Catch up on stories from the past week (and beyond) at the Slashdot story archive

 


Forgot your password?
Close
typodupeerror
×
Hardware

Playstation 3 CPU Almost Finished? 435

Posted by CmdrTaco from the more-polys-means-more-grand-theft-auto dept.
dnxthx writes "According to this ZDNet article the design of the Playstation 3 chip is nearly complete. The PS3 chip will have near "supercomputer capabilities" --- including 1 TFLOP. Reportedly, this chip is being engineered with Linux in mind."
This discussion has been archived. No new comments can be posted.

Playstation 3 CPU Almost Finished? More

Playstation 3 CPU Almost Finished?

Comments Filter:

  • Linux in mind? (Score:5, Interesting)

    by slutdot ( 207042 ) on Tuesday August 06, 2002 @01:06PM (#4019156)
    Cell's designers are engineering the chip to work with a wide range of operating systems, including Linux.

    I don't see how that sentence translates to the statement by the submitter that the chip is designed with Linux in mind. Besides, shouldn't the OS adapt to the chip, not the reverse?

    • Re:Linux in mind? (Score:2, Informative)

      by wilburdg ( 178573 )
      With chip fabrication prices dropping drasticaly, and with OS complexity increasing exponentially it is becoming much more common to design hardware around software.

    • Re:Linux in mind? (Score:4, Informative)

      by intermodal ( 534361 ) on Tuesday August 06, 2002 @01:28PM (#4019373) Homepage Journal
      Linux in mind. That means that the person/persons designing it are trying to make it easy to run linux on it. This does not make it linux-specific. If I buy a NIC with a variety of OSes listed on the box from WinXP down to MS-DOS, Win3.1, SCO Unix, and Linux, it is still designed with Linux in mind because compatibility was considered in its development and it means that it will work under linux (supposedly). The reason it was used in such a manner on this article heading (the /. one) is that most people here frankly couldn't care less about whether it'll run Windows or such. Though a teraflop PS3 as a BeBox...that'd be cool

  • Hrm.. (Score:5, Funny)

    by qurob ( 543434 ) on Tuesday August 06, 2002 @01:07PM (#4019162) Homepage

    The PS3 chip will have near supercomputer capabilities --- including 1 TFLOP.

    Wasn't the old PS2 a supercomputer, and there were export rules on it?

    Saddam was rumored to buy some to control missles or something?

    • Bah. Let them have the PS-whatever. Those boxes don't boot without a controller, which means the missle WILL HAVE A PAUSE BUTTON!

      Duh.
    • I doubt Saddam bought them to control missles after being fooled the last time. [atariage.com]

    • Re:Hrm.. (Score:3, Insightful)

      by Doomdark ( 136619 )
      Saddam was rumored to buy some to control missles or something?

      Well. Considering that 8-bit computers were enough to send Voyager and Pioneer through millions of kms of space, precisely enough to still do close encounters with planets, and considering V-2 (II world war) were able to hit targets hundreds of KMs away with no computers (but brilliant engineering resulting in sophisticated non-electronic controlling system), one does NOT really need anything resembling super computer for controlling missiles.

      Others have pointed out that the Saddam-and-superchips was mostly marketing hype, which is true enough... but there's really no need for super computers or chips for calculating missiles' flight paths. There are needs in nuclear simulations, but once again, first nuclear weapons were developed with reasonably modest computational resources.

  • export controls? (Score:5, Interesting)

    by Jucius Maximus ( 229128 ) on Tuesday August 06, 2002 @01:07PM (#4019166) Journal
    Does this mean that Japan will add export controls to this like they did with the PS2 [com.com]?

    TheJapanese government realised that the computers in the PS2s were very powerful for the time and could be networked to create a crude missile guidance system.

    • Re:export controls? (Score:2, Interesting)

      by aero6dof ( 415422 )
      Guess what, it doesn't take a supercomputer to guide a missile. There is some (flawed) logic to prevent export of supercomputers, but missile guidance isn't one of them. Think encryption.

    • Re:export controls? (Score:2, Informative)

      by Ironpoint ( 463916 )


      That was some marketing BS to promote the PS2. PC hardware was already more powerful than the PS2 at the time and far more accesible. Where, exactly, did they restrict it? Bagdad? I can't sent a piece of paper to Bagdad.

      Yea, and the Mac is a "supercomputer"

    • Re:export controls? (Score:5, Funny)

      by Lord_Slepnir ( 585350 ) on Tuesday August 06, 2002 @02:13PM (#4019725) Journal
      create a crude missile guidance system.

      Actually, the atari 2600 had this. It was called missile commander i do believe.

      • Actually Missile Commander isn't used for guiding missiles. It is used for controlling the missile defense sheild (Star Wars project). That's why all of the critics point out that it is hard to stop missiles when 10 of them show up at one all heading towards different cities.

