Slashdot Mirror
Pentium-Based Macs The Future of Apple?
seek3r writes "Found this interesting article on BusinessWeek.com regarding Apple's potential switch to Intel chips. I wonder what the implications this might have for Apple with regards to market share and software support. Have Motorola's chips really lagged behind Intel?"
when the G4 comes out... Apple has too much of a commitment to Motorola (since 1982?) and IBM (since 1992?)
Palladium/TCPA/DRM support?
I think it was the lack of competition in the Mac arena that left Motorolla high-and-dry when being compared to Intel now. I know you can't just measure MHz to MHz, but competition in any arena is better than none.
There are four boxes used in defense of liberty: soap, ballot, jury, ammo. Use in that order.
...Or am I missing something? the G4 chip has been around for a long time...
In terms of hardware site fanboy numbers, sure. But we're hitting the point where few people [*] can tell the difference between 1GHz and 2.8GHz and even hardware engineers are starting to realize this, so maybe it Just Doesn't Matter.
One thing I respect about the PowerPC chips is that the power consumption is drastically lower than for x86 chips. Drastically. It would be a shame to lose that and have everyone using 100 watt processors a couple of years down the road.
[*] Those few people are disproportionately loud.
There is a world of difference between using an x86 processor, and using an x86-based PC. I suspect Apple has closely investigated hooking up an x86-64 (or two ;) to a custom motherboard/infrastructure that would solve many of the interrupt/expansion complexities of PCs. For example, they could adopt HyperTransport, which would make multiprocessing affordable, easy to design around and most of all, leading edge, which is important to some people. Remember, Apple is expert at getting a lot out of a little - it would not surprise me if they tied a Hammer to a custom motherboard and created a whole new architecture. And that wouldn't be a bad thing.
I bought my first PowerPC-based Mac during that short, happy time when we could actually claim, without a hint of guilt or fear of reprisal, that G3 chips were "pentium crushers."
Unfortunately, despite my love for the mac platform, and my desire to claim that our hardware is "just as good"... it's not. RISC vs CISC stopped being an issue when Intel chips became RISC chips pretending to be x86's. PowerPC's still do more per clock than Pentiums, but the differences in clock speed, bus speed, and sundry other ephemerals has finally gotten to the point where for 90% of tasks, intel chips are just faster.
Don't get me wrong, I don't plan to switch until they pry my computer from my clenched, arthritic hands... but I can no longer look a computer-newbie in the eye and tell him that "Macs are just as fast". Better experiences, maybe... but as fast? No.
Of course, for most people, we're close to that point where chip-speed stops mattering... (maybe 1-2 more cycles of Moore's Law ought to do it.) How many people think about the speed of their computer while surfing, emailing, word-processing, or any such thing? (I know, I know, it's a cliche, but cliches are cliches because they're _true_.)
I think, business-wise, a switch to intel would be near-suicide for Apple. But Motorolla is dead in the water, desktop-computer-wise. Perhaps this theoretical IBM chip is the future... who knows?
The real reason is: Microsoft.
That's right folks. If OSX works on PC hardware, it has suddenly just become a competitor to Windows. What happens then? No more Mac IE, no more Mac Office. Suddenly Macs are nothing more than expensive linux boxes with a groovy desktop.
Apple can't "test" the waters by having some PPC boxes and some Intel boxes, they just have to jump head long into competition against essentially Dell for hardware and Microsoft for software. It'll never happen.
basically it'll mean we won't have to pay exaggerated prices for Macs to be able to use OS X!!
You will never see MacOS X running on a generic x86 "beige box". Apple developed MacOS X for the sole purpose of selling hardware, that's where they make all their money, despite charging for Jaguar. (Sun are the same with Solaris). In addition, the "just works" ability touted as a major Mac selling point would cease to happen once they could not guarantee with any certainty exactly what hardware their OS was running on - this is the real problem faced by Microsoft, most Windows crashes boil down to needing to have drivers for every conceivable piece of hardware supported, and being unable to prove them all.
An x86 based Mac will have sufficient custom hardware on its motherboard that you will still only be able to run MacOS on Apple hardware.
If Apple was to do Intel (read Px) based hardware, would they then go for a standard PC? Probably not as this means that their users can go to Win or *nix too easily. As they then would have to develop their own special little system, they would still have performance problems (fewer bucks spent on HW development) and expensive hardware (monopoly, or close to).
Since this rumour has been around for a long time without anything actually happening, I'd say that Apple will keep on building proper RISC based machines. We can all agree that it would be a step backwards to go from PPC to x86 from an architectual point of view, can't we.
It was pointed out that this guy was the same guy that, 5 years ago, predicted the merger of HP and Compaq for all the same reasons that they used today.
Personally I know very little about Mac's, but I can't see why moving to Intel would be a bad thing in any way.
I often found (in the old days, and were talking 8 years ago) that a Mac always appeared to run slower than the same speed PC and was substantially more expensive. I don't know if this is the same these days (having never used OSX - merely looked) but if it's true, anything that can reduce the cost and boost the speed must be good.
Avantslash - View Slashdot cleanly on your mobile phone.
For one, this has been rumored countless times before. Has it happened? No. Here's why.
One: Apple's revenue comes from it's hardware sales. If people can go out and buy plain vanilla PCs and install MacOS on them for significantly cheaper than they can buy a Mac, Apple's income will drop a great deal.
Two: As others have said, Apple's been with IBM and Mot. for a Very Long Time (tm). There have been rumors equally as valid as this one about apple developing it's *own* chip for fabbing at IBM (a company, unlike Mot., who can actually get decent yields).
Three: Again, as others have said, it's more probable that Apple will go with IBM's next-gen 64-bit desktop CPU. IBM makes good chips. They're not big in the desktop market, but the Power4 has been a big server chip for a while now, and with good reason. It was one of the first dual-chip-on-die procs that actually made public usage (afik), and did a large amount of ass-whipping.
To conclude: Apple going x86 would be stupid.
Have a nice day.
What? You want a sig?
The RISC vs CISC argument is all-but dead in the water by now. x86 chips are only CISC in the loosest sense of the word, for backwards-compatibility. They all run internal mini-RISC chips that convert CISC commands to RISC sequences via microcode.
This article chronicles some of Apple's challenges.
But on the topic. So Apple has 3 choices:
1. Wait for Motorola to get their act together. All the code optimization in the world won't make OS X as fast as it could be. Jaguar, for example, made my B&W G3 REALLY responsive compared to 10.1.5. But it occured to me, that's probably the last speed boost from software. You can only go so far.
2. Get the new IBM chip working. Hey, fine, it'll probably work. But it'll take a year or more to get it ported, documented, and in production. It won't be cheap, most likely. It will most likely be fast and powerful, but Apple walks a fine line WRT price.
3. Get Intel working. Hey, fine. Port OpenFirmware to an Intel-type mobo, then ship a computer that runs NONE of the software outside of the core OS. Wait for developers to buy one of these new machines to recompile their packages. This is where proprietary software bites you on the ass - you can't just wander between architectures with your source tarball and hope for the best. Oh, and of course, Classic won't work, and you're going to be stuck with whatever devices are already "cross platform". YOu can't just pick up a device from CompUSA and expect it to work.
The only plus I see to OSX/x86 is that the possibility for cheaper hardware might mean more people picking up an OS X box, and maybe some more drivers will be written. I'd buy one in a second, except... the majority of stuff in my Dock probably wouldn't be "ported" in the first year. So if it's under a grand, say, what good does it do me? No MacSQL, no EV Nova, no Remote Desktop... I need that stuff.
ZOMG I WOULD LOVE TO KNOW ABOUT YOUR FEELINGS ON MACINTOSH VERSUS WINDOWS, VI VERSUS EMACS, AND HOW YOU'RE NOT A DORK
Why is this so? Having never done dev work on an Apple I am pretty ignorant, but doesn't Apple release a basic API that doesn't change even when the underlying hardware changes (apparently not)? And why not release free tools into open source, so piles of developers are writing software for Apple for free?
I Am My Own Worst Enemy
Motorola lagging behind Intel is really simple market perception due to the now "standard" performance benchmark being a simple "GHz" tag. So most users (and non-technical press writers) simply assume that x86 chips are faster because they run at a higher clock rate.
As any knowledgable engineer knows this is not the case at all (as a matter a fact, in some benchmarks the PowerPC architecture beats the x86 architecture even when running at a much lower clock rate; just try photoshop on both platforms).
However, I also believe that market perception is a very important part of our society, and if you don't play the game you'll pretty much be left out unless you come with a revolutionaty technology that clearly makes a 10Ghz x86 chip feels like a snail compared to your clock-less chip. So in this regard, yes, Motorola is lagging behind x86 chips, and if I were Apple I'd be VERY worried about this. Just remember, Joe Somebody who just bought a 1.2 Ghz Mac will feel a little weird when his friend just bought a 2.5 Ghz PC, even when in real-world ussage both would perform about the same. Perception.
Oh wait, they would be taking it out.
I'm confused
The above post is an editorial, the poster cannot and will not be held responsible for all or in part for it's contents
But we're hitting the point where few people [*] can tell the difference between 1GHz and 2.8GHz and even hardware engineers are starting to realize this, so maybe it Just Doesn't Matter.
Definitely. PC manufacturers love to compete on Mhz, but a fast CPU is useless if it's starved of useful work by bottlenecks in I/O, memory bandwidth, etc. It's not unusual for a sub 1Ghz PC with good SCSI disks to handily outperform a 2Ghz+ machine with mere IDE.
Sun, SGI et al realized this years ago. Serious computing is limited not by clock speed of the CPU but by bus and memory bandwidth. That's why Sun sell systems with 300-400Mhz processors and gigaplane XB crossbar active backplanes. Nowadays with the increasing sophistication of consumer software (like the latest games), the same issues are recurring.
If you're buying a system in the near future, drop 500-1000 Mhz in CPU speed and buy faster disks or more memory with the money you saved.
Sadly, one thing that could 'break' Palladium would be the 'secret' x86 port of OS X. There are *many* people who would switch to OS X in a heartbeat (on their recent PCs). MS would at last be fighting an opponent with skill and product. Apple could put MS to bed.
I think that MS is really going for the total domination of hardware/software, and Apple is the only company that could stop it. Linux is great for many things, but Apple is *ready*.
Be careful what you wish for...
Apple makes its money on hardware, so no matter which processor is in the box, buying a Mac will be necessary to run OS X, and it will still cost big bucks.
-A
From the article:
No.
I don't want a laptop that blows hot air like hair drier or desktops that have three fans. As people realizing (as another poster mentioned) the CPUs are fast enough, I don't see much point in abandoning the PowerPCs that are small, consume little energy, and hence run so much cooler. For me, computers that are quiet and cool are much preferable to the opposite.
Another thing the author of the column seems to forget is that PowerPC is not a chip solely from Motorola. The point that IBM is also a partner and develops PowerPC chips is completely missing.
Yes, I hate it when you can tell that an x86's tappets need adjusting just by listening to it.
RISC chips are so much more turquoise, too.
Tim
I never liked CISC Prossors, I much rather have RISC chips running my systems. I find that RISC chip run smoother then CISC do
Run smoother? What the hell does that mean?
The whole RISC vs CISC argument is completely outdated. There is no such thing as pure CISC anymore. Please get with the times. You may love mac and that's fine but at least use up to date reasons for why mac is superior to pc. That's about as bad as a PC user dumping on macs "cause they only have one mouse button".
- Toby
Click here or here.
Such a move on Apple's part would complicate matters significantly. Consider that if hardware devices would STILL need mac specific drivers to meet whatever "hardware security" apple uses to make their machines proprietary--Meaning much hardware STILL won't function with OS X, whether it's on top of Intel or a PowerPC proc from Motorola or IBM.
My favorite uninformed reader was this guy:
This guy doesn't understand the term "switch." If he starts off running an Intel PC, and buys an "Intel mac" what has he really changed? Still using the same ancient hardware architecture kludged on top of a 32-bit chip sucking more juice that an a electric battleship.
Who did what now?
Reading the article gives me the feeling that the author is the sort of person who enjoys starting a flame war and sitting and watching the trolls move in. Much of what is said in the column is FUD. To quote two parts:
Here's the most compelling reason to abandon Motorola's PowerPC chip: It's falling further behind in the speed race as Intel's chips leave Motorola's in the dust.
