Congratulations to Vince

[V3]

The question I have here is for what Multimedia applications

For serving the multimedia data. According to them that isn't an easy task, when the data set gets very large.

 

[Titanio]

Alias guys are working with STI on something CELL related..

Any other details on that? I'm guessing it's just to get their software on these graphics workstations that are suggested?

 

[nAo]

Any other details on that? I'm guessing it's just to get their software on these graphics workstations that are suggested?

I don't have other details. It would be nice indeed to have something like Maya running on CELL

 

[V3]

Maya already runs on Linux, so I think it would be in optimising Maya for cell, if any.

 

[pc999]

The question I have here is for what Multimedia applications and when? In its current form its a fairly large chip, and in widescale consumer applications it will still be so in a shrink or so's time.

With the price of high end products in these market, I doubt that price would be a big worry, power/heat probably would be more.

It is like the Intel which should enter in TV market and cut the prices in half see

Intel Inside... Your Television -
http://archive.gamespy.com/ces2004/intellcos/

This is other kind of tech but shows that this market can give a lot to others, like image improving (aka WEGA), digital content like record one chanel and see 4 others at same time etc...

 

[wco81]

Sony or any other manufacturer won't cut in in the PC market anytime soon simply because whatever ship they make won't be compatible with x86 code.

Until a chip is made that can run on an OS that has as much software available to it as Windows does, no one's taking over anything.

They can go the OSX route, but then they would only take over the Mac market, which is not really huge. Good enough, and definately a good starting point, but really, there is a reason why there are billions of Windows compatible PCs in the world, and availability of software for it is the main one. How it got to that point is another discussion altogether.

The personal computer market is no longer driven by pure performance. Sure there are still gamers and people who work with digital video. But a super chip to make Office run better and enhance the surfing experience?

That's not going to have much appeal, especially when people would have to buy new software.

Nor do I see great advantages of HDTVs and other consumer electronics with the Cell. How is that going to enhance HDTV displays? Already, HDTV displays have come down a lot in price and growth is explosive right now. At the high end, there are already 1080p displays and by the end of this year, there will be dedicated silicon for decoding H.264 and VC-1. Maybe the Cell can encode these state of the art codecs on the fly?

People also talk about home servers too. Sure some enthusiasts rip their music and movies to DIY machines which they cobbled together. But there is no legal way to rip video media so you're not going to see mainstream companies offer solutions for putting your DVD collection on a server. Anyways, I don't see a need for a super chip for this kind of application either.

So right now, the killer app. for the Cell seems to be gaming.

 

[Spidermate]

Turbo Express was $300 back in 90s.

NeoGeo was ~ $400.

3DO was $700.

Do you want to know the sales figures for those?

First of all, take a good long look at how many years ago that was. Jumping to a price point that early in time, esspecially as a new comer, could completely kill your entire strategy. So your point really has no merit.

Sony, however, is leading the market right now. They can affort to slightly increase their price and still sell their merchandice. Not to say they will, but it comes to show you that a set price isn't always a guarantee that it will carry over to the next generation.

 

[Spidermate]

Instead of stating the same thing over go back and LOOK at the reasons why a 4 CELL BE is not likely right now. It has nothing to do with "Oh it is a gaming box so there is no way they stick 4 CELLs in there". Gaming box or not, you still have to deal with issues like heat and power consumption. Those are not going away anytime soon. Die size and yield (aka cost) are always a factor. 1 CELL on 90nm (200+mm^2) would be over 400mm^2 at 65nm with 4 CELLs. Die size, yields, heat, power--those issues wont go away. Put cost aside, yields could make availability very tight. And like it or not they still got to fit this all into a $300 unit. Any way you cut it there are a lot of hurdles to get to 1TFLOPs now. There is more than just wishing or saying "Well maybe they are looking bigger?" There are limits. Intel cannot slap a 5GHz chip on the market with will power alone. Manufacturing limitations and cost always catch up--even to the big boys.

