AMD's Radeon HD 6990: The New Single Card King
by Ryan Smith on March 8, 2011 12:01 AM EST- Posted in
- AMD
- Radeon HD 6990
- GPUs
Compute Performance
Moving on from our look at gaming performance, we have our customary look at compute performance. With AMD’s architectural changes from the 5000 series to the 6000 series, focusing particularly on compute performance, this can help define the 6990 compared to the 5970. However at the same time, neither benchmark here benefits from the dual-GPU design of the 6990 very much.
Our first compute benchmark comes from Civilization V, which uses DirectCompute to decompress textures on the fly. Civ V includes a sub-benchmark that exclusively tests the speed of their texture decompression algorithm by repeatedly decompressing the textures required for one of the game’s leader scenes.
New as of Catalyst 11.4, AMD’s performance in our Civilization V DirectCompute benchmark now scales with CrossFire at least marginally. This leads to the 6990 leaping ahead of the 6970, however the Cayman architecture/compiler still looks to be sub-optimal for this test. The 5970 has a 10% lead even with its core clock disadvantage. This also lets NVIDIA and their Fermi architecture establish a solid lead over the 6990, even without the benefit of SLI scaling.
Our second GPU compute benchmark is SmallLuxGPU, the GPU ray tracing branch of the open source LuxRender renderer. While it’s still in beta, SmallLuxGPU recently hit a milestone by implementing a complete ray tracing engine in OpenCL, allowing them to fully offload the process to the GPU. It’s this ray tracing engine we’re testing.
There’s no CrossFire scaling to speak of in SmallLuxGPU, so this test is all about the performance of GPU1, and its shader/compute performance at that. At default clocks this leads to the 6990 slightly trailing the 6970, while overclocked this leads to perfect parity with it. Unfortunately for AMD this is a test where NVIDIA’s focus on compute performance has really paid off; coupled with the lack of CF scaling and even a $240 GTX 560 Ti can edge out the $700 6990.
Ultimately the take-away from this is that for most desktop GPU computing workloads, the benefit of multiple GPU cores is still unrealized. As a result the 6990 shines as a gaming card, but is out of its element as a GPU computing card unless you have an embarrassingly parallel task to feed it.
Facebook
Twitter
RSS
130 Comments
View All Comments
ET - Tuesday, March 8, 2011 - link
If you're going to keep your case open all day with your ear to the graphics card, then you might get that 70dB+, which won't be too nice on the ear. On the other hand you won't get much gaming done. :)I don't know the exact distance Anandtech measured this noise at, but Kitguru measured at about 1 metre and got 48dB when running Furmark, 40dB for normal load.
bobsmith1492 - Tuesday, March 8, 2011 - link
70dB is very low and only would apply if you are essentially living next to your computer - that's a 24-hour exposure level.NIOSH recommends 85dB as the upper limit for 8 hours of exposure, with a 3dB exchange rate - that is, every time you halve the amount of time you're exposed to the sound you can increase the volume by 3dB.
http://www.cdc.gov/niosh/docs/96-110/appF.html
looniam - Wednesday, March 9, 2011 - link
*ahem*"The document identifies a 24-hour exposure level of 70 decibels as the level of environmental noise which will *prevent* any measurable hearing loss over a lifetime"
does NOT say what the maximum is.
from my experience as an audio tech:
95dB is the start of temporary hearing loss, 110dB is the start of permanent and 140dB is the threshold of pain.
and for drunk people, they can't hear anything below 150dB :)
Ninjahedge - Thursday, March 10, 2011 - link
What?futuristicmonkey - Tuesday, March 8, 2011 - link
Any chances for some memory OC benches?nitrousoxide - Tuesday, March 8, 2011 - link
Both AMD and nVidia are out of mind, they are ignorant of the consequence by putting two gigantic chips with 5+billion transistors on the same board. I can't find the point of buying such outrageous card instead of building a CFX/SLI system. At least the latter isn't that loud, isn't that hot and consumes hardly more power than those monstrosities.TSMC is to blame. They dropped 32nm so it is impossible to get 6990/590 within 300W power envelope. But neither AMD nor nVidia turn back, but keep making these non-sense flagship cards.
Figaro56 - Tuesday, March 8, 2011 - link
No, it's not.Amuro - Tuesday, March 8, 2011 - link
I will be water cooling them! :)nitrousoxide - Tuesday, March 8, 2011 - link
What a shame it will be when next-generation 28nm Single GPU flagships wipe out these monsters with ease while consumes half the power, running silent.strikeback03 - Tuesday, March 8, 2011 - link
That hasn't traditionally happened. Look at the comparison between the 6990 and the 4870x2 - 2 generations, IIRC one process shrink. The 6990 does generally put up much better numbers, but consumes a lot more power to do so. Looking at a comparison between the 4870x2 and 5870 (double GPU on a larger process size to next-gen single-GPU) they are very close, with the 4870x2 overall holding a slight lead. And of course none of these high-end reference cards have ever been silent