4.7GHz IBM Power6 Spotted

Ilgaz notes that The Register has posted benchmark results from Oracle 11i running on four 4.7GHz Power6 chips. Quoting: "The speedy chips confirm IBM's boasting that Power6 would arrive near 5GHz. They also show that IBM's customers have a lot to look forward to in terms of raw performance." Rumor has it that the Power6 chips will be announced on Tuesday.

Comments

Did Apple make a mistake?

[Whiney Mac Fanboy]

Power6 sounds like it's going to be pretty damn cool - Perhaps Apple made a mistake jumping to intel so soon...

*sighs* I for one yearn for the days of smugly ending any performance argument with some PC user with "Well, we've got Altivec & Altivec is magic."

Re:Did Apple make a mistake?

[jcr]

Nope.

If Apple had been waiting until now for a 64-bit chip they could put into a portable, they'd be in Very Big Trouble.

The PPC has a lot going for it, but Apple made the right choice.

-jcr

Re:Did Apple make a mistake?

[spribyl]

I have always been perplexed by the move from the 64bit P5 to the 32 bit duo in the first place. Though, I understand that the duo2 is 64 bit.

Re:Did Apple make a mistake?

[stoneymonster]

Huge increase in mac sales since the intel switch? Massive profits? Stock well over $100? Yeah they made a mistake. Look, sometimes business decisions are just that: business, regardless of whether they're the most exciting decision from a technical or geek standpoint.

Re:Did Apple make a mistake?

[Wesley Felter]

It's interesting to see PowerPC on the other side of the megahertz myth for a change.

Also, I bet Power6 would work great in minis and MacBooks.

Re:Did Apple make a mistake?

[mdhoover]

Hard to say. It was a good move for them going Intel considering they couldn't get their Power5 chips shipped quick enough, and they can convert Windows users easier to using their hardware (VPC ran like a dog).

Still, for me the mac did lose some of its magic with the switch... maybe someone should give Jobs a nudge to put together a limited edition, high end successor to the G5 with the Power6 for us PPC nuts... I'd buy one...

Re:Did Apple make a mistake?

[mrchaotica]

What I don't understand is, since Mac software has to be Universal nowadays anyway, why Apple doesn't just permanently keep its lineup as a mix of PPC and x86, picking whichever chip suits the particular machine they're designing at the time? Power6 Xserves along side Core 2 laptops... it sounds good to me!

Re:Did Apple make a mistake?

[l0ungeb0y]

"Did Apple make a mistake?"

No. Not at all.

The Power series was the high end server class, meant for big iron.
The PowerPC series was the vastly scaled down little brother intended for the desktop class.

IBM wasn't all the interested in making chips for Apple.
And who can blame them? Lower profit margins and less units sold.

Intel is a much better match for Apple, which is a consumer grade CPU manufacturer. And since the switch, Apple has not had the embarrasment of lower performing CPUs and long waits on CPU upgrades that IBM and Freescale saddled Apple with.

If Apple had stayed with IBM, they would have been pushed to the Cell processor. And that would be a bad PR move, running on the same CPU as your game consol runs on. And there would of course be no gaurantee switching to that processor family would result in better product cycles from IBM.

Apple made the right choice, The relationship with IBM was no longer viable.

Re:Did Apple make a mistake?

[rolfwind]

As the transition sees developers pushing out binaries for both chips, I don't see a downside of Apple straddling the fence and using both types of chips. (The people who use paralells or virtualization will know what their software runs on....)

They could also continue using that as leverage against intel or ibm when it comes time to price certain chips.

(Assuming the intel contract only forbade them getting other x86 CPU suppliers, not CPU suppliers in general.)

Re:Did Apple make a mistake?

[Short Circuit]

I'll wait until the specs come out Tuesday* before I decide if the Power6 is interesting. Sure, it's got GHz. But how many ALUs does that cover? FPUs? What about its SIMD instruction capabilities? How long is the pipeline? Is it insanely long a la NetBurst?

Not to mention that IBM didn't seem to be putting any resources at all into a low-power verion of the POWER5; What makes you think they would for the POWER6? Without a low-power chip, Apple would have a hard time making laptops with a decent battery life.

* Well, I'd rather wait until that guy at Ars Technica does one of his in-depth analyses on the POWER6 architecture. His essays on Intel, AMD and Mac-basis CPUs were awesome, and his essays on hardware in general are how I learned that stuff. When I found their Technopaedia, I spent days soaking it in. Pity they didn't carry it over with their site redesign a couple years back.

Re:Did Apple make a mistake?

[Whiney Mac Fanboy]

But Apple didn't have to switch - they could've gone intel for laptops & waited for power6 for high end.

Re:Did Apple make a mistake?

[Anonymous Coward]

"...some PC user..."

Does this mean you finally accept the utility of distinguishing PCs from Macs in casual conversation?

By the way, Jack calls Kate in flash-forwards off the island, and Claire splatters on the pavement beyond healing after falling from a window. Hope this helps, you douchebag.

Re:Did Apple make a mistake?

[Anonymous Coward]

POWERn != PowerPC

They (mostly) share a common ISA but the chips themselves have always been quite different.

Re:Did Apple make a mistake?

[KonoWatakushi]

Also, I bet Power6 would work great in minis and MacBooks.

A Core2 Duo would also work great in a Mini.

Apple now has tons of options with Intel, yet they still refuse to produce competitive hardware; whichever chips they are using is irrelevant. The future looks like more of the same: insanely priced outdated hardware, so that they can maintain their obscene margins.

I have no problem paying more for Apple hardware, but right now, they are simply gouging customers. What's worse is that none of that revenue is being used to further the Mac platform, as is evidenced by their almost nonexistent R&D investments. (Which work out to far less per machine than it costs for any PC OEM to bundle a copy of Windows.)

Re:Did Apple make a mistake?

[stevesliva]

How long is the pipeline? Is it insanely long a la NetBurst?

* Well, I'd rather wait until that guy at Ars Technica does one of his in-depth analyses on the POWER6 architecture.

IBM has confirmed that POWER6 has the same pipeline depth and roughly the same execution unit configuration as the POWER5.

Not that it's the usual 27-page article, but still...

Re:Did Apple make a mistake?

[Anonymous Coward]

The stock price is more related to the iPhone and phenomenal iPod sales.

Re:Did Apple make a mistake?

[LWATCDR]

Unfortunately true. The new Power series could work in a desktop or workstation but it will be expensive. I am sure that with enough money IBM could make a Power6 that was better than the Core 2... Except they wouldn't make a profit on it. As much as I hate loosing yet another better then the X86 ISA from the desktop Apple did the right thing.

Re:Did Apple make a mistake?

[kestasjk]

If your software has to take advantage of both chips it probably won't be optimized for either, and it's also a lot of extra unnecessary work.

Re:Did Apple make a mistake?

[Short Circuit]

OK, that does it. This processor is awesome. I haven't lusted for a processor this much since Stokes did a piece detailing AMD's Hammer architecture.

(Delicious link. Thanks.)

Re:Did Apple make a mistake?

[suv4x4]

why Apple doesn't just permanently keep its lineup as a mix of PPC and x86, picking whichever chip suits the particular machine they're designing at the time? Power6 Xserves along side Core 2 laptops... it sounds good to me!

As a software developer why you should work twice more (OSX intel / OSX ppc) to produce a piece of software that will work on roughly 2-3 % of the desktop computers out there?

If Apple would keep randomly altering their hardware and require compatibility with a range of completely different architectures, in the end it'll completely alienate the developers. As Microsoft knows very well, developers, developers, developers are you best asset in this fight.

Furthermore, no, being Universal binary is not a requirement, and I know few companies which release only Intel versions of their Mac software (example: Adobe's Soundbooth)

Re:Did Apple make a mistake?

[Nasarius]

How much software is "optimized" for a specific architecture, beyond what the compiler does? How much "unnecessary work" is there, beyond what has already been done in the creation of universal binaries? It's extra work for Apple, but essentially none for the vast majority of application developers.

Re:Did Apple make a mistake?

[suv4x4]

And since the switch, Apple has not had the embarrasment of lower performing CPUs and long waits on CPU upgrades that IBM and Freescale saddled Apple with.

Oh, embarrassment you call it. I didn't know producing ads where Pentium laptops are crushed with a steamrollr counts as embarassment.

That's why I don't like Apple quite much. They lie in your face, though apparently anyone with a clue sees the truth transparently.

Re:Did Apple make a mistake?

[datapharmer]

I agree that they should keep a mixed line-up (when advantageous). They could easily offer different platforms for different solutions; intel might be cheap and great for laptops and the mac mini, but why not offer servers and workstations with these chips?
Unfortunately if they wait very long they may permanently lose the opportunity to do this because there are already programs coming out that are intel only: Joost, Adobe Premiere, etc. These vendors claim they will have support for the PPC computers in the future, but really - why would they want to go out of their way to support a "dead" platform?

Ironically most of these intel only apps are video related which is unfortunate given the advantages of altivec.

Re:Did Apple make a mistake?

[c_forq]

At least their Macbooks are competitively priced. I am currently pricing laptops, with a focus on portable, and I have found nothing with close to the bang for the buck as the MacBook, the closest I've found are refurbished machines with Core Solo chips (however my requirements include a screen under 14" and the weight to be around or under 5 pounds. My bang for the buck analysis also takes into account student discounts, which some manufacturers don't have. I have found if I was willing to go for a 15" or larger everyone and their mother in the industry makes a laptop that could compete, but that isn't portable enough for my needs).

Re:Did Apple make a mistake?

[Cadallin]

I'm going to give the very uncommon answer: Yes. Definitely.

Why? If they were going to switch to x86, they waited way too long to do it. By the time the first Intel Macbooks shipped, IBM had had low power G5's available for months. These could have absolutely been user for a Powerbook G5. The desktops, of course would have been shipping POWER5 parts, what would have been the G6 (By the time these POWER6 machines made into Macs, they would have been the G7). The correct solution to many of the other problems was to pressure Nvidia to make CUDA multiplatform. This could have been Altivec^2. Apple would have had the expertise to really leverage modern GPUs as Stream Processors for Media use. This would have been amazing. What we got instead was x86, which gave us Windows legacy support. Useful, definitly, but nowhere near as exotic or sexy.

Re:Did Apple make a mistake?

[Anonymous Coward]

Wow dude. You're special. You were an Apple fanboi back when only uptight dickheads used Macs. Congratulations. Your outrage should be directed at your god Jobs for taking your exclusive little club mainstream... instead you troll on slashdot. Lose the turtleneck, it's restricting blood flow to your brain.

Re:Did Apple make a mistake?

[Anonymous Coward]

POWER6 chips alone will likely cost more than a whole Xserve. There's no way Apple would be able to use them in computers. That's why Apple used PowerPC chips instead. There is expected to be a "light" version of the POWER6, but Apple would be out of business if they waited for it.

dom

Re:Did Apple make a mistake?

[RockoTDF]

But in the case of high end software, it would not be all universal binaries. Just look at any *nix, they make a version of their OS and its software for LOADS of different architectures.

Re:Did Apple make a mistake?

[jcr]

But Apple didn't have to switch

Remember the Sun 386i? Nobody believed it had a future, and that was a self-fulfilling prophecy. If Apple hadn't committed to a switch, then a lot of developers wouldn't have bothered to build their apps for Intel, since they could just run under Rosetta.

-jcr

Re:Did Apple make a mistake?

[solios]

UBs are the new Fat Binaries - Apple dropped the 68k as soon as they were able but Fat Binary soft that could run on both 68k and PPC was around for quite a long while thanks to the big install base of 68ks. Now history's repeating itself - the only reason we have UBs is because of the hugehugehuge (proportionately speaking) PPC install base.

Apple has no reason to go back to the PPC. The profit margin on intel kit is much higher..... and if you don't think it's about profit, ask yourself why all of the low end PPC machines had okay ATI or NVidia graphics, while all of the new low end intel machines have totally bullshit Intel GMA graphics?

Apple : Not Passing The Savings On To The Consumer since.... what, 1984?

Re:Did Apple make a mistake?

[DurendalMac]

Nope. This is the POWER6, not a PowerPC. IBM made the G5 (aka 970) as a derivative of the POWER4. IBM told Apple that they didn't want to make a derivative of the POWER5, so they were SOL on an upgrade path. This is not the kind of processor you would EVER see outside of a top-end workstation, server, or mainframe. It's not something Apple would have used.

Re:Did Apple make a mistake?

[guruevi]

I think you're just bashing Apple, but ok. Yes their hardware is a bit more expensive, but the return is also much larger in forms of better fit-together hardware, service and when you can write off your hardware.

Especially once you go into businesses. The difference in price between Apple and Microsoft starts getting huge once you spec out an environment for >50 people. With Apple you know your clients cost you $129, your server $999, Remote Desktop for $499, no limits, everything integrated with Kerberos + LDAP. With Microsoft you got that and then you have to start calculating CAL's for Exchange, CAL's for your Terminal Server, CAL's and server licenses for your SMS and WUS, and each little piece that will make life easier as the admin costs you an extra CAL which is all included in the Server+Client+RemoteDesktop licenses on Mac. Next to that, the average Windows machine lasts 3 years before it get's old and slow. I've seen G3's running OS X 10.4 without a hitch, G4's are mainstream in many companies and most haven't even gone to Intel yet.

Then support costs. A call to Apple or registering at the Genius bar will at maximum cost you a little over a $100/year and I've always been satisfied with the answers, usually I'll hold for 5 minutes and somebody picks up and I'll either get an answer or get transferred to an engineer in the next 20 minutes. Once somebody in Cupertino actually demonstrated a problem I had in my local Apple Store through Apple Remote Desktop and iChat. Ever called Microsoft? I did recently for a server problem and just to take your call they ask $128 (we happen to have an Enterprise agreement but our rep was nowhere to be found) and put me on hold for 1 hour, then picked up again and said the queue was 2-8 hours. They called me back 20 hours later. My manager was grey, because I'm a contractor paid by the hour and as long as that server was down, I couldn't do shit (tik-tok $55 * 6 actual working hours).

