NVIDIA GTX 1080 FPS Benchmarks – Dx12 & Vulkan (& OpenGL) vs. Dx11
This page initiates our game benchmarking for the GTX 1080. We do not include any Dx11-only games on this page, but we do provide comparative data between Dx11, Dx12, and Vulkan (where present) for tested games. Some of this data is presented as a percent change value or other unique values, like ms latency. These charts provide insight as to async compute performance and overall acceptance of new APIs by the hardware.
OpenGL is also included through DOOM, freshly-launched. Note that, since our tests were performed, AMD released a hotfix driver that reportedly resolves some R9 390 & R9 390X performance issues we experienced.
OpenGL DOOM – GTX 1080 vs. GTX 980 Ti, Fury X, & 980
We posted a DOOM benchmark almost immediately upon the game's launch. Absent from those charts were our already-conducted GTX 1080 performance tests – you'll find those below.
The GTX 1080 runs close to 60FPS at 4K resolution. Some lightweight graphics tuning would push the card into 60FPS range – mostly by dropping shadows or other post FX settings by one rung. The GTX 1080 Founders Edition (stock clock) outperforms the stock GTX 980 Ti by 13.8%. The Fury X is outperformed by 21.4% in this OpenGL title. Real-world differences, when looking at raw framerates and frametimes, are largely unnoticeable at this resolution – but the 1080 is off to a good start considering its value proposition versus these two similarly-priced cards.
At 1440p, the GTX 1080's performance lead plants it at 98.3FPS, aided by the best low frametimes on our bench. The GTX 1080 is the most tightly timed card for frame delivery that we've ever tested. Against the predecessor GTX 980 non-Ti, performance gains are 30%. Against the 980 Ti, the performance difference is 13% again.
1080p produces a CPU (or other hardware) bottleneck, with the GTX 1080 and GTX 980 Ti being capped around 128-129FPS average. To be fair, you're probably not playing at 1080p with one of these cards, anyway; at least, not unless you're trying to hit 144Hz – in which case that looks like a reasonable use case.
Comparative Dx12 vs. Dx11 FPS in Ashes of Singularity – GTX 1080 vs. 980 Ti, Fury X
Ashes of the Singularity offers some of the most reliable and accurate testing data out of all our Dx12 API games. Ashes easily reproduces comparable metrics and provides data down to the number of batches being pushed through the pipe.
We rip the Satellite Shot 2 data from Ashes, which shoves large batches down the pipe and chokes components. This is somewhat of a worst case scenario for the GPU. The above chart represents raw FPS output (averaged) for Dx11 vs. Dx12 on each card, the below chart shows the millisecond latency (frametimes) on each API, and the next one shows the percent change from Dx11 to Dx12 when it comes to frametimes.
At 1080p/high, the GTX 1080 crushes the other cards underfoot. NVidia's asynchronous compute advancements have clearly worked out (at least, in Ashes) and are producing gap-widening gains versus when compared against the previous architecture. AMD's Fury X and R9 390X are still the most impressive when it comes to gains, though. These two cards are choking on some sort of Dx11 optimization issue – something that nVidia's good at, when it comes to circumventing bottlenecks with drivers – and are limited by Dx11 in the Satellite Shot 2 heavy benchmark. With Dx12, the cards can unleash their full potential and nearly double framerates – but they're still behind the GTX 1080.
In contrast to this, the GTX 980 Maxwell card ranked high among Dx11 performers, but falls to the bottom of the chart for Dx12.
The GTX 1080 has made obvious improvements to Dx12 optimization and framerate.
Here's the frametime chart with 1080p on High. Lower is better. AMD's crux is its Dx11 frame latency, which creates the stuttering seen in Dx11. The GTX 1080 has an unprecedentedly low 13.38ms average frametime.
The Fury X sees a ~120% latency reduction, the 390X sees a 76.9% latency reduction – both a reward to AMD's investment in asynchronous compute – while the older GM204 Maxwell architecture struggles to stay positive. GTX 1080 and GP104, however, combine brute force COMPUTE with Async improvements to generate a 48.65% jaunt in Dx12. That's big news for nVidia, and shows that all this asynchronous talk isn't just sabre rattling.
Comparative Vulkan vs. Dx11 in Talos Principle – GTX 1080 vs. Fury X, 980 Ti, 980
The Talos Principle was our initial Vulkan benchmark a few months ago. The game's advanced reasonably within that time, but still exhibits poorer Vulkan performance than DirectX 11 performance. CroTeam, the developers, expect to see a switch-over point where Vulkan surpasses Dx11 performance in The Talos Principle.
Our tests show the GTX 1080 leading the charge for Vulkan performance. There's about an 11% difference favoring Dx11 across the GTX 1080's comparative benchmarks.
Comparative Gains in Tomb Raider – Dx11 vs. Dx12 Performance on GTX 1080
Tomb Raider is a little wonky as a title – it's got some funny optimization in spots, and AMD tends to do best with Dx11 as a result (whereas nVidia sees tiny, but consistent gains with Dx12).
AMD doesn't come off of Tomb Raider looking too hot. The delta between the GTX 1080 Founders Edition and GTX 980 Ti Reference, in order of 1080p, 1440p, and 4K, is 28.7%, 15.7%, and 26.6%.
Let's move on to DirectX 11 benchmarking for some tried-and-true metrics, including our 1% and 0.1% low frametime analysis. Hit the Next page button for that.