Yes, if you are going per Mhz this is true, but once again Intel is a CISC chip with plenty of legacy components and the PowerPC is a RISC chip,
with plenty fewer transistors. Mhz is not an indication of work or performance. It is on the other hand a good indication of the heat that the chip will emit.
Several engineers familiar with the hardware work that goes on inside Apple wrote to say that, yes, it has quietly developed a Pentium microprocessor that could power a Mac.
It is a known fact that Apple has an internal project, known as Maklar, where MacOS X works on Intel chips. Apple is a hardware company and while plenty of R&D might be going on, only so much actually ends up as a product. It may end up being real, but any smart company has backup plans, even if they never see the light of day.
Add to all this that e-week, the same source that started this hornets nest, also mentioned that Apple is working with IBM to use the 64-bit PowerPC chip in future Macs. The truth is, Apple is likely to abandon Motorola, as Motorola is incapable of developing any chips that have a market other than embedded solutions. Motorola has really appears to be trying to get out of the desktop processor market.
These are my points of view - you are free to disagree.
Jumpstart the tartan drive.
Of course, we wonder how long it will be until some astute hacker makes this ability null and Mac OS X will be able to run on Beige Boxes. And if this happens, will it be a big problem? I mean, if Microsoft hauled off and proclaimed "you must now use Dell systems and if you don't you're not allowed to gripe about BSOD's anymore" people would have their head (again), but Apple wouldn't even have to say that - they could come out with an Intel OS and it would just be agreed/assumed that no one using a non-Apple box could go stuff themselves. Developers could have the best of both worlds - the Intel architecture they're used to and the closed nature consoles afford them (plus they can use this to make non-game applications, to boot).
Still, on the topic of similar hardware I'm shocked that it's been close to a year and we've had no XBox emulators for the PC. I mean, sure there's things to work around on the XBox (not the least of which is supposedly the fact that the data on XBox DVD's is backwards) but I figure if they can get Linux on the XBox, surely they could get XBox games to run on the PC. Perhaps the above scenario isn't so plausible after all.
Schnapple
without compromising Apple's control over its hardware.
Apple can simply continue to only allow certain hardware to work with its OS. Just because they move to a new processor doesn't mean they can't continue to do what they have always done. If Motorola and IBM can't help Apple keep up in the Mhz wars (Ghz, now), then why not contimplate a move to Intel or AMD? Use one or the other, and continue on. This could lower prices a bit, and keep the Apple moto of "It just works" intact.
They stuck me in an institution, said it was the only solution, to...protect me from the enemy, myself
The problem with Motorola's chips is that the front side bus (FSB) only runs at 167 Mhz. This means that Macs cannot truly take advantage of DDR RAM so long as they use the current line of chips, even though Intel machines have had this for two years now.
Back when the G4 was designed, things were looking bad for Apple, so Motorola retrenched into the embedded market. These processors need low power, not high bandwidth. That is why Apple laptops are so nice and Apple desktops are so lousy right now.
Furthermore, the focus on the embedded market is why Motorola does no deep instruction analysis (Again not needed in this market). Intel's investment in this area is what has helped their SPEC score over the years, not the clock speed.
There are rumors flying about a new IBM chip that fixes all of these problems, but that is all they are right now -- rumors.
This topic has been beaten to death here and on arstechnica.
:(
I personally can't see it happening for several reasons, the number one being software. Apple has commited to the intel/moto design, which includes a cpu library (altivec). Any 3rd party apps if not rewritten will need to be run in some horrible altivec->intel emulation kludge, which will be nightmareishly slow, and defeat the purpose.
Slower than the cartoon we know as XP? - probably not, but still slow.
The other thing is power consumption/heat dissipation - for mobile applications intel/amd just plain suck up too much juice and run too hot.
Apple is currently suffering because its chip suppliers have not been producing faster ships at the rate they should be, but until next month (chip conference) its all speculation as to what apples' long term plans are.
I've read this guy's writings before, and I find it annoying that his article got slashdotted. Now he is probably an even more highly regarded hack.
What interests me is that Apple hasn't said anything about this matter so far. These rumours must have their impact on Apple's sales; if I'd run a Mac-based shop and have plans to upgrade my systems, I would wait until I'm certain about the future; if they're really making the move I may postpone the upgrade. Apple must know this and must know about the rumours. Now there are 3 possibilities:
1. They're thinking about the possibility of making the move but don't know yet. In this case they will probably not say anything about this matter because it increases uncertainty.
2.They're not thinking about a move at all. They would most certainly let their customers know this to take away any uncertainty.
3.They're indeed planning to move. They don't want to make this known too soon since it will most certainly make buyers wait until the new systems are on the market.
So. We haven't heard anything from Apple yet so we're probably dealing with case 1. or case 3. here. :)
0x or or snor perron?!
The only real advantage of the PPC at the moment is that it lacks a lot of backwards compatibility cruft and, because of its RISC design, consumes less power and spreads less heat. It is a fine notebook CPU (and Apple is a fine notebook manufacturer). But Apple seems to have had no other chance but giving up this advantage by selling its newest line of desktop G4 Macs with dual CPUs, keeping up with Intel at least halfway with such a "hack".
gopher://cramer.plaintext.cc http://cramer.plaintext.cc:70
Apple will switch to a new chip when the PPC line no longer delivers what they need. First scenario: the chips don't keep up with the demands of future Mac operating systems, in terms of capability or speed. Second scenario: Apple's production rate outstrips the PPC production rate. I.e., they can't buy enough chips.
Neither of these conditions appear to apply now. That can change, though.
Mitigating against a chip switch in the near future: 1) Alienation of existing Apple customers. No one is happy when their hardware and sofftware is threatened with obsolesence. 2) Moving Apple developers to a new architecture. Apple isn't finished moving developers -- big and small -- to Cocoa. Abandoning Cocoa anytime in the next few years would risk loss of many independent developers.
Meanwhile, stop fantsizing about running an OS X on you $600 AMD boxes. Apple won't position itself as a direct competitor to Microsoft in the OS market. (And, Microsoft won't start selling a version of Windows that runs on the Mac.) Apple is in business to sell hardware. They write software to give people a reason to buy that hardware. For its part, Microsoft seems convinced that they can make money selling Office into the Mac market.
-- Slashdot: When Public Access TV Says "No"
It's not unusual for a sub 1Ghz PC with good SCSI disks to handily outperform a 2Ghz+ machine with mere IDE
If you said "a clustered array of RAID5 15,000 RPM drives versus a 5400RPM single drive", then that would have made sense, but to use SCSI versus IDE as the big differentiation is just silly: The intrinsic SCSI advantage has been disproven countless times.
Sun sell systems with 300-400Mhz processors and gigaplane XB crossbar active backplanes
That's pretty disingenuous: Sun sells systems with tens or hundreds of those "300-400Mhz" processors, disproving your "CPU power doesn't matter" BS. I guarantee you that if Sun weren't sliding behind in the CPU game (it's hard to compete with AMD and Intel with such a small niche market) they'd sell much more powerful CPUs. Instead they compensate by clustering dozens of them together.
If you're buying a system in the near future, drop 500-1000 Mhz in CPU speed and buy faster disks or more memory with the money you saved.
You'd save next to nothing. An Athlon 2200+ costs $220 Canadian here, and puts you in the upper realm of CPUs. Considering that most power PCs have 512MB of RAM (which is virtually never exhausted. Despite having several development tools open, and SQL Server running, and several different browsers, I currently have 370MB free. Adding more memory will merely increase the capacitive load of my PC). Secondly, adding a faster disk only matters if you do tasks which are heavily disk I/O intensive, which the overwhelming majority are not (especially because people have so much memory, and hence disk cache). It's like saying you'll get better video encoding performance by equipping your PC with a faster CD-ROM drive.
This BS "CPUs are faster than we'll ever need" nonsense is as tired of an argument as it was a decade ago when contrarians were assuring us that a 386 was more power than any reasonable man would ever need. History has shown their claims to be absurd, yet as they say: History repeats itself. Take a man who claims that his Pentium 667 is "faster than I'll ever need" and give him a P4 2.2 to use for a week. Put him back on his 667. 9 times out of 10 he'll be on the phone to Dell to upgrade his PC. Most people who claim that they don't need better say so because they've never SEEN better.
Additionally, try doing some video editing on your PC. While the hard drive is a factor (because massive amounts of data are read and written), the processor is massively more an influence: An Athlon 2200+ will perform the task that much quicker than a Athlon 1500+, again thoroughly reputing your claims that processors are overpowered. That's especially telling as video processing is one of the most disk and memory bound activities.
So what does this article bring us? Not much. The guy doesn't even seem to be aware of the even more frequent rumour that Apple will switch to IBM chips.
This story is actually less informative than previous articles on the subject.
Lasers Controlled Games!
Let's go over this one last time. First, Apple will never release OS X to run on a generic Intel PC. If they did, they'd sell about 100,000 copies to geeks who don't want to buy Apple hardware. When those geeks find out that there's no software for OS X/Intel, they'll gradually move back to dual booting Linux and WinXP, leaving OS X as an interesting oddity like the copy of BeOS they installed once too. I mean, you can only watch the genie effect or transparent Terminal windows on top of a screensaver running on the desktop so many times before it gets old.
Let's not even get into the nightmare that it is to support every piece of crap cheapo PC hardware combination like MS has to. Apple does not want that, period.
Why will there be no software? Look at how long it took (and is still taking in many cases) vendors to update their software for OS X. Now imagine Apple pissing them off by telling them to recompile and retest under OS X for Intel. Sure, that part probably won't be as big as moving from OS 9 (unless they've got a lot of endian or other hardware specific code), but recall how long it took vendors to switch to PowerPC native code. Ain't gonna happen. Let's imagine: OS X Intel comes out; Apple tries to convince developers to support it, but they (wisely) wait and see how it goes. Nobody buys it, and software vendors see that it's going nowhere, so they don't bother with it. No software == no point. Good luck!
Furthermore, what's the incentive to port to OS X Intel if (a) it's a relatively small, untested market, and (b) more importantly, they already have a Windows version that works fine? Along these lines, for Apple to provide any sort of VMware-like Windows emulation under OS X Intel would be suicide for the platform. Application vendors would just tell their customers to run it under Windows/VMware. What then is the incentive to develop a version for OS X Intel?
For Apple to move their own hardware to Intel would also piss off a lot of people. They pulled it off once with PowerPC, but that was truly necessary. It went amazingly smoothly, but it was really a couple of years before PowerPC native apps starting showing up in numbers and the newest PowerPCs were fast enough to emulate the old 68ks as fast as the last ones. Does anyone really want to go through all that again? It would be a couple of years before Apple would even hope to be up to par with Windows in performance! Not gonna happen.
Sure, I don't doubt that Marklar exists. It does give them that last desperation option, when there's no hope for anything else. But perhaps more importantly, it serves to improve the OS X codebase simply by making it platform transparent. The one instance where I could possibly see an Intel-based product from Apple would be XServe. Just a thought - but if you're not likely to be running PhotoShop or ProTools or Quark on a server, perhaps an Apple branded unit with Intel would work out with all Apple server software.
The only intelligent thing Haddad says is in the second to last paragraph, where he essentially acknowledges that software would be the biggest roadblock. Developers will likely balk at the prospect of porting to yet another platform, and "without software support, the Mac would truly be dead." Exactly.
Of course, the most likely scenario lies with the rumors of the Apple/IBM collaboration on a next generation PowerPC chip. That's where I'd put my money. Nobody knows if/when G5 will ever come out and Motorola doesn't seem to care about the non-embedded market. Hopefully IBM can bring Apple back to the days when PowerPC really did crush the Pentium. We'll see.
Say hello to zMac.
Some users, however, would welcome a PC version of OS X. That would enable Windows emulation software, such as VirtualPC by Connectix, to run much faster. "The ability to switch back and forth easily between OS X and Windows would be a major coup," says Sasaki. Ian Crooks, operations engineer at Pennsylvania-American Water Co., declares: "I for one would switch tomorrow if they would release a [Pentium] machine."
This is exactly why Apple should never port OS X to an Intel architecture.
Virtual PC would run much faster if it didn't have to emulate the microprocessor, true. So much faster that it would discourage companies from coding for OS X itself, because you could run their Windows products on VPC.