Re:Did Apple make a mistake?

[jcr]

How much software is "optimized" for a specific architecture, beyond what the compiler does?

Very little of it, outside of the libraries that Apple provides.

-jcr

Re:Did Apple make a mistake?

[DurendalMac]

Um, that ad was created long before the G4 and G5 stagnated, and yes, the G3 was faster clock-for-clock than a Pentium 2. IBM hit a wall with the G5 and didn't want to make a G6 for Apple. At the same time, Intel got their head out of their asses and put out chips that were actually efficient. It was great timing, and switching was a very smart move.

Re:Did Apple make a mistake?

[jcr]

At the time that ad was made, it was accurate. The PPCs were beating the pentiums quite easily.

They lie in your face,

Nope.

-jcr

Re:Did Apple make a mistake?

[mrchaotica]

As a software developer why you should work twice more...

By "work twice more" you mean "check an option box in XCode," right?

If Apple would keep randomly altering their hardware and require compatibility with a range of completely different architectures, in the end it'll completely alienate the developers.

Yeah, just like how the wide range of different architectures most UNIX software runs on alienates developers...

...oh, wait.

Re:Did Apple make a mistake?

[Anonymous Coward]

The vast majority of people, it turns out, are latent Mac users, and we have no problem with that. We do find it tedious when dweebs like WMF bitch endlessly about Macs not being designed for linear thinkers like himself.

Re:Did Apple make a mistake?

[jcr]

why you should work twice more

Building universal isn't twice the work. Most apps don't have any intrinsic byte-order dependencies, and very few people ever wrote CPU-specific code that depended on Alitvec (for example).

-jcr

Re:Did Apple make a mistake?

[nbvb]

Thanks for bringing back those awful, awful memories.
</sarcasm>

SunOS begone!

Re:Did Apple make a mistake?

[mrchaotica]

and if you don't think it's about profit, ask yourself why all of the low end PPC machines had okay ATI or NVidia graphics, while all of the new low end intel machines have totally bullshit Intel GMA graphics?

Personally, I'm okay with Intel graphics -- at least they've got Free Linux drivers. Of course, that's more of a concern on non-Apple computers, since any Mac I'd own would be running OS X...

Re:Did Apple make a mistake?

[644bd346996]

This is a troll. Universal binaries are not twice as hard to create as architecture-specific code. For almost all programs, there is no extra work needed beyond a checkbox. Apple even provides frameworks for vector code that will use SSEx on Intel and Altivec on PPC. I've yet to hear a good reason why an app couldn't be Universal. The closest I've heard is dealing with Wine based apps (ie. those using Cider). And since those aren't really Mac apps, it doesn't matter.

It could be a very good idea for Apple to release Power6 based XServes while keeping the workstations, desktops, and laptops on Intel. It wouldn't alienate any developers, because they barely have any extra work. The worst that could happen is that a few QA guys gripe.

Re:Did Apple make a mistake?

[suv4x4]

Building universal isn't twice the work. Most apps don't have any intrinsic byte-order dependencies, and very few people ever wrote CPU-specific code that depended on Alitvec (for example).

You can tell that to the Flash developers who worked their ass off to deliver the Intel version of Mac Flash quickly.

That little player has loads of ASM and SIMD instructions to be able to pull off what it does in this size and this speed.

Also you're not accurate about Altivec, multimedia apps like Photoshop make very good use of SIMD extensions, and interestingly enough, so do a big chunk of the audio instrument/processing apps.

Re:Did Apple make a mistake?

[Marillion]

Perhaps they had a sweet deal with Intel that made it the right thing for Apple.

Remember: Apple is out to make Apple shareholders happy. And, of course, the same is true for Microsoft, Sun, Intel, Oracle, IBM, Google, Amazon, and all the rest of them. The fact that from time to time they make a great product for their customers is a means to an end, not the end itself.

Re:Did Apple make a mistake?

[the_humeister]

If Apple had stayed with IBM, they would have been pushed to the Cell processor. And that would be a bad PR move, running on the same CPU as your game consol runs on.

Oh? You mean like how the XBox uses a Pentium 3? How embarrassing for the rest of us that we're just using a chip that's suited for video games.

Re:Did Apple make a mistake?

[mrchaotica]

Look at the Thinkpad X60 (and/or Tablet). They're more expensive than the MacBook, but not by all that much considering the current sale, and if you're a student the tablet is wonderful for taking notes (I love mine!).

Re:Did Apple make a mistake?

[c_forq]

The only Thinkpad in my requirements and price-range is $250 more than a Macbook, with a slower processor, smaller hard drive, and doesn't include bluetooth. (The X60 is $1,251.75 at sale price, and the Macbook is $1,000 after student discount).

Re:Did Apple make a mistake?

[PipsqueakOnAP133]

Dude, all work's been done.

Either, you're doing a mac specific app and use the Accelerate.framework which handles conversion to SSE3 or Altivec depending on the platform... ...or, you'll just pull in all the SSE work you did from the Windows Flash runtime since it's the same chip and these are all not OS dependent.

Same thing for Photoshop. The plugin architecture makes it hella easy since they should have started with plugins for all the heavy stuff anyways. Recycling! It's not just for cans.

Re:Did Apple make a mistake?

[samkass]

It's pretty simple, really: almost no one needs 64-bits yet.

Re:Did Apple make a mistake?

[Jeff DeMaagd]

POWER isn't PowerPC. I don't think it's worth keeping two architectures in parallel like that for long term, not with Apple's current volume. Apple doesn't make that many Xserves last I heard, something like tens of thousands per year, when the other server companies exceed that by as much as 20x. IIRC, HP was selling 200k 1U servers when Apple sold 12k 1U servers. I don't know what IBM's numbers are, but POWER-based workstations and servers were a lot more expensive than Apple's stuff.

Re:Did Apple make a mistake?

[yankI]

Maybe supporting two architectures require extra work for the OS X development at the moment, so they may want to transition to a single architecture as soon as possible.

Re:Did Apple make a mistake?

[Whiney Mac Fanboy]

We do find it tedious when dweebs like WMF bitch endlessly about Macs not being designed for linear thinkers like himself.

I don't bitch about macs tho' - I think they (along with os x) are quite nice. It's purely Apple's hypocricy, DRM pushing, legal threats against bloggers, that I have a problem with.

Re:Did Apple make a mistake?

[Anonymous Coward]

* Well, I'd rather wait until that guy at Ars Technica does one of his in-depth analyses on the POWER6 architecture. His essays on Intel, AMD and Mac-basis CPUs were awesome, and his essays on hardware in general are how I learned that stuff.

Link me.

Re:Did Apple make a mistake?

[Anonymous Coward]

Does this mean you finally accept the utility of distinguishing PCs from Macs in casual conversation?

Considering that was in the same sentence as "Altivec is magic", I think you're taking the sentence a little too seriously...

Re:Did Apple make a mistake?

[sodul]

It cost a huge amount of money to R&D the motherboards.

By switching to Intel Apple not only got a predictable source of CPUs, but also got most of the work done for them for the motherboard designs. If you look at the Mac Pros, their motherboards are almost a copies of Intel boards. I'm sure that having PPC board along the Intel board would at least double their development cost (for the motherboards only), and with no financial advantage.

An other big reason is that Apple will drop support for PPC in a few years, so the longer they keep PPC around, the longer they have to spend money to support it.

Re:Did Apple make a mistake?

[cmacb]

This is a troll. Universal binaries are not twice as hard to create as architecture-specific code. For almost all programs, there is no extra work needed beyond a checkbox.

I don't think it's a troll, I just see evidence that the Apple fanbase can't admit that there was any downsides to whatever Steve comes up with then, now or in the future. It would OBVIOUSLY be better to keep support for multiple processor types (I think that discipline actually makes the code more stable too), allowing Apple to use high-end PowerPC systems for "mainframe" type applications, or even newer low-end units if they indeed beat Intel in whatever area the market is hot. That flexibility is in fact what Apple boasted about with OS X, but I think all the fanbase got out of it was "Intel good, PPC bad, Og no more like PPC, Ugh."

Big computing grids will just get the chips from IBM, other integrators, or (as they have already done) gangs of Playstations. How many large installations of Apple servers are there? Not specialized applications, but whole companies, Ford, GE, GM? Not a lot I don't think and Apple hasn't shown much of an appetite for changing that. Artsy Fartsy gadgets is the future for Apple (and I'm using one right now by the way, but one of the "old" PPC models). I may even get an Intel Apple at some point, but it will have to beat head to head whatever HP Dell, and Sony are offering. I won't pay a premium for what I consider a plebeian "Windoows" machine.

Re:Did Apple make a mistake?

[Anonymous Coward]

I'm assuming you're the person s/he's talking about.
This is Slashdot, we can search and find that post. If you're looking for page rank, relax, Slashdot adds a no-follow to submitted links (last time I checked)..

Re:Did Apple make a mistake?

[jcr]

I think the Macromedia developers themselves would tell you that their work wasn't typical.

-jcr

Re:Did Apple make a mistake?

[mrchaotica]

An other big reason is that Apple will drop support for PPC in a few years

What makes you think that? After all, Apple (secretly) maintained support for x86 all the way from when OS X was still called NeXTStep!

Re:Did Apple make a mistake?

[solios]

Intel graphics are fine if all you run is Office... personally, I can't stand the fact that the chipset's VRAM is actually system RAM, and "shared video memory" doesn't mean "I have a gig of ram! I can run Doom3 with 512 megs of VRAM!" It means "I have X megs of system ram that I can never, ever use for anything but video memory."

Which is really misleading when they're billing a machine that has ~448 megs of useable ram as having 512.

Regardless of the "for" arguments, the fact that Apple - who bills their kit as high-end "luxury" personal computers - is using fleamarket video at maserati prices.

Re:Did Apple make a mistake?

[marafa]

apple uses the powerpc a stripped down version of the power cpu.

Re:Did Apple make a mistake?

[Lobster Cowboy]

Because one of Apple's new selling points is the ability for all new Macs to run Windows as well as OS X. Can you imagine the nightmare Apple would have if they mixed processors?

Re:Did Apple make a mistake?

[Spy Hunter]

The MacBook weighs 5.1 lbs. The ThinkPad X60 weighs 3.1 (small battery; 3.6 with 8-hour battery). That's what you're paying for.

When I was shopping for laptops I wanted to look at Macs, but my weight requirement threw them all out of the running. I ended up getting an X60s which weighs 2.7 lbs (3.3 with big battery), and every time I carry it I'm glad I did. IMHO the long-rumored MacBook Thin can't come soon enough.

Re:Did Apple make a mistake?

[jhol13]

Had he recommended Java there would have been zillions "write once test everywhere" comments.

But apparently Apples "universal" binary does not require any testing whatsoever ...

Re:Did Apple make a mistake?

[mrchaotica]

Intel graphics are fine if all you run is Office...

I've got Intel graphics on my X60, and I'm in the middle of installing a bunch of 3D games in Linux (Tremulous, FlightGear, Scorched3D, Neverball...); I anticipate that it'll run them just fine. It also works really well with Compiz/Beryl. Personally, I think it's a lot better than having an Nvidia or ATI chip, and not having 3D support at all.

Re:Did Apple make a mistake?

[Anonymous Coward]

The only thing more pathetic than a PC user is a PC user trying to be a Mac user. We have a name for you people: switcheurs.

There's a good reason for your vexation at the Mac community's grasp on the unidirectional arrow of time: You don't speak its language. Remember that the Mac was designed by artists, for artists, be they poets, musicians, or avant-garde mathematicians. A shiny new Mac can introduce your frathouse hovel to a modicum of good taste, but it can't make Mac users out of dweebs and squares like you.

So don't force what doesn't come naturally. You'll be much happier if you stick to an OS that suits your personality. And you'll be doing the rest of us a favor, too; you leave Macs to Mac users, and we'll leave beige to you.

Re:Did Apple make a mistake?

[suv4x4]

you'll just pull in all the SSE work you did from the Windows Flash runtime since it's the same chip and these are all not OS dependent.

The argument was about keeping PPC or not. So how do you pull that from SSE on Windows Flash?

Furthermore, if you're a startup, who writes version 1 of a software, where do you "pull" this from?

The accelerate framework is a toy, for serious work, you need to code it manually.

Re:Did Apple make a mistake?

[Anonymous Coward]

Don't need 64 bit? Sure you do!

Do you plan on using more than 3gb of ram in the near future?
Then you'll need to switch to something addressed with more than 32 bits...

(I know I know the technical limit is 4GiB, but it's more like 3.5 if you factor in the overhead, vectors etc)

Re:Did Apple make a mistake?

[Anonymous Coward]

The Power6 devices are extremely expensive and are not suited for the high volume market where x86 is. And a Power6 dual-core die consumes about 200 Watt. They play in a different arena than what Apple want.

Re:Did Apple make a mistake?

[cerberusss]

As Microsoft knows very well, developers, developers, developers are you best asset in this fight.

Nowadays I have to grin sarcastically about that, since I know what hoops you have to jump through to develop for Windows Mobile 6. Not a simple compiler and be done, noooo just download an 8 CD monster and several APIs. Makes me happy that I'm primarily a Linux developer.

Re:Did Apple make a mistake?

[drsmithy]

at I don't understand is, since Mac software has to be Universal nowadays anyway, why Apple doesn't just permanently keep its lineup as a mix of PPC and x86, picking whichever chip suits the particular machine they're designing at the time? Power6 Xserves along side Core 2 laptops... it sounds good to me!

