Sniper Elite 4 – HBCC On vs. Off
Sniper Elite 4 is our first game benchmark. At 4K with high settings, Dx12, and Async compute, we found performance to be functionally equal between the HBCC A/B tests. With HBCC On, we plotted 52.7FPS AVG, 46.2FPS 1% lows, and 45.4FPS 0.1% lows. Disabling HBCC landed us at 52.2FPS AVG, 45.9FPS 1% lows, and 44.8FPS 0.1% lows. Not only is there no discernible difference, there is no difference outside of usual deviation and error margins.
This game has traditionally shown favor toward AMD and its Vega hardware, so we would have expected that uplift would be most likely here -- but it's also not very VRAM-hungry, eliminating much of the need for HBCC.
Ashes of the Singularity – HBCC On vs. Off
Moving on to Ashes of the Singularity with Dx12, 4K resolution, and High settings, we mostly saw the same: These results had us at 66.6FPS AVG to 66.9FPS AVG with HBCC enabled, 44.3 to 43.5FPS 1% lows, and 41 vs. 40.2FPS 0.1% lows. Once again, these are within test-to-test deviation.
Ghost Recon: Wildlands – HBCC On vs. Off
Ghost Recon: Wildlands at 4K plots the HBCC tests at 36FPS AVG with HBCC on and off, with lows largely tied at 33 and 32.5 or 32 and 31.8. There is no appreciable difference here, and we are once again within variance. We also saw no difference at 1080p, though that’s less significant, given the diminished focus on memory at this resolution.
For Honor – HBCC On vs. Off
For Honor at 4K and using Extreme settings makes for one of the more memory-intensive games that we work with. Vega with HBCC enabled plotted a 43FPS AVG versus 42.7FPS when disabled, marking these scores functionally equivalent and within variance. The lows are 40.7 versus 40.3 and 39.3 versus 38. Once again, all of this is within test deviation and variance.
At 1080p, For Honor has us at 125FPS AVG for both HBCC enabled and disabled. We would expect less likelihood for differences to emerge at these lower resolutions, given the VRAM usage reduction.
Hellblade – HBCC On vs. Off
We configured Hellblade to 4K with Very High settings, providing one of the more taxing gaming tests for the Vega 56 card. Under both HBCC conditions, performance was exceptionally consistent and plotted at about 27FPS AVG, with lows at 25 and 16 for each set of tests.
Shadow of War – HBCC On vs. Off
Testing Shadow of War with the normal texture pack, 4K resolution, and high settings had us at 42-43FPS AVG on both iterations of the HBCC testing, marking equal performance. We later enabled the 4K texture pack and ran Ultra settings, with HBCC reconfigured to a 12GB framebuffer to preserve RAM for the pack. Those settings gave us, once again, about the same performance between our HBCC on and off tests.
Consulting the work of our friends over at Techgage, it looks like Rob Williams reported similar results back in August, so that hasn’t changed much with the 17.10.1 driver push. We did find uplift in Superposition, but the trouble is that this uplift begins to emerge primarily when the card is beginning to struggle for other reasons – like frequency, shader count, or ROP limitations. It’s difficult to fill an 8GB framebuffer, so AMD’s claims of 50% uplift in AVG FPS may make more sense or better come to fruition were a 4GB card to exist. At this time, we do not have a means to create a 4GB framebuffer, so we can’t validate those claims in a direct comparison. We have some ideas, but they may not work. We’ll see.
In the interim, there’s really not much reason to disable HBCC, and it occasionally provides a 0.5-4.0% performance increase. Limited system memory would be a reason, of course. It might be worthwhile to enable HBCC and just leave it in the background, hoping occasional uplift will emerge. We can only speak for the applications which we’ve tested, naturally, and there’s potential for HBCC uplift in production applications (untested) or memory hog games (user mods for Skyrim and FO4, for instance, are often VRAM pigs). In the event you encounter one such game, HBCC enablement could help; of course, if stability or system RAM issues are encountered, there’s obviously not much loss to disabling HBCC, either.
As it stands now, we’re not presently able to observe appreciable performance uplift in any of our tested games, with one exception being the synthetic applications. We observed repeatable uplift in Superposition (~4%), with similarly repeatable (but insignificant) ~1-1.4% uplift in Firestrike. That’s probably good news for competitive benchmarkers, it’s just not meaningful for most users.
Let us know if you find an application where this technology is better leveraged. We may attempt some other applications in the future, but will probably wait for more driver updates to push.
Editorial: Steve Burke
Video: Andrew Coleman