I just think too many are jumping to conclusions way too early. Even if the PS3 doesn't get four PEs, many continue to claim that Sony has already publicized the final specs for the PS3 (or at least made it obvious) even though the architecture is still at a 90nm. Those figures could have been released as a smoke screen dispite how legitimate they are, we don't know yet. Assuming it will only function off 1 PE right now is kinda silly to me and completely closed-minded.

 

[Guden Oden]

The cell unveiled at the conference has two I/O links. A 35-ish GB/s link to the GPU would be heaven. A 35-ish GB/s link merely for the I/O chip would be madness.

I think it would be much more reasonable with a scenario where one cell connects to GPU and to another cell, which in turn connects to I/O...

Yet, this scenario doesn't make sense for a workstation with 16+ cells like Sony's been talking about. A one-dimensional interface with that many chips would be terribly inefficient, instead of bandwidth scaling with chips it would for all intents and purposes decrease the more chips were added. A workstation-version cell would therefore need at least three I/O links like Opteron for example.

 

[MfA]

Or just use switches.

 

[DemoCoder]

Well, at CES, almost every new TV was 1080p, and 1080p HDTV with MPEG-4 AVC takes alot of CPU to decode and even more to encode. If you want to build TVs, HD-DVDs, or DVRs can that decode and encode MPEG-4 @ 1080p in realtime, you'll need something like CELL, or, you'll need a custom MPEG-4 chip. Encoding would be needed if you wanted to take an MPEG-2 ATSC broadcast and recode it on a PVR to MP4.

Also, scaling and image processing via software would be possible as well.

CELL would make one hell of a Home media system, if you want to sell something that allows familities to splice and dice video, sort of a Living Room version of Final Cut Pro/iMovie/Adobe Premiere. Non-linear editing on PS3 hooked up to Sony HDV cam anyone?

 

[Inane_Dork]

Well, at CES, almost every new TV was 1080p, and 1080p HDTV with MPEG-4 AVC takes alot of CPU to decode and even more to encode. If you want to build TVs, HD-DVDs, or DVRs can that decode and encode MPEG-4 @ 1080p in realtime, you'll need something like CELL, or, you'll need a custom MPEG-4 chip. Encoding would be needed if you wanted to take an MPEG-2 ATSC broadcast and recode it on a PVR to MP4.

Encode, yes. Most people only need to decode, though, which is possible with current PC CPUs.

 

[nelg]

It is like the Intel which should enter in TV market and cut the prices in half see

Intel Inside... Your Television -
http://archive.gamespy.com/ces2004/intellcos/

Intel's LCOS is DOA.

 

[PC-Engine]

Turbo Express was $300 back in 90s.

NeoGeo was ~ $400.

3DO was $700.

Do you want to know the sales figures for those?

First of all, take a good long look at how many years ago that was. Jumping to a price point that early in time, esspecially as a new comer, could completely kill your entire strategy. So your point really has no merit.

Sony, however, is leading the market right now. They can affort to slightly increase their price and still sell their merchandice. Not to say they will, but it comes to show you that a set price isn't always a guarantee that it will carry over to the next generation.

You're missing the point. Sure there's nothing stopping SONY from selling PS3 for $400. Question is will they sell more or less at that price? Will Xenon and/or Revolution follow? If MS and Nintendo stays at $300 then SONY would be the odd man out. How many people would buy a $400 PS3 instead of a $300 PS3? Why would they buy a $400 PS3? What does it offer over a $300 Xenon or Revolution?

 

[mech]

Do I get props yet for claiming that Cell would run at > 2Ghz, while Vince and PC Engine disagreed?

http://www.beyond3d.com/forum/viewtopic.php?t=3477&postdays=0&postorder=asc&start=20

I still stand by my opinion that in 2005 we're unlikely to see any CPUs running at less than 2Ghz.

1-1.5GHz is likely IMHO. With the range extending upto 2GHz, but with little chance.

I did also say that P4s would be at 10ghz by now though

 

[Geeforcer]

I did also say that P4s would be at 10ghz by now though

Yeah, that didn't quite work out.

 

[mech]

I know

Even Intel thought it would happen though...