Because the market for such machines is miniscule and costs in designing, testing and mass-producing computers are non-trivial.

(This is before even getting into the additional costs that would be heaped on developers, discouraging them from writing for the platform.)

In other words, it would cost Apple a lot of money while delivering few, if any, advantages (and a whole bunch of disadvantages).

Re:Did Apple make a mistake?

[toplus]

I think you guys are missing the point. The main reason Apple moved to Intel is not because of the performance of the processors, but because of the fresh image and appealing for new consumers and market share. Many people now buy Macs and install Windows, and would not buy the Macs if they could not install Windows on them (as in PPC)

Re:Did Apple make a mistake?

[kestasjk]

They why did it take so long for Adobe and all the other Apple software companies to release universal binaries?

Little/big endianness woes, using 32-bit and 64-bit on both architectures, double the amount of testing, etc. It's no small matter to port a large application to a whole new processor architecture.

Re:Did Apple make a mistake?

[solios]

Try running, say..... Maya. Or Motion.

Ultimately, the point I'm aiming at is that paying premium prices for bargain basement video really chafes my ass - if I'm going to lay down for kit that's twice the price of an equivalently powered wintel box, I'd like some name brand video and user access to all of the system memory.

Re:Did Apple make a mistake?

[drsmithy]

Especially once you go into businesses. The difference in price between Apple and Microsoft starts getting huge once you spec out an environment for >50 people. With Apple you know your clients cost you $129, your server $999, Remote Desktop for $499, no limits, everything integrated with Kerberos + LDAP. With Microsoft you got that and then you have to start calculating CAL's for Exchange, CAL's for your Terminal Server, CAL's and server licenses for your SMS and WUS, and each little piece that will make life easier as the admin costs you an extra CAL which is all included in the Server+Client+RemoteDesktop licenses on Mac.

Of course, the Windows environment you've just described (Exchange, Terminal Server, etc) is _vastly_ more functional than the OS X environment you described, so it's not surprising in the least it would cost more.

Next to that, the average Windows machine lasts 3 years before it get's old and slow. I've seen G3's running OS X 10.4 without a hitch, G4's are mainstream in many companies and most haven't even gone to Intel yet.

If someone is happy with a G3 or G4 running OS X, they'll be just as happy with a P2, P3 or P4 running Windows XP. It's *far* kinder to older hardware than OS X is.

Re:Did Apple make a mistake?

[c_forq]

If I had the budget for one with better specifications I would do the Thinkbook, but $1,000 is at the very top of what I have budgeted. I was looking at a smaller, lighter refurbished Gateway (MX1027) at just under $800, but decided that the extra $200 was worth it for having new hardware, having the Apple Warranty, and having the better specs. I would take a good Thinkpad over the MacBook Pro, but both are way out of my budget.

Re:Did Apple make a mistake?

[goober1473]

Why would apple have been pushed to cell? Power and Cell are two very different things.

Re:Did Apple make a mistake?

[mrchaotica]

Ah! Unlike most of the other replies, you have a really good point!

Re:Did Apple make a mistake?

[mrchaotica]

Well, Maya would still run better on my Intel graphics than it would on an otherwise-equivalent laptop with an Nvidia card using the 'nv' Free Software driver.

But I do get your point; I'm sure my girlfriend (an animator) would love to have a version of my X60 tablet with high-end graphics for running Maya.

Re:Did Apple make a mistake?

[drsmithy]

As the transition sees developers pushing out binaries for both chips, I don't see a downside of Apple straddling the fence and using both types of chips.

That's probably because you don't have to worry about their shareholders.

Re:Did Apple make a mistake?

[mrchaotica]

Did you know that it's possible to configure a $15,000 Mac Pro on Apple's website? With eight cores? Yeah, I think Apple is quite well-acquainted with "extremely expensive" and "low volume!"

Re:Did Apple make a mistake?

[Lost Engineer]

Buy a real compiler. Hell buy two.

I mean who seriously codes much in assembly these days. Even if you do, you can always hire somebody who can code PowerPC assembly like... oh say me.

Re:Did Apple make a mistake?

[Tolkien]

by mrchaotica (681592)

What I don't understand is, since Mac software has to be Universal nowadays anyway, why Apple doesn't just permanently keep its lineup as a mix of PPC and x86, picking whichever chip suits the particular machine they're designing at the time? Power6 Xserves along side Core 2 laptops... it sounds good to me!

by suv4x4 (956391)

As a software developer why you should work twice more (OSX intel / OSX ppc) to produce a piece of software that will work on roughly 2-3 % of the desktop computers out there?
Furthermore, no, being a Universal binary is not a requirement, and I know a few companies which release only Intel versions of their Mac software (example: Adobe's Soundbooth)..

by jcr (53032)

Building universal (???) isn't twice the work.

Dear God, people, my brain nearly exploded reading your posts. Your spell-check may not complain, but at least reread your posts for grammar! Those errors almost made me feel nauseous. It's careless off-the-cuff typing like that, that younger folks (equally influenced by god-awful SMS texting and junk) read and learn from. It's insignificant upon first glance, sure. There is a reason that people who read a lot of books tend to be smarter. They are subconsciously re-enforcing their language skills as a side-effect of reading the author's work (this assumes, of course, that the author is literate).

This, of course, is where you say (the lack of capitalization within the quotes is intended) "hey! language evolves with cultural influences and whatnot"
Well, to that I say: "BUSH". Do you call that culture? ... I also say:

Oh how I loathe that which the Internet has done to our language, and the care with which we speak and write it. We've gone from Shakespeare's Hamlet et al, to "omgd00dwtf!?! i so roxz0rd you're azz, im so much leeter!1! lolroflbbq" It's because of this disgraceful and careless attitude towards literacy that I have believed, and still believe that we will end up slowly degenerating (or reverting, call it what you will) back to a state of cavemen-like communication. Grunting at each other to communicate, all because grunting is so much easier to do. Over the generations, people will have become lazier and lazier with keeping literacy intact; going from long eloquent stanzas of beautiful poetry to express a simple but elegant thought, removing, combining, and abbreviating words throughout the years, until finally we can fit entire thoughts into acronyms and canonical abbreviations (like NASA or IANAL(...)). Eventually we'll get bored of speaking and typing in acronyms and such, instead using more verbose facial expressions and pointing at things with our fingers, and grunting happily or growling when the other person guessed right or wrong what we meant.

Remind you of (m)any animals?

</rant>

Re:Did Apple make a mistake?

[Lost Engineer]

That little player has loads of ASM and SIMD instructions to be able to pull off what it does in this size and this speed. I have heard of Apple fanboys, but I have never heard of a flash fanboy. You are the first person everm *EVER* to refer to flash as anything but bloated and slow.

If there's a lot of assembly in Flash, then Macromedia needs a new freaking profiler.

Re:Did Apple make a mistake?

[fritsd]

It's tricky but not as hard as you make it sound. Look at Debian if you don't believe me: this picture shows what percentage of the programs is compiled for each architecture: stats.png it's usually over 95%. This includes little- and big-endian (mips, mipsel), 32-bit and 64-bit (x86_64), and weirder (s390). Also note the x-axis on the picture runs from the year 2001 :-) And yes, I know, compiled doesn't mean it actually also works :-)

As to why Adobe can't be bothered to create a working flash player for (at least) 64-bit AMD64: I have no idea; we can't see the source so we can't see how difficult it would be to port it.

Re:Did Apple make a mistake?

[Lost Engineer]

Ya know, I have just been making arguments about how easy it is to port to a new architecture, assuming you're not writing machine code and you've been writing decent code, and you ruined it for me.

Your mention of big grids made me think multi-threaded, and porting anything multi-threaded is going to require more than a checkbox. I can hear contrary responses about good code already, but I still think this is a non-starter for Apple supporting two architectures in the future.

Re:Did Apple make a mistake?

[Lost Engineer]

If I could get something the size of an X60 tab with high-end graphics I would bow down and worship the future.

As is stands now, a decent graphics card would draw enough power to require a battery the size of that whole computer.

Re:Did Apple make a mistake?

[cmacb]

Same here. I had just about talked myself into a desktop PPC system (learned to like Apple on this here Powerbook) when they made the announcement. No way am going to buy what is essentially a glorified Windows system just to run OS X. I'd rather just get a high end generic Intel box and go back to Linux (well, I've since done that actually, I just use the Powerbook for travel these days). This will probably be my last Apple computer (unless they change their minds again, which I don' anticipate).

Re:Did Apple make a mistake?

[Anonymous Coward]

yeah, 2-3% (or so) is only 5-20,000,000 PCs... why bother.

Re:Did Apple make a mistake?

[Lost Engineer]

Interesting combination of you and the sibling post:

What do you think are the odds that Microsoft keeps a secret version of Windows that supports PowerPC?

Re:Did Apple make a mistake?

[Anonymous Coward]

It's alot easier to test on a PowerMac and an IntelMac than it is to test on 5 revisions of four Java VMs that might be deployed by any given user. Also, there are truckloads of things you can do with C/C++/ObjectiveC that you can't (or shouldn't) do with Java --like write GPU-intensive videogames. If you want to avoid cross-platform testing, you have to become a PERL, PHP or Ruby jockey just like the poor BASIC and COBOL saps before you.

Re:Did Apple make a mistake?

[sodul]

Even during the pre-10.0 times the PPC on the desktop was in trouble (remember the g4 stuck around 400Mhz for over a year?), so they had to keep an exit door open. Freescale did not deliver on the long term, then came the G5 from IBM, full of promises ("3Ghz by next summer").

Motorola/Freescale spent the money on embeded and cell phone rather than on dektop CPUs. IBM moved their money to the high end servers and the cell... which I think is a HUGE disappointment.

I'm not saying Apple will stop making OS X working with n other CPU. They will do it internally, to force developers to not code specifically for x86 and keep the OS 'open'. I'm talking about publicly supporting the PPC machines. Having a dev version that kind-of works in the back (like the intel versions of OS X 10.0 through 10.3) and a version that works well like 10.4 is a whole lot different, and it took Apple a few month to get a 'stable' OS X Intel.

Re:Did Apple make a mistake?

[pato101]

As a software developer why you should work twice more (OSX intel / OSX ppc) to produce a piece of software that will work on roughly 2-3 % of the desktop computers out there? I understand your point, but my personal experience developing software for several unices is that having several platforms is a good idea since some bugs arise at some platforms and others at other ones, so your software quality gets improved. Also, as already pointed the work is not twice the work but a bit more - and that "a bit more" consists in correcting bugs that happen only at one platform (but which could randomly happen at other one at any time).

Re:Did Apple make a mistake?

[Anonymous Coward]

Uhh, Microsoft can barely support x64. I doubt that they would waste resources on what is essentially a dead platform on the desktop.

Re:Did Apple make a mistake?

[Anonymous Coward]

grunt grunt GRUNT

Re:Did Apple make a mistake?

[rolfwind]

Perhaps, but I see a lot of value in apple having multiple architecture - it doesn't tie them down to any one supplier which was their essential problem with IBM. They won't run into the same problems with Intel (mainstream CPUs is their business) but I could see Apple using alternative architectures to get their foot in the door in niche (but lucrative) markets like servers and graphics where another chip is preferred because of speed or ease of low level programming (without the x86 hacks).

If this is the scenario, they won't even have to offer this to the mainstream market, perhaps to avoid Joe Sixpack support on why every different mac app on the internet won't run, and perhaps to stay sweethearts with Intel.

Apple has transistion sucessfully from the Motorola chips to PowerPC to Intel - they have the expertise to keep a constant "transistion" and Linux proves the viability of multiple architectures.

Who knows, this may not even be bad for Intel - perhaps Apple could have (and in the future) could give Intel an orgasm by making the server products Itanium based.

Re:Did Apple make a mistake?

[jcr]

Those errors almost made me feel nauseous.

Psychopharmacology is making great strides. Maybe they have a pill to cure you of the anxiety you feel when people don't obey your demands.

-jcr

Re:Did Apple make a mistake?

[Ilgaz]

If Apple decided to focus on portable machines and low power entertaintment devices, they didn't make a mistake. Intel really looks viable choice but they shouldn't be exclusive Intel vendor, they should always leave door open for AMD. They are way close to Intel now, too much.

I was just against the horrible level of propaganda against PowerPC by the Boss himself. PowerPC G5 is a very viable/modern processor for Desktop machines/Workstations. People got confused about what Steve Jobs means while advertising Intel switch. He was pointing the future and massive switch from Desktop to Laptop by ordinary people. If we think about G4 and people rendering DVD and soon Blu-Ray on their portables, he has a point. Motorola doesn't seem to care about laptop CPUs and IBM's focus is those server/high end workstation chips. What would happen if Steve Jobs called IBM and asked for a Intel portable competing/beating portable Power chip? IBM would politely say they are focused on high end chip wars, e.g. Sun Niagara.

I am watching the Cell Linux scene and possibilities. It seems that monster could make some amazing media focused computer. PS3 is just an example, imagine a more Desktop thing rather than optimised game console.

While on it, thanks to Slashdot editor who fixed my horrible first time submission.

Re:Did Apple make a mistake?

[TheRaven64]

Universal binaries are not twice as hard to create as architecture-specific code. For almost all programs, there is no extra work needed beyond a checkbox It's not quite that simple. For one, it's not a checkbox, you have to write 'ppc i386' in the architectures box...

On a more serious note, there are some differences. If you use the new vector extensions for GCC then you can write code that compiles to AltiVec or SSE, but if you use the old Apple AltiVec stuff then it's not quite compatible with this. Beyond that, there are some very slight differences between the behaviour of the linker on PowerPC and i386 which can bite you if you're not careful. There are also some subtle differences (read: bugs) in the implementation of Foundation and AppKit.

