Cell Workstations in 2005

yerdaddie writes "The cell processor will be introduced in graphics workstations before release in the Playstation 3, according to press releases by IBM and Sony. As previously discussed, IBM will be releasing more details in February 2005. However, apparently prototype workstations have already been "powered-on" and will be available in 2005. Since Windows on PPC was scrapped back in 1997, this leads to speculation that perhaps Linux, AIX, or BSD will be the operating system for cell workstations."

Comments

I may be wrong...

[wcitechnologies]

I may be wrong, but to me this sounds like hyper threading with a new name. Can anybody enlighten me?

Re:I may be wrong...

[Anonymous Coward]

Sounds more like some kind of multi-core processor where the number of cores can vary greatly.

Re:I may be wrong...

[ponos]

I may be wrong, but to me this sounds like hyper threading with a new name. Can anybody enlighten me?

It's not the same. Hyper threading divides processor units (e.g. a multiplier or an adder) in order to keep most units of the single core busy. This happens because Intel processors have very long processing pipelines (thus the very high frequency compared to AMD), so stalling them can be quite costly. In order to avoid this, Intel simply keeps track of two "virtual" processor states, essentially 2 copies of all registers, and schedules instructions from any of these two execution threads in ways that keep most units busy. By chosing from 2 threads instead of one it has greater chances of finding an instruction that can be computed by an idle (at that time) unit.

Cell architecture, on the other hand, seems to rely on multiple simple cores, each of which is complete. A central Power processor core keeps them working together. I assume (but I do not know!) that the benefit of this architecture is : (a) adding multiple cores is easy and increases cost linearly (b) software that works for a 16-core chip will also work for a 2-core chip, but slower (therefore the same processor can be adapted to different needs, just like multi-unit videocards, without expensive redesign) (c) an inherent understanding of parallelism (on the chip) allows chaining them together in an easy fashion. Maybe we will start counting cores instead of MHz in a few years, when all cpus will have peaked at some--obscenely high--MHz limit. Details on the cell chip are very vague and ridden with marketing buzz-words, but it appears it will be able to execute many more parallel threads than an Intel HT processor (2 threads maximum in parallel).

What worries me most is the fact that Sony (which also sells music/movies etc) says it'll have on-chip capability to protect copyrighted works. I don't know what this will mean for the GNU/linux crowd.

Disclaimer: All the above is wild speculation. I am not an engineer.

P.

Re:I may be wrong...

[Doctor Memory]

all cpus will have peaked at some--obscenely high--MHz limit

Speaking as someone who started out with a 1.774 MHz processor, current CPU speeds are already obscenely high. Hell, my disk drive has more memory (2MB vs 16K) than my first computer...

Massive Parallel Processing

[Umbral Blot]

Actually it sound more like parallel processing to me, where many CPUs are connected together to form one larger CPU. Perhaps you can remover CPUs from the network while active?? Or maybe it is just easier to expand. Their page seems to be full of hype (in my opinion), but no description of concrete benefits from this technology. Also why is this in the games section ... seem more like hardware to me.

What is a cell processor?

[Sanity]

This article provides some background.

Maybe...

[Spruitje]

Well, knowing IBM and Sony there is a great change that it will run linux.
At the moment it seems that linux is the choice for development on the PS2 and I think it will be with the PS3.

Re:Maybe...

[Build6]

i'm more curious as to whether there'll be two separate chassis/machines (one from IBM, one from Sony... or even more per Cell-partner?), or of it's just going to be one basic machine that may/may not have different corporate logos slapped on it?

(i'd think it'd almost certainly be linux, no uncertainty there :-)

hrm. actually, an even bigger question... will there be blinkenlights! *memories of the BeBox*

If this really _DOES_ come out,

[9-bits.tk]

then we probably would be seeing Linux for Cell or similar. Reading that reminds me of the XBOX-Linux and the GameCube Linux projects.

I wonder what the average speed of the processors would be? And if they'd include HyperThreading?

Distributed Processing

[Halcyon-X]

It has been stated before that the PlayStation 3 is expected to be capable of distributed processing due to the capabilities of the Cell architecture. Whether or not that will indeed be the case remains to be seen, it is certainly a lofty goal for the current market penetration (not to mention speeds) of broadband in the home. Does Sony expect these PS3s to cooperate with their Cell-based television sets?

Re:Distributed Processing

[drinkypoo]

The stated goal is for some future playstation (maybe the fourth generation) to use the cell processor and yes, to cooperate with cell devices in televisions, dvd players, et cetera. If we end up with cell PCs they'll be candidates too. They could run linux, of course. To be honest, that's the Xbox, if it were clustering, and it could have been if there were any reason for it to be. Sony will probably use some kind of IEE1394 (i.Link in Sony's parlance) possibly including 800Mbps in order to connect Cell devices. 1394 allows significant cable lengths and near-gigabit speeds today; it is intended to support 1.6Gbps and later even 3.2Gbps (over fiber.)

Re:Distributed Processing

[DigiShaman]

There is a reason they called it Cell. Put enough of them togeather and you get multi-cell neural network. Maybe a super computer with enough cells can be called a "brain"

And please, never install one of these on the planet Zebs and call it Mother Brain. Ok?