  • Thats it? (Score:3, Funny)

    by bytor4232 ( 304582 ) on Tuesday August 06, 2002 @01:08PM (#4019169) Homepage Journal
    By the time 2005 comes around, everyone will have a Terraflop of processing power in their toaster. Comon Sony, cant you do better than that?

    • Re:Thats it? (Score:5, Funny)

      by Lord_Slepnir ( 585350 ) on Tuesday August 06, 2002 @02:23PM (#4019799) Journal
      Have you played the new Unreal 2003 demo yet? People already have toasters in their computers, so computers in their toasters isn't that far off.

      You saw it here first: The FIRST TOASTER VIRUS

      if(Toast_Present == 1)
      {
      Turn_On_Coils();
      While(Toast.OnFire() = 0)
      {
      if(Lever.Manual_Eject() == 1)Lever.Jam();
      if(Power.Unplug() == 1)Power.Source = reserve_battery;
      }
      Eject_Flaming_Toast_At_User();
      }

  • engineered with Linux in mind
    Perfect for dropping off inconspicuous items [slashdot.org] in the workplace!

  • lara? (Score:4, Funny)

    by edrugtrader ( 442064 ) on Tuesday August 06, 2002 @01:09PM (#4019175) Homepage
    until they can get a 3D lara to give me a lap-dance, i'm not impressed.

    • Re:lara? (Score:2, Funny)

      by IvyMike ( 178408 )

      until they can get a 3D lara to give me a lap-dance, i'm not impressed.

      Now there's an interesting tie-in with the "force feedback controllers" article.

  • What kind of a processor is that and where did they get it? And if so, how many millions will that PS3 cost?
    • i thought that too, but maybe they are talking about triangles per second or something... 1 trillion triangles is high end nowadays.

    • Re:1 TFLOP? (Score:5, Insightful)

      by unicron ( 20286 ) <unicron AT thcnet DOT net> on Tuesday August 06, 2002 @01:21PM (#4019303) Homepage
      It's not about the fact that the PS3 can push 1 TFLOP, it's about the fact that the PS3 can push 1 TFLOP while being small and relatively affordable that makes it bleeding edge technology.

      I for one am getting kind of tired of all these technology pushes in gaming consoles while the games continue to go down hill in terms of enjoyability. Now, it may just be my age at the time, but when I remember back to being a kid and playing Nintendo, I remember more than half the games I ever played were REALLY, REALLY fun to play. I'm 23 years old and I can talk forever about old school Nintendo with friends that can remember the days. Too often these days we judge games based on their technological feats, giving a game credit for crap like "volumetric fog" and "real time shadows", etc. but we hardly ever just say "That game is just plain fun".

      I think it may be time to pick up a Gamecube, especially with 3 old school classics getting a revamp(Metroid, Zelda, Starfox). Maybe then I can relive that joy from childhood.

      • Re:1 TFLOP? (Score:5, Insightful)

        by Skyshadow ( 508 ) on Tuesday August 06, 2002 @01:50PM (#4019527) Homepage
        Of course, you were probably more easily amused as a kid. I seem to remember that a lot of games were always fooled by perfecting a single trick or strategy, then repeating it over and over.

        To me, Crash Bandicoot is every bit as fun as Super Mario (not to mention that it has great attitude), Morrowind kicks Phantasy Star's ass and Grand Theft Auto III... well, there's nothing that really compares.

        So, basically, I completely disagree with the idea that games aren't as good as they used to be. *Some* games are worthless tech showcases (I call these "Jurrasic Park games"), but then those were always around, weren't they?

      • Yeah, you fit the age bracket. I'm 23 too. I remember all the NES classics (and my 2600). The only thing I've found that amuses me in the way of console games anymore are the Zelda games for the N64 and Smash Brothers for the N64 and GameCube. Super Smash Brothers has to be one of the best games *ever*. Nintendo seems to be the only company that cares about fun anymore, instead of just eye candy. Somebody around here has a .sig that says something about Nintendo being about quality rather than quantity. That guy's got it right. I've yet to find a PS2 game other than Devil May Cry that can hold my attention for more than a day.
      • I think it may be time to pick up a Gamecube

        Do it. You'll have no regrets.

        I bought one with Pikmin, Super Monkey Ball and Rogue Leader for exactly the reasons you describe. Rogue Leader has turned out to be a disappointment but the other two are simply fabulous.

        I own a PS2 as well, but with the honourable exception of SSX and, to some extent, GT3 (not GTA3) I really haven't had all that much fun out of it. Worms is superb, but that's a PS1 game. Bomberman is good, but that's due out on Nintendo too (and that's a PS1 game too).

        I had the choice of getting either an X-Box or a Gamecube, and I plumped for Gamecube because all the X-Box stuff just looked too serious. Getting back to your point, I personally believe the X-Box to be more powerful than the Gamecube and so buying on specs alone I should have bought an X-Box. The reason I didn't was the games line-up: nothing was just straightforward, bright-coloured fun. You're an ex-Nintendo gamer - you know what I mean.