Re:Did Apple make a mistake?

[TheRaven64]

POWER isn't PowerPC This keeps coming up. Since the POWER3, all POWER chips have been fully compliant with the PowerPC spec. Not all PowerPC chips, however, implement the full POWER instruction set (although it is possible to trap-and-emulate the few remaining ones in software).

Of course, the price of the chip alone for a POWER6 is higher than any Apple made system. They go in real computers (where 'real' is defined as 'more expensive than you can probably afford').

Re:Did Apple make a mistake?

[Ilgaz]

Flash does show signs of horrible quality control which you can see at my Versiontracker post:
http://www.versiontracker.com/php/feedback/article .php?story=20070416075853952

The issue is, Internet Plugin arch is archaic on Apple too as told very nicely and openly by Webkit developers
http://webkit.org/blog/96/background-music/

It was designed in a time while a plugin designed to do vector graphic based UIs and Apps wasn't abused as a media plugin. That is my observation of course and sorry, I might be old fashioned but I still miss the times when Realplayer plugin opened its own UDP ports, switched bandwidth based on user latency and Quicktime extends itself with new codecs.

I am not very sure about the CPU optimisation since even Photoshop Elements V 3.0 comes with its own Altivec support plugin. Yes, the cheap Photoshop and horribly outdated.

Re:Did Apple make a mistake?

[TheRaven64]

OS X (specifically XNU) uses a 4GB/4GB memory split, meaning that the kernel and applications both have separate 4GB address spaces. This means that any 32-bit application can access up to 4GB of memory. Most operating systems use a 1GB/3GB split, which limits the amount of space available for the application to 3GBs, but makes system calls cheaper since they don't need a full context switch (the kernel's address space is mapped into every process's address space, but marked as no-access for unprivilieged code).

Very few individual applications need anything like this much. Video editing and 3D authoring are about the only two I can think of outside highly specialised systems. If you want to use more than 4GB of RAM, you still can on a 32-bit system. Modern x86 chips (i.e. anything newer than the Pentium Pro) support PAE, giving a 36-bit physical address space. This lets you have up to 64GB of RAM, although individual processes can only have 4GB of it each. I used a G5 for a while which had 8GB of RAM (sadly, it wasn't mine, and they wouldn't let me steal it), and it was really hard to get RAM usage above 50%. Even counting the disk cache, the amount of memory in use rarely went over 4GB, so I think it will be a few more machine upgrades before 64GB stops being enough.

By the way, the current generation of Apple laptops are limited to under 4GB of physical memory (3.5GB, as I recall) due to a problem with the Intel north bridge chip, not due to any limitations of the CPU.

On PowerPC, it makes sense to keep as much code 32-bit as possible, since that makes loads and stores of pointers cheaper, and results in faster code. On x86, going to 64-bit gives you a load of extra registers and a few other nice things, so it's not quite so clear-cut.

Re:Did Apple make a mistake?

[Anonymous Coward]

lose is the opposite of win. loose is the opposite of tight.
losing is the opposing of winning.
loosing is what that goatse guy is doing with his butt

Re:Did Apple make a mistake?

[TheRaven64]

But how many ALUs does that cover? FPUs? What about its SIMD instruction capabilities? How long is the pipeline? Is it insanely long a la NetBurst? It's been a few months since I last looked at POWER6 in detail. As I recall, it has more execution units than POWER5 (obviously). It has a VMX SIMD unit (i.e. Altivec-compatible, but without licensing a trademark from Freescale). The pipeline is something like 14 stages; shorter than NetBurst. The biggest thing to keep in mind is that this is an in-order CPU. These days, Intel and AMD are the only people still thinking out-of-order is a good thing; everyone else has decided it doesn't have a good cost/benefit ratio.

Oh, and it has a 4MB L2 cache and 32MB L3. Presumably someone at IBM looked at Itanium and got cache-size envy.

Re:Did Apple make a mistake?

[solios]

NURBS modelling with a tablet? I think my brain just exploded. o_o

Re:Did Apple make a mistake?

[Wolfrider]

--64-bit Linux is a bit faster in my experience. Tested 32- and 64- Ubuntu 6.06 LTS on the same hardware and stayed with the 64.

--Won't have to switch kernels if I upgrade RAM (2GB currently) and haven't run into the OOM killer yet. ;-)

Re:Did Apple make a mistake?

[Kuad]

screen under 14"

You are aware that the 13.3" Macbook is pretty much the size (and weight) of most 14.1" notebooks, yes? For example, A Dell D630 is larger by 5mm x 10mm x 12mm and you get a full 14.1" display at 1440x900. Most importantly, it's 300g lighter.

There are a lot of things to like about the Macbook, but it is not particularly small or light.

Re:Did Apple make a mistake?

[evilviper]

Yeah, just like how the wide range of different architectures most UNIX software runs on alienates developers...

No. Open source *nix software developers write and test on x86/Linux, and release it. That's all. If it's open source, there are others who are interested enough in getting it working on their non-x86 machine that they'll do all the work for you, and send back patches.

Nothing out there, that is even remotely complex, can just be recompiled everywhere. Projects that do support multiple achitectures, and need performance on all of them, like ffmpeg and MPlayer, put tremendous amounts of work into optimizations and compatibility for each architecture.

Re:Did Apple make a mistake?

[evilviper]

Personally, I think it's a lot better than having an Nvidia or ATI chip, and not having 3D support at all.

We're talking about OS X here. It's all closed-source binaries. Complaining about license incompatibilities between binary drivers and X11 is complete crap.

Re:Did Apple make a mistake?

[evilviper]

And yes, I know, compiled doesn't mean it actually also works :-)

More than that, the fact that 95% compiles doesn't mean there aren't an army of 3rd parties going through the effort of making significant code changes, and submitting them back upstream. I've done it myself many times. Complex code is never just a recompile to get it to work elsewhere.

Re:Did Apple make a mistake?

[redcane]

They pull it from the windows version of flash they already wrote. The OS library calls will be different but they should be able to share large chunks of assembly programming they have for x86 on windows with MacOS on x86, assuming they did any assembly in the first place..... If your a startup, and you wrote version 1 of your software without thinking about cross platform cross-compiling or porting, and you wrote it for MacOS, your obviously only after the Mac segment of the market.

Re:Did Apple make a mistake?

[Anonymous Coward]

You guys are using a copyrighted 8-stripe IBM logo for this story. As far as I know, from working in IBM, this is not allowed.

Re:Did Apple make a mistake?

[Mike89]

I'm in the middle of installing a bunch of 3D games in Linux

Like Tuxracer?

Relax, relax, I'm teasing not trolling ;)

Re:Did Apple make a mistake?

[Brickwall]

You can't carry 5.1 lbs? Is this the ultimate definition of a "girly man"? Go to a gym, you wimp.

Re:Did Apple make a mistake?

[Anonymous Coward]

And you don't think being able to use a right mouse button on an airplane makes the Thinkpad worth it?

Re:Did Apple make a mistake?

[JonLatane]

I'll agree with you that UBs are the new fat binaries, and that the Intel chips are probably increasing their profit margins, since they're cheaper. But your assertions about graphics are rather... ridiculous.

First off, first-gen Intel GMA graphics are way better than the 16-32MB ATI 7xxx series graphics in the old low-end PPC Macs. And current-gen GMA blows them out of the water. I think nostalgia is clouding your judgment here. I mean, seriously, you can play WoW and whatever else you want with the GMA 950 included in a Macbook. Maybe no Doom 3 or Quake 4, but that's reasonable to me.

And as far as passing the savings on goes... only the Macbook and the cheapest iMac have Intel GMA graphics. Go ahead, try to price a Dell E1405 (the closest thing to a Macbook) configured similarly. I built one like the high-end white Macbook (1GB RAM, WXGA display is near 1280x800, 120GB 5400RPM HD, 2.0GHz C2D) and even without the least-crippled version of Vista, it costs more than the thinner, lighter Macbook. If you want Vista Ultimate, it's more than the black Macbook.

So actually, I would say Apple is passing the savings on to the consumer. More so than even Dell, apparently.

Re:Did Apple make a mistake?

[Junks Jerzey]

Ultimately, the point I'm aiming at is that paying premium prices for bargain basement video really chafes my ass - if I'm going to lay down for kit that's twice the price of an equivalently powered wintel box, I'd like some name brand video and user access to all of the system memory.

Higher-end name brand video = more power consumption + more heat. Neither of these are things you want in a small form factor notebook.

I own a MacBook and I've honestly had no issues with video speed. Remember that the Intel motherboard chipset we're talking about here is NOT the same as motherboard chipsets from 6 years ago.

Re:Did Apple make a mistake?

[nurb432]

I totally agree.

Thanks Steve, for screwing things up for us and giving us 'yet another PC'. I will be keeping my G5 until it melts down.

Re:Did Apple make a mistake?

[Anonymous Coward]

And if your application doesn't use Xcode? Depending on what build system you use it can be a big pain in the ass to just set up the cross build, plus all libraries you use must also be universal.

There's also lots of software that builds fine on different architectures, but don't actually work correctly on anything except whatever it was written on. Finding and fixing all the issues can be very time-consuming, and is definitely more involving than setting an option in Xcode.

Re:Did Apple make a mistake?

[mrchaotica]

Nah, I skipped that one -- there's not actually any "competitive" mode to it, so it's boring. Also, it annoys me that Ubuntu apparently renamed it "Planet Penguin Racer."

Re:Did Apple make a mistake?

[Tolkien]

It's not about people obeying my demands as you put it. It's simply about caring for the English language and promoting its well-being. Do you want the next authored epic poem to be written in SMS-speak?

Honestly!

Re:Did Apple make a mistake?

[shelterpaw]

For several reasons. One, they want universal applications so people who's print shops and other long-term hardware investments don't become useless (lawsuit, class action) in the short-term; that's 4 to 6 years. Second, their intentions are to leave PPC out of the mix completely, which they've done with hardware. Third, confusion, if they kept both in the mix customers and reseller employees would have a tough time explaining the difference and why some applications work on one platform and not the other.

Over the long haul they'll drop PPC support and state that it's too expensive to maintain both builds. The reality is they know that old hardware will need to be upgraded and people will have one choice. Plus, IBM can't be relied upon to keep up with the pace of Intel/AMD when process upgrades are concerned.

Re:Did Apple make a mistake?

[darkwhite]

Sorry, but just because you can run some tiny games that use a laughably miniscule fraction of modern GPUs' capacity doesn't mean you have decent 3D. X60 rocks, and Intel's CPUs rock, but Intel has yet to produce a decent GPU... and nVidia/ATI do have drivers for their high-end GPUs, even if they're closed source and ATI sucks all around lately.

Re:Did Apple make a mistake?

[c_forq]

And again besides wieght the MacBook has better specs for the price than the Dell. After adding the 2.0 Ghz Processor, 1 GB of RAM, a bluetooth module, and putting the harddrive to 80 GB it comes out to be more expensive than the MacBook, although not by much. And after all the included Dell Reminder and Trail crap that came on my last Dell I am a bit hesitant to go with them again. Also the MacBook has added perks of the Apple Remote and a built-in webcam (not things I am to concerned about, but that tip the scale if it is exactly equal with another model).

Re:Did Apple make a mistake?

[RockoTDF]

They did make a version of NT back in the day that would run on PPC.

Re:Did Apple make a mistake?

[RockoTDF]

Uhh, I run an ibook with a 1.33Ghz G4 and a gig of ram and I know that it runs smoother than say, a PIII at the same clock speed. What apple does is they make the new OS X releases more RAM and less CPU intensive, so older models can last a lot longer since a RAM upgrade is easier than buying a new machine or upgrading the CPU.

Re:Did Apple make a mistake?

[drinkypoo]

totally bullshit Intel GMA graphics?

Can you explain what is totally bullshit about integrated graphics? Yes, they're slower. If you're buying a low-end machine, you don't need the latest and greatest.

Re:Did Apple make a mistake?

[mrchaotica]

nVidia/ATI do have drivers for their high-end GPUs, even if they're closed source...

Right, so they actually don't have drivers, then.

Re:Did Apple make a mistake?

[drinkypoo]

This keeps coming up. Since the POWER3, all POWER chips have been fully compliant with the PowerPC spec. Not all PowerPC chips, however, implement the full POWER instruction set (although it is possible to trap-and-emulate the few remaining ones in software).

AFAIK the only PowerPC to support the then-current POWER instruction set was the original, the PPC601.

PowerPC chips have tended to have higher clocks and less functional units than the POWER processors upon which they are based. Be interesting to see what happens with the next gen.

Re:Did Apple make a mistake?

[RockoTDF]

Cell is still related to POWER in the same sense that PPC is....kind of like AMD64 still being an x86 processor, etc.

Anyway, there is a video out there somewhere of the Cell in a workstation at a trade show demonstrating 3d rendering and stuff, and its quite clear that it is running OS X. So from that there is probably a behind-the-scenes binary sort of like there was with x86 for all those years.

Re:Did Apple make a mistake?

[Nossie]

Excuse me but do you mind if I ask whose English you speak? :)

Re:Did Apple make a mistake?

[aliquis]

I guess because it's harder to optimize for any particular arch.

Re:Did Apple make a mistake?

[Anonymous Coward]

why Apple doesn't just permanently keep its lineup as a mix of PPC and x86

There is no technical reason why they couldn't. The Rosetta software (really QuickTransit, which Apple licences) translates in the opposite direction too, and does it well -- IBM's own literature touts the ability to run x86 code on their POWER ISA chips.

Most apps will want to support legacy PPC machines anyway -- there are a lot out there -- and new apps from scratch built with Xcode are usually pretty much ready to be built as a Universal Binary as they are.