Re:Distributed Processing

[TommyBear]

The primary goal of IBM with regards to the Cell processor was that it be scalable first as a rack mounted solution. Therefore the Cell Processor Workstation (CPW) will be available first up as individual form factor boards, containing a CPP, several DPPs and other small components for I/O etc.

Re:Distributed Processing

[drinkypoo]

If the API is written well enough the developers will have little to do to take advantage of the Cell processor network, they will however have to intentionally break their code up into relocatable threads that get all their data through the API and not through calls to the disk or memory or whatever. However, this is the kind of power that would make games like Fable actually work as they were envisioned. Other Cells can be precomputing graphics, meshes, et cetera. They could be used for car AI, to allow cars in racing games to precompute a line to follow based on what they saw/experienced in the last lap. Esoterically, simulation games could use them to compute weather effects. I'm sure I'm forgetting plenty of possibilities. If they pulled it off properly, and DID implement a compatible Cell-powered PC, you'd also gain benefits from having other Sony crap in your house when using that platform, which could be a big motivator. If Sony came up with a Cell-powered PC running Linux and ported OpenMosix to their kernel/architecture then I'd pretty much be sold :)

Re:Distributed Processing

[badboy_tw2002]

Do people actually believe this (tr)hype? Were you the same people actually getting giddy about the awfully named "Emotion Engine" allowing realistic hair or somehow providing better human reactions to characters in 1999?

Console games work and develop well because of one thing: standardization of platform. If you put your game in any console of the same type, it will run the same (besides various regional differences (PAL, NTSC) and maybe some hardware changes later on in a production run, ala XBox's two DVD drives)

You do not design for "potential extra processing" from someone's TV, toaster, aibo, or whatever. You design for the LCD, which is the unit that everyone buys. You might be able to take advantage of extra hardware like voice headsets or harddrives, but even then your game has to work well without it. (Example: Xbox allows you to precache data from the DVD on the harddrive, but you still need to be able to meet loading time standards without it. i.e. you can do better than 15 seconds with the harddrive, but no worse than without).

Can you imagine the testing nightmare of "better AI" if someone has a Sony DVD player nearby? Do you test every level with every combination of chip configuration out there?

This of course has been written with the thought that this is at all possible. Well, sorry, it isn't, and the super IBM cell processor isn't going to make it so. Console games work off extremely hard deadlines, and that's the refresh rate on your TV. Every 16 or 32 ms you need to have a new frame rendered and ready to go. You can't schedule a few frames for processing on the microwave and ask for them back whenever. What your drawing depends on the real state of user input, ai, physics, lighting, scripted events, etc. The state of the game at any point in the future is unknown, and thus in those 16 ms you have to figure out what needs to be updated, how the world should change, and finally render that to the screen. The actual rendering time might not even be half of the time you have for a frame. Do you have the bandwidth to send that data out and expect it back in the same frame? If so let me know so I can get some of that!

I could see remote AI processing, MAYBE, but that still has to be able to be done on the console anyways for the LCD case. AI is one of the worst things to debug in game development as a lot of times it can be non-deterministic. You do not want to throw another variable into the testing, especially not when its hardware.

Sony has a very good marketing department for continuing to push this crap. They've said "we will use this cell technology in other products besides the PS2" and "In the future the PS platform will interact with other Sony brand components", thus meaning that maybe your PS2 can start popping popcorn or something, but that has nothing to do with processing, its just networking. But somehow the two get combined on fan sites to mean "OMG, buy 28 PS3s and Jaxter and Dax runs at 6000FPS!!!"

What you will see with cell processing is a continuation of the mulitprocessor platform the PS2 had, but in a more generic sense. This should allow very interesting stuff to be done, and while games will be initially harder to develop, there's going to be some really cool stuff coming out of this. But don't believe you're going to suddenly see a sentient household that's drawing a few extra pixels in GTA VI: The Quest for More Money.

Re:Distributed Processing

[badboy_tw2002]

These are console games, not PC. Players of console games have much bigger expectations than PC gamers as far as QC. If I play HalfLife 2, I might expect that there are some driver issues I would have to deal with, or problems if I don't have enough memory or hard disk space.

And you're completely wrong about level loading and the such. Console games run off disks. That means you cache the data in a preprocessed state in the exact order it will be read off the disk. When you game MUST load in 15 seconds (no exceptions or Sony refuses to publish your game), level loading is as completely optomized as it can be.

Where is the win with more than one processor? The bottleneck in level loading is not the processing time, its reading the data off the DVD, which on modern consoles is an asycnh process given to an IO processor with DMA. Generally the format read off the disc matches pretty closely whats in memory, because there isn't any type of LOD considerations, or considerations for how powerful the player's machine is. You can't increase texture levels because everyone displayes the same textures.

Usermods (mainly just maps), are fairly new in the cosole world. Right now I can only think of a few instances, and those are made with ingame editors and submitted to a common clearing house for approval by the publisher.

And like I stated, yes, you could offload AI, but what would be the poitn. Not graphics. The deadlines are too hard, and the bandwidth too intensive.

Re:Distributed Processing

[DeadScreenSky]

A problem with your optimism is this:
Let's say you use an 'outside CPU' to do AI computations. The game already had to do this as if you only had 'one' CPU (the PS3 itself), so it probably ended up using less than 20% of its CPU time (and probably a lot less than that) for AI. So you end up with an extra 20% of CPU time for better graphics - big deal. It's not worth the extra program complexity (and make no mistake - this capability will make games significantly more complicated).

