NVIDIA's 1.4 Billion Transistor GPU: GT200 Arrives as the GeForce GTX 280 & 260
by Anand Lal Shimpi & Derek Wilson on June 16, 2008 9:00 AM EST- Posted in
- GPUs
Folding@home Now on NVIDIA
Folding@home, for those who don't know, is a distributed computing app designed to help researchers better understand the process of protien folding. Knowing more about how protiens assemble themselves can help us better understand many diseases such as alzheimers, but protien folding is very complex and takes a long time to simulate. the problem is made much easier by breaking it up into smaller parts and allowing many people to work on the problem.
Most of this work has been done on the CPU, but PS3 and AMD R5xx GPUs have been able to fold for a while now. Recently support for AMD's R6xx lineup was added as well. NVIDIA GPUs haven't been enabled to run folding@home until now (or very soon anyway). Stanford has finally implemented a version of folding@home with CUDA support that will allow all G80 and higher hardware to run the client.
We've had the chance for the past couple days to play around with a pre-beta version of the folding client, and running folding on NVIDIA hardwarwe is definitly very fast. Work units and protiens are different on CPUs and GPUs because the hardware is suited to different tasks, but to give some perspective a quadcore CPU could simulate tens of nanoseconds of a protien fold, while GPUs can simulate hundreds.
While we don't have the ability to bring you any useful comparative benchmarks right now, Stanford is working on implemeting some standard test cases that can be run on different hardware. This will help us actually compare the performance of different hardware in a meaningful way. Right now giving you numbers to compare CPUs, PS3s, AMD and NVIDIA GPUs would be like directly comparing framerates from different games on different hardware as if they were related.
What we will say is that NVIDIA predicts that the GTX 280 will be capable of simulating something between 5 and 6 hundred nanoseconds of folding per day while CPUs are going to be two orders of magnitude slower. They also show the GTX 280 handily ahead of any current AMD solutions by high margins, but until we can test it ourselves we really don't want to put a finer point on it.
In our tests, we've actually seen the GT200 folding client perform at between 600 and 850ns per day (using the timestamps in the log file to determine performance), so we are quite impressed. Work units complete about every 20 to 25 minutes depending on the protien and whether or not the viewer is running (which does have a significant impact since the calculations and the display are both running on the GPU).
Hardware H.264 Encoding
For years now both ATI and NVIDIA have been boasting about how much better their GPUs were for video encoding than Intel's CPUs. They promised multi-fold speedups in performance but never delivered, so we've been stuck encoding and transcoding videos on CPUs.
With the GT200, NVIDIA has taken one step closer to actually delivering on these promises. We got a copy of a severely limited beta of Elemental Technologies' BadaBOOM Media Converter:
The media converter currently only works on the GeForce GTX 280 and GTX 260, but when it ships there will be support for G80/G92 based GPUs as well. The arguably more frustrating issue with it today is its lack of support for CPU-based encoding, so we can't actually make an apples-to-apples comparison to CPUs or other GPUs. The demo will also only encode up to 2 minutes of video.
With that out of the way however, BadaBOOM will perform H.264 encoding on your GPU. There is still a significant amount of work being done on the CPU during the encode, our Core 2 Extreme QX9770 was at 20 - 30% CPU utilization during the entire encode process, but it's better than the 50 - 100% it would normally be at if we were encoding on the CPU alone.
Then there's the speedup. We can't perform a true apples-to-apples comparison since we can't use BadaBOOM's H.264 encoder on anything else, but compared to using the open source x264 encoder the performance speedup is pretty good. We used AutoMKV and played with its presets to vary quality:
In the worst case scenario, the GTX 280 is around 40% faster than encoding on Intel's fastest CPU alone. In the best case scenario however, the GTX 280 can complete the encoding task in 1/10th the time.
We're not sure where a true apples-to-apples comparison would end up, but somewhere between those two extremes is probably a good guesstimate. Hopefully we'll see more examples of GPU based video encoder applications in the future as there seems to be a lot of potential here. Given how long it takes to encode a Blu-ray movie, we needn't even explain why it's necessary.
Facebook
Twitter
RSS
108 Comments
View All Comments
skiboysteve - Tuesday, June 17, 2008 - link
FANTASTIC write up on fine-grained TMT. I was unaware about this threading technique and was always thinking of this in class or whenever someone would talk about hyperthreading. this technique was literaly in my head for well over a year and I didn't know what it was called or that it even had a name. I always thought there had to be a more elegant way than hyperthreading to do multithreading down at the chip level without doing the OS style time slicing.i was sitting there wondering how the hell the schedule and run these SPs and then bam whole page about it
really appreciate the effort that goes into researching the core of these chips. i know not everyone likes it but for guys that are educated and work in the field its really interesting
DerekWilson - Tuesday, June 17, 2008 - link
remember though that this type of fine-grained TMT only has payoffs in systems running millions of threads concurrently.on an OS you'll see hundreds or even thousands of threads on heavily used systems, but there still wouldn't be enough concurrent action to justify this type of architecture for general purpose computing.
of course, as developers push towards an effort to thread their code as much as possible, who knows what architectures might be worth exploring on the desktop ...
coder0000 - Tuesday, June 17, 2008 - link
Very well written! A couple of points:1) Last week at WWDC Apple announced OpenCL as an alternative to CUDA. It's a C99 based HLL for creating compute kernels that can be deployed to GPU's and CPU's. Today Khronos officially announced a working group for this, and NV is a part of the committee. As such, your wish for an industry standardized compute language similar to CUDA that runs on all platforms and vendors HW may not be so far off.