        Cheers,
        Ian

        • agreed, if you want to have fun, go with nintendo, they've never steared me wrong. As long as you don't have a fetish for running people over you'll be happy you did. Super Monkey Ball is one of the best games I've played on the current systems, and Super Mario Sunshine is looking to be fantastic (or so I've read on import review sites).
  • you haven't seen anything yet.

    In terms of scalability, the uber-parallel-processing-pipelined PS2 makes a lot of sense, and will continue to get more powerful in the future as its software improves. In terms of usability though, the PS2 has irked a lot of console developers because it's an entirely different beast and doesn't behave like a PC when you get down to performance bottlenecks.

    The PS3 and beyond can only continue this trend. Sony hopefully won't make the same mistake ease-of-use wise, but the PS3 will be getting tantalizingly close to the "do everything you ever cared to do in a game" performance.

    The future of this technology is hugely dependant on software capability to make sense of and utilize it. This will be the biggest hurdle, and clearly nothing like it really exists today.

    One of the next big steps may be carbon-nanotube based computing, because it will enable architectures with massive hierarchical processing power and near limitless involatile stupidly fast memory, all embedded everywhere. Carbon (and other) nanotubes will be used for both logic and memory (as well as actual display surfaces), and ultimately be laid out more like a brain than a serial system.

    I look foward having a complete system in a display where you push morphing procedures in one end which ultimately get streamed into content on the output side.

    The networked aspect will be important too, but not how it's colored in this article. Your games will ineveitably run graphics processing on your local machine, with non-realtime and background tasks offloaded to others on the network. However, distributed simulation of gaming environments will only really make sense when players become the content producers and the worlds expand procedurally to simulate whatever ideas of interest their imaginations have conjured.

    Then I just have to ask, when game consoles power the realization of our imaginations, whose world are we going to be living in? [hint: this is rhetorical, don't answer, just think about it]

    • Then I just have to ask, when game consoles power the realization of our imaginations, whose world are we going to be living in? [hint: this is rhetorical, don't answer, just think about it]

      But if we think about it, won't it become the game? Our video gaming experience will be like some sort of existential nightmare about whether or not we're playing a game. You'll be trapped, unable to beat the level until you answer the question, but unable to answer the question until you beat the level. And we thought games were addictive now...

    • Re:If you think the PS2 architecture is weird (Score:4, Insightful)

      by Phil Wilkins ( 5921 ) on Tuesday August 06, 2002 @05:56PM (#4021582)
      > the PS2 has irked a lot of console developers because it's an entirely different beast and doesn't behave like a PC

      Noooo, the PS2 irked a lot of ex-pc developers, because it wasn't a PC, and the poor lickle PC developers got very worried when they discovered they weren't in Kansas anymore, and big unka Bill wasn't holding their hand.

      Existing console developers were already used to strange machines. You think the PS2's weird, you should have seen the Saturn, or the SNES (especially when you added in the SuperFX).

      Load balance 16 parallel cores? BRING IT ON!

  • Must be a good thing (Score:5, Funny)

    by JohnCC ( 534168 ) on Tuesday August 06, 2002 @01:10PM (#4019191) Homepage
    A large international company trusted by millions can only be a good thing for the linux community...

  • Late 2004? (Score:3, Interesting)

    by qurob ( 543434 ) on Tuesday August 06, 2002 @01:11PM (#4019202) Homepage
    At this rate, commercial production of Cell could come as soon as the end of 2004.

    The article states they've merely got the pen and paper design almost complete. No working hardware, and it 'could' end up in the PS3

    Toshiba and IBM have had more than their share of flops.

    Remember the Toshiba MPACT [vxm.com] chipset that was supposed to take over the 3D Graphics/Sound/Video market in the PC world?

    • Re:Late 2004? (Score:2, Interesting)

      by King_TJ ( 85913 )
      Yeah, that was my thought too.... If they're saying it won't be finished until at least the end of 2004, and expected for official "launch" in 2005 - that's too long a wait for the next Playstation.

      Microsoft will already have at least 2 more X-Box upgrades on the shelves by then.... virtually guaranteeing it will dominate over the outdated PS2.

      If this "cell" gets used in a Playstation, I'd bet more on it being in a PS4 - with some other upgrade in-between as the PS3.

      I already sold my PS2 (at a considerable loss, even) due to lack of interest. When I first saw Gran Turismo 3 - I thought I had to have it. After owning it a while and buying 14 more games for it, I realized that Gran Turismo 3 was about as good as it gets. Most games have considerably worse graphics, and some have worse gameplay too. I get much more out of my Pentium 4 system. In another year, PS2 will look pretty pathetic next to the current crop of PCs.