All of this is ignoring whether the latency for network AI is even acceptable. It isn't for a lot of games. I have been playing a lot of the fighting game Dead or Alive 2 Ultimate lately. Many player actions in that game are measured in how many frames they take, with 60 total frames in each second. Many moves have periods of well less than 5 frames. The AI has to be able to respond at that kind of speed. The same is true for other popular genres like racing games.

I can't honestly think of any game task other than very non-responsive high-level AI (such as city traffic patterns or something) that would be suited to the latencies we are talking about. Maybe some types of physics calculations? That can be useful in some games, but you still go back to the problem that you aren't going to be able to realistically gain more than 30% or so of your CPU time back, without having game features in the first place that require more CPUs. That would be commercial suicide (at least in the forseeable future). Most of these high latency tasks just won't work using a 'low res' version.

Re:Distributed Processing

[badboy_tw2002]

I'm not being a luddite, I'm offering you a view from the marketing and technical standpoint of consoles. The console market is MUCH different from the PC market. While the PC is a very "open" platform, consoles are closed and done so by the manufactuers, who impose a very strict guideline on the content produced by publishers and developers for their systems. This means establishing a system of technical certifications every game must pass. These are things like loading times, stability, user interface, etc. All these guide games made for a platform towards a standard that the manufactuer wishes everyone to follow. And since they actually make the discs, you have no choice in the matter. So yes, I see that technically you COULD offload AI to enhance graphic performance, but like another poster replying to this message said, you have to do the game anyways at the same level without the extra help. Not doing so would risk fragmenting your market. Console players do not want to have to worry about having installed two extra CPU units, they just want to play the game on the back of the box. They aren't used to not getting a uniform experience. The dynamic LOD point was a case for "some platforms use lower poly models, some can perform with the full poly versions" Nothing to do with a console using dynamic LOD, and many games do. I was saying that because you have a standard platform, you can optomize all your data once for a specific set of hardware and be done with it. Even user created content can be preprocessed exactly as the memory and hardware would expect it to be by the creating console or publisher. The console is NOT a PC. The hardware remains consistant because that's one of its strong selling points - award winning and excellent games still come out 5 years after you've plopped down $300 for the system, unlike PCs where 3 years later you can't play any new games. Adding an upgrade curve back into the equation will only serve to fragment your market, and that means less dollars, and THAT my friend is why you won't see it happen....the bottom line, and that's it.

my favorite quotes

[mxpengin]

For all practical purposes, the PowerPC has been relegated to a Mac-only solution while high performance NT users have turned to Digital's Alpha....

This move puts Apple Computer in another awkward position: the company had been planning on using Windows NT in its Web servers.

And my favorite actual fact is that microsoft is going back to Power PC with the new Xbox . But Im sorry that Alpha has been erased from the map.

Re:my favorite quotes

[hypnotik]

But Im sorry that Alpha has been erased from the map.

As am I. I've always thought Alphas were some of the cooler architectures out there. And it's rather amusing to think that Microsoft had NT ported to a 64bit processor a long time prior to the introduction of the Opteron. Granted, there are alot of architectural differences between the Opteron and Alpha, but that's why the HAL existed. Too bad that Microsoft did away with a lot of the HAL to gain video speed. I bet they're regretting that now.

Anyway. Back to the Alpha, with a bit of searching you can pick up a fairly decent Alpha machine on the cheap. Look at the Personal Workstations (PWS - codenamed Miata) for some good performing Alphas. They run from 433mhz to 600mhz and will take PCI cards. More importantly, they're well supported by linux and have builtin sound, NIC, IDE and SCSI. You'll probably have to get a new graphics card for them though, as the TGA2 card that they came with isn't supported. A PCI Voodoo3 works nicely as a replacement (that's what I have in mine).

And the coolest thing about having one is that you know you'll have one of the earliest (and best) 64bit workstations around.

Girls will love you. Other geeks will fear you.
You will be (appologies in advance for this) the Alpha geek.

Cell Processor Architecture: Graphic

[amigoro]

The graphic

Re: Cell Processor Architecture: Graphic

[Alwin Henseler]

That would be "Fig.1", taken from patent #6,526,491 "Memory protection system and method for computer architecture for broadband networks" as filed with the US patent office. This describes the architecture in fairly good detail, but to what degree actual machines will match this description, remains to be seen.

BTW. the figure illustrates "the overal architecture of a computer network in accordance with the present invention"

Previous /. article provides link to this description.

Memory Requirments

[Umbral Blot]

The article says that each chip is running its own kernel. That seems like a lot of wasted energy to me. I agree that it could give a serious boost to performance. However what about the memory requirements (RAM specifically)? It sees to me that each micro-kernel is going to need some RAM of its own, and to get the promised performance you would need many of these micro-kernels. This technology may end up more limited by memory requirements than the speed of the chips.

Re:Memory Requirments

[nacturation]

It sees to me that each micro-kernel is going to need some RAM of its own, and to get the promised performance you would need many of these micro-kernels.

