The DirectX 12 Performance Preview: AMD, NVIDIA, & Star Swarm
by Ryan Smith on February 6, 2015 2:00 PM EST- Posted in
- GPUs
- AMD
- Microsoft
- NVIDIA
- DirectX 12
Frame Time Consistency & Recordings
Last, but not least, we wanted to also look at frame time consistency across Star Swarm, our two vendors, and the various APIs available to them. Next to CPU efficiency gains, one of the other touted benefits of low-level APIs like DirectX 12 is the ability for developers to better control frame time pacing due to the fact that the API and driver are doing fewer things under the hood and behind an application’s back. Inefficient memory management operations, resource allocation, and shader compiling in particular can result in unexpected and undesirable momentary drops in performance. However, while low-level APIs can improve on this aspect, it doesn’t necessarily mean high-level APIs are bad at it. So it is an important distinction between bad/good and good/better.
On a technical note, these frame times are measured within (and logged by) Star Swarm itself. So these are not “FCAT” results that are measuring the end of the pipeline, nor is that possible right now due to the lack of an overlay option for DirectX 12.
Starting with the GTX 980, we can immediately see why we can’t always write-off high-level APIs. Benchmark non-determinism aside, both DirectX 11 and DirectX 12 produce consistent frame times; one is just much, much faster than the other. Both on paper and subjectively in practice, Star Swarm has little trouble maintaining consistent frame times on the GTX 980. Even if DirectX 11 is slow, it is at least consistent.
The story is much the same for the R9 290X. DirectX 11 and DirectX 12 both produce consistent results, with neither API experiencing frame time swings. Meanwhile Mantle falls into the same category as DirectX 12, producing similarly consistent performance and frame times.
Ultimately it’s clear from these results that if DirectX 12 is going to lead to any major differences in frame time consistency, Star Swarm is not the best showcase for it. With DirectX 11 already producing consistent results, DirectX 12 has little to improve on.
Finally, along with our frame time consistency graphs, we have also recorded videos of shorter run-throughs on both the GeForce GTX 980 and Radeon R9 290X. With YouTube now supporting 60fps, these videos are frame-accurate representations of what we see when we run the Star Swarm benchmark, showing first-hand the overall frame time consistency among all configurations, and of course the massive difference in performance.
Facebook
Twitter
RSS
245 Comments
View All Comments
Jeffro421 - Thursday, February 12, 2015 - link
Something is horribly wrong with your results. I just ran this benchmark, on extreme, with a 270X 4GB and I got 39.61 FPS on DX11. You say a 290X only got 8.3 fps on DX11?http://i.imgur.com/JzX0UAa.png
Ryan Smith - Saturday, February 14, 2015 - link
You ran the Follow scenario. Our tests use the RTS scenario.Follow is a much lighter workload and far from reliable due to the camera swinging around.
0VERL0RD - Friday, February 13, 2015 - link
Been meaning to ask why both cards show vastly different total memory in Directx diag. Don't recall Article indicating how much memory each card had. Assuming they're equal. Is it normal for Nvidia to not report correct memory or is something else going on?Ryan Smith - Saturday, February 14, 2015 - link
The total memory reported is physical + virtual. As far as I can tell AMD is currently allocating 4GB of virtual memory, whereas NVIDIA is allocating 16GB of virtual memory.trisct - Friday, February 13, 2015 - link
MS needs a lot more Windows installs to make the Store take off, but first they need more quality apps and a competitive development stack. The same app on IOS or Android is almost always noticeably smoother with an improved UI (often extra widget behaviors that the Windows tablet versions cannot match). Part of this is maturity of the software, but Microsoft has yet to reach feature parity with the competing development environments, so its also harder for devs to create those smooth apps in the first place.NightAntilli - Friday, February 13, 2015 - link
We know Intel has great single core performance. So the lack of benefits for more than 4 cores is not unsurprising. The most interesting aspect would be to test the CPUs with weak single core performance, like the AMD FX series. Using the FX series rather than (only) the Intel CPUs would be more telling. 4 cores would not be enough to shift the bottleneck to the GPU with the FX CPUs. This would give a much better representation of scaling beyond 4 cores. Right now we don't know if the spreading of the tasks across multiple threads is limited to 4 cores, or if it scales equally well to 6 threads or 8 threads also.This is a great article, but I can't help feeling that we would've gotten more out of it if at least one AMD CPU was included. Either an FX-6xxx or FX-8xxx.
Ryan Smith - Saturday, February 14, 2015 - link
Ask and ye shall receive: http://www.anandtech.com/show/8968/star-swarm-dire...NightAntilli - Tuesday, February 17, 2015 - link
Thanks a lot :) The improvements are great.0ldman79 - Monday, February 16, 2015 - link
One benefit for MS to have (almost) everyone on a single OS is just how many man hours are spent patching the older OS? If they can set up the market to where they can drop support for Vista, 7 and 8 earlier than anticipated they will save themselves a tremendous amount of money.Blackpariah - Tuesday, February 17, 2015 - link
I'm just hoping the already outdated console hardware in PS4/Xbone won't hold things back too much for the pc folks. On a side note... I'm in a very specific scenario where my new gtx 970, with DX11, is getting 30-35 fps @ 1080P in battlefield 4 because the cpu is still an old Phenom 2 x4... while with my older R9 280, on Mantle, the framerate would stay above 50's at all times at almost identical graphic detail & same resolution.