Measuring the general-purpose compute performance of multi-GPU solutions is a challenge because not every app knows how to exploit more than one graphics processor at a time. We also have to strike CUDA- or Stream/APP-only software from our list. That doesn’t leave many options, which is why we’re limiting our search to OpenCL-accelerated applications.
The most obvious benefit to OpenCL is that both vendors’ cards compete on a playing field that is as level as we can make it. Besides, a comparison using real-world metrics covering floating-point (FP32) and double-precision (FP64) math is much more interesting than a huge field of synthetic benchmarks. As usual, we also include a number of current workstation-class cards to see how they fare relative to their consumer siblings.
Rendering
We chose two different renderers that take almost opposing approaches to optimization. On one hand, we have the well-known LuxMark benchmark based on the LuxRender engine. On the other, we use the integrated benchmark of RatGPU, an application that tends to favor Nvidia cards but isn’t really optimized for either architecture. LuxMark reports its result in samples per second, while RatGPU measures the time per run.

There’s really not much to say about LuxMark that the chart doesn’t already tell us. AMD’s GCN architecture dominates, and an OpenCL-optimized application able to exploit two Tahiti GPUs simply screams.

Meanwhile, RatGPU shows us what many CUDA-enabled renderers have proven in the past, namely none of the Kepler-based GeForce cards can keep up with the Fermi-based GeForce GTX 580 in compute-heavy software. It’s a little strange that the VLIW4-based Radeon HD 6970 is faster than Radeon HD 7970 GHz Edition, though.
Encryption
The software we’re using for this test treats the multi-chip cards as if they have one GPU, so performance scales very well. AMD’s Radeon HD 7990, which seems to excel in integer-based hashing operations, performs really well, followed by a number of other GCN-based boards.


Financial Analysis Performance (Float/FP32)
We see the same sort of near-ideal scaling from the Radeon HD 7990 in our four financial analysis benchmarks (two benchmarks with two levels of precision each). Indeed, AMD’s flagship almost delivers two times the performance of the single-GPU Radeon HD 7970 GHz Edition, despite slightly lower clock rates. Meanwhile, the GeForce GTX Titan and 690 can’t even compete.


Financial Analysis Performance (Double/FP64)
Repeating those two benchmarks using double-precision math makes the differences even more apparent. While Nvidia’s other cards struggle with FP64, the Titan actually does quite decently, especially compared to the GK104-based GeForce GTX 690 and GTX 680. The trick is to activate CUDA’s dual-precision mode in the card’s driver, which also extends functionality to OpenCL. Although this negatively affects clock rates, the card is faster overall in FP64-based workloads.
Meanwhile, the Radeon HD 7990 doesn’t need any tweaking to achieve its impressive and chart-topping performance.


- AMD's Malta Becomes The Radeon HD 7990
- Much-Improved Acoustics, With One Nagging Issue
- Test Setup, An Explanation Of FCAT, And Benchmarks
- Results: 3DMark
- Results: Battlefield 3
- Results: BioShock Infinite
- Results: Borderlands 2
- Results: Crysis 3
- Results: Far Cry 3
- Results: Hitman: Absolution
- Results: The Elder Scrolls V: Skyrim
- Results: Tomb Raider
- Radeon HD 7990 Vs. GeForce GTX 690: The Pepsi Challenge
- Noise Measurements And Fan Speed
- Noise Analysis: Frequency Spectrum And Videos
- OpenCL: General-Purpose Computing
- OpenGL: Synthetic Gaming Performance
- Can The World’s Best Bundle Save Radeon HD 7990?
If you're looking to game at 1920x1080, I can save you a ton of money by recommending something less than half as expensive. This card is for folks playing at 2560 *at least.* Next time, I'm looking to get FCAT running on a 7680x1440 array
Thats some nice gains from the prototype driver.
Power is the one thing I didn't have time for. We already know the 7990 is a 375 W card, while GTX 690 is a 300 W card, though. We also know AMD has Zero Core, which is going to shave off power at idle with one GPU shut off. I'm not expecting any surprises on power that those specs and technologies don't already insinuate.
If you're looking to game at 1920x1080, I can save you a ton of money by recommending something less than half as expensive. This card is for folks playing at 2560 *at least.* Next time, I'm looking to get FCAT running on a 7680x1440 array
People mostly buy Intel (I3/i5 a lot more than i7) just because Intel can provide top of the line CPUs in the i7 Extreme range. Same goes here, if some hears that AMD has a better 1000$ card than Nvidia, they will probably spend 100-200$ for an AMD card and not Nvidia.
Power ... unless you`re not a guy who saves 2 years in a row for this card to have a 6 months nerd gaming glory you won`t care that much how power hungry this card is.
Is just like asking Ferrari or Lamborghini how many mpg their cars do.
2. Regarding the fan noise and the hum : It would be interesting to know how much noticable is the fan noise and the hum with increaseing listner distance. IOW, which noise is more noticable at near/medium/far distances ?
Drivers still are AMD's biggest weakness. I would have expected AMD to havetop-notch , A-one drivers to go with the HD7990. After all, this is AMD's halo product. The first impression is what matters. The conclusion is basically "Card is good. Drivers are poor, with better coming in future". So ultimately its selling a promise, which may/may not succeed. It appears to me that AMD doesnt value its own products.
Also, has the Video Conversion Engine in AMD taken off ?
Unfortunately I'm really not sure the whining issue is something that can be fixed with a driver update. I think it has more to do with the hardware on the board than anything else. But it's good to see that AMD has finally recognized the frame time variance and micro-stutter problem, and are actively pursuing a solution. Although the test in the review was limited, I think it's telling that every gamer tested was able to recognize the difference between AMD and Nvidia cards, and even the difference brought by AMD's own prototype drivers.
I know and that's what I meant by hopping that they would have addressed the whining with this card. It happens to all their cards, well the ones that I have owned especially the XFX and if they knew what causes then they should have fixed it.
Let's hope that the prototype driver will also translate to better drivers for all their GPUs and address the frame rate issues. Other than that, it is a good card but I think, for my personal use since I was waiting to see what this can offer, I will just get the GTX 680 or the GTX 780 next month and will definitely go back to AMD if they address those issues.
For everyone seeking power and heat results:
http://www.anandtech.com/show/6915/amd-radeon-hd-7990-review-7990-gets-official/16
It consumes a lot of power under load, substantially more than the GTX690, but like Chris said that's to be expected. The big difference with the 7990 seems to be acoustics in relation to temps at load. It's a massive improvement over the 6990, and pretty much on par with the GTX690. Unfortunately the coil whine seems to undo a lot of the improvements made to the stock cooler, but all things considered it's pretty impressive what AMD was able to do in this area, especially in comparison to unofficial solutions from other vendors (dual slot, only requires 2 8-pin).
Sorry, that's a reading fail on my part. Thought you said, 'hope they'll address the whining' or something to that effect.
No problem.