An Overview of Intel's Teraflops Research Chip

[kenc51]

It will apply to whole new computing concept - not only games...

True, but I think it will be severely limited to ONLY business applications...

~Ibrahim~

It sounds like you still think this is something they are going to bring to the market! It ain't gonna happen.

It's for research.......nothing more!

[T-Shirt]

but something based on this will research will reach the market in 5 years or so.
Only the dozens (or hundreds) of cores won't be the simplifed ones on the research core, but top end server cores 3 or 4 generations beyond what we have today.
OR
maybe the end results from this research product in quantum computing.
http://gizmodo.com/gadgets/gadgets/quan ... 236486.php (don't let the title fool you this is a big step forward)
no you won't see it on your desktop (it needs supercooling) but it might mean the weatherman can tell you today the weather on a day next fall AND BE 100% CORRECT (well, 99.5 per cent of the time )

[ikjadoon]

It will apply to whole new computing concept - not only games...

True, but I think it will be severely limited to ONLY business applications...

~Ibrahim~

It sounds like you still think this is something they are going to bring to the market! It ain't gonna happen.

It's for research.......nothing more!

I know this particular prototype won't get to the market, but what is the point of research if you never use it in a real product??

I'm sure sometime down the future, it's architecture might start to show up in Intel's chips, but I don't think it will appeal to gamers, even then. It takes at a bare minimum 1 year to make a game, at least 3 for a good game, and possible more for a GREAT game. No way that they will code games for 80 cores, hell, we can't even do two!

I'm beating a dead horse, by now: multi-core is intended for business...Helps slightly in single threaded games and a tad more in dual-core games (like the two ever made), but they are otherwise useless for the gamer.

I don't mean to say that they are bad, heck, I have one! It helps with the few encoding tasks I need to do and I think it helps in AutoCAD, not sure, but not intended for the gaming community. But, as always, we will buy the fastest ones, even if they are only faster on paper.

~Ibrahim~

[liqnit]

When Game developers will start working with multi core
we will see amazing and very fast games
each new card (such as the 8800) is in its based multi core

[ikjadoon]

When Game developers will start working with multi core
we will see amazing and very fast games
each new card (such as the 8800) is in its based multi core

I disagree that they will become faster. I think they will just become much more realistic.

A lot of people forget that GPU's are extremely multi-threaded compared to a CPU, even a quad-core CPU.

~Ibrahim~

[liqnit]

Image Quality & realism will be better but speed is always important
and most important is much better AI

[ikjadoon]

Image Quality & realism will be better but speed is always important
and most important is much better AI

AI and physics will take a huge leap when we get multi-threaded games. Can you imagine your enemy being smarter than you? Today, AI = a dummy running into the wall. It'll be enormous.

Physics: Ageia has made a product that (sucks) shows us where we can go, but certainly not the limit of our potential.

~Ibrahim~

[BenT]

Good lord, it must take a monster to cool that thing..

[DMB2000uk]

Not much more than a standard cooling setup. Its suprisingly cool running.

Some quotes from a presentation toms hardware attended that just state again what this processor is for:

At this time, the CPU is built to explore floating point capabilities on processors that could open the door to new types of applications.

...during a presentation, Intel dampened our hopes that such a processor would be in PCs or servers anytime soon. The company said that the 80-core chip is just a research chip that will not become a product for the commercial market. But, technologies that are developed within its terascale project could trickle down into mainstream products.

The similarities with AMD's future roadmap go even further, as the company conceded that there may be a limit to how many processing cores on one die make sense. With the arrival of the first dual-core computers Intel often mentioned that the future will bring dozens or even hundreds of cores on one processor; now the company says that there could be a "sweetspot" for the number of cores on a chip. "At some point, the cores are getting into each other's way," an Intel representative said. "It's not just about adding cores. Other improvements are needed as well."

Specifically, Intel indicated that, in the current environment, processor will increasingly gain from the simple addition of cores until 16 cores are reached. After that, the baseline performance of processor will benefit less from the addition of cores and other enhancements will become more important. According to Intel, cache improvements will take the center stage, followed by thread scheduling and new instructions.

Dan