      • Re:Late 2004? (Score:2, Insightful)

        by InfernoBlade ( 472055 )
        Microsoft will already have at least 2 more X-Box upgrades on the shelves by then.... virtually guaranteeing it will dominate over the outdated PS2

        Consoles dont work like that. The SNES had about a 7 year long run, the Original Playstation had about 6 years. You dont upgrade consoles every year like MS would like, people wont blow $300 every 18 months on a console when their old one still works fine.

        MS _WONT_ be releasing upgrades for the XBox til the next line of console upgrades in 2005, and thats if they want to seriously piss off their customers, because that would be about 4 years with their console.
      • > Microsoft will already have at least 2 more X-Box upgrades on the shelves by then

        Which will completely fracture the X-box market, and make X-box development as much fun as pc-development is now. X-box 2, slightly better than the old one. Yeah, that's really going to sell.

      • Re:Late 2004? (Score:3, Interesting)

        by tgibbs ( 83782 )
        Yeah, that was my thought too.... If they're saying it won't be finished until at least the end of 2004, and expected for official "launch" in 2005 - that's too long a wait for the next Playstation. Microsoft will already have at least 2 more X-Box upgrades on the shelves by then.... virtually guaranteeing it will dominate over the outdated PS2.
        I wouldn't be surprised if Sony's goal in releasing this early info is to panic Microsoft into doing just that. The big appeal of consoles to consumers and developers is that they have long product lives relative to computers. Companies that rush the next generation to market too fast get a bad reputation with consumers and developers, as Sega discovered.

        If all goes well for the XBox, it may catch up to the PS2 in sales by the end of the year, and maybe in userbase by the end of the following year, so Sony has plenty of time. And if they panic Microsoft into releasing Xbox upgrades, they may have even longer....

    • Re:Late 2004? (Score:5, Informative)

      by slyfox ( 100931 ) on Tuesday August 06, 2002 @01:25PM (#4019335)
      The article states that they have "taped-out" the design. However, when I visited IBM-Austin last December (I gave a presentation on my research) they were still in the high-level idea phase. There is no way they could have decided on the design and completed it so quickly. My guess is this is a "test chip", like the one they did for Power4. Power4's test chip tested some of the critical circuits and such, but it was not the final design.

      That said, it seemed like they were considering some pretty wild ideas. However, I remember hearing about plans for the Playstation 2 chip a couple of years before it shipped; at the time it was hard to fathom, but when it arrived it wasn't as big a leap as I thought it was going to be. (Though still quite impressive.)

      I expect the Playstation 3 will be just as impressive, but not earth-shattering. They key will be how easy it is to write programs that take advantage of the raw computational power.

  • Chip With linux in mind eh? (Score:3, Interesting)

    by t0qer ( 230538 ) on Tuesday August 06, 2002 @01:11PM (#4019203) Homepage Journal
    Why didn't they just buy out transmeta? I know they just had a big round of layoffs, lost some big contracts, and can really use the cash right now.

    The main benifit of course would be having linus. Throw in the transmeta technology after that.

    The really scary thing about the whole sony/linux relationship is the parent company Sony is also Sony Records, one of the biggest supporters of DRM and the DMCA. It's kind of odd that they would support an open O/S that will never have DRM in it, makes me wonder why?

    --toq

  • Learn from the last Sony hype-fest. (Score:5, Insightful)

    by Performer Guy ( 69820 ) on Tuesday August 06, 2002 @01:12PM (#4019207)
    Come on, we've heard the hype from Sony before with their PS2, which was a nice system but not all it way hyped to be. OK PS3 will be an interesting piece of cheap hardware but do we have to see a round of flawed comparrisons that measure a single metrics as Sony try to promote themselves to an audience only too eagre to lap it all up. Take it all with a pinch of salt.
  • Within 3 paragraphs:
    "It will have the ability to do north of 1 trillion mathematical calculations per second, roughly 100 times more than a single Pentium 4 chip running at 2.5GHz."

    And

    "I just don't see that Cell is revolutionary, except in its marketing impact, Glaskowsky said "

    If the first statement is true, I would say that's quite revolutionary.

  • Slashdot in mind (Score:5, Interesting)

    by ucblockhead ( 63650 ) on Tuesday August 06, 2002 @01:13PM (#4019222) Homepage Journal
    Reportedly, this chip is being engineered with Linux in mind."
    Translation: the marketing guys mention Linux to get slashdot coverage.

    • Re:Slashdot in mind (Score:2, Insightful)

      by wilburdg ( 178573 )
      Don't mean to burst your bubble, but I imagine Sony marketing is chasing bigger fish than the fickle Slashdot crowd.

