• Hey Guest. Check out your NeoGAF Wrapped 2025 results here!

IBM's Cell processor could reshape entertainment

pcostabel

Gold Member
Forbes

Some techies say PlayStation 3, which may debut by midyear and could end up in 100 million homes in five years, will usher in the next microchip revolution. The Sony system owes its prowess to a microprocessor called Cell, which was cooked up by chip wizards at IBM (with help from Sony and Toshiba) at a cost of $400 million over five years. The Cell chip, based on a design inspired by supercomputers, runs at least ten times as fast as Intel's most powerful Pentium. More important, Cell boasts a staggering fiftyfold advantage in handling graphics-intensive applications that will define the next generation of visual entertainment--blindingly fast and seductively immersive games, virtual-reality romps, wireless downloads, real-time video chat, interactive TV shows with multiple endings and a panoply of new services yet to be dreamed up.

BM reckons Cell, potent and versatile, can do a lot more than just play games. It sees a role for it in mobile phones, handheld video players, high-definition televisions, car design and more. Scientists at Stanford University are building a Cell-based supercomputer. Toshiba plans to use the superchip in TV sets, which one day could let fans watch a football game from multiple camera angles they control. Raytheon is set to use Cell in missile systems, artillery shells and radar. Other companies envision new high-definition medical imaging. "Cell is the next step in the evolution of the microprocessor. It's a peek into the future," says Craig Lund, chief technology officer at Mercury Computer Systems, which makes medical and military systems and is taking orders for Cell servers.

IBM is already at work on beefier versions of Cell, and it has launched an allout campaign to woo a new generation of code-crunchers and game boys to write software for its futuristic chip. In an extraordinary move IBM disclosed hundreds of Cell's design secrets on the Internet, releasing a developer's guide that 10,000 programmers have since downloaded. IBM, with annual sales of $94 billion, says Cell could power hundreds of new apps, create a new video-processing industry and fuel a multibillion-dollar buildout of tech hardware over ten years.

"We think this is going to spawn the next generation of growth in the industry," says James Kahle, 45, the renowned chip designer and IBM Fellow who oversaw the creation of Cell. "This chip will give you performance that is not achievable with any other architecture." Adds H. Peter Hofstee, an IBM scientist and the chief architect of a key part of the Cell chip: "We're talking about everything from making TVs to shooting things up into space to building huge supercomputers." He and Kahle spend much of their time on the road, running mind-blowing demos and proselytizing prospective licensees and geek groupies.
 
It really looks to be a great proccessor for Multimdia purposes. I can't wait to see this as a coprocessor to regular computers or maybe a fully Cell based desktop computer.
 
187900.jpg
 
the current Cell is teh OLD now that Sony-IBM-Toshiba are working on the NEXT generation of Cell Processors....

......both smaller versions of the current Cell architecture and next-generation Cell architecture, as well as mini and micro Cell processors.


omg PS3 obsolete already. just kidding :lol
 
Really nice article, with a fair number of new tidbits in there.

The good news: Some designers say creating games for Cell is far less complicated than writing for PlayStation 2. "Anyone who worked on the PlayStation 2 is jumping for joy," says Jeremy Gordon, chief executive of Secret Level, a gamemaker in San Francisco that is remaking a classic 1980s Sega videogame for the new Sony box.

Maybe my memory is foggy, but do we know what this is yet?
 
AfterBurner Climax, perhaps ? a PS3 port of the Lindbergh arcade game ...

that's my guess. they (Lindbergh and PS3) both have NV4x GPUs
 
The whole article is really worth reading. A fair bit of new comment in there.

We're seeing stuff that goes dramatically beyond what we can do with the current generation [of games]," says Andrew Goldman, chief executive of Pandemic Studios, a Los Angeles outfit that wrote a series of popular Star Wars games for PlayStation 2. "And what you will see over time is going to be even more amazing." He says it will take years to fully exploit Cell's capabilities.