Keeping in mind that there are various distros which fit on a 1.44 MB floppy disk *with* userland utilities, I don't think the size of the kernel will prove to be the limiting factor on a modern workstation.

Re:Memory Requirments

[TheRaven64]

From what I've read about the Cell line, each core can run its own kernel (i.e. it doesn't have to). This provides some interesting possibilities, for example a general purpose kernel running on one, while a real-time kernel runs on another and handles things like sound. Current systems have to make a choice when it comes to scheduling algorithms:

1. Make one that works for all (or, at least, most) cases but is hideously complicated, or
2. Make one that focusses on one class of application (e.g. throughput-oriented, realtime, etc).

Most monolithic kernels choose 1. Several micro-kernels implement the scheduling algorithms in user-space, allowing them to be swapped easily. Having a large number of cores available to the system would allow this to be dynamically tweaked.

This approach seems more in line with the exokernel project than any microkernel I've looked at. If you've got some spare time, exokernel is well worth a look.

Re:Memory Requirments

[gilesjuk]

But it will be a micro-kernel that will be rather small. You'll have cell processors for doing processing work, then other cell processors acting as I/O controllers with their own kernel.

Think outside the box, equating the cell design to existing PC architecture is silly.

Besides, you said it was wasteful? aren't many clusters built of entire computers where you have display hardware, floppy drives, hard disk, RAM etc...?

Re:Memory Requirments

[Gopal.V]

> I don't think the size of the kernel

The old UNIX SYSV kernel took a whopping 54kb of memory !. I'm now running the same kernel in user space and playing around with it.

Hehe, it's a fun project for CS Majors to play around with.

Re:Memory Requirments

[brianosaurus]

The floppy example was to demonstrate that a kernel fits in well under 1.5MB, given that an entire operating system, including userland, can fit on a disk.

[pure speculation follows, as i haven't read any of the cell processor articles ;) ]

If you have, say, a 64-cell graphics workstation, you probably have it loaded with Gigs of memory, "sacrificing" a meg or two per processor for the kernels is pretty negligible.

If 2 meg/kernel (on the high side) is a significant chunk of the overall system memory, the system is probably misconfigured. Is there a practical use for a 1024-CPU computer with only 4G of RAM? Does it even make any sense (financial, technical... any?) to have such an arrangement?

Re:XBOX2 + Cell = Windows

[binary42]

No. You are mistaken. Xbox will have a PowerPC derivative. Not a POWER derivative. Also, I should note that Cell, although part is derived from the POWER4, is not really anything like the POWER4 architecture.

Platform showdown?

[Halcyon-X]

What's interesting is that how Sony and Microsoft handle their product launches may have an impact on the amount of games we see for these systems. With Linux gaining ground on the desktop (bear with me here), it is concievable that it might be a larger target for games, if not gaming development on a 64-bit workstation. Epic have already committed to an Unreal Tournament development platform on Linux (Windows 64-bit taking its time is probably also a factor).

The most interesting part, however, is that MS may be putting up .NET as the development environment for the X-Box 2. It makes sense that MS would try to leverage their gaming platform to lure developers onto the .NET platform and commit their engines to that API.

On another note, could Linux and Mono play much of a role in this if the Cell does indeed provide a Linux environment for development? If Sony is able to provide a less expensive development environment, development costs may ultimately go down and the consumer would benefit.

This could be either by the increase of choice since the bar of entry would be lowered for smaller software houses, or by cost if the games are indeed cheaper as a result; Existing engines and software could be ported or would be compatible, or due to the the ease of coding on a familiar platform.

Effects on the future of entertainment

[hussar]

From one of TFAs: The Cell workstation is designed to deliver tremendous computational power, helping digital entertainment content creators generate higher quality content with richer and more dynamic scenes, much faster than current development systems.

This points at more than just game consoles. This looks like Sony is looking ahead to a future in which they can dispense with actors entirely and rely on realistic computer generated characters. Should be a good bit of money to be saved if you don't have to pay an actor millions to star in your film. Could be other applications too: Animated news announcers with features finely tuned to inspire trust in the viewer, human-like avatars in intelligent appliances, human-like answering machines and customer service line responders, etc.

So, how far are we from the footage ala William Gibson's Pattern Recognition and the "live" entertainment ala Neal Stephenson's Diamond Age?

Re:Effects on the future of entertainment

[drinkypoo]

Movies and video games are growing closer together all the time anyway. Spider Man 2 the video game made almost as much as Spider Man 2 the movie. More and more video games are turning into movies, and sooner or later that's going to become a regular driving force behind a whole genre of moviemaking. Video games are finally getting the recognition they deserve... anything that sucks up that much time from the world deserves recognition :)

Re:Effects on the future of entertainment

[Snart Barfunz]

An actor (as opposed to a 'star') can create subtleties of expression that may be beyond CGI. Think about it - intelligence, experience and talent, directly controlling facial muscles. As opposed to a CGI-jockey with a mouse shifting polygons around. Our brains are hard-wired to decrypt those facial signals and quickly notice when they are 'off' in some way. So, yes, this might replace some actors, but only the bad ones! Oh - and porn of course.