      • Re:Slashdot in mind (Score:5, Insightful)

        by Daniel Dvorkin ( 106857 ) on Tuesday August 06, 2002 @01:58PM (#4019605) Homepage Journal
        You know, at this point, I wouldn't be terribly surprised if Slashdot coverage actually does register on marketing radar. Sure, the number of Slashdotters is pretty small compared to the total target market of the PS3 (or any other major piece of geekware) but we're early adopters, a big enough crowd to provide a spike in early sales figures; we're also, more importantly, the sorts of people others come to for advice on what geekware to buy.
    • Translation: the marketing guys mention Linux to get slashdot coverage.

      Yeah, because otherwise, this place would never discuss a next-generation video game system.

    • Maybe I'm wrong, but aren't PS & PS2 only linux-consoles out there? As in you can actually buy a linux distro for them and the dev tools are actually linux?

      More power to them, why reinvent the wheel etc.

  • A terraflop? (Score:5, Interesting)

    by wilburdg ( 178573 ) on Tuesday August 06, 2002 @01:16PM (#4019242)
    The terraflop statistic is a little hard for me to swallow.

    The NERSC IBM SP RS/600 (the fifth most powerful computer in the world, according to top500.org) located in Berkeley consists of 2,944 processors. The processors are distributed among 184 compute nodes with 16 processors per node. Each node has a common pool of between 16 and 64 GBytes of memory.

    This machine is a 3 terraflop system. Although, I guess three PS3's could do the same...

    • Re:A terraflop? (Score:3, Informative)

      by quantaman ( 517394 )
      Yeah, I'm thinking it's a typo and they meant gigaflop. I'm not sure about Intel and AMD but I know G4's have run at over a gigaflop for a few years, right now they have a peak of 15 (dual processor). So a cheap processor for the console market hitting a gigaflop sounds about right. That would also explain the "supercomputer on a chip," as one of the big things about the G4 was that the 1 gigaflop barrier meant it qualified as a supercomputer (and a military weapon:).
      • ...one of the big things about the G4 was that the 1 gigaflop barrier meant it qualified as a supercomputer (and a military weapon:).

        Interestingly, the GAO has just completed an investigation into what constitutes a "supercomputer" these days and what the US is doing (or not doing) to control exports.

        According to reuters [cnn.com], the GAO report is critical of the Bush administration's decision to increase the limit last January "from 85,000 Millions of Theoretical Operations Per Second, or MTOPS, to 190,000 MTOPS." Not sure how MTOPS convert to FLOPS, but the article states that the average PC is about 2,100 MTOPS and that Unisys currently produces the only systems that exceed the 190,000 MTOPS limit.

        The article also mentions that the State and Commerce departments believe that the limits on processor power needs rethinking to address networked systems of less powerful computers(imagine a beowulf cluster of these, etc.).

      • Several things (Score:5, Insightful)

        by Sycraft-fu ( 314770 ) on Tuesday August 06, 2002 @03:30PM (#4020315)
        First yes, basically all current Intel and AMD chips can pull a Gflop. More or less and P3 or Athlon chip above 850mhz can do 1 Gflop in real world tests (specifically according to SiSoft Sandra).

        Second the classification of a G4 as a "wepaon" or a "supercomputer" is not correct. The way that is done is based off of theortical operations per seconds (be they interger or floating point). In 1998 that was 2,000 MTOPS (million theoritical operations per second) or 2 Gflops if you want to look at it that way. That has since changed and currently the US can export up to 190,000 MTOPS computers to "Tier-3" countries (countries judged unsafe in terms of non-proliferation of mass destruction weapons) which are places like China, Russia, and most of the Middle-East.

        Finally, Sony probably is telling the truth about Tflop perofrmance.... Sort of. I'm betting that the chip wiill have a theoritical max of 1 Tflop, which is not unheard of, provided we are talking about speical DSP operations for graphics type stuff. The GeForce 4 4600 gets about 1.23 Trillion ops per second according to nVidia. Thing is, the GeForce 4 is a graphics DSP, all it does is push pixels. It's subunits do things very fast, but can do only that one thing (ie vertex shaders ONLY do vertex transforms, not general work). A P4/G4, on the other hand, can do anything. It can do all the same kinds of calculations a GeForce 4 can, but can also do all the calculations any other DSP or system can, given enough time.

        For a long time we've had the ability to design specialised chips that ar much faster, but more limited, than general purpose CPUs. That's the whole reason for ahaving a 3d accelerator. You just can't make a CPU that fast yet, it would take hundreds of CPUs working together to equal the power of a GPU, HOWEVER that GPU is good only for graphics. You still need a CPU for general purpose calculation.

        In a video game console, the lines often become a bit more blurred. One chip may do many different things. Some of the functions traditonally on the GPU in computers might be on the same chip that happens to do CPU work as well.

    • Re:A terraflop? (Score:4, Funny)

      by guttentag ( 313541 ) on Tuesday August 06, 2002 @01:50PM (#4019531) Journal
      You misunderstood. A "terraflop" is a finishing combo in a top secret street fighter game that will be released with the PS3. This is different from the "teraflop" which refers to floating point ops.
    • Keep in mind that the PS2 consists of a whole bunch of processors. I'm in no way a PS2 expert, but I believe there's the main processor, the vector unit, one for sound (?), and maybe one or two hidden somewhere else.

