PDA

View Full Version : Apple shows interest for CELL technology


mustique
02-17-2005, 02:45 PM
Donno if this is old news but here's the link

http://www.totalvideogames.com/pages/articles/index.php?game_id=&article_id=7216

Shogmaster
02-17-2005, 03:18 PM
You can chock this on to a tech idiot anaylist @ Merrill Lynch that like to make huge leaps of faith: "Duh~ Cell uses a IBM G5 VMX inside so it should be perfect for Apple computers!".

Cell is not exactly a chip architecture that's ideal for desktop usage (heavily depenent on parallel processing for performance), while G5, A64 and P4 are all OOOE (out of order execution) monsters. Cell is great for doing things that call for massive paralleism (like calculating multiple vertices and rendering individual pixels), but will be far from optimal for doing things that will depend on OOOE, like everything else we do on our computers.

Another things is, Cell's FLOP performance comes mostly from it's 8 SPE (Synergetic Processing Element) units working under the command of one PPE (Power PC Processing Element), and not from the ALTIVEC perfromance of that PPE part. So it's 256GFLOPs can't be utilized until Apple and it's 3rd parties rewrite all the apps to specifically take advantage of those SPEs. Yes, a snowball's chance in hell of that one.

Really, Apple is way too invested and dependant on the ALTIVEC SIMD instructions to even care about the Cell. If Cell's performance with ALTIVEC was incredible, that's one thing, but Cell's ALTIVEC performance is not all that hot, so there's no incentive for Apple to go through all that work to use Cell in their computers.

mustique
02-17-2005, 04:43 PM
... So it's 256GFLOPs can't be utilized until Apple and it's 3rd parties rewrite all the apps to specifically take advantage of those SPEs. Yes, a snowball's chance in hell of that one...

Given the fact that Apple doesn't rely much on 3rd parties as MS does, I think it wouldn't stop Jobs from trying to make use of CELL. He did more surprising things in the past.
Also G5 doesn't look like a promising cpu for Apple. So why not. I'd like it :)

Shogmaster
02-17-2005, 04:58 PM
Given the fact that Apple doesn't rely much on 3rd parties as MS does, I think it wouldn't stop Jobs from trying to make use of CELL.

So you expect Apple to go through all their apps and recode and compile them all over again, and then flip a big birdy to the likes of Adobe (who's not in the best of terms with Apple right now anyways) and say "you're on your own suckas!"?

Let's smell the coffee and wake up, shall we? ;)

He did more surprising things in the past.

Surprising? Yes. Downright stupid? No.

Also G5 doesn't look like a promising cpu for Apple. So why not. I'd like it :)

If you won't listen to me, then listen to this guy (a cool article I just found on engadget).

http://www.jonpeddie.com/index.shtml

What the Cell Isn't


No, it's not your next PC or server.

Information on the new cell processor from Sony and IBM has been somewhat available to the geekily inclined public for quite a while now. So, I was very surprised by the questions I was being asked after the Cell's more official debut at the ISSCC conference. And, I was even more surprised by some of the reports that came out of the conference. Reporters have been consistently trying to make the Cell a competitor to the venerable X86 architecture and the Pentium in particular.
For example, here is the sort of quote you might encounter:
"Analysts say that the difference is likely to shake up the contemporary model of computing that Microsoft and Intel have fostered. [because] The new chip also has been designed to handle multiple operating systems and programs simultaneously."
The Cell is not a Pentium killer or a competitor—it's a SIMD supercomputer, and just because it has an IBM Power processor in it, which has been demonstrated to run X86 code in an emulation mode, that's not the use or the future of this Cell.
Here's another one:
"Later this year, Intel and Advanced Micro Devices each plans to launch their own 'multicore' chips, which similarly up the number of commands that can be executed at once. IBM and Sun Microsystems already sell such multicore microprocessors, largely for business servers."
The Cell has an array of SIMD RISC processors, not multi-cell X86 processors. Nor will it be a super server chip—it's just not built that way. It's built to process the same kind of data, which is highly predictable, not changing data like a GP CPU in desktop or a server deals with. Mix up the data stream and types and feed it to a Cell and you'll see it screech to a slow crawl. It costs 18 cycles any time a branch mis-predict— that's not server-quality processing.