Re:Effects on the future of entertainment

[Forbman]

So test audiences become instead screeners/raters for parametric computer beings. "Is this one seem happier, or sadder? 1 or 2?" blah blah blah, just like for getting a pair of eye glasses. Get 100 people of a certain demographic pigeonhole, and let them rip. Or, maybe it will be even more meta than that? A website, ala "Hot or Not" (whatever it's called), where people will sort of generate character appeal parameters w/o knowing they are doing it.

The trick, if I remember reading correctly, is to not try to be TOO human. Given the amount of appeal of and loyalty to some anime characters, though, it's probably not as hard as we think it might be.

Re:Effects on the future of entertainment

[drinkypoo]

They're comparable mediums because both are moving towards being two things at once: computer generated, and photorealistic. Neither genre has yet achieved completion in both at once, but both are sneaking up on it.

Games are going to reach the point where in terms of visual quality you will not be able to tell them apart from movies. In general some types of camera angle will not work in games, while others will not really work in movies. However, you do sometimes see movies with scenes in the first person, so while Half-Life 2 isn't the game, there may one day be a shooter that could be a movie.

Video Games also have been getting higher and higher quality writing over time, though some argue that has already peaked in some of the more inventive and imaginative RPGs. As they gain more acceptance and thus greater exposure, more people have been able to become excited about video games specifically because they have become more polished.

I've played my way up through video game history (I'm just old enough to have been there for essentially all of it, so I guess there is something good about the time at which I was born) and I've definitely noticed video games and movies becoming more like one another. Most of the changes in movies (besides video games made from movies, only one of which have I ever bothered to watch) are subtle and I might be imagining most of them, but the changes in video games are pretty obvious and undeniable. Meanwhile, it is clear that video games are gaining wider acceptance and especially among groups into which they have traditionally had poor penetration.

Ultimate workstation...

[CaptainPinko]

I've always wondered this --I mean it's so obvious that since it's not done it must mean it's flawed-- why doesn't Transmeta release a mobo with it's chip and a blank code for emulating the processor. Hobbyists emerge and write multiple emulator.

You'd boot into something like Grub and choose your processor. That way you could run a UltraSPARC workstation, MIPS, Itanium, or something as small as a PIC. It'd be great for cross-platform development especially for embedded users.

I'm sure processor hobbyists would spring up to fill every niche of emulator. Probably be a great proving ground for design theory.

Considering the low heat output you could have a dual/quad-processor box.

Maybe someone would figure out how to run multiple translators at the same time so you could run x86 and PPC and 68K at damn-near native speeds

To me that'd be the ultimate workstation.

Re:Ultimate workstation...

[Steveftoth]

You do realize that the whole reason that Transmeta's processor works well at all is because it's hopelessly optimized for emulating x86 instructions. And their software took years to write and it still is not 100% correct. (they still have some bugs in the x86 emulation) It's not going to be easy to do such a thing and at the end of the day what would be the advantage of running emulation at that level when you can just run a user level process to emulate a PIC, or ultrasparc or whatever you want?

I don't see the point of being able to boot into a random chip because you also have to emulate the entire computer, not just the cpu.

Even if you could emulate an ultrasparc cpu, you can't just throw it into a PC case and boot solaris, you have to use an actual SUN computer that has the right video, network and ide cards in it otherwise you'll have a broken machine. There are lots of little things that will cause the machine to break. The cpu is the heart of a computer, but it's not the only piece. They all have to fit together or it won't work. Just like you can't go and install a copy of OSX on a motherboard for the MorphOS (you can, but it's through an emulation layer, Mac on Linux) It's not at the kernel level.

Cool... "Beowulf" on steroids...

[advocate_one]

The more of these you have in your house, the faster the game/app you're playing/using will run as it will automatically use spare capacity on the other machines networked together in your house... I for one am most certainly looking forward to getting my hands dirty coding for these beauties... Bring on the Cell Processing Overlords... I'm ready.

Re:XBOX2 + Cell = Windows

[Henriok]

In all fairness, we really don't know from what processors the CPU in Xbox2 nor Cell will be derived from, but the most reliable information we have says that the Xbox2-CPU will we quite similar to PowerPC 970, but with three cores. As such it is indeed a PowerPC processor but it is also POWER4 derived. Cell on the other hand is stated to have a 64 bit Power core, and that's quite different from saying that it's POWER4 derived. IBM uses the term "Power" for both PowerPC- and POWER-processors, so it very well could be, and probably is, PowerPC-based and not POWER-based.

The core in Cell is probably an highly evolved PowerPC 440 based core since that is a quite proven, capable, lean and have a very modular design. I think it would be unwise to build Cell around a massively complex design like POWER4. It would suffer immensely from compelxity, power consumption and its monolithic design.

Real-time applications

[DCstewieG]

I'm still wondering about the real-time uses of this, i.e. PS3. Latency becomes a huge issue when you're trying to render a frame every 16ms.

Re:Real-time applications

[Anonymous Coward]

I may be completely wrong, but I would like to think that IBM and Sony have already thought about this. I very much doubt that they'd design this chip, release it, and then find out if the chip to chip latency will cause timing problems in games.

I'm going to make a wild guess here: I think that, generally speaking, one local dedicated cell processor will be used for renderinging. Any extra distributed processors (in toasters and whatnot) will be used for the AI's threaded/asynchronous world domination planning and such. Imaging playing a RPG where your Evil Opponent is in fact your toaster - and it's working on overtime to destroy you!