Apple might not want to do all the work of maintaining their own motherboard designs for the POWER architecture for cost reasons, and since it's likely that x86-64 ISA chips in the marketplace will have better price-performance and performance-per-watt for the foreseeable future anyway, that work could be wasted. These are cost and risk management issues, though, not technical barriers.

Re:Did Apple make a mistake?

[stephentyrone]

What do you need that Accelerate.framework doesn't provide?

Have you filed a feature request? http://developer.apple.com/bugreporter/
You can use a free developer connection account to do so. If it's a feature that could be useful to multiple developers, there's a decent chance it will be added.

Re:Did Apple make a mistake?

[aliquis]

It's not like it would be two times as much work thought. Probably once they have made it compatible one time it would more or less be no extra work at all. Thought harder to optimize for anything.

Re:Did Apple make a mistake?

[Anonymous Coward]

Do you want the next authored epic poem to be written in SMS-speak?

Why not? It worked for e e cummings!

Re:Did Apple make a mistake?

[Jeremi]

But apparently Apples "universal" binary does not require any testing whatsoever ...

It requires testing two places... once on an Intel Mac, once on a PowerPC Mac. And 99% of the time, that's just to double-check ... most things "just work".

So no, it's not completely zero-overhead, but it's not the n-platform testing nightmare that Java was (is?) either.

Re:Did Apple make a mistake?

[treeves]

I'm not a big [J.R.R.] Tolkien fan, but others may know whether it is indeed ironic that someone with the nick Tolkien should be modded Troll.

Re:Did Apple make a mistake?

[kimvette]

That would be ideal because then it would give Apple leverage over the chip producers. Right now they are practically married to Intel. In the inevitable divorce, whom are they going to turn to for the rebound relationship? AMD? I think you're right, they could have different products at different levels using different processors, because if they keep the kernel and drivers current on both platforms, they're better-equipped to deal with supplier issues and to apply leverage when necessary.

Re:Did Apple make a mistake?

[Anonymous Coward]

They did wait, and there was nothing in the PPC/POWER space which was suitable. Thus the (delayed) introduction of the Mac Pros following the Power Mac Quad G5s, which were released after the Macbooks.

They also waited on Core 2 so that they could use x86-64 (EMT64T) and never have to use or support IA32. Unfortunately that didn't work out either. Despite the pain of having to use a third architecture after all, adopting Core (IA32-only) was judged to be smarter than yet more waiting (on Intel this time, instead of IBM or Freescale).

You better believe that Apple will do everything in its power (no pun intended) to avoid ever having to wait on a CPU vendor again.

Re:Did Apple make a mistake?

[guruevi]

What exactly is more functional? Each and every item you will cite, I can give you a working and most of the time already shipped alternative for. Of course it's not going to be Exchange but it's going to have the same functionality.

Re:Did Apple make a mistake?

[drsmithy]

What exactly is more functional?

Exchange, Terminal Server (Apple hasn't even *vaguely suggested* that they might have a competitor to TS), SMS, WSUS (WSUS is free, by the way, so you don't need to worry about "CALs" for it).

Of course it's not going to be Exchange but it's going to have the same functionality.

Nothing that comes with OS X Server, as far as I know, has the functionality of Exchange. At least, not yet. Exchange *is* more than just sending and receiving email, you know.

Re:Did Apple make a mistake?

[drsmithy]

Uhh, I run an ibook with a 1.33Ghz G4 and a gig of ram and I know that it runs smoother than say, a PIII at the same clock speed.

Funny, because I've got a 768MB, 1Ghz G4 iBook and it's painful to do anything more than, say, watch a DVD, check email, *very* basic web browsing or pull photos off my camera. Which is ok from the perspective of that's all I bought it to do, but I also have an old ~550Mhz P3 laptop that absolutely stomps all over it in terms of responsiveness. I could grudgingly use the P3 full time (with the exception of games), whereas even the light usage I subject my iBook to frequently becomes an exercise in frustration of beachballs and UI pauses.

Heck, my Mum's G5 iMac, with a gig of RAM, still has far too many annoying UI pauses for my liking - and it's easily twice as powerful as my iBook. An equivalent costing (at the time) PC, however, runs XP blazingly fast.

OS X is just slow. I've used a lot of Macs and a lot of versions of OS X (since a Beige G3 running "Rhapsody"), and the only ones that have even come close to being able to run it well are the dual G5 and Intel machines.

Re:Did Apple make a mistake?

[bulliver]

* Well, I'd rather wait until that guy at Ars Technica does one of his in-depth analyses on the POWER6 architecture.

You mean Jon Stokes? He's got this writeup posted earlier today, though it isn't exactly indepth. Apropos, I'm halfway through a new book of his right now: "Inside the machine". Pretty good for a non-EE/CS duffer like me...

Re:Did Apple make a mistake?

[fatphil]

There's still very little need for assembly in most fields. Instruction sets are fairly compiler friendly, and compilers are fairly good nowadays. (Though XCode's codegen always fails to use 1 FPU register, which is a right pain.)

The last bit of FPU-intensive C code I wrote, when I threw it at Freescale's analysis tools, absolutely saturated
every single floating point pipeline so that hand-coded asm code _could not have been faster_ than my fairly trivial, and fairly portable C.

I was in some ways very happy, but I had also been looking forward to actually learning how to code non-trivial PPC assembly loops, so was also disappointed.

Re:Did Apple make a mistake?

[guruevi]

Exchange IMAP integrated into the domain. And yes, you can use calendars and share calendars, e-mail and more using IMAP (that's what the protocol was made for, that's how Microsoft did Exchange). There's also other good alternatives to Exchange that are free if you really need all that functionality, but I have seen few implementations that are more than sending/receiving e-mail and appointments (which currently Mail and iCal work very nicely together in a 'domain', sending out invites, accepting appointments into the calendar). Next version, they should have an even better alternative. The current Apple Server also has iChat server and check out this for calendaring: http://trac.macosforge.org/projects/calendarserver . The difference between Exchange and the rest is that Exchange packs everything, even the stuff you might not need all together as does the client (Outlook). The rest of the world keeps their stuff separated.

Terminal Server VNC/ARD/X (there's even alternatives to that, again for free)

SMS Workgroup Manager and ARD have all that functionality (remote control, patch management, software distribution, and hardware and software inventory), you can also patch your local images for netboot without actually booting them and then use netboot for all your clients instead of having a local disk.

WSUS Update Server sits in Apple Server

Re:Did Apple make a mistake?

[jcr]

Do you want the next authored epic poem to be written in SMS-speak?

That's not for me or for you to decide. An author's free to write whatever they want, and your approval is not required.

-jcr

Re:Did Apple make a mistake?

[Chanc_Gorkon]

You are correct. IBM had Power 5 CPU's for their Bladecenter chassis. Although with virtualization, Blades may be come a thing of the past. It's now increasingly common to see pSeries shops running many LPARS on a p570 or 590 chassis. Anyway, there's no reason Apple could not have put one of these blade p5's in a MacBook Pro or MacBook. Actually, if they did that, it would still be a Powerbook. On the other side of the coin, however, the price would be much higher then MacBooks and MacBook pro's currently are.

Re:Did Apple make a mistake?

[drsmithy]

Terminal Server VNC/ARD/X (there's even alternatives to that, again for free)

I don't mean to be rude, but you don't really appear to know what you're talking about (the above being the most obvious example in your reply to demonstrate this).

Re:Did Apple make a mistake?

[Lost Engineer]

Instruction sets are fairly compiler friendly As a guy who works on compilers frequently (as in it's my job), I got a kick out of that. You're right though -- they're certainly more friendly these days to the compilers than they are to us people. If you want a good example, dump some assembly from a cross-compiler targeted at Analog Devices Blackfin series (gcc supports it.) One look and I thought, when I am replaced by a robot, this is like the language it will speak.

Re:Did Apple make a mistake?

[fatphil]

I think the syntax used can be abominable, but the actual instruction sets are human friendly too most of the time, particularly on RISC architectures. x86 would have to be the ultimate totally-bollocksed-up architecture, yet funnily enough it's the one I most frequently asm in.

I've never understood the gcc -S or objdump output from HPPA, Spark, Power, or Alpha, but consider myself familiar enough with the workings of the processor that I'm a pretty nifty coder for them. (#registers, approximate latencies, load/store issues, existance of conditional operations, branch overheads - if you know those you can code very tight C, with just the right amount of loop unrolling etc. .) No doubt a blackfin would look just as noisy to me!

Z80 was the last instruction set where it was bloody obvious what everything did (apart from the wacky nybble-based rotate/swap instruction).

Re:Did Apple make a mistake?

[Kuad]

Latitudes are the "professional" line and don't come with that crap. (By "crap", I mean both the Dell software and the Apple remote and webcam) Having used both, I feel it's *well* worth it to have a larger, higher-resolution screen and less weight. Having a keyboard that's not annoying is a fringe benefit.

Naturally if your requirement is running MacOS, the Dell is useless and the Apple represents infinitely better value. I merely objected to the notion that the Macbook is particularly small or light.

Yeah, but...

[wcspxyx]

...does it run Vista?

Re:Yeah, but...

[Anonymous Coward]

> ...does it run Vista?

Barely.

Re:Yeah, but...

[mdhoover]

Heh, what does?
/me runs

Re:Yeah, but...

[joek1010]

No, it doesn't meet minimum system requirements.

nice

[wizardforce]

an IBM p570 server showed an average response time of .625 seconds when handling requests from 2,100 users. That compares to a p570 with 2.2GHz Power5+ chips that handled .983 requests per second for 2,000 users.

not bad, they were about due for a speed increase from those standard 3Ghz CPUs.

4.7? sounds familiar

[Anonymous Coward]

This shit does as many cycles per second as there are bytes on a single-layer dvd.

Re:4.7? sounds familiar

[TheRaven64]

It's an interesting number for another reason. The original IBM PC had a 4.77MHz CPU. Now IBM have one that cycles 1,000 times faster (and does a huge amount more work per clock).

Re:4.7? sounds familiar

[sensei moreh]

4.7? As in 4.77MHz? Those were the days.

I want one

[urban_warrior]

feel the power!

THANK GOD NO

[Vandilizer]

enough said.

Power isn't PPC

[Sycraft-fu]

Despite the similar name, and somewhat related architecture, IBM's Power line are not PPC chips and aren't suited for desktop use. That's not to say that some technologies from them can't go in to other chips, but drooling over what is essentially a minicomputer/mainframe chip is silly.

The reason Apple switched is because, despite all the hype, Intel continues to make really fast chips for a good price. When Apple was on PPC I saw never ending arguments as to how much faster the chips were. All those never seemed to pan out in actual operation. Why that's the case isn't important from Apple's standpoint, they just want fast chips for low cost.

I suppose if you want to long for the days of Altivec and talking about tech stuff you don't fully understand, that's great, however Apple has to be a bit more pragmatic and realise that while Altivec might sound cooler than SSE3, SSE3 is an API for a damn fast vector unit and that's all that really matters. Most people don't care about contrived benchmarks, they care about the wall clock benchmark, meaning how fast does the system do what they want, and further how cheap can they get that system for.

Re:Power isn't PPC

[DreadSpoon]

"IBM's Power line are not PPC chips and aren't suited for desktop use"

Yes, yes they are PPC chips. In terms of core instruction set, they're the same. The PPC970 that Apple used for a short while were derived directly from the Power design, as I recall.

The PC in PowerPC doesn't mean "Personal Computer." It means "Performance Computing." PPC is an instruction set, and Power is an IBM brand/product name. Many companies make PPC chips besides IBM, and the majority of those chips are embedded chips not at all designed for usage in a PC.

Re:Power isn't PPC

[Anonymous Coward]

But that still doesn't change the fact that they're not the same thing. YFI.

Re:Power isn't PPC

[timeOday]

Despite the similar name, and somewhat related architecture, IBM's Power line are not PPC chips and aren't suited for desktop use.

Meaning what? For years, Sun and Silicon Graphics tried to convince everybody that their inherently "high end" hardware was in no way comparable to mere PC hardware, no matter the benchmarks. Eventually everybody woke up and realized it was hogwash.

Re:Power isn't PPC

[1lus10n]

I'm still waiting on the 128 way Intel/AMD ... or anything greater than 16 way that can keep up with the RISC, Sparc or POWER based systems.

When your running apache it doesnt make a difference. If you can get onto Oracle RAC then it will matter less, but for right now there is still a ton of business to be done on the high end of things. Sun's T1 chip is also a metric fuckton better and running web apps. Especially java. 32 threads, low power and so on.

x86 has always been designed for mass use, and as such will usually lose to specialized chips. (see: cell, power6, niagra/t1 and so on)

Re:Power isn't PPC

[PipsqueakOnAP133]

SSE3 damn fast?
Even Apple's dev docs mention to developers who are planning on using SSE that there will be plenty of problems:
http://developer.apple.com/documentation/Performan ce/Conceptual/Accelerate_sse_migration/migration_s se_translation/chapter_4_section_8.html

From what I recall, there were discussions which mention that moving from well-optimized Altivec code to well-optimized SSE3 code will result in a significant performance drop.

Re:Power isn't PPC

[Usquebaugh]

No they are not!

Power != PowerPC

PowerPC has a subset of the Power op codes.

PowerPC is seen as a embedded/desktop platform
Power is used in AS/400 and RS/6000 boxes

Power code does not run on the PowerPC, lack of certain op codes

I'm sure a better explanation is available on wiki, but they are not the same.

Re:Power isn't PPC

[Anonymous Coward]

Right. The wikipedia explanation says: The POWER3 and subsequent microprocessors in the POWER series all implement the full 64-bit PowerPC architecture. The POWER3 and above don't implement any of the old POWER instructions that were removed from the ISA when the PowerPC ISA came out or any of the POWER2 extensions such as lfq or stfq.