Not only that, but eventually somebody -- not Apple, certainly -- would release a project similar to WINE that would allow Windows programs to co-exist with OS X programs. It won't be completely compatible, of course -- especially as Microsoft changes the APIs -- but it would give companies another excuse not to develop for OS X.
A third factor is the cost of porting existing Macintosh OS X software to this new architecture. Facing that cost, why not port to Windows and let the Mac run your program through these emulation options?
As time goes by, Macintosh users would have to depend more and more on Windows software. Sure, they'd prefer software designed specifically for their platform, but developers won't be selling it, because it's easier and cheaper to code for Windows. Eventually, the users would just switch to Windows because Windows programs will run better on Windows computers.
What I find more intersting is why is the slashdot crowd so obsessed with this idea? If apple did the switch to intel processors do people on slasdot really belive that:
I find it quite ironic that the crowd that touts open source software seems to have an ongoing dreams of running proprietary software. The funniest thing is, most elements for building an open source variant of OS X are around:
- The code of darwin, the kernel, is available and compiles on IA32, albeit in a limited fashion.
- The Cocoa libraries have an open source clone: Gnsustep.
- Re-implementing Carbon according to the spec, which is publicly available and quite clear would clearly be less work than project WINE.
So what gives?I work in a lab where we produce a very widely used piece of scientific software, and we do benchmarking on everything from old 68k Macs to new Dual G4s to AMD and Intel boxes running both Windows and Linux. The fastest benchmark we have on record, despite the fact that we dropped over five grand on our dual G4, was an $1100 dual Athlon XP 1800+ using Intel's C compiler version 6. It's not just faster than the fastest Mac benchmarks, it's WAY faster. We haven't tried any higher dual Athlon systems, but I suspect they'd be faster still. I'm not saying that an Athlon system would be faster than a Mac in all circumstances (I don't know one way or another), but the benchmark I've got the most experience with has got the Macs losing in a landslide.
That being said, I think OSX beats the crap out of Windows as an OS, and I'd really love to see such a great OS on a cheap, fast box. Can't have everything, I suppose.
I wonder what the implications this might have for Apple with regards to market share and software support.
The implications are that Microsoft will destroy them. Duh. This will never happen unless Steve Jobs is blinded with his own ego and his shareholders let him.
After spending so much time and effort bashing the Megahurtz Myth, there's no way they'd go with Intel P4 chips and their performence killing 20 stage pipeline.
OTOH, they might go x86-64 on the AMD Hammer series. Gobs of memory bandwidth, excellent FPU, high clockspeed and VERY high performence. Plus, by targeting x86-64 as their starting point, they get both optimized performence AND by definition don't run on 32-bit chips, so there's less whining from users about not running on their 32-bit generic PCs. They can go 8-way multiprocessor economically with the Opteron series too.
*sigh* I guess it works. You present a few facts, then use them as the launching point for unreasonable claims.
PPC != Apple. You start by attacking the XServe, which may be deserved, and expand the attack to the rest of the PPC family. It doesn't wash.
G4 does not compete with Xeon. POWER4 (itself a wholely compliant PPC chip) does, and you know what it Smokes Xeon as a server chip. Xeon scales to what 8 way, with a contorted memory bus structure? POWER4 scales to at least 24 way, probably higher if IBM cared to offer something bigger and integrates onto a modern server crossbar switch.
If Xeon is so good, why aren't companies converting their Sun/Oracle installations to it rather than RS/6000 POWER4 machines?
Please spare us the classic bait and switch strategy of arguments.
What if it is just turtles all the way down?
yeah they might like the fact x86-64 is all shiny and new (mac people are attracted to this and mr jobs loves it )
BUT
1/ you would have to get adobe to port photoshop all over again
(photoshop is a carbon app and has lots of PowerPC asm still in the mac version)
2/you would have to have an emulator not only for PowerPC but all the OS interfaces much like running VMware with the whole OS
(although VMware approach is of emulating the whole machine you could shortcut it as you only have limited amount to emulate)
3/ the back catalog of hardware that you have like the apple system controller + gigabit NIC ASIC would have to have serious work not just a tweak
so whats really going to happen then smarty pants ?
apple tweaks the system controller for either RapidIO or IBM interface depending on supplier
(you get the real thing which matters in computing BANDWIDTH )
they have a seperate level 3 cache that apple can mess around with to get extra performance and so sell differant machines at differant price points
apple use's MOT chips for laptops and IBM chips for servers
regards
John Jones
Does your typical mac user even care if his/her CPU is 13.4% slower than it's intel counterpart? I am willing to bet that the reasons why they chose Apple in the first place will way outnumber a slightly slower CPU. (Plus with all the licensing and proprietary closed source layers -- I doubt you will ever see Macs make an inroads in the server market no matter the CPU speed...)
(+1 Funny) only if I laugh out loud.
Where did you get the data to plot this?
It took an entire day of web searching, chasing down news releases, Mac and PC time-lines, and other tid bits of information to determine what was the fastest system actually released (and not just announced) in that year.
Can you give us a link to that data and the methodology used for the comparison?
I don't publish scientific articles about research undertaken for my personal enlightenment. I did not bookmark the sources, but I did keep the collected data:
I tried to post the data but the lameness filter disallows it!
Maybe then we could all elevate this debate above "is not/is too."
Do you have any reason to doubt it, or you just like being difficult?
While it is important not to take statements at face value, it is also important to run a quick mental check and see if they are ok. This partial suspension of disbelief is what allows a young scientist to reach the frontier in a short period of time. Think about it, if you had to doublecheck every experiment in history you would never get anywhere.
A pair of Athlons (or whatever they are called these days) using their hypertransport, to give huge memory bandwidth, and put firewire, serial ata, and all the ports on board.
This meshes with the graphics and video themes that Apple is popular with, and gives them a path to 64bits via sledgehammer. And it would make a fair server, if it fit in a 1u case.
I keep seeing replies to this story that go along the lines of "Apple is a hardware company and will thus design a proprietary system around an Intel processor so that it won't run on clones".
While I would not be surprised if this were the case, the fact that the CPU would be the same would eliminate a huge roadblock in the way of emulation. It should be possible, without too much effort, to produce an 80%+ speed emulator for a beige box that would run the OSX made for the Apple-on-Intel box.
I'd love to see Apple port to x86, because then it would only be a matter of time before I could run OSX on my commodity hardware without paying the Apple Hardware Tax.
This highly depends on the application. A single SCSI drive against a single IDE drive performing a single task may show the same performance. However, when you add multiple tasks and a lot of disk access , SCSI beats IDE hands down. As you add drives (don't even bring up RAID yet), tag command queing and parallel data paths blows away IDE no question. Now, add RAID into the equation, especially looking a the huge caching controllers available for SCSI with no IDE counterpart and you see that SCSI is certainly the way to go. Computer manufactures aren't idiots; IDE is cheaper and if it were on equal footing with SCSI no one would offer SCSI solutions. That having been said, no high-performance workstations or servers use IDE.
Sun, HP, etc., have for years sold small MHz machines that outperform the GHz machines available mainly because they use RISC technology and aligned instructions. Clustering has not been a large part of Sun's business -- ever! And, as far as multiple CPU's in a single box, yes, all these systems offer and endorse this, but then so does Intel if you read their journals. Intel ran themselves into needing GHz clocking because of poor chip design (backward compatible to x86). Sun and others don't design chips in those ranges because of the cooling requirements and heat failure rates. It is far easier for them to make lower MHz machines with multiple processors because they run OS's and software that can work UMP or SMP, where Intel has issues in the common market environment (example: Windows 95/98 unable to work SMP).
Click here or here.
I also work on single processor Sun, SGI, and IBMs, all of which at lower Mhz are MUCH faster than my PC (except maybe the slower SGIs, like the Indigo R10000s; at 150Mhz, they're showing their age but STILL keep up with the PC in rendering speed). Sun's problem is not technology, it's sales. IBM is just killing them in marketing. I talked to a guy the other day that's getting ready to begin replacing their 1800 Sun servers with AIX boxes. He concedes the Suns are superior, but they have been convinced from the confidence bestowed by IBM's superior marketing skills. It's widely known that Sun has superior tech, inferior business sense.
I totally agree with you that it's BS the people that say 'current CPU speed is all we'll ever need', but it's equally BS to assume that the 'faster' Intel chips are actually the 'fastest' chips out there because of some marketing-driven clockrates. Superior architecture trumps clockrates any day of the week, and Intel is still lacking in the former. Incidentally, I'd take a single processor Ultra Sparc III box at 1.05 Ghz over a 2.0Ghz PC, even running *nix, any day of the week. As a matter of fact, I usually do.
You're right. They'll make oodles of money, just like Be and NextStep did when they gave up on making their own hardware and instead tried selling an alternative x86 operating system.
It's not unusual for a sub 1Ghz PC with good SCSI disks to handily outperform a 2Ghz+ machine with mere IDE
Doing what? And what kind of disk subsystem? Put in a single IDE drive and a single SCSI drive, both running at the same spindle speed, and I doubt there will be a measurable difference. Modern IDE is not as godawful bad as it used to be. And yes, I used to be a SCSI advocate, and I definitely agree that SCSI has it's place. But it's not on the desktop.
Nowadays with the increasing sophistication of consumer software (like the latest games), the same issues are recurring
Which is why there's absolutely no increase in framerate or other performance benchmarks as you increase CPU speed, right? Oh wait, there is...
If you're buying a system in the near future, drop 500-1000 Mhz in CPU speed and buy faster disks or more memory with the money you saved
Yes, that $30 will go far.
Looking at Newegg, the cheapest AMD Athlon processor I can buy is a 950 MHz Athlon. For another $13 I can buy an Athlon XP 1600 (1400 MHz). The most expensive Athlon XP available right now is a 2200 (1800 MHz) for $155. And an Athlon XP 2000 (1667 MHz) is $100. The reality is that CPU prices aren't as high as they used to be. In fact, you're likely to spend more on memory than you do on the CPU. Heck, the motherboards can be more expensive than the CPUs now.
And the MHz does still matter. Virtually everything still winds up being CPU bound - pop in a faster CPU, everything gets faster. The same can't always be said for memory (512MB is sufficient for most purposes currently), and improving disk performance is freaking expensive (compare prices for 15k SCSI drives to 7.2k IDE drives. Don't forget to factor in the necessary SCSI card and equivalent storage space).
If you're a gamer then the best place to put money is the video card... they still scale based on CPU speed, but the difference between a $50 video card and a $150 one is far greater than a $50 CPU and $150 CPU.
Are there bottlenecks still? Sure. But despite the horrible, evil numbers that float around university EE/CompE courses it's not really that bad. If it were then we'd still be stuck back at a couple hundred MHz trying to change laws of physics to get the HD, memory, and network subsystems up to CPU speeds.
Oh, if you really want to think about just how wide the disparity between CPU speed and other systems are -- that 3 GHz P4 is actually running it's ALUs at 6 GHz. And yet it still manages to get enough data to show a marked improvement over a 2.4 GHz P4.
As any knowledgable engineer knows this is not the case at all (as a matter a fact, in some benchmarks the PowerPC architecture beats the x86 architecture even when running at a much lower clock rate; just try photoshop on both platforms).
Motorola lagging is not perception it is real. The "market perception" is overblowing it, not inventing it. Apple being forced to go dual G4 to remain competitive is real, the gap could not be "spinned away" by marketing as in early iMac days.
When PowerPC/Mac and x86/PC systems of the same clockrate are compared against one another the PowerPC/Mac advantage is usually around 20-30%. You can find specific tests where one platform grossly outperforms the other, and if you look harder you can find such tests that aren't rigged (486-optimized code, Altivec vs. MMX not SSE, etc.), but these are the exception not the rule. These rare exceptions tend to be very CPU centric and don't stress the system architecture. Looking at the CPU in isolation is of limited value. A system wide looks shows some serious shortcomings on the Mac side, namely memory. Example: DDR not really utilized by the CPU. And this is a Motorola failing not Apple's.
Apple has never really even hinted that they were considering going to X86, other than by the internal port of OS/X to X86 (which probably took all of 2 man-months). The idea that they should or will make this move is largely an invention of morons who don't understand that Apple is a hardware company.
The big potential losers if Apple should switch chips would be software developers. They would be forced -- perhaps for the second time in two years -- to rewrite their programs, this time to make them work with a Pentium-based Mac. That's no small task -- and could be a disaster for the Mac community, since many of its developers are small shops. And without software support, the Mac would truly be dead.