Seriously; being able to dedicate an entire modern CPU for an AI could make games much more interresting (although we've reached the point where lots of RAM is probably more important than CPU time..).

Windows

[MustEatYemen]

While PPC support was dropped, if I recall correctly back in the Win NT 4.0 days, NT was amazing because it was designed from the ground up because it could basically be compiled for any endian chip/any aritecture.

Since it is the core of the current and future lines of windows, the windows base should be portable to a cell based system, basically it requires some new drivers and probably tweaking of the HAL abit. The problem is that all the applications (that we all consider part of the windows os but are really just applications running on top) would need to be redone.

Microsoft would have one of these machines in house by now for they're windows teams to work on supporting. That I have no doubt, what I do doubt if microsoft will consider this important/the future and if they'll support it during the inital release (w/ longhorn maybe?) or if they'll come late and lose a large section of the market as we all jump and have to use a *nix as the desktop.

If this whole cell thing is more then hype, and is the wave of the future, Microsoft will support it.

Re:Windows

[TheRaven64]

NT was amazing because it was designed from the ground up because it could basically be compiled for any endian chip/any aritecture

I don't think NT supported any big-endian platforms. Even on PowerPC it ran in little-endian mode. Porting to a new platform was not quite a straight recompile, but it did only require porting the HAL, not the entire system. OS X works in a similar way - the Mach microkernel is used as a HAL (which is how NeXTStep ran on so many architectures with such relative ease).

Since it is the core of the current and future lines of windows, the windows base should be portable to a cell based system, basically it requires some new drivers and probably tweaking of the HAL abit

Unfortunately, since NT 3.5, Microsoft have been moving things closer to the kernel for speed reasons (meanwhile, everyone else has been moving things out of the kernel for stability reasons).

I suspect that processor independence is one of the major reasons why Microsoft are plugging .NET so much[1] (and delaying Longhorn). .NET code is CPU-independent, and so can easily be run on a new architecture. By encouraging people to start developing .NET code now, they probably hope that it will be possible to emulate any legacy x86 code at a reasonable speed on whatever chip the user happens to have (note that they recently bought the company that makes a very good x86 emulator for PowerPC). I would not be surprised if Longhorn ships on Itanium and PowerPC and comes with a transparent x86 emulator (as MacOS did for 68k code).

[1] Don't forget that .NET was conceived at about the time Intel were saying Itanium is the future and that x86 is going to die soon.

Re:Windows

[VCAGuy]

Actually, they didn't. Windows XP comes with several HALs out of the box:

"Standard PC", Non-ACPI PIC HAL (Hal.dll)
"MPS Uniprocessor PC", Non-ACPI APIC UP HAL (Halapic.dll)
"MPS Multiprocessor PC", Non-ACPI APIC MP HAL (Halmps.dll)
"Advanced Configuration and Power Interface (ACPI) PC", ACPI PIC HAL (Halacpi.dll)
"ACPI Uniprocessor PC", ACPI APIC UP HAL (Halaacpi.dll)
"ACPI Multiprocessor PC", ACPI APIC MP HAL (Halmacpi.dll)

cell phones/PDA - gaming handheld - desktop

[S3D]

The same processor powering cell phone, PDA, gaming handhel device, gaming console and general porpose workstation can be a way out of porting-emulators hell which is handheld development is for now. However there will be different OS for handhelds still probably - for examle Nokia unlikely drop Symbian in favor of Linux...

How does CELL solve the software problem?

[erice]

What'd I'd like to know is what IBM's solution to the software problem is. Software has always been the achilles heel of multiprocessor systems. Most existing programs and even most existing programmers can't use the resources efficiently. That's why we have gargantuan superscaler, out of order processors. Expensive in terms of hardware but it suits the software better.

So, why is Cell going to be easy to program, when other parallel systems aren't? The bits of that i've seen about the architecure suggests that programming might be an absolute bear.

Re: How does CELL solve the software problem?

[Alwin Henseler]

So, why is Cell going to be easy to program, when other parallel systems aren't? The bits of that i've seen about the architecure suggests that programming might be an absolute bear.

That's likely *the* key to success of this architecture. As far as I can tell, it isn't really new in a fundamental sense, parallel/distributed architectures have been around for some time. What IS new, is that this would be the first time that a) this new architecture and b) associated computing potential, hits the mass market, getting into ordinary people's hands.

You're absolutely right here. The real problem is not building such machines, but easy/effectively programming them. I suspect the success of this platform will depend on how easy IBM/Sony will be able to make that. If it's an incredible machine, but hopelessly difficult to program, I'd guess it would be a flop. But if programming it turns out relatively easy, it could be a huge success, and start a whole new era.

We'll see, time will tell...

Re:How does CELL solve the software problem?

[anothy]

while programming in a multithreaded/multiprocessor environment takes a bit more thought than programming otherwise, it's not nearly as hard as it used to be - or rather, it needn't be so. many modern languages (like my favorite, Limbo) can give you multithreading support (with or without multiprocessors) effectively for free. as long as that goes with light-weight threads (like Inferno and Plan 9 give, or with the stupid "special light-weight process" junk present in many unixes), you've got most of the battle won (there's still some design questions to answer, but all your crap work goes away). even the older languages have a plethora of thread models that work (some better than others), at least enough to make it so that you don't have to think about threading more than the problem you're actually trying to solve. in these languages it's certainly not "free", but it makes the cost/benefit tradeoff much more reasonable than it used to be.