Re:Power isn't PPC

[Anonymous Coward]

Have you really done any serious computing with PPC chips?If you would like to know more,see this:http://www.pixelglow.com/stories/macintel-fas ter-than-altivec/

Re:Power isn't PPC

[raftpeople]

Actually the benchmarks show a different picture. Hardware designed for server computing is not the same as hardware that is designed for commodity use (e.g. home/low cost).

Re:Power isn't PPC

[Anonymous Coward]

No they are not!

Are too are too! Uh-huh! (Aah, second grade was so much fun...)

I'm sure a better explanation is available on wiki, but they are not the same.

Did you even try looking it up before saying this?

Starting with the POWER3, released in 1998, "POWER processors implemented the full 64-bit PowerPC and POWER instruction sets, so that there were no longer any IBM processors that implemented only POWER or only POWER2".

So claiming "Power != PowerPC" was true 10 years ago, but not today.

PowerPC is seen as a embedded/desktop platform

Exactly: "seen as". It's just perception. By saying "POWER6" they're trying to give you the impression that this will handle your enterprisey needs, but that wimpy rack of PowerPC servers won't. I guess it's working.

Re:Power isn't PPC

[TheRaven64]

SSE3 isn't so bad. The biggest thing AltiVec had going for it was the vector permute instruction, which had a lot of general uses. SSE3 introduced something similar, although slightly less flexible. The biggest problem I've found with SSE3 is that you can't cheaply branch from a vector comparison. All vector comparisons on AltiVec set a condition register and could be used for branching directly. Doing the same on SSE seems to require copying the result to a scalar register and branching on a compare-with-zero on that.

SSE4, rumour has it, will include scatter-gather instructions. This will make auto-vectorisation a lot simpler. Currently, the compiler can't do a lot of potential optimisations due to the way data is laid out (note: this is only true of C-like languages. Sane language which allow the compiler to re-arrange data don't have this problem). With SSE4, you should be able to load four scalars from anywhere in memory, operate on them, and then store them back.

Re:Power isn't PPC

[timeOday]

I would hope so. I am simply asking, where are those benchmarks?

Re:Power isn't PPC

[644bd346996]

So, if the IBM Power6 chips implement the full POWER ISA (which has the PowerPC ISA as a strict subset), then doesn't that make Power6 chips PPCs (and then some)? Since the IBM Power6 chips can be used to run any software that is compiled for PowerPC, it isn't actually wrong to call them PPCs or PPC compatible. It isn't any more wrong than calling a square a parallelogram.

Re:Power isn't PPC

[raftpeople]

The short answer is the benchmarks are all over the internet. For example, look at the OLTP results at WWW.TPC.ORG, you will notice that the top 10 only have Itanium and Power processors. If x86 processors performed the same as Itanium and Power, don't you think you would see systems with x86 at the top of that list?

If you dive into the detailed specs of the various processors you will see their strengths and weaknesses, there is no "one size fits all".

Re:Power isn't PPC

[Chanc_Gorkon]

What is this RS/6000 you speak of? IBM hasn't called POWER based servers RS/6000 for years. It's pSeries now.

Re:Power isn't PPC

[mink]

You forgot "System P".

watts

[DreadSpoon]

The Power6 uses "under 100 watts in performance sensitive applications."

WAAAY too much for a notebook or a mini.

Re:watts

[twitter]

The Power6 uses "under 100 watts in performance sensitive applications." WAAAY too much for a notebook or a mini.

So how slim can you make a Vista premium ultra ready dual core hamburger grill? Don't those things suck down 300 W at idle before you turn on that clumsy enlightenment knock off GUI? Power has always sucked less power per flop, I'm not sure why this one would be different, except for the screaming clock speed.

Re:watts

[mrchaotica]

So how slim can you make a Vista premium ultra ready dual core hamburger grill? Don't those things suck down 300 W at idle before you turn on that clumsy enlightenment knock off GUI?

Well, Aero Glass works on a Core Duo Mac Mini, so I'd have to say the answer is "pretty damn slim."

Re:watts

[The Bungi]

hi twitter. I just replied to your FUD here, but I wanted to make sure you get a direct link to the PDF file that contains the final CERT report on the outage:

https://reports.energy.gov/BlackoutFinal-Web.pdf

Again, page 133-134.

I'm really looking forward to watching you rationalize this one away.

pipeline length

[Anubis350]

How long is the pipeline? Is it insanely long a la NetBurst?

Isn't IBM one of the proponents of hypertransport, a la AMD64...? In that case, I seriously doubt a long pipeline here

Re:pipeline length

[Wesley Felter]

Everyone except Intel is a proponent of HyperTransport, but that doesn't mean that Power6 will use it. And what does that have to do with pipeline length?

Yeah, but...

[llamalicious]

... does it blend?

That's old school

[Burdell]

I had a 4.77MHz IBM years ago. Oh wait, you said G, not M.

Re:That's old school

[olman]

Yeah. Just think about it. Took more than 25 years to get from there to here. I'll be disappointed if I can't buy my 4.7THz processor for my 50th birthday (in 17 years)

can you imagine...

[zyrorl]

a beowulf cluster of these??!?!?!?! omg.

Re:can you imagine... Yeah. It's called PERCS...

[Anonymous Coward]

Yeah. It's called PERCS. - http://domino.research.ibm.com/comm/pr.nsf/pages/n ews.20030710_darpa.html.

Ho hum

[Disharmony2012]

Why are consumer-end processors still stagnating under 5 GHZ?

Re:Ho hum

[Anonymous Coward]

AMD and Intel are focusing more on optimizing their processors and not bumping up the clock speed.

This is illogical, however. What happens when your mHz get stolen?

Re:Ho hum

[SeaFox]

Because we can do no Moore at the moment.

Re:Ho hum

[ricree]

I know that was just a joke, but I would like to point out that Moore's law is still continuing just fine at the moment. Most consumer processor designers have decided that instead of using the extra transistor density to increase speed, to use it for all these multi core chips that have been produced the past couple years.

Re:Ho hum

[Icarus1919]

Because most of us who have desktop pcs and stuff aren't running weather simulations or fragging at the highest possible FPS. My desktop runs at a little under 3ghz and it's just fine for me thank you. Most other people I know don't need that much power either.

Re:Ho hum

[Anonymous Coward]

Yeah I believe the business logic is as follows:
1) Compilers still suck at finding things to do in parallel, so why bother adding another 20 FPU's to the CPU?
2) So instead add a bunch of new cores to the computers
3) Convince your customers they need to RIP one movie, download another, play their mp3's and run Oblivion with 200fps simultaneously
4) Rinse and repeat as necessary, 2->4->8 cores

Re:Ho hum

[Just Some Guy]

Why are consumer-end processors still stagnating under 5 GHZ?

Why are consumers so hung up on only one facet of processor design at the expense of ignoring others? Clocks per second is only one part of the throughput equation. Instructions per clock is equally important, and new CPUs are vastly better here than their predecessors. Who cares if your CPU is only running at 100MHz if it can still retire 10 billion instructions per second?

Re:Ho hum

[FuturePastNow]

Because we're not willing several thousand dollars for a processor.

Re:Ho hum

[FuturePastNow]

*to pay* /preview

Re:Ho hum

[DrDitto]

Why are consumer-end processors still stagnating under 5 GHZ?

Cooling. Power. Cost.

Re:Ho hum

[PipsqueakOnAP133]

Cuz we can't cool them off fast enough without a refrigerant or peltier stacks?

Seriously, it's like high rpm engines and friction. Clock it higher, heat it faster, wear it out sooner.
Torque is good. It's all about the IPC.

Too bad for you, it's easier to lube a engine than a toggle a transistor with less waste heat.

Re:Ho hum

[Anonymous Coward]

Because a fair amount of them can tell one number is higher than another. Too many numbers or options are way over their heads.

Re:Ho hum

[wannabgeek]

3) Convince your customers they need to RIP one movie, download another, play their mp3's and run Oblivion with 200fps simultaneously

Nice troll, but almost every computer now runs more than one process. The instant messages, email applications, browser at the very least. And not to mention, anti-virus. So I do think dual core chips are useful.

Re:Ho hum

[twistedcubic]

and noise

Re:Ho hum

[Lost Engineer]

So I do think dual core chips are useful. You're right, but how far can we scale this? I really think the 8-core chips of the future are worthless for the desktop. It's just so much easier to write efficient programs than it is to exploit that sort of parallelism. Then again, you could argue we don't even need all the cycles we have.

Re:Ho hum

[Anonymous Coward]

Why are consumers so hung up on only one facet of processor design at the expense of ignoring others?

Because Intel's "Megahertz is king!" marketing for the past 10 years or so worked very well. Now they dug themselves a hole and have to spend the next 10 years reeducating their customers.

Re:Ho hum

[mattcasters]

That answer doesn't sound right.
An Intel Core 2 Duo consumes 65W at 2.66Ghz.
The Power6 consumes 100W.

As such I'm sure an extra fan will be needed at the most.

Re:Ho hum

[Anonymous Coward]

I agree with your point, just remarking that we cannot really compare these two wattages on equal footing yet.

The 100 W figure is just an estimate thrown around by marketing. We'll have to wait and see what some 4.7 GHz chip actually draws under full load -- and I want a techie to measure and report it.

Moreover, IBM and Intel calculate the "thermal design power" somewhat differently, as was thoroughly discussed over at Ars Technica forums in the endless "G5 versus..." threads; Intel's figure is bound to be slightly higher for the same consumption (more "worst case") than IBM's.

However, as said, I agree with your point. The Power6 won't be far off from the worst iterations of Pentium 4 regarding power (and cooling and noise). Warmly remembering early Prescott steppings...

However, the in-order design is intriguing -- how much of a weakness is it in the end? And was it at all inspired by the simple core design in PS3, XB360, Wii?

Re:Ho hum

[yfarren]

Honestly, from what I understand, what you just said, sounds like it makes sense, but doesn't.

Processor SPEED is a determined by the (maximum length of a normal pipeline stage (not including things like floating point handled by a separate unit or stalls etc.)) * (number of pipeline stages) . So, when you increase the transistor density, you are making the transistors smaller/faster. How many cores you put on a chip, shouldn't really effect the speed of the chip (unless you are making the pipeline stages longer, by putting on more cores -- which, shouldn't really be happening, unless the cores are designed to interact somehow, but I don't think they are, at that level).

So, yes, people are choosing to put more cores on a chip. But that, in and of itself, shouldn't make each core slower/faster. That is still determined by the longest pipeline stage. Which, all other things being equal, should get faster, as transistor density goes up (i.e. the longest pipeline stage has a certain number of transistors. Each transistor has a certain speed at which a signal propagates. As the transistors get smaller (and transistor density goes up) it takes less time for a signal to propagate through that transistor. Making that pipeline stage take less time. Making the chip faster).

That is, the extra density, in and of itself, should make the chips faster. The fact that you can now do more things, with those extra transistors, shouldn't slow the chip down.

Re:Ho hum

[thommym]

You're right, but how far can we scale this? I really think the 8-core chips of the future are worthless for the desktop. It's just so much easier to write efficient programs than it is to exploit that sort of parallelism. Then again, you could argue we don't even need all the cycles we have. 8 core is not future. They've been here a year or so in production. Ask Sun... Maybe not for desktop, but anyway.

Re:Ho hum

[Anonymous Coward]

Because Intel cancelled Tejas when Dell/HP/etc. decided they didn't want to spend big $$$ on CPU cooling?

Re:Ho hum

[jozmala]

The shrinking transistors decrease transistor latencies, shrinking lines increase line latencies for given length.
From the beginning of microprocessors to pentium 4 era we have been able to trust newer processes improving the frequency by good margin, but nowadays the transistor latencies are far less important in determining the frequency. At same time when frequencies increased there was also increase in power consumption. In very early days those where under 1 watt chips.
We have hit wall there by capping the amount of power majority of people are willing to accept from their computer. Then there was also by reducing amount of transitors per pipeline stage. But now we are in part where shortening pipeline stages doesn't bring real world performance. As we can see in late P4 desings which were with overly long pipeline.
We have hit LOTS of walls all the same time. Now adding more parallerism brings more performance than hunting the last bits of sequential performance.

Re:Ho hum

[Hatta]

People have been saying the exact same thing since at least the 500mhz days. What's changed?

Re:Ho hum

[darkwhite]

That is all nice and well, and Moore's law as stated by Carver Mead still holds and will hold for a long time, but the main upshot of it no longer holds. Namely, you can no longer expect a single, procedural thread to run twice as fast as it did two years ago. And, for instance, in scientific computing, there are many things that are either not parallelizable or take months of work to parallelize.

Re:Ho hum

[Anonymous Coward]

'Cos speed in the CPU just isn't where we need it, the problem is getting data from the disks to keep the CPUs busy is. Having 4.7GHz is pointless without massive memory banks, very wide memory busses, and stonking great big I/O busses. But unless IBM come out with a radically new and therefore massively expensive chipset with a new form of interconnect, they'll be stuck with a fast CPU sitting idle most of the time waiting for data. PCs have it even worse, to make those PC motherboards cheap they need to keep the bandwidth narrow, so that's why you're still running a slower PC chip - Intel and AMD realised they could do more by looking at the infrastructure around the chip.

Oh, and Apple is a complete irrelevance.

PS: Kudos to /., it's so much easier to start a flame war here than at The Register!

Re:Goat$e!

[Anonymous Coward]

Perhaps IBM pulled the Power6 chip out of their ass?

Power

[suv4x4]

They apparently want to emphasize this chip's gonna make quite the power bills.