      They're probably adding up all the processors. The NV30 is supposed to be floating point all the way through the pipeline. You'd have to assume PS3 would be floating as well if it wasn't already. However, the 1TFLOP number still sounds like a whole bunch of poop.

    • Re:A terraflop? (Score:2, Insightful)

      by stlc8tr ( 598851 )
      Because everyone's definition of a FLOP is different. The top500.org uses Linpack. But none of the manufacturers do because a Linpack FLOP is much harder to achieve than a "paper design" FLOP. Look at Apple's claim of a 15GFLOP G4. There is no way that a G4 with PC133 memory can even sustain 0.25 GFLOPs under Linpack but is that stopping them from claiming multi-GFLOP performance? Of course not! Sales & Marketing always has the last say in these types of things.

  • Moore's Law (Score:2, Interesting)

    by archnerd ( 450052 )
    I'm having some trouble believing that in two years there will be a consumer chip 100 times as fast as the ones today. Moore's law would say that it will be twice as fast. I'd believe 5 times and maybe even 10. But not 100. ZDNet is way too gullible.

    • Moore's Law is not a law (Score:4, Interesting)

      by Space cowboy ( 13680 ) on Tuesday August 06, 2002 @01:27PM (#4019367) Journal
      Moore's law is NOT a law, at best it's an observation that has so far been consistent.

      The cell is a highly parallel chip, it is outside the bounds of Moore's "law" because it doesn't follow the same design methodology. If I designed an FPGA today that had 1000 FPU's, and a simple CPU to control them, I could easily best a P4 in FLOPS. Trivial. Sony has done/will do in hardware what I have suggested, and given that they've been working on it for a couple of years, I think there may be more than just a couple of extra FPU's.

      All it takes is a little thought....

      Simon

      • Also, the PS2 used vector processors, and I would assume that the PS3 will as well. This means that in a single clock cycle on one of these processors, four addditions can be performed (but only one divide I believe). They are likely using something more advanced, but just adding a few of these will reduce the clock cycle needed to do this enormously.
        • 4 multiply-accumulates, 1 divide. ...and with good reason. Transforming a vertex through the perspective transform takes 16 multiply-accumulates, and 3 divides. So a ratio of 4 fmacs to 1 fdiv unit is pretty optimal.

  • Sony is pretty funny (Score:5, Insightful)

    by quantax ( 12175 ) on Tuesday August 06, 2002 @01:17PM (#4019265) Homepage
    Wow, same song, different year. Last time Sony acted like the PS2 chip was 'God-on-a-PCB'. They even claimed that they could make highend 3D dev systems that could blow the machines of that time away with super realtime rendering, etc. And now, they say they have a supercomputer-like chip. Maybe for the PS4 they can tell us about the NASA beowulf-cluster-like chip which can predict the stock market's picks up to 1 year in advance. Oh, and also create a 1:1 model of the universe, complete with infinity. Seriously, I understand that these chips are powerful, but Sony hypes this crap like its god-in-a-can. Lets not buy into it.

  • "Could" ? (Score:2, Insightful)

    by sfennell90 ( 562864 )
    "...could enter production in 2004" "...could end up inside the PlayStation 3" Doesn't sound too definitive that the PS3 will get this chip does it? This is just more marketing and hype from what I read. I also really doubt that Intel will be standing still for the next two years, so the comparison to today's processors is completely worthless.
  • You just connect a bunch of them together and you can do anything! Realign warp fields, degauss tachyon emitters, and render fighting games with big bouncy breasted women. Now THAT's a good use of a teraflop or two -- accurate breast bounce.

  • From the article:

    "It's going to take an enormous amount of software development...to really make it get up and dance." - Richard Doherty, analyst with Envisioneering

    The chip will not only perform the heavy computational tasks required for graphics, but it also will contain circuitry to handle high-bandwidth communication and to run multiple devices, sources say. Ultimately, Cell will provide a "much more interactive way of delivering content, including advertising, sports and entertainment such as video," to a wide range of Internet-ready devices. - Jim Kahle, director of broadband processor technology and a research Fellow at IBM.

    From earlier threads on here, even if it is geared towards Linux, I wonder if the impending inclusion of Palladium and other DRM would make it into a processor like this? It initially sounds like this would be an ideal candidate, since having different processes would make it easier to program just that one part to exclude your copied DVDs or your non-WMAs.

    That, in itself, might derail this from being a powerful addition to the Linux arsenal, but then again, wouldn't that be exactly what M$ would want?

  • Here's a thought. The idea behind these chips is that they combine several smaller chips (Cells) into one large one, then use multilple processor cores to control the information. Want to make it small? Just one processor core and a few Cells. Need more power? Add more cells and more processor cores.