Mercury, which sells modules for medical gear made by General Electric, Philips and Siemens, says Turismo could make a CT scanner so fast that it will be able to paint a 3-D image in four seconds versus five minutes on an Intel Pentium. Mercury is even pushing Cell to firms that create computer-generated special effects for movies. "This chip is opening doors for us," says Joel Radford, a Mercury vice president.

The PlayStation hook inspires confidence at Raytheon, the Waltham, Mass. defense contractor, which has studied Cell for 15 months and plans to use it in scores of next-generation systems. "Sonar, infrared sensors--there are hundreds of products that Raytheon designs that could use this type of technology," says Peter Pao, chief technology officer. "Current chips are going to run out of steam. We always look to the future."
 
CELL'S GENESIS FIVE YEARS AGO BEGAN WITH AN AUDAcious challenge. Sony's
new PlayStation 2 had just debuted, and Sony videogame chief Ken
Kutaragi was already looking ahead to the next version. He told IBM he
wanted a thousandfold increase in power.

By April 2004 the first working chip came off the line at IBM's silicon
factory in East Fishkill, N.Y. The new Cell didn't deliver the 1,000X
gain that Sony wanted--but it did deliver 50X.


Kutaragi's dream: 1,000x increase
full Cell Processor:roughly 50x increase
PS3's watered-down Cell: roughly 35x increase


to keep this Sony-positive, I will say that Kutaragi will probably reach his dream for PS3, with the PS4 - which would be quite an acomplishment.
 
choplifter said:
Kutaragi's dream: 1,000x increase
full Cell Processor: roughly 50x increase
PS3's watered-down Cell: roughly 35x increase

Aim high, and you'll probably hit high even if you don't make it all the way. And they did. Anyway, it's arguable what 1000x more powerful means - 1000x the fp capability? 1000x faster in a particular task? Just a guess, but I'd say you could find some code that runs 1000x faster on PS3 Cell than on the EE.
 
choplifter said:
Kutaragi's dream: 1,000x increase
full Cell Processor:roughly 50x increase
PS3's watered-down Cell: roughly 35x increase


to keep this Sony-positive, I will say that Kutaragi will probably reach his dream for PS3, with the PS4 - which would be quite an acomplishment.

XBOX 360 Xenon: 1.5X increase?
 
MassiveAttack said:
XBOX 360 Xenos: 1.5X increase?


heh, Xbox Intel CPU to IBM Xenon CPU is something like a 76x increase if you go by Microsoft's spec (115 Gflops vs 1.5 Gflops)


Xbox Nvidia NV2A GPU to Xbox 360 ATI Xenos GPU is a 3~5x increase in raw power and a larger increase in pixel shader power, but not as much of a leap as Xenon CPU is over Intel Xbox CPU.
 
gofreak said:
Aim high, and you'll probably hit high even if you don't make it all the way. And they did. Anyway, it's arguable what 1000x more powerful means - 1000x the fp capability? 1000x faster in a particular task? Just a guess, but I'd say you could find some code that runs 1000x faster on PS3 Cell than on the EE.

Eh, seeing how the EE handled most things in PS2, PS3's Cell isn't that alone this time with the RSX handling a good chunk of graphics workload. Overall system performance should shot past 35x easily. EE handled most things like texture and lighting while GS was a very fast rasterizer.
 
oh and I think that Sony's leap (with Toshiba's help) from PlayStation to PlayStation2 was a much more impressive acomplishment than PS2 to PS3.

PS3 is what it is mainly because of IBM and Nvidia -- without them, PS3 as it is would've been impossible. PS2 was mainly a Sony effort on the graphical side and a collaboration with Toshiba on the Emotion Engine side.
 
HomerSimpson-Man said:
Eh, seeing how the EE handled most things in PS2, PS3's Cell isn't that alone this time with the RSX handling a good chunk of graphics workload. Overall system performance should shot past 35x easily. EE handled most things like texture and lighting while GS was a very fast rasterizer.