Oh, yeah. That's why.
Imagine running an x86 Mac that has no native version of Office or Photoshop and runs PPC-based versions like molasses, but runs Windows versions at native speed. Imagine trying to convince developers to write for OS X instead of Windows at that point. Why should they bother?
--- Work, worry, consume, die. It's a wonderful life. -- Bill Griffith
1) Pentium chips run hot and comsume so much power the fan on the thing will be huge and whirrrrring at 7000 rpm minimum.
2) With Intel(TM) DRM(TM) you won't be listening to your music, so you'll hear that fan loud and clear.
I'd love to see the numbers that show that the "the intrinsic SCSI advantage has been disproven", even once.
Sure, in raw throughput and seek times, IDE drives can perform on a par with SCSI. Certainly, the bus speed is up in nearly the same realm now... but you forget the part that's important to USEFUL speed. A SCSI subsystem will generally operate at about 10% CPU utilization during saturated disk I/O, the same load on an IDE subsystem will suck up closer to 80% of your CPU cycles.
I don't know about you, but I'd rather have my CPU doing something more useful than copying bits back and forth between a bus and a controller all day.
IDE is the best way to go if you're building a network file server -- it's MUCH cheaper, just as fast, and the file server by definition has nothing better to do than sit and do disk I/O (and network I/O), but I'd rather have a SCSI disk in my workstation any day.
As far as the CPU arms race... of course faster is better. But it's a question of what you're doing. Do I need a 1GHz+ cpu for moving windows and icons around on my desktop? No (well, I *shouldn't*, if Bill f***ing Gates didn't write such crappy code!).
The things that need fast CPU's are the same things that have always needed fast CPU's. Scientific calculation, Graphics manipulation, and of course Games. I suspect the last one is the only REAL reason most people "NEED" faster CPU's. Of course, I also think game designers lost their way back in the early 90's. I certainly think many of the games from the 1MHz C64 were more original and indeed more entertaining than much of what's on the shelf now. There are exceptions, but most of the "innovations" in gameplay today involve glitter without substance (see Quake 3 vs. Quake 1 -- much better graphics, but gameplay? atomsphere? storyline?).
Show me a game today that has the same level of immersion as Zork, and I'd happily go buy it and whatever hardware it needed to run. UT2003 looks great, and I'm sure will be fun... but it and the 2GHz CPU and $300 graphics card it wants won't engulf me the way a text game I can play on my PDA can.
No matter how fancy the hardware and how clever the graphics rendering, it'll never be as fast or as natural as your own imagination.
The OS that is running on the hardware has more than a little bit to do with comparitive speed as well.. This is blatantly obvious when using rendering programs such as rhino3d.. My home computer running win2k can render things exponentally faster than the lab computers here which are 3x times as fast but run 98.
I am ashamed I didn't think of putting that in the first place. I must be becoming a PHB. :-)
Tim
Folks, it ain't gonna happen. Apple was only able to get through the 68K->PPC change because it was able to write a good emulator, and a decent PPC on X86 emulator is still yet to appear, and I think unlikely ever to appear. And they're just now getting through the Classic->Unix change. Apple is much more likely to go with an IBM chip now that they seem to have successfully evangelized Altivec to IBM.
--
"Open source is good." - Steve Jobs
"Open source is evil." - Microsoft
Yeah - look at the way Jaguar and Porsche suffer from being confined to a tiny part of the overall car market.
That analogy is flawed because Jaguars and Porsches are a lot faster than the average car.
Mac people are like the guys who buy jags, mgs and so on. Sure it will always be a small part of the market but that doesn't mean Apple can't make money doing it.
To carry the analogy further, Macs are like niche cars that can't use the same fuels, oils, or tires as "normal" cars. That's the problem that Apple has: In order to be successful, they have to convince software publishers to create Mac titles. Those companies have to be convinced that it's a financially sound decision to hire Mac software engineers, Mac support staff, and to buy Macs to be used for development, testing, and support.
Apple is always on the hairy edge. If there were fewer Mac titles, they'd lose market share. Then there would be fewer Macs and the incentive to develop Mac titles would be less -- which would mean even fewer titles. I think you see where this is going.
I wish Apple well, but the only way that I think they have a chance in the long run is to bit the bullet, change CPU families, and create Macs that perform as well as PCs at similar price points.
If they try to become a software house like Microsoft by selling OS-X for generic x86 PCs, they will probably be destroyed by Microsoft. If Microsoft actually viewed Apple as a competitor (rather than a faux competitor that keeps the FTC off of their backs), life would get ugly at Apple. Microsoft would likely not produce a version of Office for OS-x86 (clever name, eh?). Microsoft would discourage Windows developers from creating titles for OS-x86. Microsoft could withold support or even actively sabotage titles with "service packs" to punish software publishers who released OS-x86 titles.
Just my $.02 on the subject.
I'm a software guy, so admittedly I only partly understand the issues involved in any hardware port. However, since, as I understand it, the Athlon was mimicing x86 instructions and breaking them down into a more Risc-like set, couldn't it do something similar with the PPC instruction set?
:) ?
I'm sure 64 bit brings its own set of problems, and my money's on the IBM Power4 mutant, but in theory at least, couldn't something from AMD be *tweaked* to emulate PPC (or would that kill any performance advantage and make for a really warm Crusoe
just my blog and pix
The main reason I hope Apple don't switch to x86 is because PPC is different (different is good) and frankly, x86 is a fucked up architecture - it's a 20+ year old architecture that's been kludged over and over.
In almost every discussion that involves PowerPC vs. x86 sooner or later this sentiment comes up. Ok, it may be right or not - but HOW DO YOU TELL ? Are you programming assembler on any of these platforms ? Or do you tell by looking at the chip ? */me opens up his G4 tower, pries away the cooler.. hm... does the same to his Duron...* looks quite the same... a rectangular piece of something looking technical on some kind of socket...
Are you really knowing about a relevant difference or are you just babbling some marketing fluff you heard somewhere without having a clue ?
I mean - Unix (used as a basis for OS X) is a 20+ year old architecture that's been kludged over and over - does that mean that OS X sucks ? What about Linux ? Different is good - so why not run Plan 9 - that's as different as it can get, I think. But both are not really valid arguments for or against any platform IMO.
I don't know about you, but I'd rather have my CPU doing something more useful than copying bits back and forth between a bus and a controller all day.
IDE has had busmastering for about 7 years, removing the CPU from most disk control operations. Furthermore most modern IDE controllers also offer command queueing (i.e. Promise). Of course, it's of marginal use as controlling a hard drive takes such a micro-iota of actual processor time nowadays.
A SCSI subsystem will generally operate at about 10% CPU utilization during saturated disk I/O, the same load on an IDE subsystem will suck up closer to 80% of your CPU cycles
When I'm copying at full bore between partitions, or over a high speed network, the drive is completely saturated and the CPU usage is about 2%. During most copy operations, the most CPU intensive aspect of it is Windows drawing the "Now Copying..." dialog.
I can't agree more. I've been overclocking my PC since I got my 486-33 10 years ago. Anything I could do to squeeze more horsepower out of it to play the newest games.
- I upgraded memory
- I learned that 256k cache is worth it, and 512k didn't help as much..
- I learned how to use mem
/d to put TSR's in certain places in memory so I had plenty of conventional free. (7th guest in 4MB - on a SCSI CD - baby!)
- I did the basic 'up the bus speed'
- Ran my VLB video at 50mhz
:)
- Optimized bios Ram speeds..
I kept all that up until a couple years ago..Overclocking and optimizing has mostly'deteriorated' into CPU upgrades, revving up the MHZ on the MB. I now find I don't need that as much, because the blockage is in the Internet connection, not my CPU. The games run fine, but lag is (always) intermittant. Playing Descent on a 486 over a 28.8 with Kali WASN'T just the line speed, but also your PC speed. I have a PIII 800 now, and games run just fine. Now it's the line speed.I think this finally set in with me a couple weeks ago when I broke down and bought my first notebook (and new PC since the 486 :). I was concentrating on screen size, memory, and HD. After I purchased the Notebook, I was surprised to discover it was a 1500 Celeron. I only checked because I couldn't remember if I got an AMD or Intel, and I was thinking it was around 1200.
So many years of tweaking, and now I can't remember what CPU BRAND I even bought.
On a side note, While Mandrake 9.0 RC2 installed PERFECTLY (not sure about Winmodem, don't need it) on this Toshiba 1405S171 (from Circuit City), I have one annoying issue. If I type 'rick' fast - at my normal typing speed. It comes out rriicckk. The same thing happens on my home system mdr8.0-8.2. I thought I just had a shitty Mb ;)
I can't think of a good search term for that kind of an issue, so I havn't found a solution, any ideas?
"I can't give you a brain, so I'll give you a diploma" - The Great Oz (blatently stolen sig)
Actually a VMWare-type solution might be a better idea assuming that Windows could run on an Intel-based Mac box.
Basically he thinks that he could get the best of both worlds on a single box. I am not so sure that he is right.
Lasers Controlled Games!
Go back to law school. Or alternatively, read the DMCA.
autopr0n is like, down and stuff.
I'm not sure why every single /. contributer assumes that a company can only sell a single family of CPU, but has *anyone* considered for a moment that Apple may be ready to put both x86 and PPC in the same lineup?
Consider:
Apple nuked OS 9 booting effective 1/1/03.
Mac OS X is clearly being supported internally for x86.
Apple is concerned about Mot/IBM pulling through with decent CPUs.
My guess is that Apple is going to put both PPC and x86 in boxes next year. The x86 boxes will run Windows, but OS X won't run on a Dell or HP box, etc.
Apple can sell the superior performance (let's all hope...) 64 bit chip that IBM is working on to the pros, and sell either G4 or x86 to the masses. If you really are concerned about Windows compatability, there's an easy out for you - you don't need to decide between Mac OS and Windows - you can have both.
There's some risk in a strategy like this, but by and large it could be a big win for Apple. The 'switch' campaign gives consumers a way to ease into Apple products - buy a 17" iMac x86, run XP on it (maybe like VPC?). If you want to use iTunes, you can do that, and eventually you may upgrade your Windows software to Mac OS X versions.
Of course, if the PPC performance isn't there, nobody will want those boxes and developers might as well chuck OS X development altogether. That's the biggest risk.
for the rest of us to move to Itanium.
Amazing magic tricks
Never mind that that useless pipeline easily outperforms the current best offerings from AMD and Motorola (though Intel and AMD are playing leapfrog, Intel's on top at this moment.) Do you even know what processor pipelines are for? Do you know that Apple's past comments comparing pipeline depths of powerpc processors to the pentium 4 was complete and utter FUD? Have you even looked at fair and reasonable benchmarks?
The plain truth is that powerpc processors and Macs have been lagging behind in performance for a long time. Top of the line G4s use 1.25Ghz processors. Even if they were twice as wide superscalar (I don't believe they are) AND the majority of programs could take advantage of all the extra execution units most of the time (which is not often the case on any superscalar CPU), they would still not match the performance of a top-of-the-line P4. Not to mention the fact that the Apple hardware would STILL be much more expensive.
How long has Apple been demonstrating performace superiority by relying on artificial benchmarks that consist of a select group of Photoshop filters? Preciesely as long as they've been lagging behind in performance. They've even given up on the performance edge lie completely now (though plenty of Mac cultists think comparisons made five or ten years ago are still relavent.)
Unfortunately, Apple's current marketing campaign sucks. Instead of showing some snob talking in vague ambiguous terms about how OSX is so much better than Windows, actually SHOW OFF THE OS. Demonstrate how easily you're able to open you're co-worker's MS Office documents (the Mac version of Office is much better than the XP version IMO). Then start minimizing and maximizing crap. After they cream their shorts, lots of PC users will be lining up to pay for overpriced Apple hardware.
This post is not a dig against Hammer. OSX running on Hammer would be pretty damn sweet. If I could run OSX on commodity PC hardware, I'd do it in a hot minute (or at least dual boot to it). In fact, there's nothing stopping Apple from dressing up PC hardware nice and pretty and running OSX on it. Unfortunately, they'd almost certainly make it proprietary hardware using an x86 processor (and probably still nVidia graphics hardware, which would be nice). Anything else would probably be suicide, even if they decided to just be an OS company.
