GeForce + Radeon: Previewing DirectX 12 Multi-Adapter with Ashes of the Singularity
by Ryan Smith on October 26, 2015 10:00 AM ESTAshes GPU Performance: Single & Mixed 2012 GPUs
While Ashes’ mutli-GPU support sees solid performance gains with current-generation high-end GPUs, we wanted to see if those gains would extend to older DirectX 12 GPUs. To that end we’ve put the GeForce GTX 680 and the Radeon HD 7970 through a similar test, running the Ashes’ benchmark at 2560x1440 with Medium image quality and no MSAA.
First off, unlike our high-end GPUs there’s a distinct performance difference between our AMD and NVIDIA cards. The Radeon HD 7970 performs 22% better here, just averaging 30fps to the GTX 680’s 24.5fps. So right off the bat we’re entering an AFR setup with a moderately unbalanced set of cards.
Once we do turn on AFR, two very different things happen. The GTX 680 + HD 7970 setup is an outright performance regression, with performance 40% from the single GTX 680 Ti. On the other hand the HD 7970 + GTX 680 setup sees an unexpectedly good performance gain from AFR, picking up a further 55% to 46.4fps.
As this test is a smaller number of combinations it’s not clear where the bottlenecks are, but it’s none the less very interesting how we get such widely different results depending on which card is in the lead. In the GTX 680 + HD 7970 setup, either the GTX 680 is a bad leader or the HD 7970 is a bad follower, and this leads to this setup spinning its proverbial wheels. Otherwise letting the HD 7970 lead and GTX 680 follow sees a bigger performance gain than we would have expected for a moderately unbalanced setup with a pair of cards that were never known for their efficient PCIe data transfers. So long as you let the HD 7970 lead, at least in this case you could absolutely get away with a mixed GPU pairing of older GPUs.
Facebook
Twitter
RSS
180 Comments
View All Comments
SuperVeloce - Sunday, December 6, 2015 - link
X360 and PS3 were already much more "to the metal" than we were used to from DX9. In this respect those consoles have more in common with DX12 than any older DX versionsCreig - Tuesday, October 27, 2015 - link
How long before Nvidia sabotages this the way they stop PhysX from working if an AMD card is detected in your system?0ldman79 - Monday, November 2, 2015 - link
I think Nvidia has already disabled Physx from working if AMD is detected.I couldn't run Physx with my Radeon 5750 and any Geforce card as a Physx co-processor. There was a massive workaround that I got going once for about a half an hour, then I reboot and couldn't get it working again.
I spent around 4 hours of my life trying to get paper to flutter around more realistically in Batman only to have it fail on reboot.
Nvidia really has the wrong idea in this situation. It worked just fine in the past.
Samus - Monday, October 26, 2015 - link
You know...nothing rhymes with orange ;)rituraj - Tuesday, October 27, 2015 - link
flange?AndrewJacksonZA - Tuesday, October 27, 2015 - link
Please suck on a lozenge while rhyming with orange.uglyduckling81 - Sunday, November 1, 2015 - link
Nvidia are not going to like this. They will patch their drivers to make sure this no longer works.lilmoe - Monday, October 26, 2015 - link
What about Intel's iGPU?Ryan Smith - Monday, October 26, 2015 - link
Since this early release is limited to basic AFR, there's little sense in testing an iGPU. It may be able to contribute in the future with another rendering mode, but right now it's not nearly fast enough to be used effectively.DanNeely - Monday, October 26, 2015 - link
What about pairing the IGP with a low end discrete part where the performance gap is much smaller? I'm thinking about the surface book where there's only a 2:1 gap between the two GPUs; but any laptop with a 920-940M part would have a similar potential gain.On the desktop side, AMD's allowed heterogeneous xFire with their IGP and low end discrete cards for a few years. How well that setup works with DX12 would be another interesting test point.