DRM

[Joseph_Daniel_Zukige]

naw, Jobs wouldn't do that, would he?

Stupid impatience on the UWB business ...

build your own

[Anonymous Coward]

uh....can't you just wait a bit and get the IBM machine and run yellow dog on it, if you really have to have the G(Power)6? I mean, the whole deal is IBM will be offering these for sale, so you really can buy one. Go ahead and do that, then just get an apple sticker and slap it on the box. Even if all you can get from IBM with that chip is a rackmount or a blade,so what, get that, then I bet you could still cram it in an old mac server tower case like a 9500. And there ya go, no worse than putting a built 350 chev small block in some 1930s frame, a hotrod is a hotrod, just have fun with it. We're geeks, we *don't have to* do what the suits tell us to do.

DB is i/o bound

[benow]

Putting the db on solid state drives would do much much more than running on faster processors. I'm all for more processing power, but reduce the worse bottleknecks first. Heck, raided iRAMs are cheap (comparatively).

Re:DB is i/o bound

[Heembo]

Huge write operation applications such as a database are not viable in terms of drive longevity when compared to old-school hard drives - at least for now.

Re:DB is i/o bound

[Anonymous Coward]

They didn't post the AWR report for this test, but assuming it's the same as the others from that page, no it isn't. The top wait event at 89% of total time waited is CPU time (which is logical I/O, scanning through RAM, and processing results), followed by db file sequential read, which is optimized index scans (physical I/O) at 7%. So it's accurate to say that this test was bound by either CPU speed or memory bandwidth, but certainly not physical I/O.

Re:DB is i/o bound

[Anonymous Coward]

You can't really do general purpose databases with flash RAM because of the very limited number of writes that flash supports. I tried that with Linux systems with ITX motherboards connected to 8GByte Compact Flash drives. They were a little faster than hard drives for only ~$100 more than a small IDE hardrive, but they all quit within six months. Most flash now claims to support wear leveling, but in practise they do so very poorly. We were runing MySQL which has several hotspots, like blocks where the count of the rows in the table is stored and the next ID # to use with autoincrement ints, so the flash died quickly. We're still using flash. We just had to write a customer circular log-based storage file system to make writes as even as possible. From a few tests, it appears we can now get 12+ years out of the flash before it dies due to writes. We lost the ability to use an SQL database so our solution is not appropriate for most people.

Re:DB is i/o bound

[TheRaven64]

Last time I talked to someone from IBM, they told me that they were selling POWER5 systems with 1TB of RAM. I think a lot of databases would fit in the disk cache on systems like this...

For a database application you are probably better off going with something like the T1, and having it run other threads while it's waiting for the disk to come back.

Re:DB is i/o bound

[sonofagunn]

With a nice enough RAID system and a large enough RAM cache, disk I/O may not have been an issue.

Re:DB is i/o bound

[Anonymous Coward]

For a database application you are probably better off going with something like the T1, and having it run other threads while it's waiting for the disk to come back.

Um, if you are waiting for the disk, why a 32-way CPU to run threads simultaneosuly? You'll be fine with a one-way CPU switching between threads -- since you are waiting for the disk...

Re:DB is i/o bound

[TheRaven64]

If you're waiting for a disk a lot, then you are going to be context switching a lot, so you want something that can context switch fast. SPARC is good at this, largely due to the TLB architecture (context switches on SPARC don't require a TLB flush). SMT also helps, since you have other threads that can run while your blocking thread is being context-switched out. I'm assuming you'd want to be running a thousand or so threads on your T1, one per client, with most of them in a blocking for I/O state at any given time.

Re:DB is i/o bound

[benow]

Here's some benchmarks including db for the gigabyte iram (dram over sata). It would be hard to believe that db access is not i/o bound. Raid'ed atlas' are fast drives, and slugs in comparison. AFA ram duration goes, dram does not suffer the decay issues of flash, but is more expensive. The iram is the cheapest of the ram based sshd's, but not the fastest... something like the Texas Memory Systems Tera-RAMSAN is much faster.

Re:DB is i/o bound

[benow]

Hmmm, and here's a comparison of SATA RAID vs SSD for Oracle.

"Overall, based on elapsed time, the SSD array performed the queries 276 times faster than the ATA array. Note that this was using the 30-hour limitation on the queries. Had I waited for those long-running queries to complete, the difference may have been much greater."

"In summary, the SSD technology is not a silver bullet solution for database performance problems. If you are CPU-bound, the SSD drives may even make performance worse. You need to carefully evaluate your system to determine its wait profile and then only consider SSD technology if the majority of your wait activity is I/O related."

IBM probably did choose a cpu bound operation in order to show off their procs, but it is not the case that cpu increases will help out in all situations.

Re:DB is i/o bound

[kuleiana]

Sure, if you insist on not indexing your data, using the wrong column types, and storing large strings directly in your db, but I guess only "real" software engineers would know that, eh?

OS X Server = PPC/Intel, OS X = Intel

[Kadin2048]

I think that you basically mentioned the only real place where there's a market for PPC: on servers. Although I've always been a big fan of the Power architecture (I have a dual-G5 spaceheater sitting under my desk that I'm writing this on, right now), I don't think that offering G5 PowerMacs along side Intel PowerMacs would really do anything besides confuse customers and potentially make the platform less appealing for developers who don't realize how easy Universal code is to produce. So I think that's a non-starter.

However, keeping OS X Server (which under the hood really isn't that different from regular old OS X, but it's marketed as a totally different product) Universal, and producing PPC XServes in addition to Intel boxes, might not be a bad idea. PPC XServes have always had a fair bit of popularity in the HPC and scientific-computing segments over x86, and for servers, a lot of the software in use is OSS anyway and is architecture-agnostic by design. So they wouldn't really be confusing any developers there -- most of the software that runs on OS X Server is either supplied by Apple, or is OSS, or (in the case of custom HPC code) may have been written/optimized specifically for Power/Altivec in the past already, so they'd be saving their customers work by offering a PPC product.

I think there could be a lot to gain by keeping a PPC model around. They might not even have to do too much hardware design; if they didn't burn too many bridges with IBM on the way out, they could probably use one of IBM's Power-based blade-server boards in a 1U case...particularly with the way Cell hasn't been selling, IBM would probably be happy for the microprocessor sales.

Re:OS X Server = PPC/Intel, OS X = Intel

[Amiga Trombone]

I think that you basically mentioned the only real place where there's a market for PPC: on servers. Although I've always been a big fan of the Power architecture (I have a dual-G5 spaceheater sitting under my desk that I'm writing this on, right now), I don't think that offering G5 PowerMacs along side Intel PowerMacs would really do anything besides confuse customers and potentially make the platform less appealing for developers who don't realize how easy Universal code is to produce. So I think that's a non-starter.

I'd agree with that assessment. Also, consider that desktop/laptop CPU's have different requirements than server CPU's. One of the reasons Apple dumped PPC was that IBM wasn't earning enough on chips optimized for desktops to invest in the necessary R&D to keep them competitive with x86.

That is not an issue with servers, however, Power6 is already optimized for that purpose. Apple could probably offer a very attractive XServe indeed based on that chip. It would give them an offering that would outperform anything based on x86, making OS X a more attractive and versatile platform in general. I'd like to see them go for it.

Re:OS X Server = PPC/Intel, OS X = Intel

[Ilgaz]

They really burned bridges by labeling our CPUs as some outdated low mhz monsters to millions of people, anyway, what has been done is done.

Speaking as a Quad G5 owner, I don't think Xserve Power6 would be viable choice since there is already AIX on that scene with some amazing IBM support. Look what they offer to PowerPC 970 (yes G5) Workstations and imagine what server owners get:

"Service Agent: The Service Agent is available at no additional charge. When installed on an IntelliStation POWER 185 workstation, the Service Agent can improve the level of maintenance service from IBM.
The Service Agent
Monitors and analyzes system errors, and if needed, can automatically place a service call to IBM without customer intervention
Can help reduce the effect of business disruptions due to unplanned system outages and failures
Performs problem analysis on a subset of hardware-related problems and, with customer authorization, can report automatically the results to IBM service"

It sounds like Apple Crash reporter but with a difference. The difference is, right after you even notice the problem, actual IBM guy comes to your workplace and fixes things if not remote fixable via service center.

For HPC, there is already Linux/Power solution which is said to be one of the best
http://www.terrasoftsolutions.com/products/y-hpc/

Re:OS X Server = PPC/Intel, OS X = Intel

[Anonymous Coward]

Back then, I posted that Apple wasn't really "switching" to x86 because once developers got the hang of shipping fat binaries they could just as well ship a new line of PPC machines if those happened to be better at the time. And therefore that the one and only reason Apple made such a fuss about The Switch was to get devs to ship fat binaries in the first place instead of just waiting out for the next PPC generation, which would have been a disaster.

Of course that was totally ignored by both the Apple "Steve said so" fanboys and their PC "we were right all along" alter-egos.

I think the fact that IBM added an Altivec (sorry, VMX ;-) unit (that POWER5 didn't have), which is all but useless on servers, is a rather strong hint that at least IBM would like to get POWER6 in Macs.

Or perhaps they just had die space to waste...

I guess we'll see now.

Re:OS X Server = PPC/Intel, OS X = Intel

[Brickwall]

WTF?? The article is about a major technical improvement by IBM, and half the comments are about Apple's decision to switch processors?

I'd be much more impressed if people were talking about what new applications would be possible with this much processing power, what this means for IBM customers, and how this shows that North American companies aren't losing every technical advantage as some people purport.

Re:OS X Server = PPC/Intel, OS X = Intel

[mrsteveman1]

If Apple comes back out and says "oh yea we're actually bringing PPC machines back, should have mentioned that last time eh?" people will not put up with it anymore, even on servers.

They have switched arch multiple times now, each time showing a complete lack of planning for the future, and every time they end up having to code a VM into the next OS to run older applications. This time they actually switched to an arch that they spent the last 12 years bitching about. So either Apple fabricates marketing, or they are completely wrong on one side or the other.

It's sad really, OS X is a good operating system and most of its problems stem from being owned by Apple, insisting things be exactly a certain way, it is ruining their chances for a lot of opportunities.

Re:OS X Server = PPC/Intel, OS X = Intel

[Anonymous Coward]

IBM's work on Cell may also have enlightened them on the usefulness of SIMD in server and workstation CPUs.

Re:OS X Server = PPC/Intel, OS X = Intel

[RockoTDF]

Well, the majority of /. posters probably have more experience dealing with PPC Macs than IBM mainframes, so whenever POWER comes up apple is what comes to mind. It only makes sense that whenever IBM does something cool with power that people will question whether the Intel switch was necessary.

Ars Technica Technopaedia

[Short Circuit]

Link me. Ars Technica's Technopaedia may be found here. Of particular interest are the Technical Blackpapers (indexed at the bottom of that page) and the section on CPU Theory & Praxis.

Re:Power isn't PPC, but don't forget iPods

[bubbaD]

I thought the last Word was that it was Intel's ability and willingness to provide iPod chips that made the switch to Intel in all Apple hardware so desireable. Considering iPod is really Apple's cash cow now, that made sense to me. Apple makes Intel a good deal- give us cheap iPod parts, and to sweeten the deal, we'll move the less profitable computer line to Intel CPUs as well. Remember that there must have been a huge overhead in transitioning from PPC to x86, and alienation of some big Apple software vendors Freescale with Code Warrior and Terrasoft Solutions YDL comes to mind. This meant porting Everything, not just a matter of getting cheaper chips. They could have dealt with AMD after all.

TOASTY!

[Z80a]

thats thing must be hot :3

Re:Did Apple make another mistake? Son of Woz

[bubbaD]

I wish they had kept producing Motorola 65xx-family Apples like the Apple II and III for a low end and Macintosh 680x0 for high end back 20 years ago. I think the IBM clones with Microsoft were able to overwhelm Macs partly because Apple abandoned its established base of Apple customers while x86 makers and M$ worked with some really ugly kludges, just in order to maintain backward compatibility. Jobs likes to burn his bridges, though and I expect Apple will continue the tradition.

XXXServe?

[bussdriver]

Apple could consider enterprise customers by porting OS X to IBM's hardware (or just rebrand it like they used to do when Apple sold laser printers.

suggested names:
XXXx2 Serve
Xx6 Serve
Apple POWERServe (just makes you think of a ball doesn't it?)
ZServe
iPOWER
XPOWER
POWER X (advertise using a comic book theme)
Enterprise G6-07
G6 Cube
NeXt Cube
X007 (2007)
SuperMac (cant use BigMac)

X-Frame (as in mainframe with virtualization; my favorite)

iVapor (apple will never seriously target enterprise)

POWER6 is interesting for servers

[raftpeople]

I'll wait until the specs come out Tuesday* before I decide if the Power6 is interesting

The Power6 is interesting in the server processor arena because IBM has been the leader in this market since Power4 and every indication is that the Power6 will continue this. Keep in mind this is not a desktop or laptop processor, this is a processor designed for large business servers.

Not to mention that IBM didn't seem to be putting any resources at all into a low-power verion of the POWER5

The POWER5 was also a business server process, not desktop or laptop. You are thinking of the PowerPC portion of the Power architecture which is what Apple was using.

220 watts is not slim.

[twitter]

Aero Glass works on a Core Duo Mac Mini, so I'd have to say the answer is "pretty damn slim."

How much of the 220 W that computer uses is the processor and why were you complaining about the 100 W that Power 6 is using at more than twice the clock speed?

Re:220 watts is not slim.

[ralmin]

Aero Glass works on a Core Duo Mac Mini, so I'd have to say the answer is "pretty damn slim." How much of the 220 W that computer uses is the processor and why were you complaining about the 100 W that Power 6 is using at more than twice the clock speed?