2) I believe your interpretation of how multiple threads simultaneously execute in an SM is incorrect. Per thread context switching is not free, and you would never be able to execute a different thread every cycle in the manner described. There is far too much context that needs to be swapped out, and there would be significant power implications for doing that, in addition to the latency. Instead, I believe what NV is claiming is that any given SP executes a single thread. All threads in the SM can all be a single warp, but you can also have multiple threads (one per SP) all executing simultaneously in an SM.
DerekWilson - Tuesday, June 17, 2008 - link
1) I haven't had a good chance to look at OpenCL, but I certainly hope that if it's everything everyone is saying it is in the comments here that it takes off in a bigger way than CUDA :-)2) it does not context switch per thread -- warps define a context, and you have 32 threads grouped together. these threads all share the same instruction stream, which is why if threads in a warp take different directions on a branch all 32 threds must follow both paths.
NVIDIA has flat out stated that every schedule clock a new warp is scheduled and that it takes 4 clock cycles to process one warp on an SM. For both of these to be true, we conclude that the scheduler alternates scheduling SPs and SFUs on altenating clocks which means the SPs would be scheduled every 4 clocks relative to itself.
On 8 SPs per SM, you some how need to execute 32 threads in 4 clock cycles. This makes sense if you execute 4 threads per SP in some way. The details at this point are fuzzy though.
regardless, if an SP executes 4 different threads from the same warp, there is no need to context switch to execute any of these threads -- again, threads in the same warp share context.
skiboysteve - Tuesday, June 17, 2008 - link
could be a large explanation of the 2x register file size. and remember that the SP doesn't have to worry about the context switch, the SM handles having the data in the right placeanandtech02148 - Monday, June 16, 2008 - link
From this conclusion, Amd seems to be the shrewd player, let nvidia and intel duke it out in the high voltage, heat, meaningless speed gpu while Amd can pull something like its first dualcore or athlon64 for the win.this new beast from Nvidia will have how many developers making games for it right away? i'm guestimating maybe 2yrs-4yrs down the road we'll see a decent title that take full advantage of this hardware.
by then Amd will have something of a midrange that can more than handle the games.
2 things nvidia could work on that it already has, the ps3 market, and small graphic devices to improve profits. shrink the ps3 gpu further so Sony can shrink it's machinel and sell more.
PrinceGaz - Monday, June 16, 2008 - link
The GT200 core may be a technical masterpeice in terms of actually making something that big which is fully functional on GTX280 cards, but it seems to me the penalty of fabbing it at 65nm negates much of the benefits of such a wide GPU.They've had to drop the clock speeds throughout presumably because of the ridiculous amount of heat such a large core generates, which means the ~60% performance advantage in current games over the G80 core at similar clock-speeds is somewhat reduced.
Given that ATI are not producing their 55nm cores in AMD's fabs but instead are getting them churned out reliably elsewhere, nVidia have made a mistake this time around in having their high-end product rely on previous-generation fabrication as it makes it run too hot to allow the clock-speeds needed for it to be the product it should be. There is always a risk in transitioning to a smaller fab technology, and nVidia suffered badly in the past by doing so too early, but with a chip the size of the GT200, they really should have gone to 55nm even if it meant a delay of a month or three, whilst the smaller cut-down derivatives were rolled out first.
ekpyr - Monday, June 16, 2008 - link
Great article, but what about the microstuttering issues present in Nvidia's 9800GX2 cards (both SLI and Quad-SLI)? There is very little discussion on this, but I've seen some benchmarks where the FPS floor is 4fps with the 9800GX2s. Can you add a subjective review of whether or not the actual gameplay is smoother with the GTX280s across these games? Aggregate numbers may say one thing, but I've returned a 9800 GX2 Quad-SLI setup because it was unable to handle the incredible amount of texture loading that was done in Age of Conan (2560x1600 4xAA 'High' settings = 4fps). The 8800 GTX Tri-SLI configuration I am currently using is more resilient to microstuttering with its increased bus and memory capacities, but I'm very curious about the GTX280s and their increased memory and bus on texture-heavy games like Age of Conan.DerekWilson - Monday, June 16, 2008 - link
the only game that came close to having this issue with quad sli for us was oblivion.in that game at high res lag and stutter are unbearable and the game is unplayable.
we didn't notice any stuttering issues with a single GX2.
i'm working on some analysis tools to show details like this better in future articles.
TheJian - Monday, June 16, 2008 - link
I find it humorous that nobody discusses the fact that the shrink has already taped out and will likely be out in two months or just after. This humongous chip was only released so that when AMD releases in the next few weeks they will be behind still in single GPU cards. This is basically what Intel does to AMD every time AMD has a better chip. For all intents and purposes this is a PAPER release of what will come in 2-2.5 months (In Intel's case they just show you what will be out 6 months from now, and a large portion of people don't buy an AMD because Intel might be ahead by xmas...LOL - works like a charm every time AMD is ahead). THE DIE SHRUNK CHIP! Most likely with faster speeds. I suspect they'll come with "ULTRA" version first (and stick it on top of the price heap, so as to not kill all FAT cards in the channel already) and then filter down as these big suckers leave the channel. That's if they even plan to sell more than a few of these to begin withat 65nm. It's only out there so AMD won't look any good in two weeks.MIND SHARE is everything, which is why Intel's KING of the paper launch when behind strategy. They've even went to doing it for all chips no matter what now. Nehalem scores 6 months before availability. AMD's marketers have no clue an should be fired. You have to play the same DIRTY game as your enemy or you've already lost. If AMD had half a brain in their head they'd paper launch an ultra or 2x4870 version for the same reason...LOL. Then claim "our 4870x2 makes nvidia look like crap for $600"...ROFL. Who cares when it's available, just say it. Having said that, Nvidia will wipe the floor with them in 2 months anyway on a 2xGTX280 that's die shrunk. Which is all they are doing today...BUYING TIME!