Re: How does CELL solve the software problem?

[LWATCDR]

Of course we may be looking at a whole new way of programming. Just as objects have really replaced structured as the prefured metaphor maybe IBM is going to create a new programing language maybe c++plusP. You create several objects that all run in parallel with messages flying back and forth. Cell could be the system 360 of the 21st century. A system that can scale from a PDA up to mainframes all running one OS, and all talking to each other. Microsoft should be very afraid.

Re:XBOX2 + Cell = Windows

[Henriok]

All POWER-processors have been fully compliant (32 and 64 bit) PowerPCs since POWER3, and before that the RS64-procesosrs were too fully PowerPC compliant. So.. you are wrong in saying that most POWER-processors isn't PowerPCs since they have been since 1999, and they have been even more PowerPCs than "clean" PowerPCs since they until the 970 didn't have the full 64 bit ISA.

The ISSCC papers state that Cell is Power based, not POWER based. There's a significant difference here since IBM in its marketing use the "Power" moniker to encompass both PowerPC and POWER processors. If you have seen different papers than I have, please provide me with an URL of PDF that proves me wrong. This is important stuff :)

Probably OEM

[Henriok]

I think IBM, Toshiba and Sony eventually will license and sell Cell technology to those who are interessted. One of the core ideas is that they want to spead this technology as far at they can since every Cell based machine tap on the computational power from all other Cell based appliances in its vicinity. The more the merrier!

Cell isn't one processor, it's a class of processors. The one that will go into the workstation is more powerful than one that will fit into a PDA, or a HDTV. I think that IBM will make one workstation, and Sony will make another. They will use different boxes and logos but they probably will use a common "Cell based" lable yet unseen, just like "Intel Inside".

Windows for Power exists

[dan_sylveste]

The development kit for Xbox 2 is Windows NT4 for PPC with Xbox 2 extras.

Re:Windows for Power exists

[forkazoo]

Just so we are clear, Power4 is a PowerPC chip, so I'm not sure exactly what distinction you are making...

you are right

[Henriok]

In essence Cell is just that, but it doesn't stay there. Cell technology can distribute it's load to other Cell processosrs nearby. It's built from the ground up to use grid technology transparently. Quite revolutionairy.

Re:XBOX2 + Cell = Windows

[shplorb]

Actually, CELL is based around the 970. Expect about 80-90% performance compared to an equivalently clocked 970. Where it goes nuts is that there's a number of vector units attached that are basically "VMX on steroids" to quote one of the main guys at IBM behind this. The vector units (or Data Plane Processors as they're calling them) can also communicate between each other as well as with the central core. The workstations are actually headless server blades, each of which will have 2 CELL's on them and they'll be running Linux.

This stuff isn't bullshit, it was all disclosed Thursday at the Australian Game Developers Conference. I didn't sign a NDA so it's all good. I also fondled a PSP =]

Good for Gentoo

[should_be_linear]

Cell workstations will be 8-way tipically, which many programs (like GCC) is able to use. If claims of Power5/Cell performance are true, it means that it will compile linux kernel under 5 sec. (8-way). All system, including KDE/GNOME and standard set of apps will take less then hour. Sounds too cool to be true.

Re:XBOX2 + Cell = Windows

[TommyBear]

XBOX2 will be based on parallelized PowerPC G5 processors. It is not Cell tech. Game dev studios are using PowerPC G5 macs to develop next gen titles for the XBOX.

On-chip DRM worries

[avocade]

I'm still a bit worried that I've not heard much about the seemingly built-in DRM management of this new platform (that seem to be able to spread to all facets of technology, including toasters). According to a clause in the pressrelease by IBM and Sony from Nov. 29, the Cell processor will have:

- On-chip hardware in support of security system for intellectual property protection.

Is this the end of tampering-capable hardware (e.g. machines where you can modify the kernel, bypass DRM-systems etc) that some people have long foreseen? Anyone more in-the-meat of the technical details care to elaborate on this?

Re:On-chip DRM worries

[SagSaw]

Is this the end of tampering-capable hardware (e.g. machines where you can modify the kernel, bypass DRM-systems etc) that some people have long foreseen?

Not necessarily. There is no indication of what is meant by "hardware in support of security". It could be instructions to speed-up asymmetric encrytion, a processor serial number, a special unit that must cryptographically activated for certain instructions to function, or something else entirely. It does not imply that only signed bootloaders/kernels/OS's will be able to be run, nor does it imply that use of the support will be manditory. The ability to reject unsigned OS's or programs would be useful for a game console, but would generally not help sell *nix workstations.

Just what IS a cell processor?

[zippity8]

Nothing's official just yet, but this is WAY more interesting than studying for finals, so here we go:

Processor instructions are broken into an 'apulet', which contains data as well as code to perform an operation. This is probably why its claimed that if more processing power is needed, then its a simple task to add a new workstation and the work can be offloaded.

A cursory read suggests that its like creating a cluster of highly efficient yet simple nodes.

Corrections are welcome.

