DirectX 12 Multi-GPU Performance

Shifting gears, let’s take a look at multi-GPU performance on the latest Ashes beta. The focus of our previous article, Ashes’ support for DX12 explicit multi-GPU makes it the first game to support the ability to pair up RTG and NVIDIA GPUs in an AFR setup. Like traditional same-vendor AFR configurations, Ashes’ AFR setup works best when both GPUs are similar in performance, so although this technology does allow for some unusual cross-vendor comparisons, it does not (yet) benefit from pairing up GPUs that widely differ in performance, such as a last-generation video card with a current-generation video card. None the less, running a Radeon and a GeForce card together is an interesting sight, if only for the sheer audacity of it.

Meanwhile as a result of the significant performance optimizations between the last beta build and this latest build, this has also had an equally significant knock-on effect on mutli-GPU performance as compared to the last time we looked at the game.

Ashes of the Singularity (Beta) - 3840x2160 - High Quality - MGPU

Even at 4K a pair of GPUs ends up being almost too much at Ashes’ High quality setting. All four multi-GPU configurations are over 60fps, with the fastest Fury X + 980 Ti configuration nudging past 70fps. Meanwhile the lead over our two fastest single-GPU configurations is not especially great, particularly compared to the Fury X, with the Fury X + 980 Ti configuration only coming in 15fps (27%) faster than a single GPU. The all-NVIDIA comparison does fare better in this regard, but only because of GTX 980 Ti’s lower initial performance.

Digging deeper, what we find is that even at 4K we’re actually CPU limited according to the benchmark data. Across all four multi-GPU configurations, our hex-core overclocked Core i7-4960X can only setup frames at roughly 70fps, versus 100fps+ for a single-GPU configuration.


Top: Fury X. Bottom: Fury X + 980 Ti

The increased CPU load from utilizing multi-GPU is to be expected, as the CPU now needs to spend time synchronizing the GPUs and waiting on them to transfer data between each other. However dropping to 70fps means that Ashes has become a surprisingly heavy CPU test as well, and that 4K at high quality alone isn’t enough to max out our dual GPU configurations.

Ashes of the Singularity (Beta) - 3840x2160 - Extreme Quality - MGPU

Cranking up the quality setting to Extreme finally gives our dual-GPU configurations enough of a workload to back off from the CPU performance cap. Once again the fastest configuration is the Fury X + 980 Ti, which lands just short of 60fps, followed by the Fury X + Fury configuration at 55.1fps. In our first look at Ashes multi-GPU scaling we found that having a Fury X card as the lead card resulted in better performance, and this has not changed for the newest beta. The Fury continues to be faster at reading data off of other cards. Still, the gap between the Fury X + 980 Ti configuration and the 980 Ti + Fury X configuration has closed some as compared to last time, and now stands at 11%.

Backing off from the CPU limit has also put the multi-GPU configurations well ahead of the single-GPU configurations. We’re now looking at upwards of a 65% performance boost versus a single GTX 980, and a smaller 31% performance boost versus a single Fury X. These are smaller gains for multi-GPU configurations than we first saw last year, but it’s also very much a consequence of Ashes’ improved performance across the board. Though we didn’t have time to test it, Ashes does have one higher quality setting – Crazy – which may drive a bit of a larger wedge between the multi-GPU configurations and the Fury X, though the overhead of synchronization will always present a roadblock.

DirectX 12 Single-GPU Performance DirectX 12 vs. DirectX 11
Comments Locked

153 Comments

View All Comments

  • CiccioB - Sunday, February 28, 2016 - link

    The so called ASync Compute implementation AMD has in HW IS NOT PART OF DX12 SPECIFICS.
    I hope that is clear written that way.

    DX12 describe the use of multiple threads flying at the same time. nvidia does support them, with some limitations in number and preemption capabilities with respect to what AMD HW can.
    This however does not mean that nvdia HW does not support Async compute or it is out of specs. AMD just made a better implementation of it.
    Think it as it was for tessellation: nvidia implementation is way better than AMD one, but the fact that AMD can't go over certain values does not mean they are not DX11 compliant.

    What you are looking here is a benchmark (more than a game) that stresses the multi-threaded capabilities of AMD HW. You can see that AMD is in a better position here. But the question is: how many other games are going to benefit from using such a technique and how many of them are going to implement such a heavy duty load?

    We just don't know now. We have to wait to see if this technique can really improve performance (and thus image quality) in many other situations or it is just a show off for AMD (that has clearly partnered to make this feature even more heavy on nvidia HW).
    When nvidia will star making developers using their HW accelerated Voxels we will start to see what feature is going to hit worse one another's HW and which is going to give better image quality improvements.

    For now I just think this is a over used feature that like many other engine characteristics in DX11 is going to give advantage to one side rather than the other.
  • anubis44 - Thursday, February 25, 2016 - link

    That's because it never will be. You can't enable missing hardware.
  • xTRICKYxx - Wednesday, February 24, 2016 - link

    I hate to be that guy, but I think it is time to dump the X79 platform for X99 or Z170.
  • Ryan Smith - Wednesday, February 24, 2016 - link

    Yep, Broadwell-E is on our list of things to do once it's out.
  • Will Robinson - Wednesday, February 24, 2016 - link

    NVidia got rekt.
    DX12 lays the smak on Chizow's green dreams.
  • Roboyt0 - Wednesday, February 24, 2016 - link

    Do you have 3840x2160 results for the R9 290X per chance?
  • Ryan Smith - Wednesday, February 24, 2016 - link

    No. We only ran 4K on Fury X and 980 Ti.
  • Stuka87 - Wednesday, February 24, 2016 - link

    Really hating the colors of the graphs here. All grey, legend has one blue item, but no blue on the graph....
  • Ryan Smith - Wednesday, February 24, 2016 - link

    It's something of a limitation of the CMS. The color bar is the average; the grey bars are in the same order as they are in the legend: normal, medium, and heavy batch counts.
  • Mr Perfect - Thursday, February 25, 2016 - link

    I was wondering what was up with that. Maybe someone could do a little MS-Paint bucket fill on the images before publishing? :)

Log in

Don't have an account? Sign up now