My dream would be if Apple made OSX more conformant to unix standards (i.e. the unix standard filesystem layout). Imagine running the Aqua gui on your *nix of choice. I'd drop X11 like the dirty slut that it is.
... "Give me a woman who loves beer and I will conquer the w
Why do we assume that if Apple changed chip manufacturers, they would also change platforms and architecture? It seems to me a much more likely senario that if Apple were to change processor vendors, they would either
A) develop a new architecture
or
B) continue development on the PPC architecture, just with a new company.
After all, IBM makes x86 chips, but they're developing PPC chips for Appple too. It seems to me that if Apple could provide them with the correct tools to do the job, AMD or another manufacturer would be happy to take on the extra revenue that the PPC chips could bring in. Assuming they can justify the R&D costs.
On a side note, if Apple does switch, it seems highly unlikely that they would switch to Intel. Maybe IBM, maybe AMD, but they've spent too much time bashing Intel that to switch over to them would be a worse PR move than the M$-Apple alliance.
T Money
World Domination with a plastic spoon since 1984
It does when you count the price of that spiffy LCD monitor. It's a bit unfair to compare the price of a Mac with a Cinema LCD monitor to the price of a headless Athlon box using a monitor that was lying around the office. Macs have supported standard VGA monitors for at least ten years now, and there's no reason why you have to buy a new monitor with a new Mac.
--
"Open source is good." - Steve Jobs
"Open source is evil." - Microsoft
(Personally, I don't think Palladium will ever see the light of day. It looks pretty clear that AMD won't sign on for it. That means Intel won't do so either. Otherwise, they're dumping quite a few future sales in AMD's lap. Intel is known for holding out until the last minute when negotiating big changes to their products.... I think you'll continue to see them talking about "considering Palladium" up until the last minute, but it won't really be in there.)
Therefore, I believe that assuming Palladium isn't in new Intel CPUs, Apple has no reason not to use their processors in a new line of Macs.
I don't think Apple would want to try to silently switch over an existing product line to Intel chips though. Instead, they should design a whole new Mac (with another cool new case design and all), and trumpet it as the "Mac with Intel inside".
As long as they can build enough supporting hardware on their motherboard to prevent PC users from "lifting" the Intel version of OSX - they should be just fine. I don't see why any Mac user would care which chip powers their Mac, unless they're being an unreasonable zealot about the whole thing? In my scenario, everything else stays the same. You still need to buy a Mac if you want to run OSX. OSX still runs all the same software, only faster with the latest Intel CPU behind it.
I suggested Apple needs to do this yesterday in a thread and was called a troll and stupid....
Put OSX on a pentium and watch XP die a quick death. Even if it costs apple the office suite, given a year that will be all M$ has to offer and they will be porting it for anyone willing to pay.
errr....umm...*whooosh* *whoosh* Is this thing on ?
Common sense tells me the x86 PCs are faster than Macs just because so much more time and money goes into R&D for Intel and AMD's CPUs.
Benchmarks where the playing field is fairly level support that conclusion.
For the Mac cheerleaders: Photoshop filters heavily optimized for the G4 do not constitute a level playing field, but even then the G4 barely keeps up with current x86 tech.
By level playing field I mean, for example, all the machines running Linux and compiling emacs with a Spark target. In those kind of benchmarks you quickly find that a 500MHz CPU is a 500MHz CPU, regardless of manufacturer.
Under capitalism man exploits man. Under communism it's the other way around.
It might be a selling point for more people than just you, if people were made aware of it. Show a side-by-side comparison of how many dollars' worth of electricity will be consumed over the next five years by an iMac and by a typical Wintel system.
Show another comparison where the savings are even more dramatic if the Wintel system is connected to a CRT (as opposed to the iMac's LCD).
Show a third comparison where 30 such computers are used in an office in Phoenix (where the air conditioning is always running), and the thermal output of the Wintel machines drastically increases the operating costs of the HVAC system.
That that is is that that that that is not is not.
Yes, IDE has had busmastering for a long time. I think 7 years is even pessimistic, it's been bus mastered for a very long time. However, with IDE the bus mastering seems to just interrupt the CPU less for disk transfers, not totally absolve it from those duties. This is why SCSI has historically had a 2-3% CPU utilitization with the bus maxed.
While you make a good argument for purchasing an aftermarket IDE controller (which can perform tasks with the CPU utilization of SCSI), the reality of the matter is that virtually zero OEMs ship a system that way, they use whatever is built in on the motherboard. Which almost always consume a large amount of CPU time when performing disk I/O.
This is why the only people who build enterprise-class database servers with IDE drives at their core are idiots. That or they're penny-wise and dollar-stupid.
Moof!
RISC vs CISC stopped being an issue when Intel chips became RISC chips pretending to be x86's
If Apple moved the Mac platform to Intel processors, then they will need to have a PowerPC emulation layer so that all the code written for existing Mac hardware will continue to work (mostly, hopefully).
Does this mean that Intel chips would be RISC chips pretending to be CISC pretending to be RISC?
I'm a Lightwave user. Lightwave works on both Mac and PC (Windows). I'd love to go to Mac, but all of LW's coolest plugins are compiled for the Intel architecture.
If I could run OSX on my Intel/AMD, that means I'd finally have some choice between Windows and Apple, since the playing field would become level.
I am a bit naieve, though: If OSX were to suddenly run on an Intel processor, would that mean that Intel compiled plugins would instantly work? Would Newtek have to massage the code to make them work?
Yeah - I think you just made a very valid point that many people forget.
Premium services are typically only purchased by the upper-class people that can ride out economic downturns, relatively unscathed. (Oh sure, they might cry in their beer about their stock values plummeting - but if they're pretty "well to do" to begin with, they likely didn't invest more in stocks than they could "afford to lose".)
The people literally living "paycheck to paycheck" aren't shopping for luxury goods. It's a "luxury" when they can afford a new (or someone's second-hand) computer in the first place.
I think where confusion comes in with Apple is their rather quiet refocusing of their target market over the years. After all, they got their start pushing their computers into schools and trying to tell everyone they were the only option worth buying, at any price, because they were easier to use. Nowdays, they've really gone to much more of a high-end, artsy customer-base. Sure, educators still buy them - but their students often don't. I know a number of PC users who never owned a Mac, but they "lust after" a Titanium Powerbook, or a G4 with the cinema display. Honestly, most of them will never actually buy one - because they don't have the cash for it, and in the "real world", it just makes better economic sense to buy upgrades for their existing P4 system, one piece at a time. But the fact they perceive a Mac as "cool to own", despite owning a high-end PC already, is rather telling.
I've chimed in on this myself with the following story on Forbes.com:
Will Apple Put Intel Inside?
August 9, 2002
Rumors are buzzing that Apple computers may one day be stamped "Intel Inside." It won't happen.
http://www.forbes.com/2002/08/09/0809apple.html
" (see Quake 3 vs. Quake 1 -- much better graphics, but gameplay? atomsphere? storyline?).
"
Quake1 had a story line, but quake 3 was never about a story, it was about getting online and blasting your friends. It does that very well.
Now, I won't get into what you should like, that is just stupid, and I wouldn't presume to know what you will enjoy. I would wager that Zork was one of your first games, and so it will always be great in your (and mine!) mind.
The Kruger Dunning explains most post on
Also, while the guy is right that the transition would be a big pain for the developers, in the long run it might make things easier for them, because most of them keep a seperate branch of x86-optimized code because they also sell it for Windows. Post-transition, these two branches would be able to have much more in common. That might make things easier in the long run.
Alright--here is a reason for not making the transition: the upcoming desktop Power4's from IBM. I am almost certain these will be in Macs sometime in 2003, and when they are, most of our beige pc keyboards will be covered with drool.
An Athlon 2200+ costs $220 Canadian here, and puts you in the upper realm of CPUs.
Thats the problem with all these discussions. When you get right down to it, Intel and AMD fans really do believe that the performance rating of the CPU is its clock rate.
What's the Athlon's performance with 8 watts of power? Will it even run with 8 watts? How can you have a laptop otherwise? How many clock cycles go by on average before it executes an instruction? 20? 30?
Yeah, and you guys panned the ipod too: http://apple.slashdot.org/article.pl?sid=01/10/23
Less backfires, doesnt leak oil? Who knows?
OK, lets say they'll be doing that! switching to the x86 architecture
- thing-as-the-lower-priced-PC does.
:) ...)
Why close it of? Use special moded CPU's with special motherboards...etc etc? Only to remain a closed of platorm? In doing so, not letting users form doing what they what with a pice of hardware they bought (it really doesn't sound that attractive to switch to... more like they would be doing-a-$ms--thingie, "cough up the money, and WE'll control you! You can thurst us!")
All I _really!!!_ wish for in a computer, is that after I buy it, I may do what the heck I want to it, with it, on it! And run what software (OS) I want (may it be Linux, unix, Mac OS X, BeOS.. etc), and how I want it. And no further ridiculous cost like "to run that, you have to use exactly this, you may not/can't reuse your working-old-one".
Now that would be comprimized, if I bought "the new Apple AMD x86-64 MAC" and found that I can't play with it. Can't run my stuff, can't fiddle with the hardware, have to buy super-expensive-ultraDitt&datt-that-does-the-same
And as for the "porting Mac Os X software to the x86 architecture". It really would make this harder in the closed platform approach, cause it would make developers rewrite their software. On the other hand in an open platform, I think, the Mac would gain a lot (both in new avalible software choices & reuse of already cross platform avalible software). Take Adobe, (and x-other developers) already have almost all their software running on both x86 and PPC, so it wouldn't bother them if Apple would switch. Maybe it would even cut cost for them (goodbye to the Apple PPC Mac-department
[The end]
Having 5 guys/girls doing the same thing, is both time & energy consuming.
Having 5 guys/girls cooperating on the same thing is less time cosuming and energy saving.
I don't claim I know more than I know, and if you know you know more than I know, then by all means, let me know.
I made a recent post saying this was in the works. No one believed me. Believe me next time, for I know things!
Question
http://www.ironfroggy.com/
That's not an article, it's an opinion column. No fact, sources, even credible rumors - just wild assumptions that anyone here could make with no less authority. C'mon people, next we'll be calling Springer a current events show!
Think outside the... Hey, where'd the friggin' box go?
However, what about Intel? Some might say this would give Intel a monopoly position in PC hardware, and lead to calls for their breakup. They could spin off the StrongARM division, for example. I wish I was a corporate lawyer, I'd need a forklift to get my bonus check to the limo..!
(this is not a
Apple and Motorola introduced several incompatibilities into their PowerPC designs which made their instruction set incompatible with IBM's POWER architecture.
Eventually, Apple is either going to have to drop the usage of these extensions, or pay a lot of money to have them maintained as a separate architecture.
Whatever induced them to diverge in the first place? Was Motorla trying to lock them in as a customer?
I find that RISC chip run smoother then CISC do.
Run smoother!?
So, what you're saying is you can detect 'roughness' in a process cycling billions of times per second, using a display device that runs dozons of times per second (60-120 or so).
I'm impressed!
autopr0n is like, down and stuff.
Interesting account of previous x86 work at Apple:
Please help!
I am in the roofing business and recently my boss took away my hammer and gave me a handsaw instead. Now it takes MUCH longer to pound in my roofing nails. Saws suck! Who designed these things anyhow? There are sharp bits all along one edge and I often cut my hands as this crazy saw flaps back and forth. =(
GIVE ME BACK MY OLD HAMMER!
- Toby
I've been hearing this stupid argument for YEARS. It's WRONG WRONG WRONG, and it's been wrong every time I've heard it over the past decade. Every time a new CPU comes out, people go ON AND ON about how most users have no use for such a fast computer, or how Moore's law is nearing it's limits. It's ALWAYS BULL. It's a beautiful example of human shortsightedness in action. Yeah, sure, the average user can do their word processing and web browsing and e-mail just fine on a two or three year old computer. When was the last time this WASN'T true?
It's a good thing we've got games pushing hardware faster and faster, otherwise no progress would be made. Nobody really knows what sort of software we'll be running in the future. Back in the day when people were saying "you won't need anything more than a 386", did they think about the real time spell checkers available in most word processors? High-quality voice recognition?
Every time somebody says we're going to run into the limits of such-and-such technology (the end of Moore's law), there's ALWAYS some sort of breakthrough or advancement to keep it plugging along. How many years ago was it that they predicted that we'd have to get rid of magnetic media and go to optical? How many times has Slashdot posted some story about vaporware holographic memory or something to replace hard drives? Remember bubble memory?