I think you'll find that computer uses 110 W, being a Mac Mini, not an iMac. The iMac's 220 W includes the built-in screen!

Need Software

[Anonymous Coward]

It's a shame that the amount of software that can run on them may fade.

But 110 W is...

[cnettel]

You provide the wrong link, this is the proper one. 110 W, for the complete machine, and that's AC (so even if the CPU was the only DC component in the machine, it would end up consuming more from the mains).

For comparison, I think the Core Duo TDP in that machine is something like 30 W, maybe a bit more.

Power6?

[Jaewon Choi]

Is this the successor of Power PC G5 chip? As you all knows that G5 chip was powerful, but not suitable for mobile platform because of it's big heatsink size. Actually if the heat problem couldn't be solved, Power6 processor will be used only for middleframes or workstations. This is the reason why Apple has moved from G5. But I still believe the power of RISCs.

Re:Power6?

[larien]

Nope, it's the successor of the Power 5 chips used in Intel pSeries servers (primarily running AIX, but you can run linux). The PowerPC chips had some similarities with Power, but they're not the same chip by a long chalk. Power 6 is unlike to appear on anything smaller than a 2 CPU server.

Re:Power6?

[Ilgaz]

PowerPC G5 is a lightened Power4 chip. Yes that multi core, 64bit RISC CPU. That could give idea about the monsters mentioned on that story.

Not saying for you of course, telling for people who thinks about buying a Desktop based on Power6. I don't think such thing could happen except if IBM decides to update their "Real" Power Workstation http://www-03.ibm.com/servers/intellistation/power /285/ and CTIA is not essentially entertainment I guess. :)

If I were you, I'd really folllow http://www.terrasoftsolutions.com/ and http://www.power.org/ , I suspect very interesting things on way with PS3 (!) Cell Processor. Soon or later I will have to switch YDL Linux anyway so keeping eye on stuff. I even begged for PowerPC Linux Flash from Adobe already :)

Re:Power6?

[Anonymous Coward]

I think you meant to say "Nope, it's the successor of the Power 5 chips used in IBM pSeries servers"

Guess it was just a slip of the brain :-)

Re:Power6?

[Jaewon Choi]

Thanks for your inform. I thought it was same. :)

Re:Power isn't PPC, but don't forget iPods

[TheRaven64]

Intel doesn't provide any of the chips for the iPod, and they just sold the part of their organisation that made ARM-compatible chips. They make the chips for the Apple TV, but Freescale also make a number of parts that would fit those needs.

The main reasons for the switch was that Intel provided the whole setup - CPU, chipset and GPU (at least for the low-end) - and that Intel had a track record of supplying volume. Apple are less than ten percent of Intel's market for desktop chips. They would have been over half of AMD's and were almost all of IBM and Freescale's. This meant that Apple got the R&D benefit from Dell's CPU purchases (and vice versa).

Apple could virtualize the O/S

[master_p]

Perhaps Apple should think about virtualizing its O/S, i.e. XCode could output bytecode and then the kernel could translate that to the native instruction set, thus freeing themselves from any dependency from a particular CPU architecture.

Re:Apple could virtualize the O/S

[joto]

What a marvellous idea. They should write all their crap in some slow-as-molasses bytecode. Surely the way to go if loosing further marketshare is your thing!

On the other hand, x86 is a non-native instruction set. They stopped making x86 processors somewhere back in the early 90s. Every x86 processor since, has emulated the x86 instruction set with their own microcode. Which, from a performance standpoint, is far better than doing the same thing in software.

Re:Apple could virtualize the O/S

[red_dragon]

Yeah! They could get their inspiration from the AS400/iSeries/System i servers, that'll surely show IBM who's better!

Re:Apple could virtualize the O/S

[master_p]

On-the-fly translation technology has progressed quite a lot. Java apps are almost as fast as native apps, for example. And the O/S could keep the translations around.

Re:Apple could virtualize the O/S

[joto]

Correction: Unoptimized java apps are usually within the same order of magnitude regarding speed as unoptimized native apps; if your benchmarks happen to overlook the effects of slow startup due to just-in-time compilation, and huge virtual memory overhead (global system slowdown) due to garbage collection. In the real world, people still optimize their native apps.

I'm not against people using java where it makes sense. But imagine a system where every little command required launching a java interpreter (e.g. "ls"), and you see why it's not such a great idea. Java works great for writing moderate to large programs on budget and within time. But there's a reason suns "java os" never saw the light of day (and that reason was not related to failure in marketing it)

Re:Apple could virtualize the O/S

[master_p]

The kernel would be the virtual machine, so your comment is valid but not in this case.

For servers...

[Anonymous Coward]

As I understand it from an IBM source, the POWER6 chip should appear in the System p and System i lines in the model 570.

Re:For servers...

[mink]

For now. Configuration seems to have been hastily tacked on to the config program rather then the expected POWER 6 based systems tab I was expecting. It's not like they had advance notice they were going to release a new processor platform.

Why Apple/Intel was the right move

[zerofoo]

The shop I work in right now is a mix of Dell and Apple hardware. We now buy all our Desktop machines from Apple - why?

Intel CPUs.

We can now run Windows and Mac OS on the exact same hardware. Dell has lost all our desktop business as the result of Apple's move to Intel. One hardware platform is very nice from a purchasing and management perspective.

I'm sure we aren't the only shop with that strategy - and that's why Apple's conversion to x86 was a good decision.

-ted

Re:Why Apple/Intel was the right move

[laffer1]

We are transitioning to all Apple hardware at work. Previously we have a 50/50 mix of iMacs and dell optiplex systems. What I'm curious to see is what OS ends up as the most popular. Apple's move is risky . If OS X does not fare well, they did sell hardware but why buy apple boxes long term of OS X is dead? I like OS X better than Windows, but I realize some people prefer Windows.

Power 6 - Server Based Processor

[nyclinix]

How on earth did the announcement of Power 6 turn into a debate about Apple and small consumer electronics? The Power 6 is designed to populate IBM's heavy-hitting AIX servers. They have large amounts of on-board cache and are designed to work in virtualized server environments - both hardware virtualization that IBM calls LPARs (logical partitions) and software virtualization (similar to Solaris zones/containers). A mid-sized server is capeable of running 50 or more AIX partitions and to copy one partition to another with a mere few seconds of interuption. The technology is very similar to the well known features of IBMs mainframes. IBM has strongly hinted that the P6 (and it's successors) will be the chip that will power future mainframes, AIX and I5 (as400) systems someday. The new chips use way too much power and are too large to fit in portable consumer electronics and I doubt any consideration was given to hand-helds during design. As for Apple - they have experts that can perform cost/benefit analysis on chip prices and this chip is going to cost a lot more that Intel (mobile).

Re:Power 6 - Server Based Processor

[Bright Apollo]

mod parent up +1 informative, this is exactly what the CPU announcement means. Apple has nothing to do with a CPU replacement for mid-range computers (RS/6000 and AIX). Apple and Microsoft have nothing that can take advantage of all that it offers.

-BA

Sounds familiar...

[Anonymous Coward]

That's 4.7 GHz unless you press the Turbo button on the front panel, then it goes up to 8.

Intel wouldn't let them

[Anonymous Coward]

anyway, why Apple doesn't just permanently keep its lineup as a mix of PPC and x86, picking whichever chip suits the particular machine they're designing at the time?

The same reason Dell couldn't touch AMD chips until it's customers forced it to.

Intel's "volume pricing"/extortion deal with Apple probably insists on near 100% Intel chips or they'll get screwed.

No I believe it goes to 11.....

[MoronBob]

For when you need that push over the edge.

Database requests?

[Anonymous Coward]

Okay, I know that database transactions require a lot of processing power, but since such transactions generally also involve a lot of disk read / write accesses, not to mention network overhead, I somehow feel that database transactions do not constitute the best environment to show what a new CPU is capable of. Where are the cold, hard benchmarks showing the number of FLOPS and/or benchmarks related to high-performance applications that mainly rely on CPU power, such as 3D rendering applications?

Re:ATTN: SWITCHEURS!

[somersault]

Do you mean Claris? ;) And my Mac Classic was beige.. bandwagon jumpers are the ones that only started using Macs when they were funky colours.. I stopped liking Apple so much when they brought out the iPod and concentrated on that, but it's good that it made them some cash to get a bit more into the mainstream with their computers. I decided to get a Macbook Pro as my main work machine to indulge in a bit of nostalgia (though unfortunately there no longer appears to be even emulated support for 680x0 processors in the OS :[ ). Also nice to be able to browse the net without worrying about spyware (beyond cookies).

Re:ATTN: SWITCHEURS!

[Anonymous Coward]

If you don't know what Cmd-Shift-1 and Cmd-Shift-2 are for, GTFO

What if you're too young to know what a floppy drive *is*, kinda like the kids in your other pictures?

just bring back...

[Anonymous Coward]

the laptops running AIX. I'd pay top dollar. ahhh memories

Did you notice

[rrohbeck]

that is has a BCD unit?
Finally I can run my COBOL code at optimal speed. Yay!

IBM's own announcement

[Anonymous Coward]

http://www-03.ibm.com/systems/p/news/features/2007 /annc_0522.html

Apple Certified Servers?

[bill_mcgonigle]

IIRC, HP was selling 200k 1U servers when Apple sold 12k 1U servers.

That's an interesting statistic - I'm surprised they keep making them. People really like OSX Server, especially in schools - but nobody I've worked with really loves their XServes (XServe RAID is another story).

I bet the servers will be the first Apple Certified hardware that gets OSX on a non-Apple brand. Yeah, yeah, yeah, Apple is a hardware company - but at 12,000 units you're having trouble staying profitable.

Re:Apple Certified Servers?

[Jeff DeMaagd]

I don't think those unit numbers are terribly small, but I think maybe their volumes have increased since then.

not so much

[sentientbrendan]

>Apple could probably offer a very attractive XServe indeed based on that chip
Not really... Power is different than PPC. This would essentially require that a *third* platform be supported.

Also power is for mainframe... Server's these days generally use cheap X86 boxes. Switching to the power platform would increase costs for your average xserve significantly, which would be bad since they're already pretty costly (their *min config* is 3k).

Aside from that, there would be extra R&D costs associated with putting together apple branded hardware with an IBM power architecture, and there might be problems making the servers fit in a standard rack. The power architecture has traditionally run pretty hot (how do you think they got up to 4.7ghz?) and would need extra cooling.

Re:ATTN: SWITCHEURS!

[Anonymous Coward]

No, I meant Clarus. What are you waiting for? GTFO.

Re:Power6 is a PPC

[Anonymous Coward]

Guys, I work developing compilers for PowerPCs.... All of POWER[345] chips are PPC, which is not a strict subset of the old POWER architecture. In fact, each new chips has introduced new instructions are the PPC architecture is revised. There haven't been non-PPC POWER chips since the POWER2.

Apple's closed-ness is their biggest advantage.

[Kadin2048]

It's sad really, OS X is a good operating system and most of its problems stem from being owned by Apple, insisting things be exactly a certain way, it is ruining their chances for a lot of opportunities.

Actually, and I say this as a Mac user, OS X / Darwin is an unspectacular (really -- there were some really painful DB and Apache performance benchmarks vs Linux that came out a while ago) operating system, the only saving grace of which is that it's owned by Apple, which can keep it basically under control. If it wasn't for that, it would probably end up like Plan9, or any number of other interesting OSes that very few people actually use.

The reason I've always bought Macs, and will continue to buy them (although I do keep a Linux workstation around at home now too) is because they're a closed platform. You want a wireless card? There's no screwing around -- you go to the Apple store, you buy a wireless card. It costs a lot of money, but it's worth every goddamn penny, because you know it's going to work without any screwing around. There's only one, the drivers are all built-in, it's well documented (from a user's perspective) and usually well QAed. Linux can't say that, and Windows certainly can't either; it's a direct consequence of the Mac being a closed platform. (You do get it on other high-end closed platforms; e.g. some mid-range and higher IBM gear -- you want something, you call your IBM rep with your checkbook in hand, they send someone to plug it in, end of story.)

I'm as much of an open-source and open-standards guy as anyone -- I wouldn't want Apple dominating the computer field, any more than I enjoy Microsoft dominating it (not at all). But to say that Apple's strength isn't derived at least in large part from the fact that they control everything, from the hardware up to the userland apps, is naive. (And they do a good job at it, too -- but they wouldn't have the opportunity to excel if they didn't have that platform to work with.) Perhaps Apple could open everything up and still succeed by selling a guaranteed-to-work "reference platform" ... but why mess with something that works?

Re:ATTN: SWITCHEURS!

[somersault]

What is 'Clarus' then, oh wise and learned Mac user..? Nothing turns up on google apart from a greek god. Claris, on the other hand used to make a productivity suite for Macs, and I used to use it to do my schoolwork..

Re:ATTN: SWITCHEURS!

[somersault]

Haha http://macslash.org/comments.pl?sid=6626&cid=11708 5 . Hypocritical bandwagon jumpers!!! Very sad :P

Re:ATTN: SWITCHEURS!

[Anonymous Coward]

Google harder, you fucking switcheur.

Re:ATTN: SWITCHEURS!

[Anonymous Coward]

Stop. Just stop. You're making an even bigger fool of yourself. If you think "Clarus" is just a misspelling of Claris, you really, REALLY need to GET. THE. FUCK. OUT. RIGHT. NOW.

What are you waiting for?

Re:ATTN: SWITCHEURS!

[somersault]

So knowing about OS X hacks is cooler than having used Macs for 20 years? Hmm.. and I've never really 'switched' from Macs, since I was never really primarily a Mac fanboy, I've used them alongside my Amigas and PCs occasionally though. I still prefer Macs to PCs, but you seem like a bandwagon jumper to me.