[b]Is it your next workstation?

IBM Chairman, President, and Chief Executive Samuel Palmisano said that this year it'll sell a workstation containing Cell. That's true, it will be built at IBM's Boeblingen facility near Stuttgart (see Tech Watch, November 29). Such a system will be used for software development for the future Sony console. It's possible it may be used for scientific work, but special compilers and applications will have to be created to make use of it. That comes under the "someday" heading.


Is it a superomputer, on a chip?

Well yes, architecturally, it's exactly like the multi-processor SIMD vector machines that were built in the mid-eighties, and to some extent are still being built (only 4.2% of today's supercomputers are vector based). However, at 256 GFLOPS it's not, by itself, in the supercomputer class. (The slowest machine on the current list of the Top 500 supercomputers does 624 GFLOPS http://www.top500.org/lists/plists.php?TB=5&M=06&Y=2004.) But you can envision an array of Cells being made into a supercomputer, and given the expected low cost of this soon-to-be-mass-produced chip, that's entirely feasible, except that it might wrinkle the feathers of IBM's BlueGene group a bit.


It's not a Grid computer

The other popular misconception, which IBM and Sony are doing nothing to dispel, is the notion that the Cell will hook up to some type of grid (like Butterfly net) and either tap into more resources or allow millions of user to simultaneously play a game. Yes, online gaming can take advantage of the Butterfly net, and yes, IBM and Sony said they would use it. But it's not to share Cell resources—it's to play against others just like any other online network.
Take a look at this quote from a story published by CNET:
"Cells can even roam over a network, allowing the processor to perform a type of distributed or grid computing, an increasingly popular enterprise technique in which demanding tasks are divvied up among a gang of networked computers."
Clearly, there are some major misunderstandings about what networked and Grid computing are all about. Grid is used for collaboration and uses the Internet, not the fastest data exchange network in the world and certainly not fast enough for distributed for real-time computer, the type of computing that epitomizes a game.
A Playstation 3 could borrow unused processing power from other consoles on a network, for example, to complete a demanding task such as delivering streaming video. Sony has done a masterful job at creating a science fiction scenario that ignorant journalists have latched onto. The only problem is some of those ignorant journalists have good memories, and in a year or two when the next-generation Playstation is delivered and it doesn't do all these suggested whiz-bang things, those same journalists are going to feel used, and then they will write searing stories about how Sony and IBM have failed to live up to the expectations and promise of the lauded Cell. This misdirected PR BS is going to haunt IBM and Sony. Lest we forget, when Sony introduced the Playstation 2 Emotion Engine it was promised to power a wide variety of consumer electronics, and so far it's failed to live up to that promise.


Is it your next TV?

Well, Toshiba has been quoted as saying it plans to incorporate it into high-end TVs. Given the crashing ASPs on TVs these days, it's not likely the Cell, in its current configuration, will show up inside a TV, high-end or otherwise. The vector processors could be useful for MPEG decoding/encoding, but you wouldn't need eight of them, so a stripped-down version, with maybe two SPEs and a Power CPU, might be made for high-end TVs sometime in 2006, maybe.


So what is a Cell?

It is an eight-IMD processor array with a powerful master controller and program execution GP CPU. Each SIMD has local memory, high-speed interconnects, and a segmented memory structure ideally suited for graphics and codec applications. It has a super-high-speed chip memory interface and an equally fast flexible I/O for the graphics processor. It is currently running Linux and could run other OS's on the GP CPU.
It will be a really great game console processor and general multimedia accelerator, and should cost out at a reasonable consumer price range.—Jon Peddie

chadtheartist
02-17-2005, 04:59 PM
What if Apple decided to use it as an on board graphics enhancer for Core image/video? Would that be too far fetched?