Well indeed. If you're comparing system to system, including graphics chips, the difference is much wider. Simply comparing EE FP to Cell FP really understates the gap in so many other ways too, like the memory heirarchy (which is as important to Cell's performance IMO).
 
choplifter said:
*random crap*

The 1000X comment was allways in context of the PlayStation3 holistically; it's likely to be on the order of this mark over the PlayStation2 when you look at the amount of computation in the device. Granted such a metric is hard, if not impossible, to objectively decide on, but the appreciable gain is there.

An example with preformance as a corollary is that the Graphic Synthesizer had 7.35M logic transistors, the RSX has over 300M. And note that as logic budgets increase the amount you can devote to computation increases nonlinearly and outpaces a linear approximation.

choplifter said:
oh and I think that Sony's leap (with Toshiba's help) from PlayStation to PlayStation2 was a much more impressive acomplishment than PS2 to PS3.

Utter bullshit; I'm sorry but it is irreconcilable with the real world

And, as an aside, personally, anytime I hear someone say "Kutaragi's dream" my eyes roll...
 
Eh, seeing how the EE handled most things in PS2, PS3's Cell isn't that alone this time with the RSX handling a good chunk of graphics workload. Overall system performance should shot past 35x easily. EE handled most things like texture and lighting while GS was a very fast rasterizer.


Well indeed. If you're comparing system to system, including graphics chips, the difference is much wider. Simply comparing EE FP to Cell FP really understates the gap in so many other ways too, like the memory heirarchy (which is as important to Cell's performance IMO).


well I don't know if I can agree that PS2 to PS3 is going to be MUCH more of a leap in graphics then several dozen times more. certainly not 100x more. even Nvidia's CEO said that PS3 CPU is in the ballpark of 50x over PS2 GS, and he is not known for underestimating his own GPUs.
 
gofreak said:
Well indeed. If you're comparing system to system, including graphics chips, the difference is much wider. Simply comparing EE FP to Cell FP really understates the gap in so many other ways too, like the memory heirarchy (which is as important to Cell's performance IMO).

I say, if people really wanted a huge number to play with regarding system power, maybe pixel shaders performance? Haha, that right there is big number right? Sony went from no real pixel shading to SM3.0.
 
fear not, Vince, I think PS3 is pretty damn impressive. plus I will wait until later this year before making ultimate judgements, but I still think the leap from PS1 to PS2 will still seem larger than the leap from PS2 to PS3.
 
choplifter said:
Kutaragi's dream: 1,000x increase
full Cell Processor:roughly 50x increase
PS3's watered-down Cell: roughly 35x increase


to keep this Sony-positive, I will say that Kutaragi will probably reach his dream for PS3, with the PS4 - which would be quite an acomplishment.

The numbers are not accurate comparison, as it basically just going by FLOPs.

Due to the parallel nature of CELL, certain tasks are several magnitudes faster as evidenced by the CELL white papers and other articles posted around here. Specifically, many of the tasks mentioned in the Forbes article are the ones that have serious gains versus competitors’ CPU’s.




HEY SHOG –

DOES THIS SOUND LIKE THE EMOTION ENGINE ALL OVER AGAIN???!?!?!?

:D
 
HomerSimpson-Man said:
I say, if people really wanted a huge number to play with regarding system power, maybe pixel shaders performance? Haha, that right there is big number right? Sony went from no real pixel shading to SM3.0.

Exactly, the question is of metrics and how you quantitate these things. What comes to mind based on what you said is Mikkelsen's paper on DOT3 approximation using the Graphic Synthesizer and VU1. It takes 4 passes to even attempt to approximate a contemporary hardware solution on the GS and how do you measure the 'oppertunity cost' - if you will - of using VU1 cycles. Now extrapolate this out to every surface in a 1080p/720p next-generation game of vastly more complex enviroments. Nevermind lighting models, newer mapping techniques, physics, it's just huge.
 
Top Bottom