... "Give me a woman who loves beer and I will conquer the w
just try photoshop on both platforms
At the picture sizes I work with, just about any task in pshop (except loading, rrr) is just about instantaneous. Photoshop is also more heavily optimized on the mac platform then it is on the PC.
Even then, I'd be willing to bet Photoshop on a high end dual Athlon system would be faster then a high end Mac. Perhaps you have some actual benchmarks to prove otherwise? I doubt it.
autopr0n is like, down and stuff.
Is this all the mac users have left? Power consumption? I mean, some people want quiet PCs, but spending thousands of extra dolars for a mac is not a very viable solution to many people, especialy when you consider the low power consumption PC chips out there (VIA etc) you can get low cost/low noise/low price trifecta if you really want.
And my laptop is really more like a iron or hotpad then a hair dryer. Just a high temprature brick.
autopr0n is like, down and stuff.
The original design goal of the RISC paradigm was to make computer chips less complex which frequently has the side effect of making the chips more efficient in terms of how much work can be done in a single clock tick and in how much power is drawn.
Nope, the original design of RISC was to make computer chips less complex, requiring fewer transistors and more instructions to do the same thing. But since, they were simpler, they could run at higher clock speeds.
RISC does less per instruction then CISC, but it runs more instructions per second. That was the original idea, although it's gotten twisted around by Mac advocates now that their RISC chips run slower then Intel "CISC" chips that are actually RISC anyway.
The x86 instruction set is complex, but since the p-pro, all intel chips have been RISC. AMD has been making risc chips since the k5.
autopr0n is like, down and stuff.
there seems to be a push by motorola into the imbeded and kiosk market. Motorola cpus appear all over the in lots of devices other than computers. The market being what it is/has been it might not have been such a bad idea...
errr....umm...*whooosh* *whoosh* Is this thing on ?
If Apple REALLY wants to make a jump, they should partner up with Sun and use UltraSparc chips on proprietary motherboards.
sometimes "the enemy of my enemy is my friend"
not always of course, but in this case I think it would be a great combo
Old age and treachery almost always overcome youth and skill.
Ummm 3Ware sells 12 drive IDE caching raid controllers that can saturate a pci-64 33Mhz bus. AMD and Intel have in fact caught up to the big boys in raw performance per cpu as the 2.8Ghz P4 beats anything but the largest Power4's and HP PA8700's which cost about 10-20 times as much. Now we just need someone to come along and build boxes with large numbers of them (barring the fact that there isn't yet a Xeon based on the same process as the 2.8Ghz chip.) The price/performance of the comodity chips and there related subsystems is so far beyond what Sun, SGI, IBM et al can hope to achive that I think they have some serious problems ahead. In fact for many computationally expensive applications people can't wait for the Opteron because it brings very large memory support to chips that are so much cheaper and in some cases faster then the competition that it is silly. For instance 32 1GB DDR modules would cost only $11K at an above average price of $350/module, whereas a similar amount of ram from Sun costs $40K, then there is the projected cost of Opteron chips which should be around $500 versus $4-8K per CPU from Sun depending on which Ultrasparc 3 you want.
There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
I meant 'low heat/low noise/low price'. I'm very tired.
autopr0n is like, down and stuff.
Besides, if Apple only uses Intel CPU but implementing proprietary chipset designs, it loses the advantages of switching architecture that makes the new x86 Mac platform so attractive
No. By using an Intel CPU and an Intel PCI chipset the GHz gap disappears, they weak DDR support disappears, etc. Being able to run MacOS X on generic clone hardware is much less important, and it would be suicide for Apple anyway. Apple is a hardware company.
Don't misunderstand my data. I'm not claiming Pentiums are twice as fast as PowerPCs. I claim their clock rates are usually 2x appart. This can easily be checked.
/. will take it. Look at the last column, the clock difference has oscillated around 2 (up and down) throughout the last 13 years
Let me see if I can get around this stupid lameness filter. The data is year, max Mhz for Intel max Mhz for Moto in that year and the ratio of performance.
In 89, Intel 25, Moto 16, ratio 1.5
in 90, I 25, M 16, R 1.5,
Y91, I 25, M 16, R 1.5,
Y 92, I 66, M 33, R 2,
Y 93, I 66, M 33, R2,
Y 94, I 100, M 80, R 1.2,
Y 95, I 200, M 100, R 2
96 200 100 2
97 300 150 2
98 450 233 1.9
99 733 450 1.6
00 1500 550 2.7
01 2000 880 2.2
02 2800 1250 2.2
Ok, pardon the formatting but this is the only way
Apropos to another of today's articles, wouldn't Richard Stallman and the FSF insist it be called a "GNU/Linux worm" ?
"FLAKINESS." Here's the most compelling reason to abandon Motorola's PowerPC chip: It's falling further behind in the speed race as Intel's chips leave Motorola's in the dust. It's true that this race, to a large extent, is baloney, since Pentium and PowerPC chips process information differently. Yet, as Steve Townsend of technology consultants EMA Inc. puts it, "the megahertz myth is a difficult one to overcome." In particular, it creates the impressions that Macs are somehow less powerful than PCs.
OK, so because the general public thinks they're slower, Apple should abandon them? I think Apple needs the PowerPC to be one more thing that sets it apart. Macintosh computers are great and extremely thoroughly designed. The PowerPC is efficient and robust and I'd take a 1G PPC over a 2G Pentium any day. It's not the processor speed alone that makes these computers tick, it's the overall computer design along with the way the processor can handle inputs from all the different channels.
~ now you know
You do have to doublecheck every experiment.
Correction: somebody has to doublecheck every experiment. You personally should doublecheck very few: only those ones that (a) seem suspicious and(b) you have the expertise and means for it (or can readily acquire them).
Otherwise you are just needlessly bogging down your own progress.
Common sense tells me the x86 PCs are faster than Macs just because so much more time and money goes into R&D for Intel and AMD's CPUs. You can try and prove me wrong, but you can't just say PPC is faster and expect me to take it on faith.
You are misreading my claim. The original statement is: The Intel clock speed is usually twice as fast as the Moto clock speed. How that translates to actual execution speed I did not measure.
what will keep AMD, or someone else, from making a overdrive type chip mod that will allow me to connect my AMD chip to an Intel MB.
1) Benchmarking is Difficult.
2) Which version of OS X were they using?
3) What version of RedHat did they have running?
4) Who's software were they running on what?
5) What kind of "software and hardware" performance?
6) If I use Photoshop every day, Photoshop matters more to me than any other benchmark.
6a) More users have photoshop on desktop machines than have XServes.
7) RAM Configuration of the two systems. Were varying RAM Configurations attempted?
8) Hard Drives.
&c &c
Benchmarking across two platforms is *incredibly* difficult.
Integrate Keynote and LaTeX
(Exceptions include benchmarks that bypass the virtual file system, and networked disk arrays that are accessed directly by several different systems without any intervening OS).
'twould be odd. Apple had 225 MHz PPC 604e machines when the fastest Pentium (possibly -Pro or -II) was only 200. Ca. 1996
"You mean the same Apple that's stuck with PC133 SDRAM now?"
Take a look at The Apple Store. Every single PowerMAC uses DDR RAM.
While this is an easy mistake to make because Apple took a while to go DDR, please at least take a quick look before posting.
Computer Science is no more about computers than astronomy is about telescopes. --E. W. Dijkstra
Apple and IBM Team Up on 64-Bit Processor
Um, yea, I doubt that's for sh*ts and giggles...
I can't stand when journalism is blind to the fact that the SOLE reason the x86 architecture is popular is due to PRICE.
CAn'T CompreHend SARcaSm?
IBM's latest PowerPC press release: "IBM today announced it has delivered 10 million PowerPC processors to Nintendo for its award-winning Nintendo Gamecube system. ... The IBM PowerPC architecture has been selected for a variety of applications in networking gear and network-attached products, including base stations, routers, modems, Internet-access phones, digital TV boxes, laser printers, optical switches, RAID controllers, and personal computers."
In 1996 the fastest Intel was 200Mhz, the PowerPC 200Mhz was announced that year, but didn't ship from Apple, as far as I can tell (although Umax did ship a system that year).
However, it does seem that my moto figures are a bit off for that period (96-98). As I explained elsewhere they come from whatever web pages I could find announcing the actual release of systems, both for Intel and Moto.
Your data for 1994 through 1998 for Motorola are wrong.
Source: http://docs.info.apple.com
1994 110 (not 80)
1995 132 (not 100)
1996 200 (not 100) -- parity with Intel
1997 350 (not 150) -- better than Intel
1998 350 (not 233)
1999 450
...if Porche made cars the average person could afford (or was willing to pay), they might find themselves suddenly occupying a far larger niche than they do now, on name alone.
Likewise, I can tell just from reading Slashdot that there are a lot of people out there with "Mac envy," who'd really like a chance to use a Mac, but who either won't or can't pay for the hardware. Me, I'd join up in a split instant.
I'm not a geek, I'm just a clever script.
Did you check actual release dates, or just the announcements? In one case there was a system announced by Apple which didn't ship until almost a year later.
(However, it does seem that 96-98 data is a bit off).
If Apple make an OS X that runs on any Intel hardware, Linux on the desktop is dead.
Apple hardware sales will suffer, but they will more than make up for it in OS X sales.
Go Darwin!
Wrong! Yes they come with ddram but, the chipset slows the speed to the cpu back down to 133!
Why? Because the g4 was not really designed to last this long and go up to speeds high enough where the chip runs faster then what the ram can provide. These new dual 1.2ghz powermacs are cippled by motorolla chips that can't take the i/o of modern memory to keep the cpu's satuarated. It totally sucks. Motorolla is a greedy company that cares more about charging an arm and a leg for obsolete processors rather then upgrading their chip fabrication plants. $400 dollars for a chip that performs as fast as a pentiumIII 800?? Come on! I would be ferrious if I were Steve Jobs. Motorrolla is bringing down apple and its time for Steve to look elsewhere for chips. Rumour has it that IBM may be the next powerpc provider.
http://saveie6.com/
I used the introduction dates from Apple's web site. See PowerMacintosh 8100, 8500, and 8600 models.
It looks like a lot of people have already challenged your assertion that IDE is just as good as SCSI. However, no one brought up the particular issue that plagues my experience with IDE - hardware write caching.
Most modern IDE drives have write caching enabled by default. However, under every OS I've tested this configuration can lose data, even with a journaling filesystem. The problem is that the filesystem thinks that the data is successfully written to disk, but it's actually in the drive's cache buffer. If you lose power at the wrong moment, you lose that data. I've reproduced this problem with Western Digital, Seagate, and Maxtor 7200rpm 4MB buffer 80GB IDE drives under both Linux 2.4.X kernels and Microsoft 98/2000/XP platforms.
I've written in to each of those drive manufacturers and they have confirmed that the cache buffer isn't backed by some battery or other type of power reserve, and that data can be lost when power is removed.
Apparently this isn't an issue in SCSI land because SCSI drives respect a flush command, while some IDE drives do not.
The bottom line is that if you want a reliable system with IDE drives you need to disable write caching, which drastically increases disk access latency and results in reduced throughput for many tasks.
I'd love it if a kernel hacker can provide some more details as to why journaling filesystems can't forceably flush the IDE disk's buffer... I've found many older threads on the issue on the linux kernel list but haven't found any definitive resolution or action items recently.
As the situation stands now, my iozone benchmarks show a 15k RPM 80GB SCSI drive performing 2x to 3x better across all tests than a 7.2k RPM 80GB IDE drive with write caching disabled, DMA turned on, and all other hdparm options optimized for maximal performance. That is a pretty large difference. Yes, I did verify that the hdparm tuning options were working correctly.
And yes, the 3ware IDE RAID controllers have the exact same problem. They have an on-board raid cache, but it's not battery backed, so it is not a good idea to enable write caching in most cases. The 3ware cards are great and cheap, but they don't perform as well as their scsi equivalents.
Before someone tries to flame me, yes I have heard of a UPS, but for the machines I'm trying to protect I can't trust that the UPS will be properly maintained, not overloaded, strong enough to survive a long outage, or that the customer won't hit the power button themselves out of ignorance when they think that the system has "hung".