Shogmaster
02-17-2005, 05:00 PM
What if Apple decided to use it as an on board graphics enhancer for Core image? Would that be too far fetched?

Still would require Apple and 3rd party to recode and compile their apps to take advantage of it.

chadtheartist
02-17-2005, 05:03 PM
Don't they have to do that already?

Shogmaster
02-17-2005, 05:05 PM
Don't they have to do that already?

For what? They just went through such an expensive re-code-athon with switch to OSX. I don't think they want to do that again anytime soon.

mustique
02-17-2005, 05:45 PM
I don't understand why it'd be a stupid decision to rewrite your apps in a way
that can make them up to 10 times faster (given the hype turns out to be true)

Apps will be rewritten to take advantage of the emerging 64 bit era anyway.
Its only a question of time till apps will be recoded to take advantage of new procs.

All the naysaying will be forgotten then.

Also, if the CELL proc will be used for future IBM workstations, what will it run?
IBM's inhouse DCC apps?

richcz3
02-17-2005, 05:55 PM
Shogmaster, Thanks for that info.
The Cell PR machine of the last month has been trumpeting PC archtecture threats while offering obscure real world use. Case in point was mentioning IBM's proposed PC alternative. There is no doubt the Cell is powerful and forward thinking design, but the whole x86 archtecture will slip under the waves because of it is unfounded.

I don't expect the suits in PR to let up any time soon either. Sony's team will continue to push superiority over the Intel/AMD/X86 indefinately. Sony has the PS3 to push and the stakes are high. It doesn't matter if it means comparing Apples to Lemons. The general public and establishment haters will eat it up.

chadtheartist
02-17-2005, 06:06 PM
http://www.apple.com/macosx/tiger/coreimage.html

I already mentioned Core Image. Developers are going to have to rewrite some of their code in order to take advantage of it. If Apple writes Core Image to be used on something like the Cell, developers may not even have to worry about writing specifically for it.

shehbahn
02-17-2005, 06:07 PM
>I don't understand why it'd be a stupid decision to rewrite your apps in a way that can make them up to 10 times faster

because you don't really understand the implications of SIMD code. the assumption that it's going to be 10x faster is unfounded : few algorithms lend themself to efficient fine grain parallelization, fewer of those are "average desktop" tasks. did i mention that there are very few programmers around who have the skills to do this kind of work to start with ? the concept of vectorized computing is over 30 years old - the reason these types of architectures never spread isn't entirely Intel's marketing fault...

mustique
02-17-2005, 06:09 PM
I am not so sure of this.

X86 architecture and wintel will of course struggle to dominate. But we live in a transition time. 32 to 64 bit. This is the best time for others to play their cards.

The main issue is, things will develop fast, competition will soar, and we as consumers will benefit. No need to be a fanboy or naysayer. At the end of the day everybody just wants our money, don't they?

sedric
02-17-2005, 06:43 PM
i thought the G5 cpu was supposed to be the fastest thing around? remember the hype when it first came out? why are there now questions about its performance?

chadtheartist
02-17-2005, 06:47 PM
I don't think anyone is doubting the G5. I was just speculating on how Apple could use the Cell, since it's not really designed for everyday desktop use.

The G5's still stack up well against most PC systems that are similar in Specs. I'm sure it will get better.

shehbahn
02-17-2005, 07:22 PM
>X86 architecture and wintel will of course struggle to dominate.

they are around 95% of the market to date - and in case you missed the train, Intel announced a few months back that they backed out of the 4Ghz Pentium (mostly because they rightly think they can't sell liquid cooling to the masses yet) and the push for multi-core architectures. this has the exact same problems and implications : it is very hard to write parallel code.

>But we live in a transition time. 32 to 64 bit.

