Comments on: Intel details future graphics chip at GDC

Engineers are ready to spell out the inner workings and target markets for Larrabee, Intel's first graphics chip in over a decade.

[thenet411]

Knowing Intel, I'll believe it when I see it.

[3tire]

? I'd argue that Intel has been delivering on what they promise than any other tech company.

[coryschulz]

AMD is just falling behind. Will they ever catch up?

[slickuser]

they won't be able to catchup because they are running the other way...

[pithenumber]

they have caught up for now
its called Phenom II and they plan to release 32nm in Q4

we have to wait for Core i5 to see if
Intel will regain the lead in budget and midrange

[dixonpete]

So Larrabee will effectively add 'X' number of cores to my present dual core machine?

[texaslabrat]

No it doesn't work like that. Certain applications will be able to use Larrabee in a GPGPU environment to do work, but your normal run-of-the-mill threads won't see all the extra cores. This might change in the future with future convergence of cpu and gpu technologies.

[pithenumber]

So, did they confirm the 300watt power consumption rumours?

[Tod Smith]

The Sony Cell died because of this new chip!

[Dango517]

A sleeping giant awakens! I can hear the grown from NVidia and AMD from here. Question is will this be a major excursion for the chip giant or merely scremish in the realm of the GPU. Major player have entered other arenas before some have looked around, thought otherwise and left nearly as quickly as they have appeared. Other times these encroachments by major corporations have meant the end of minor league players. Only time will tell.

Definitely the future is tilling toward graphics across the computer landscape. Graphics is the future of PC and perhaps server hardware.

Wonder what took then so long?

[Dynastius]

What took Intel so long was that last time they tried to compete in the discrete graphics arena they failed badly. I think it took this long for them to get the bad taste out of their mouth from spending millions and getting blasted by the press and having dismal sales numbers. I don't think that Intel took the graphics market seriously enough in terms of engineering efforts back then. They knew that graphics were important, but back then they only saw it as a PC gamer thing.

Since that time, non-gaming applications have increasingly made use of graphics acceleration and also, Nvidia and ATI have been tapped to make graphics processors for all 3 console platforms. (Sony, Microsoft, Nintendo) Back in the 90's, that wasn't the case. And of course, 3D acceleration is also becoming an issue in mobile phones. Add it all up, and its just too hard for Intel to ignore anymore even though they were thoroughly embarrassed last time.

It's hard to tell if Intel is more serious this time or not so far. During the 90's when they tried competing with Nvidia and ATI, they made a lot of noise too...but when the chips became available, they weren't even close to on par with Nvidia and ATI.

[belal12]

Will this translate to developers having to redevelop old games and just released games to make use of the different set of cores? I am a bit concerned that we will have this released but developers may not all take advantage of it because their budgets are already tight and having to reprogram the game to take advantage of multiple cores (like Cell) will require more investment.

[texaslabrat]

I doubt it, as long as the games are written to OpenGL or DirectX api's..the hardware should be mostly abstracted requiring very little, if any, adjustments to run as expected. Especially speaking of old games...any inefficiencies will be more than overcome by the sheer amount of horsepower present on the chip. Going forward, it will be on the shoulders of the driver writers, microsoft, and the openGL maintainers to ensure that the full power of the chip is exposed to the game programmers of the future.

[odubtaig]

Not really, OpenGL only covers realtime graphics; anything else is not only not their problem, but the ARB may well tread on a few toes if they should step outside that domain.

OpenCL ( http://www.khronos.org/opencl/ ) is designed to handle this, although the remit is somewhat broader than just GPGPUs.

[jimlynch]

We need Nano Technology to accomplish the majority of the dreams we have in electronics. If it could only move along faster.

[mnemonic]

I still very skeptical. Game graphics (geometry, setup, texture/shader ops) are best accelerated with massively parallel processors (Nvidia and AMD/ATi GPUs for example) that have a very wide, shallow memory interface.
Intel's determination to stick with in-order cores means they are going to have to run them extremely fast to keep up with a more conventional GPU, which means much high power and cooling requirements compared to a current video card (can you imagine an over-sized aftermarket CPU heatsink/fan stuck on your current video card? 'Cause that's the kind of cooling that would be required).

[3rdalbum]

Larrabee is a joke. We're looking at about 64 threads of execution, versus over 200 that you'd get on a real GPU from Nvidia or ATI.

How fast will each thread be? Not very fast, really; they are general-purpose CPU cores packed together. In terms of being able to immediately program them in the same way you've been programming CPUs, it's great. In terms of graphics speed, it's going to be disappointingly unimpressive.

Remember that there's only 32 cores with two threads each, and the second thread of each core only runs 30% of the time (hyperthreading).

If this is going to replace the integrated GPUs that Intel currently sells, then this will probably be a step up and worthy of some attention. But I really don't see how we can expect good things from this GPU.

[BigGuns149]

Agreed. I think that at best this will compete with the current low end discrete graphics(eg. Radeon 4350, Geforce 9500, etc.). The proof will be in a real world demo, but this will probably be a big improvement over their current integrated graphics at least although I don't expect this to cut much into the sales of discrete graphics.

[texaslabrat]

don't forget about the SIMD portion of each core which can run 16 operations per clock. Look at the performance of modern applications when they are SIMD-enabled and when they are not. I'm not going to predict the performance of Larrabee...but I think the comments here have been over-simplifying the issues here and overlooking where the REAL performance in the chip will likely come from. At the end of the day, it's all about FLOPS and throughput. If a programmer can figure out how to keep the SIMD units busy, there's 16 GFLOPS per core available @ 1GHz clock not counting the scalar units. I don't know what they are planning on clocking these chips at..but I'd be surprised if it was as low as 1Ghz given the tech that Intel has developed for Nehelem. Some quick back-of-the-envelope guestimation shows that Larrabee *should* be comparable to current video cards in raw power..and probably a LOT easier to program for in the GPGPU arena due to the x86 ISA it's based on. So, I wouldn't be so quick to dismiss them...better to see what Dreamworks thinks of the chips after they've played with them a while....

[odubtaig]

Also worth remembering that from SSE3 upwards there have been instructions available which perform different operations on different components of a scalar, they don't all have to have the same operation performed on them just because it's one instruction. I'll be quite interested in what these new instructions are for the 'vector' unit.