Yes, because the rate of software innovation has slowed so.
However, a paradigm shift will come (other than Quake that is;) and suddenly we'll need all the extra performance. Natural speech recognition, for instance, might be such a shift. Ironically, the last demo I saw was on a G4 Powerbook, but it looks to me as though the voice dictation product are finally getting usable for most people, with a *fast enough machine*.
I'd love (as I've stated in previous posts) for Apple to support at least the new x86-64 chips from AMD. That would differentiate them from everyday 32 bit systems, and provide world class performance to boot. Plus, Apple would have a new "64 bit" ad campaign.
Cool, eh? ;-)
(I know, I know, it's a cliche, but cliches are cliches because they're _true_.)
Except, of course, when they're wrong. ;-)
I'm reminded of an old statement out of IBM (paraphrased): "The worldwide market for computers is around ten machines."
Galileo: "The Earth revolves around the Sun!"
Score: -1 100% Flamebait
Well, in terms of removing bottlenecks, Apple is behind everyone; they have only just begun to support DDR memory in their top of the line machines. I'd say that bus speed and memory i/o are where Apple is limited the most.
"Upon attaching the waterblock to my penis, I began to notice that I know nothing about computers." -- JRockway
> Secondly, adding a faster disk only matters if you do tasks which are heavily disk I/O intensive, which the overwhelming majority are not (especially because people have so much memory, and hence disk cache).
I think a fast disk IS important for anyone who wants to improve speed. Modern Windows is disk intensive, as are many applications at startup. My computer went considerable faster when I upgraded from a 5400rpm drive to a 7200rpm RAID without upgrading the CPU (1.3GHz P4) or memory (384MB of RDRAM). I think I have about as much disk access as anyone (I am not running my machine as a server except for the occasionaly Kazaa, which I do not leave open 24x7), but my computer went noticeably faster. I say upgrade the hard disk before you upgrade the CPU.
Disproven by who? Show me someone who thinks they're disproving the intrinsic advantage of SCSI over IDE and I'll show you someone who's deluded.
FACT: SCSI has a faster peak bandwidth than IDE. IDE's peak is 133MB/sec per channel, and the current SCSI implementation's is 180MB/sec per channel. As a bus approaches saturation it becomes less efficient, therefore SCSI should be more efficient at the same load.
FACT: SCSI supports tagged queueing. This means that in heavy multitasking situations or situations in which you are looking all over the disk (such as a RDBMS), SCSI can be dramatically faster than IDE. Before you get your back up, though, it is true that this offers little to no benefit on the average workstation.
FACT: IDE is still problematic after all these years. There are still situations in which one disk will work as the master with another disk but not as the slave. This is pathetic.
FACT: IDE requires one set of resources for each channel; IRQ, DMA, IO port, and memory address. SCSI requires one set of resources for each controller which might have four channels.
FACT: IDE supports only two devices per channel. This ties nicely into the above paragraph. If you want more than four devices you need more than two channels, which means you need devices/2 sets of resources. While on one hand we do have interrupt sharing, and it usually works fine, it doesn't always, and you still have to consider IO ports and memory addresses. Most devices have a half-assed implementation of plug and play which only allows a small selection of those addresses, presumably to cut down on the cost of the address decoder. In addition most IDE buses will not let you mix modes on the same channel (PIO and DMA) so your PIO devices drag everything down. Thankfully, they're disappearing, but they aren't gone yet.
So tell me again how SCSI is not intrinically superior to IDE?
Eventually SCSI on the desktop should be replaced with firewire, though I (and many others) am still waiting for actual firewire hard drives, not a disk in an external box with a IEEE1394 to IDE converter. Right now 1394 is only 50MB/sec (at 400mbps) but the new faster standards are supposed to be doing 800mbps (100MB/sec) right now and 1.6Gbps (200MB/sec) by sometime in 2003. Then we are supposed to get 1394 over fiber sometime after that at 3.2Gbps (400MB/sec). 127 devices per channel, which is more than we're likely to need any time soon. THAT has the potential to replace SCSI, again people actually make real firewire drives which we can put inside our PCs. 1394 is also nice because you can run it as a local bus, IE synchronized with the CPU. This makes it actually useful for things like putting ALL your hard drives on.
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
I know a number of PC users who never owned a Mac, but they "lust after" a Titanium Powerbook, or a G4 with the cinema display.
Exactly. I heard a joe-user type talking about (a) how hard it is to use a Mac to edit video. (Huh!?!) and (b) how insanely cool the pulsing sleep light is on an ice book and how great that is.
Clear, Dark Skies
Good grief, yes.
The "News of the Weird" syndicated column publishes little notes when it "retires" a formerly weird story because it's happened too often to make the cut. (Stupid criminals who pose for cameras while destroying them -- stuff like that.) It's past time for Slashdot to retire "Apple -- could it switch to Intel chips?" stories, at least until someone hears Steve Jobs mention it in a keynote speech.
The pros and cons have been gone over maybe sixty-thousand times. Leave it alone, jeez, this is starting to be like an OS wars topic.
"Fundamentalism" isn't about divine morality. It's about human authority.
"Destroy science and religion. Science would re-emerge exactly the same; but not religion." - Penn Jillette, paraphrased
Perhaps that's because it may not be a myth. The fact that Apple has never submitted SPEC benchmarks is also disturbing. I think Apple is significantly behind in terms of performance and they know it. The question is whether they are going to fix it by moving to IBM or Pentium.
You can get all the scores with benchmarks going back to CPU 92 and up to CPU2000, as well as other benchmarks, web, web w/ SSL, and others from Spec.org. Incidentally the current kings of the roost (integer and floating point) are IBM's Power4 and Itaniums. Alpha's were ahead and are likely to regain the speed advantage when the EV7 comes out.
Degaussing scares the bad magnetism out of the monitor and fills it with good karma.
I looked into this once tryong to convince myself to buy an LCD panel instead of a CRT for my next computer.
For an 18" LCD vs a 19" CRT, I'd have to keep the LCD for something like 17 years before it paid for the price difference in power costs, compared to a CRT running 9 hours per day. (Home computer, where I'm at work or asleep most of the day and so the CRT is turned off or in powersave mode when I'm not in front of it.) $200 for the CRT, $600 for the LCD, and 7cents per kilowatt-hour. Don't remember where I got the power usage figures, I think it was from NEC CRT and LCD monitor spec sheets.
I did not include power costs for running the AC extra in the summer. Bear in mind, though, that you run the heater less in winter too, so it is possible that you will balance this out. I didn't look into this, but it may be a wash.
This let me know that power consumption/cost alone was not a reason to get an LCD monitor for a desktop computer.
You don't realize how little power really costs... An extra 50watts, used 24 hours per day, increases a power bill about $30 per year. Takes a long time to make up for $1000 difference in system price at $30 per year.
This is my sig. There are many like it but this one is... Oops. Frank, I've got your sig again! Where's mine?
Given that you posted pointers to "benchmarks" that only measure clock speed and not performance, I think you have made my point. (Which was a claim about what people like you think performance is.)
Sure, all modern processors are superscalar. That doesn't mean a lot of them sit idle a lot of the time due to poor architectural decisions, forced upon the designers by the requirement that they maintain backwards compatibility going all the way back to the 4004. Not 4004 instruction compatibility, but the compatiblity-compromise-chain goes that far back.
Yeah, and you guys panned the ipod too: http://apple.slashdot.org/article.pl?sid=01/10/23
http://saveie6.com/
do you think Porsche would last? If they try to scale their manufacturing capability up to Ford levels, do you think their QA would remain at Porsche levels?
how long after Apple cut prices before they failed the way so many other bottom-tier PC vendors have. Hell, the way so many TOP tier PC vendors have. Of the PC companies that were formed in the same decade Apple was, how many are left? Of the PC companies formed in the 1980s - how many are left?
Clear, Dark Skies
Right. After all, no body ever hears of computer owners tweaking their machines, adding aftermarket performance boosters or even building their own. No body hears of computer uses tricking out their computers so they look cool. Nobody gets emotionally attached to their computer. Nothing like cars at all.
Clear, Dark Skies
I'd go one step further and argue that Apple buy into VMWare and use what they have already done to provide X86-on-X86. It's not "emulation" as much as it is a virtual machine. It's probably only a philosophical argument whether you'd expect x86 Apps to run in an x86 environment "window" or whether you'd expect to double-click on an application and have that app window appear natively alongside OSX applications. I'd argue for the total Windows/x86 VM environment, since that way you're just needing to emulate hardware bits and not try to do the WINE-style OS call translation (which is hard to do, and leaves you farther behind when people start wanting to run newer OSs whose function calls aren't known).
I'd also argue that the VMware capability be a permanent addition to the operating environment, not a temporary kludge to satisfy some interim changeout period. A native-OS supported VM mechanism could also support 68k or even PPC applications (much harder, I acknowledge) or people wanting to utilize a seperate x86-specific OS or environment.
While it's true that the new G4's don't have a DDR FSB, the FSB is 167MHz on all but the low-end tower, not 133.
Apple is supposedly working on a new system controller, named ApplePI.
-- if it was so, it might be; and if it were so, it would be; but as it isn't, it ain't. That's logic - Lewis Carrol
How do you figure? The branch-prediction demo was pretty straightforward, if you ask me.
The plain truth is that powerpc processors and Macs have been lagging behind in performance for a long time. Top of the line G4s use 1.25Ghz processors. Even if they were twice as wide superscalar (I don't believe they are) AND the majority of programs could take advantage of all the extra execution units most of the time (which is not often the case on any superscalar CPU), they would still not match the performance of a top-of-the-line P4. Not to mention the fact that the Apple hardware would STILL be much more expensive.
You're not taking into account a gigantic range of things with your simplistic argument. AltiVec, for example. Yes, you need instructions that are highly parrallel to make it work. Instructions like the ones Photoshop asks for. See, that's the thing: Apple computers are designed for specific kinds of tasks, namely multimedia. You big-iron-overclockers just do not understand that. I doubt you can understand that, without going to work as a designer.
Of course you can buy something that's faster than a PowerPC. Yes, that chip exists. You know what? There are even chips that out-run your Hammer chip. Yes! It's true. Who cares?
As for your 'expensive' comment, that has been destroyed over and over again and I won't take the time to further refute it here. They cost more up front, less over time.
Ok, I'm on fire now, but I don't care. I don't understand the animosity towards Apple stuff. It's just another choice. It's not trying to take away your PC, man. Apple is not a threat to you. Just see it for what it is: another vendor. They've got some interesting stuff. When AMD come sup with a new chip we're all 'ooh, nice new chip, that'll drive cometition!' but when Apple does it, you try to rip it down. No, it is not superior stuff in every single way. But all the speed in the world doesn't make a goddam bit of difference when your operating system makes you want to kill yourself.
Again, sorry for the flameage, but fucking hell, I get tired of some of the rhetoric. Just take a pill and relax. You don't like Apple? Ignore them. But save your own FUD, willya? It makes me tired all over.
If Jesus wants me it knows where to find me.
Not true at all. How many Mac users do you know?
-- if it was so, it might be; and if it were so, it would be; but as it isn't, it ain't. That's logic - Lewis Carrol
"You mean the same Apple that's stuck with PC133 SDRAM now?"
Me: "Every single PowerMAC uses DDR RAM."
You:"Wrong! Yes they come with ddram but, the chipset slows the speed to the cpu back down to 133!"
What exactly was I wrong about? Are Power Macs using DDR memory or not? I didn't say anything about the bus being DDR, dual channel, 266MHz, or anything implying that the system could make full use of DDR RAM.
This is not unlike NVidia Nforce Athlon systems. They may not be able to make use of all of their memory bandwidth (particularly if using a non-integrated video card), but the motherboard still supports dual-channel RAM regardless.
Computer Science is no more about computers than astronomy is about telescopes. --E. W. Dijkstra
Yes, my point exactly. You measure clock speed and call it "performance".
Apparently it is you who is ignorant of processor architectures, and it is not my responsibility to educate you.
The idea that Apple would trade a high performance processor architecture for a low performance (At absurd power requirements) one is laughable.
Keep wishing for that validation-- but you're going to be disappointed.
Yeah, and you guys panned the ipod too: http://apple.slashdot.org/article.pl?sid=01/10/23
Of course, MOSX is never running on commodity hardware. But Windows runs on an accepted standard set of hardware, right?