SGI has had 64 bits machines since the days of the Onyx machines. Your PS2 has 64 bits chips. Dec Alpha's have been native 64 bits for nearly a decade. there is nothing new there. one of the main reasons this finally breaks into the mainstream market is because games are slowly breaking the 1Gb of RAM requirements, in good part because of normal mapping. 64 bits is very old news in the CG industry.

richcz3
02-17-2005, 07:22 PM
Cells are going to be used in TV's and hopefully eventually in other electronic appliances. I imagine the basis is that a number of devices will be able to be hooked up together in a network. It doesn't make to much sense now, but if it takes off with other electronics manufacturers it could be big...IF they can substantiate the costs to claims and benefits.

mustique
02-17-2005, 08:32 PM
>X86 architecture and wintel will of course struggle to dominate.

they are around 95% of the market to date - and in case you missed the train, Intel announced a few months back that they backed out of the 4Ghz Pentium (mostly because they rightly think they can't sell liquid cooling to the masses yet) and the push for multi-core architectures. this has the exact same problems and implications : it is very hard to write parallel code.

>But we live in a transition time. 32 to 64 bit.

SGI has had 64 bits machines since the days of the Onyx machines. Your PS2 has 64 bits chips. Dec Alpha's have been native 64 bits for nearly a decade. there is nothing new there. one of the main reasons this finally breaks into the mainstream market is because games are slowly breaking the 1Gb of RAM requirements, in good part because of normal mapping. 64 bits is very old news in the CG industry.



OK I'm back to defend bold IBM, cool Sony and sexy Apple in court!

1. Holy Wintel might struggle to dominate in the possible future.
Till then please feel comfortable with stats like %95 market share.

2. Tiny Intel is doing no innovation at all in the 64 bit ball game.
If you want to praise somebody then please refer to AMD.

3. AMD outperforms Intel cpu's with much less MHz since years. So MHz means nothing.
(in case you missed the train)

4. We're talkin about the 32 - 64 bit transition for mainstream PC's. Where have you been?

5. I never had a PS2.

JDex
02-17-2005, 08:42 PM
OK I'm back to defend bold IBM, cool Sony and sexy Apple in court!

1. Holy Wintel might struggle to dominate in the possible future.
Till then please feel comfortable with stats like %95 market share.

2. Tiny Intel is doing no innovation at all in the 64 bit ball game.
If you want to praise somebody then please refer to AMD.

3. AMD outperforms Intel cpu's with much less MHz since years. So MHz means nothing.
(in case you missed the train)

4. We're talkin about the 32 - 64 bit transition for mainstream PC's. Where have you been?

5. I never had a PS2.

Meh!

Um... why are you playing the fanboi card... you are accusing others, but you are the one showing the signs.

The CELL tech is very interesting, and should Apple decide to adopt it, they will be in for a lot of potential power, but also a lot of potential headaches. Recoding the entire Apple catalogue (again) will be a massive undertaking and will agravate some of the non-apple application developers who just spent millions to recode to the OSX requirements.

As to your clear prejudice against the x86 platform... well maybe you should see someone. No doubt that x86 is reaching the end of it's development life. The move to multi-core x86 will no doubt cause the same potential increases in power, but also a lot of potential headaches.

As for whether one will over-take the other... who blippin cares... we'll all get the technology that works best for us when the technology comes available.

Anywho thanks for posting the link.

richcz3
02-17-2005, 09:22 PM
If Apple uses the Cell I see it in the form of a sub component of a G5 system. About the only clear (if you can all it that) thing about the Cell is that it runs on several different OS's and Linux . And it runs fast. Apple could theoreticaly use hardware components that feature the Cell chips. Any Sony camera and Toshiba Monitors with Cell based technology would be ideal to use with a Mac that utilizes a Cell based hardware/component.

To me, that is the kind of threat that is being hoisted in their PR blitz. Cell based hardware working with Cell enabled computers. This begs the question. Would IBM, Sony and Toshiba forsake the PC market?......>>>> NOT the Linux Server and PC market running 64Bit multicore processors.

CGTalk Moderation
02-17-2006, 10:00 PM
This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.