (Reduced Instruction Set Cpu).

Yes, it exists. You translate external CISC into internal RISC, and thus are able to use all the tricks in the RISC book plus lots of chip real state, energy consumption and heat, less speed. Still better than pure CISC.

> if you look at all the other CPU except the 80x86, they are all RISC CPUs, or VLIW.

Wrong. Ever heard about ColdFire, AKA M68K?

> The fact that 80x86 is CISC is irrelevant

Not for me. I actually enjoy having a fanless, energy-efficient iBook that runs quite fast for its clock, and not helping to hold the whole CPU field back for backwards compatibility I don't need.

> 80x86 has won by being 1) cheap

Cheap to buy, but expensive to design, manufacture, and costly to the environment.

> being able to run the cheapest OS (Microsoft OSs)

So is now MS giving away its source code?

Re:Clean Design

[leandrod]

> the RISC crowd was primarily right, they were just targeting the wrong area

What do you mean? That they should've focused on the internals and left the CISC ISAs alone? This would've been prohibitively expensive. Only Intel's wages of monopoly and scale made it possible, and this benefited no one but Intel not even its customers, which were given the option of avoiding better systems.

ya

[DreadSpoon]

I've read it. What does the gigahertz have to do with any of it? Obviously _you_ need to do some more reading. ;-)

Re:Clean Design

[renoX]

>You translate external CISC into internal RISC

Usually this internal RISC form is more like and "expanded form" and it is used also by RISC: it isn't truly an instruction set as its intructions has no real length..

>> if you look at all the other CPU except the 80x86, they are all RISC CPUs, or VLIW.
>Wrong. Ever heard about ColdFire, AKA M68K?

OK, my point is: each "new CPU" is RISC or VLIW, there is no "new CISC": both x86 and Coldfire exists because of backward compatibility..

>> The fact that 80x86 is CISC is irrelevant
>Not for me. I actually enjoy having a fanless, energy-efficient iBook that runs quite fast for its clock

It would be interesting to compare the power consumption of Pentium-M vs PPC, it is not obvious at all that the PPC would win: Intel has one of the best process here.

>> being able to run the cheapest OS (Microsoft OSs)
>So is now MS giving away its source code?

No, but back then MS-DOS was cheap and the BSD OS didn't manage to get popular, I'd say because of
1) a lack of marketing --> lack of commercial applications (spreadsheet, games)
2) the need of more expensive CPU to run *BSD instead of MS-DOS which would run happily on a 80286..

Now it is a different games, but MS has an enormous installed base, we'll see..

Re:Clean Design

[leandrod]

> Usually this internal RISC form is more like and "expanded form" and it is used also by RISC: it isn't truly an instruction set as its intructions has no real length..

Your point being?...

> It would be interesting to compare the power consumption of Pentium-M vs PPC, it is not obvious at all that the PPC would win: Intel has one of the best process here.

Again, you are talking about something that has nothing to do with architecture, but with economies of scale both in production and design. If Intel didn't got an unfair advantage RISC would be even better than Intel is today. If Intel was RISC, it could do better. Ceteris paribus.