Reference: EETimes

Re:Just what IS a cell processor?

[master_p]

And how apulets are going to be extracted from serially executed code produced by a C compiler? Will the applications need to be written explicitely for Cell?

The idea behind the Cell processor is a good one...it is not entirely different than what the Transputer did 15 years ago. Transputer CPUs could be connected into a grid, and the processing power multiplied accordingly, but with one assumption:

code should have been written in a special programming language that allowed easy parallelization of code.

The idea of Transputers failed because it is highly difficult to extract parallelism from code. Special development tools were not available.

The PowerVR architecture also promised 'infinite' 3d graphics speed by just adding new GPUs, since it used tile rendering, but that failed, too.

POWER train a rollin

[Anonymous Coward]

The POWER train seems to be in full motion. No more wondering why IBM is canning its x86 desktop crap.

I infer this means a full shift into Power based architecture from IBM, they will only retain x86 server products because customers may want them, but they will not play a large role in their roadmap.

And that could be a Very Good Thing. The Power architecture is superior to all x86 implementations, including AMD64, in every way. The sooner we can break out into full uncrippled 64 bit computing the better.

An Opportunity for Apple

[ezavada]

This seems like an excellent opportunity for Apple to license Mac OS X.

I'm assuming the intruction set for the cell processor is a superset of the existing PowerPC processors, or that the missing instructions could easily be emulated. If so that would make this is a graphics workstation that could run Photoshop, Final Cut Pro, Shake, and other top notch professional software immediately. The existing user base wouldn't have to buy new versions -- their old versions would run.

As discussed many times on slashdot and elsewhere, Apple won't license their OS unless they believe they can do it without cannibalizing their existing user base. Doubtless there would be some cannibalization of the high end, but if it makes OS X the clear platform for high-end graphics workstations it could still be an overall boost to Apple. I don't really know how the current high-end graphics market sees OS X. My impression is that a surprising amount of it is on Windows, and that Apple is just holding on to its market share in this area.

Anyone with more current knowledge of the high-end graphics market care to comment?

Re:An Opportunity for Apple

[Johnny Mnemonic]

hink it's more likely that Apple will license the cell technology from IBM and Sony than license Mac OS X to them.
Buy a G5, get a PS 3 Cell co-processor on-board for free?

Maybe the inclusion of the chip costs Apple $20/unit--but they suddenly go from being the OS that games go to die, to bleeding edge; every eMac and iMac includes the ability to run PS 3 games via embedded Cell processor (and, oh yeah, you need to buy a controller).

Not knowing that much about game development, would the inclusion of the Cell CPU on the Mac MLB enable it to run PS 3 games, or is there more to it than that, like video card etc?

Finally, being a Mac Gamer would no longer be an oxymoron. I think that's one of Apple's biggest holes--and they're likely to know it too.

New Ways For Wives to Nag Their Husbands

[ngyahloon]

Wife: Honey, can you turn down the TV volume, you're stealing too much processing from the microwave and my chicken wont bake nicely.

Husband (sniggers): Yah, as if it'll make it taste better

STI Cell

[bitswapper]

Too bad 3M didn't get involved.
Then it would have been the STIM Cell processor.

Cell will be a 4.6Ghz eight-core chip initially

[doctor_no]

Here is a powerpoint and article describing more information on the Cell chip that will be shown at the ISSCC (International Solid-State Circuits Conference) conference next Feburary at San Francisco.

Technological Features for "first-generation" Cell chips:
4.6Ghz Clock Speed
1.3V operation
85 degree C operation with heatsink
6.4Gb/s off chip communication

from the article:
eight cores on a single chip
90nm SOI process

Link to Powerpoint

Link to Original Article in Japanese

Re:Effects on the future of [Actors and actresses]

[drewmca]

Well said. I think a lot of people watch ET and hear about the huge salaries and think, damn, all those actors are overpaid! What they don't realize is that about 1% of all actors make enough from acting to live on.

More to the point, it's not as if acting is the biggest expense on a movie. Most movies, the film stock alone costs more than most of the actors. When a film does have a huge actor salary, it's for a reason. The producers sign Julia Roberts for $20million because they know that her name alone will make them more than that at the box office. So bringing in a whole CGI team to replace the actors doesn't exactly sound like a cost-effective measure to me, since you'd need a few people plus a lot of equipment to do the work of one actor. I'm not even going to get into what acting actually involves and how you can't just program it, because if I have to argue that point with anyone, it's a waste of time.

32-bit Windows on 64-bit alpha

[Tony]

And it's rather amusing to think that Microsoft had NT ported to a 64bit processor a long time prior to the introduction of the Opteron.

They never did port MS-Windows to 64-bit alpha; it only ran in 32-bit mode. Compaq was involved in the 64-bit port, but announced in 1999 that it was foregoing 64-bit development in favor of IA64.

Dave Cutler *did* get some early versions of 64-bit Win2k to boot on an AlphaServer, but since Compaq lost interest in developing Win2k for the Alpha (both 32-bit and 64-bit versions), MS decided to pull the plug on Win2k for alpha entirely.

This was right at the time (late fall of 1999) that Intel sent out the first of the Itanium chips.

Anyway, MS never did finish development on a 64-bit version of MS-Windows on Alpha.