Details trickle out on CELL processor...

[V3]

So how do stream processors stack up to general purpose cpus in the realworld?

Just look at the current programmable GPUs, they should be pack with stream processors.

 

[PC-Engine]

So how do stream processors stack up to general purpose cpus in the realworld?

Just look at the current programmable GPUs, they should be pack with stream processors.

Sure but how will it perform calculating AI physics?

 

[V3]

Sure but how will it perform calculating AI physics?

Its what its good for.

 

[PC-Engine]

Sure but how will it perform calculating AI physics?

Its what its good for.

GPUs are good at calculating AI and physics?

 

[V3]

GPUs are good at calculating AI and physics?

Yep, they're very fast with large array. The only problem with PC though, the slow AGP bus, should change though with the new PCI-Express and the future.

 

[London Geezer]

This TFLOPS is starting to sound like 75 million polygons/sec all over again.

What would be fundamentally wrong with that?
At the end of the day, Sony didn't lie, from the beginning they said PS2 is capable of rendering 75 (or 66?) million flatshaded poligons per second when nothing else is being processed. And it's true. Useless to know, but it's true.
They also said real world numbers would be much lower, and in the region of 10-20 million maximum in a real game, which is also true.

Or shall we stqart on the numbers MS gave before launching the Xbox? The 300million polys, then downgraded to, what was it, 150 or something...

Everyone does it. The only ones not to brag about "theoretical" numbers this generation was Nintendo, Kudos to them.

 

[Brimstone]

So Microsoft, Nintendo, and Sony all have the same basic processor? They all have POWER.

IBM, Sony, Sony Computer Entertainment Inc. and Toshiba Unveil Cell Processor


Companies Released First Details of Multicore Chip Comprising Power Architecture and Synergistic Processor

CELL is just a codename name like Nintendo's Gekko CPU. POWER is used explicitly.

 

[one]

So Microsoft, Nintendo, and Sony all have the same basic processor? They all have POWER.

IBM, Sony, Sony Computer Entertainment Inc. and Toshiba Unveil Cell Processor


Companies Released First Details of Multicore Chip Comprising Power Architecture and Synergistic Processor

CELL is just a codename name like Nintendo's Gekko CPU. POWER is used explicitly.

So the system software which controls hardware is much more important in the next-gen. I bet Cell is useless without smart OS.

 

[London Geezer]

So the system software which controls hardware is much more important in the next-gen. I bet Cell is useless without smart OS.

Everything's useless without smart software system behind it.

 

[McFly]

So Microsoft, Nintendo, and Sony all have the same basic processor? They all have POWER.

Not at all. Power basicaly is just there to control/feed the APU's where Power is really doing the computing in Xenon.

Fredi

 

[PC-Engine]

This TFLOPS is starting to sound like 75 million polygons/sec all over again.

What would be fundamentally wrong with that?
At the end of the day, Sony didn't lie, from the beginning they said PS2 is capable of rendering 75 (or 66?) million flatshaded poligons per second when nothing else is being processed. And it's true. Useless to know, but it's true.
They also said real world numbers would be much lower, and in the region of 10-20 million maximum in a real game, which is also true.

Or shall we stqart on the numbers MS gave before launching the Xbox? The 300million polys, then downgraded to, what was it, 150 or something...

Everyone does it. The only ones not to brag about "theoretical" numbers this generation was Nintendo, Kudos to them.

First why bring MS into this? Second we don't even know if 1TFLOPS is theoretically possible from a single CELL chip. You know why supercomputers aren't ranked by peak numbers???

 

[one]

The current 1st-gen 90nm Cell processor runs at 4.60 GHz @ 1.3V / 85°C !!!!!!

Japanese article about ISSCC 2005 highlights

 

[London Geezer]

First why bring MS into this?

Because you always seem to single out Evil Sony for things that other companies are also doing.

Second we don't even know if 1TFLOPS is theoretically possible from a single CELL chip. You know why supercomputers aren't ranked by peak numbers???

Your point?
You said, the 1Tflop thing is becoming like the 75Million polygons thing.

I said, might be, it's becoming like the 75million polygons thing and the 150 million polygons thing, and the 100million xboxes sold projections thing, and the huge third party lineup for Nintendo thing...

Shame that out of all those, only one is actually correct.
so, if it's becoming like the 75million thing, it might actually be true. Theoretical, but true, like everyone has been predicting for months (1Tflop being a useless theoretical number).

 

[Brimstone]

OMG INCREDIBLE and AMAZING

Shock and Awe


Beyond3d needs to get some responses from industry titans.

That clock speed is stunning.

 

[j^aws]

The current 1st-gen 90nm Cell processor runs at 4.60 GHz @ 1.3V / 85°C !!!!!!

Japanese article about ISSCC 2005 highlights

WHOA!...Just need die sizes now for 65nm, BroadBand Engine/ PS3 projections!

 

[Bohdy]

6.4GB/s?

So that is the same bus to mainram bandwidth as the current Xbox. :?

Someone explain how this won't be a huge bottleneck, as I'm not a Cell expert.

 

[London Geezer]

Errm... How can this thing run and not blow up after 0.00001 secs at 85C??

 

[j^aws]

6.4GB/s?

So that is the same bus to mainram bandwidth as the current Xbox. :?

Someone explain how this won't be a huge bottleneck, as I'm not a Cell expert.

IIRC, off-chip XDR/Yellowstone Rambus DRAM bandwidth per 16bit channel @ 3.2 GHz

 

[Gubbi]

It's probably 6.4Gbit/s per pin. Even though that does sound a bit high.

Cheers
Gubbi

 

[j^aws]

It's probably 6.4Gbit/s per pin. Even though that does sound a bit high.

Cheers
Gubbi

Yeah...prolly, just looked at the fig. again and it's Gbit/s not Gbyte/s!