    If this system works out, there could be a lot of power here. Now, here's the kicker: if they're really working to make this run with Linux and the like, what's to stop some other applications? X86 emulation, for example, done on the hardware level? Or, even better, PCC emulation - now Apple has access to powerful chips that were made from the ground up for graphics processing, something they're moving OS X into big time. It been thought that Apple might move from the PPC to something else (unless Motorola has some plans nobody knows about to make a faster chip) - this could be their ticket to both high power and economy of scale.

    Could this technology be used to challenge Intel/AMD? Probably not, and we'll have to wait until they announce more details. But since I'm working on some database programming, my mind is wandering a bit.

  • Great ... (Score:5, Interesting)

    by UnknownSoldier ( 67820 ) on Tuesday August 06, 2002 @01:25PM (#4019342)
    From the article:
    "It's like a beehive -- cell components can also be ganged together," he said.

    Just when I thought programming the PS3 couldn't be any *worse* the then PS2 (lots of fun debugging the EE, VU0, VU1, GS, SPU, IOP all running simulatenously on the PS2 :), along comes 'linked' cpus. Sure parallelization rocks for performance, but it's a head ache for game design & implementation. This is one thing the X-Box got right - port your PC game over in days, not months. Ok, enuf k'vitching.

    How long do we have to wait for Gran Turismo to show-case the PS3 ? ;-)

    • Programming for the PS3 (Score:4, Insightful)

      by el_benito ( 586634 ) on Tuesday August 06, 2002 @02:03PM (#4019642) Homepage

      "It's going to take an enormous amount of software development...to really make it get up and dance."

      *groans* Here we go again. One of the primary mistakes that these guys keep making is that every time they reinvent the wheel, we have to remake the cars, the highways, driver's training, etc! Having to relearn coding for the umpteenth time is going to actually shoot the PS3 in the foot severely.

      Non-ADD suffers should remember that when the PS2 originally debuted, there were significant problems with it's anti-aliasing abilities. Every two-bit flamebaiter was crowing the latest 'clever' pun like "Tekken Jag Tournament." These problems eventually diminished when software companies discovered a poorly-documented workaround in the PS2 phonebook of "Programming 101 (again!)" The second generation of PS2 games that hit just before this last Xmas was friggin incredible (Devil May Cry, FF10, GTA!). This was because programmers had finally wangled out of the system the ability to make it do what they want. This allowed them to concentrate resources on that crucial element: Gameplay.

      Moral of the story? Buy your PS3 a year after it comes out. That'll be when the games finally start getting good.

  • ...translates into some serious processing power, and it's a synergistic gain, not just an additive gain; it's possible that the combined abilities of multi-core chips will lead to some serious innovations in software design which is sorely needed as the advancement of software has lagged behind advancement in hardware in a big way. Indeed, it's the singular linearness of processors which have defined software development to date, so having processors with multiple core capabilities could lead to more capable software design and implementation.

    Think systems on a chip vs. processors on a chip and the possibilities start poping up.

  • 1 TFLOP CPU, 0 Tb/s memory bus (Score:3, Insightful)

    by tshoppa ( 513863 ) on Tuesday August 06, 2002 @01:39PM (#4019461)
    What good is a 1 TFLOP CPU if you don't have a memory bus to support it? The mark of a true supercomputer is not CPU power, but truly massive memory bandwidth. (Low-latency memory doesn't hurt, but for big vector problems it doesn't always help.)

    Caches help for little problems, but you don't put a 1 TFLOP CPU onto a little problem.

  • For example:

    While the processor's design is still under wraps, the companies say Cell's capabilities will allow it to deliver one trillion calculations per second (teraflop) or more of floating-point calculations. It will have the ability to do north of 1 trillion mathematical calculations per second,

    This was obviously from Zdnet's Division of Redundancy Division. It happens to be listed twice on the organizational chart.

  • Hey guys,

    The original Pentium had a floating point operation, FXCH [intel.com], which could be executed in zero cycles, since it only required that a register be renamed. Examples of FXCH code can be found here [df.lth.se]. Zero cycles per operation makes an awful lot of teraflops...

    I think we can therefore agree that the original Pentium whips the PS3 back into the depths of finite numbers from whence it came.

    Cheers!

    • which could be executed in zero cycles

      Huh, how is this possible. It would take some time to fetch and decode right? And the renaming would take some time right? Read the Intel docs it seems that the instruction swaps a register with memory?

      I think I'm missing something here, how can this possibly execute in zero cycles. The sample code simply says "no cost", which I don't necessarily equate to zero cycles. Please enlighten me, I'm confused.
  • Wasn't the 486's tag line "super computer on a chip" way back when in 1990....

    just two coppers worth of humorous memory...

  • Terrible article (Score:5, Interesting)

    by Animats ( 122034 ) on Tuesday August 06, 2002 @01:53PM (#4019570) Homepage
    What an awful article. The hype on the PS3 is very badly written. Something is wrong in Sony's PR department.