Imagine, for a moment, MacOS X86, with QuartzGL on high-end Apple-branded hardware. Given Apple's nice play with BSD/LGPL licensed software, what feature could they add to their operating system? Wouldn't they be able to port WINE? Apple could have it's very own Win32 runtime, without making consumers buy WinXP.
It didn't work well for OS/2, and it might not be the best idea for Apple, but they've got an existing user base and (lately) a cohesive corporate vision. Apple would have to burn that bridge when they crossed it. Not like IBM.
There are no trails. There are no trees out here.
The Mhz != Mhz argument is valid, but Apple's stretched it waaay beyond reality. The PPC chips *do* run cooler and *do* have a more sane instruction set, and for a lot of people that is valuable, but as for raw computational power? Nope...x86 chips are way out in front.
Oh, a better quote from the article:
Dammit, Apple, if you're moving to x86 you simply aren't going to be a competitive systems manufacturer any more. You're going to have to accept that and either sell add-ons or only sell software.
May we never see th
That extra RAM will get used for caching.
May we never see th
RTFA.
...I'd rather they go to AMD's upcoming Hammer, provided...
Couple of things...first, the IBM chip is at least a year away from complete development
We know this. It's also in the article. No one said anythign about expecting it tommorow.
Secondly, as I understand it, this particular PPC will have four processor cores on each die with a scalability of eight--meaning it could be run in dual cpu configurations.
Up to four. First runs will be (single and?) dual from what I unserstand. This will make them nice for dual processor applications.
How this will fit into Apple's plans is beyond guessing at this point. Just because something is called "PPC" is no reason to think it will easily work within the Mac architecture.
Maybe it will fit in nicely because Apple just redesigned ther system controller/bus scheme, and all G4 Towers are dual processor now? And yes, is something is called PPC it can be made to work with a Mac. You're thinking of Power4. Power4!=PowerPC. Power4 was developed by IBM, and was the chip Apple "zealots" where oogling over last time Motorola began to fail us. PowerPC was developed jointly by Apple, IBM, and Motorola. It was based heavily on IBM's POWER RISC designs, but size, speed, and lower cost where added, along with AltiVec.
Last time it was size, expense, heat and lack of AltiVec which nixed Apple's use of IBM's PPC.
No, It was Power4.
What will it be this time?
Well since they're WORKING TOGETHER ON IT, it should probably be nothing, right?
"Just because something is called "PPC" is no reason to think it will easily work within the Mac architecture.", yet Apple will with much more ease, convert its entire motherboard specs over to x86 and convince developers to make ANOTHER giant platform leap right after we're just getting near the end of this OS9->OSX fiasco? That's pretty much suicide.
It ought to easily outperfom a P4 and it would give Apple 64-bit capability (great for the corporate server market) when and if Apple might want to use it.
Or they could use the new chip which you STILL havent' read about, which will be 64-Bits.
I hate the blind x86 fans who are just like the journalists mentioned in my previous post. It's not a good move for Apple to use the x86 architecture, or any of it's near-future alternatives. Even if you had Mac-only boxen, someone out there would somehow hack Mac OS X to run on a PC. Then it woudlbe over, no one woudl buy Apple hardware. People would buy cheap boxes for Windows, Linux, or the "cracked for the other 95%" Mac OS X. And then you can say goodbye Apple.
Plus the PowerPC architecture is much more reasonably organized than x86 is (Contiguous memory which is soon to be 64-bit, big-endian, better piplining, TONS of registers, both general purpose and floating, less pwoer, less heat, etc.)
I'm not a zelot because I won't jump on your bandwagon.
CAn'T CompreHend SARcaSm?
Yeah, one of the two. (actualy modern intel chips can run either way)
autopr0n is like, down and stuff.
Since when is white a pastel color? Silver, gray and white are also the colors of an aircraft carrier. So your point is? You don't have a valid point, that's the problem. It's all just your opinion. That's the extent of Apple's colors these days. The reason they have iMacs at the salon is because they are stylish looking computers, so as long as they are going to have a visible computer, make it something that looks good.
I used to work in PC sales, and the fact is that women purchase PCs much more based on how they look. Hewlett Packard sells a TON of PCs and overpriced monitors to match them because their PCs aren't an eyesore in the living room. The color scheme is subdued, front drives and ports are covered, and some even have CD storage built into the top so you don't have media laying around.
And that's exactly why Apple made the original iMac in colors. My sister-in-law bought a green rev B iMac because it matched her art room.
Eventually Apple moved away from the colors to a nice clean white motif.
Most EVERYONE wears black.
Depends on where you live.
I would never buy a DVD player that had the same design sytle as the new iMac. The last thing I want while watching a movie is for the player to be vying for my attention.
That's you. Not everyone else. Actually with the iMac you mostly see the monitor, since it's floating out in front. The base more or less fades into the desk.
Try FalconNorthWest, Voodoo, or Alienware for some cool looking PCs. A PC should either look damn cool, or try to draw as little attention to itself as possible.
"Damn cool" is a matter of option. And looking "damn cool" while drawing as little attention to its self is kind of an oxymoron. Plus it's also just your opinion that it should be one way or another.
Your ugly iMac may seem hip yesterday, but it's going to be an eyesore soon when even your grandmother with dementia knows it's an ancient machine.
I have a G4 tower. I'm not even a big fan of the new iMac, I think the Cube was a lot better looking. Either way I think they are fairly interesting designs and should hold up pretty well. At least Apple is doing something different. A lot of people love the new iMac. To each his own.
I suppose you drive a pink Cadillac? Eh, Mr. In-Touch-With-His-feminine-Side?
Nope, I drive a black BMW.
Unlike a car, the case of a PC is just a shell. Nobody is going to walk by and go 'Hey, that 6050Z is one sweet-ass machine! Can I try it out?'
According to who? People do that all the time. You even said your self that people buy PC's based on how they look.
The appearance of your PC either needs to fit your personality, or fit into the background.
Once again, this is just your opinion.
-- if it was so, it might be; and if it were so, it would be; but as it isn't, it ain't. That's logic - Lewis Carrol
Hmm.. I do see your point a little clearer now. There are a ton of variables to this however, such as will Apple boxes run Windows? If they dont' then the argument about having tons of availible software goes out the window (no pun intended). Unless you feel like running something like WINE on OS X, a Mac doesn't have the API nessisary to run the thousands of software titles availible, and so then, back to square one: What's the point of buying a Mac?
Furthermore, a switch to an x86 chip means for any apps to be decent (i.e. run without emulation), they'd need to be recompiled and reworked, the registers are different on x86 and there are fewer of them, there are big- vs. little-endian issues, etc. Software developers just made the leap to Mac OS X, slowly I might add. How long did it take to get decent programs running on Windows XP? Not that long. Major apps didn't begin to show face on Mac OS X for quite some time, imagine if now after all that hard work and re-training of Mac software engeneers to bring their products to OS X, Apple again slams them and says "Ok, you have to work with x86 now." I think that alone would do Apple in.
Also, supposing Apple boxes could run Windows, that would pretty much make Mac OS X need to comply to a standard x86 motherboard spec, which in turn would allow Mac OS X to run on any commodity PC. There go hardware sales, as Apple's prices remain high as they need to fund this 3rd major time-of-transition. Going head-to-head wth Microsoft might not exactly be the best idea until Apple is a little more firmly settled with their current offerings.
Yes, I admit, it's all speculation at this point, knowing exactly what IBM comes up with will be a great factor. I really wish there was some kind of official word on this, they must know everyone is itching to know about it.
I just feel that while it may be a gigantic boon to Apple to move to x86, it may also be complete suicide. They'll have Marklar to back them up if it ever becomes a nessesity, but if it happens I'm sure we won't see it for at least 5 years.
CAn'T CompreHend SARcaSm?
I caught that after I hit submit... :-/
-- if it was so, it might be; and if it were so, it would be; but as it isn't, it ain't. That's logic - Lewis Carrol
RISC chips are so much more turquoise, too.
It's true, you know.
± 29 dB
charles haddad (dvorak, etc.) is in the business of getting people to flock to the bw site/mag.
he has no more interest in intel than i do in toads.
"Win treats sysadmins better than users. Mac treats users better than sysadmins. Linux treats everyone like sysadmins."
If they can't get the IBM working reasonably fast make the jump to Itanium 2. If you look at the price of the Itanium 2 they cost about the same as the Xeons with the same amount of cache. That is you subtract off cache costs Itanium 2 is comparable in price to pentium 4s.
Further I could see Intel giving Apple a sweetheart deal on the Itanium 2's as a way to pressure Microsoft to start agressively pushing the 64 bit CPUs for the Wintel platform.
Again IBM is the path of least resistance but Itantium 2 should be a fairly easy switch.
No, I think the issue is that you can't comprehend english.
I know what specint and specfp are. I was around before they were around.
They measure clock speed and call it performance. They don't measure performance.
It gets tiring dealing with half wit idiots such as yourself. The net was much better before AOL. And why is it the average poster on slashdot seems to be a twelve year old who just managed to successfully install linux and so he thinks he's 1334?
Come to me when you've built a computer from scratch-- and I mean, designed the PCB, wrote the bios and burnded yourself with a soldering iron. Then you can talk about who knows what you fuckwit.
Its really become quite clear, as I said originally: Those such as yourself, think clock rate is performance. And a "benchmark" that measures clock rate is what you will then use.
you're real happy
Yeah, and you guys panned the ipod too: http://apple.slashdot.org/article.pl?sid=01/10/23
By deferring it to "the motherboard chipset", it seems to me AMD is leaving it more open than by putting it inside the CPU itself.
How many times did we have a CPU in the past with only 1 motherboard chipset available for it?
Even if they release their own "recommended" chipset to use with it, you can be almost sure some smaller firm in a foreign country will engineer up boards using an alternative that doesn't have TCPA in it.
You probably can't, however, disable it inside a CPU just by cutting a couple pins or whatever.
As far as performance goes Apple simply is not in that game. They do not have a machine that competes in the front ranks of the data center game.
OK some Macs get sold for data center use but to date Apple does not even figure in the typical market share pie chart for that market sector.
Adopting Intel's fastest processor and bringing a ready-made constituency of users would change this significantly.
Looking for an Information Security student project suggestion?
Try http://dotcrimeManifesto.com/
Its really become quite clear, as I said originally: Those such as yourself, think clock rate is performance. And a "benchmark" that measures clock rate is what you will then use.
Uh huh. And that would explain why a 1Ghz Itanium2 comes close to the P4 2.8Ghz in specint, and surpasses it in specfp. You're a moron, and claims of your great historical knowledge are ridiculously misplaced.
Yes I'm aware of that, and I replied:
Goes with his remark about not wanting to watch a DVD of a player that was vying for his attention. I think most people look at the screen, not the player, unless they have ADD, and the same goes for how your computer looks.
It's a computer, and it either looks like "a computer" or it doesn't. But it's still a computer. It doesn't effect how it works. Also it doesn't have to either blend in or stand out anymore than your sofa or TV. But if you like the way it looks that's good too.
-- if it was so, it might be; and if it were so, it would be; but as it isn't, it ain't. That's logic - Lewis Carrol
It means when I am Compiling 30 Programs at the same time even though the system is noticably slower it is still usable. Unlike on Intell systems with Both Linux and Windows when they go under heavy load they will not respond to your input Untill they are done.
I have not seen this happen to Sun and Mac Hardware.
If something is so important that you feel the need to post it on the internet... It probably isn't that important.
It means when I am Compiling 30 Programs at the same time even though the system is noticably slower it is still usable. Unlike on Intell systems with Both Linux and Windows when they go under heavy load they will not respond to your input Untill they are done.
I have not seen this happen to Sun and Mac Hardware.
That's more likely to be a function of the kernel's scheduler rather than the processor architecture.
If you'd talked about a correlation between CISC and higher context switch times (altho' there actually isn't one) then maybe you'd have a point, because it would mean that CISC performance deteriorates under multitasking loads, and would benefit from larger time quanta. That could conceivably be called "less smooth" but the original poster didn't have a clue what he was talking about.
Yep, measuring clock speed is a stupid way to try and measure performance.
You are the moron here. I don't mind ignorant people, its ignorant people who run around and insult those who actually know what they are talking about that is the bane of slashdot.
Yeah, and you guys panned the ipod too: http://apple.slashdot.org/article.pl?sid=01/10/23