    What have we really got? One statement:

    • Cell will likely use between four and 16 general-purpose processor cores per chip.
    Now that's reasonable enough. IBM has built 2 and 4 CPU chips, and there was an 8-CPU DEC Alpha prototype, so it's about time to try this.

    But a teraflop from 16 CPUs? Not with anything like current fab technology. And there's no indication of a breakthrough at the gate and fab level; this is just architecture. There's a way to do it, though.

    Suppose each CPU has a 4x4 matrix multiplier built in (reasonable enough in a game machine), and each multiply-add unit can do one multiply/add per nanosecond (the PPC G4 does slightly better than that.) So we've got 256 multiply/add units (16 CPUs, remember) cranking away. That's a peak speed of 0.5 teraflop. And that's with current technology; a 2x improvement on that in the next year or two is quite possible. So a teraflop graphics engine isn't totally out of reach.

    It's not exactly general purpose, but it's a teraflop.

    Is 256 multiply/adders on one piece of silicon, running at 0.25ns/cycle, within reach?

  • Cell will provide a "much more interactive way of delivering content, including advertising, sports, and entertainment such as video to a wide range of Internet-ready devices," said Jim Kahle [. . .].

    Bah.

    So advertising is equavalent to sports and other forms of entertainment? Since when is advertising comparable to entertainment or a positive experience for the user? Ugh.

    Loomis

  • in mind? (Score:3, Funny)

    by tswinzig ( 210999 ) on Tuesday August 06, 2002 @01:57PM (#4019603) Journal
    I feel they should be engineering this chip with this chip in mind, not Linux. That just makes it hard to concentrate.
  • Now if that statement were actually true about Linux compatitbility, I could might be convinced to believe it. After all, MS and Sony are now direct competitors. If things were that spiteful between the two, making their platform Linux compatible would just be that extra dash of flavoring as they twist the knife...

  • Interesting implications for other apps? (Score:3, Interesting)

    by rutledjw ( 447990 ) on Tuesday August 06, 2002 @02:07PM (#4019680) Homepage
    Especially:

    "It's like a beehive -- cell components can also be ganged together," he said

    So at the risk of sounding dumb, could this have implications for Artificial Neural Network technology? It sounds like that could be an application of the "Cell technology". As I understand, there are currently 3 major applications for Artificial Neural Networks:

    • Games (CPU controled opposing characters)
    • Voice recognition (or character, image recognition, etc)
    • Fraud applications (Credit Card, etc).
    It sounds like the "Cell" chip could clearly be used for games, but what about the other two? Could it also maybe have commercial applications in mulitprocessing for servers / workstations? Now we have CPUs designed to work together instead of having multiprocessor support at the motherboard/OS level.

    • heh, I don't know why, but it that reminds me of that episode of Malcolm in the Middle (you know, that show on fox besides simpsons?), where they had a new smart kid in the class.

      Malcolm: "When I have an idea it's like all this little explosions going off in my head."
      New Kid: "My mind is like a behive with millions of bees, each one with a mind like yours."

      Or....something like that

  • Advertising... (Score:3, Interesting)

    by stuffman64 ( 208233 ) <stuffman@gm a i l . c om> on Tuesday August 06, 2002 @02:08PM (#4019687)
    From the article:
    Ultimately, Cell will provide a "much more interactive way of delivering content, including
    advertising , sports and entertainment such as video," to a wide range of Internet-ready devices, said Jim Kahle, director of broadband processor technology and a research Fellow at IBM.
    Great, does this mean we get pop-up ads during our games? Or how about pop-unders that appear when game is turned off?

    Most likely, though, they mean some sort of dynamic ad generation, like billboards in games for advertising that change to target certain audiences. Anyone remember when games didn't have ads at all?
  • Sony makes some awesome hardware, but don't make the mistake of thinking it will be general purpose like an Athlon or Pentium IV. The way this works in the PS2 is that vector instructions can process four values at a time, and there are multiple, almost completely independent, CPUs with these instructions. So we're talking about a custom, multi-processor, highly parallel system. It's going to take special case code to exploit it, it's not like off the shelf Linux will start getting 1 TFlop performance.

  • Almost ready? I think not... (Score:3, Insightful)

    by Thai-Pan ( 414112 ) on Tuesday August 06, 2002 @03:01PM (#4020099) Journal
    The only thing even close to being almost ready about the PS3 is that the processor has been taped out. This means that they have the plans on paper for the chip -- that's it. There's no working chip, no fab process figured out yet, no software, no sound or off-core GPU (if there is one?) or anything. Claiming the PS3 is almost ready is like a real estate agent claiming your new house is almost ready when all he has is a blueprint with no lot, and no materials.

Slashdot Top Deals

The moon is made of green cheese. -- John Heywood

Close