Ashes of the Singularity Coming, with DX12 Benchmark in thread.

Kaapstad · 26 Feb 2016 at 09:07

2160p Maxed
2 x TX @1420/2002
5960X @4.0

==========================================================================
Oxide Games
Ashes Benchmark Test - ©2015
C:\Users\Keve\Documents\My Games\Ashes of the Singularity\Benchmarks\Output_16_02_26_0852.txt
Version 0.90.17445
02/26/2016 08:56
==========================================================================

== Hardware Configuration ================================================
GPU 0: NVIDIA GeForce GTX TITAN X
GPU 1: NVIDIA GeForce GTX TITAN X
CPU: GenuineIntel
Intel(R) Core(TM) i7-5960X CPU @ 3.00GHz
Physical Cores: 8
Logical Cores: 16
Physical Memory: 16253 MB
Allocatable Memory: 134217727 MB
==========================================================================

== Configuration =========================================================
API: DirectX 12
==========================================================================
Quality Preset: Custom
==========================================================================

Resolution: 3840x2160
Fullscreen: True
Bloom Quality: High
PointLight Quality: High
Glare Quality: High
Shading Samples: 16 million
Terrain Shading Samples: 16 million
Shadow Quality: High
Temporal AA Duration: 0
Temporal AA Time Slice: 0
Multisample Anti-Aliasing: 8x
Texture Rank : 1

== Total Avg Results =================================================
Total Time: 60.003143 ms per frame
Avg Framerate: 51.023167 FPS (19.598940 ms)
Weighted Framerate: 44.610073 FPS (22.416462 ms)
CPU frame rate (estimated if not GPU bound): 58.454727 FPS (17.107256 ms)
Percent GPU Bound: 0.000000 %
Driver throughput (Batches per ms): 4019.209229 Batches
Average Batches per frame: 18737.292969 Batches
==========================================================================

Kaapstad · 26 Feb 2016 at 09:23

1080p Maxed
2 X TX @1439/2002
5960X @4.0

Harlequin · 26 Feb 2016 at 09:27

struggling to change screen res - how can you do it? no matter what I change in the ini file it defaults to 4k

Kaapstad · 26 Feb 2016 at 09:29

Harlequin said:
struggling to change screen res - how can you do it? no matter what I change in the ini file it defaults to 4k

Change the desktop resolution to 1080p before you start.

In the game settings select fullscreen and bench.

Kaapstad · 26 Feb 2016 at 09:30

Am I the only one using 8xMSAA ?

fs123 · 26 Feb 2016 at 09:45

We need a new AMD vs Nvidia video comparison. This older comparison video shows significant lighting differences between a 390X running the latest AoTS benchmark vs a 980 running (allegedly) an older version. Will be interesting to know if that difference still exists in the latest version.

Besty · 26 Feb 2016 at 09:46

In-coming news story on theonion.com 'AMD Driver team sends chocolates, flowers and letters of affection to Microsoft's DX12 team'.

Harlequin · 26 Feb 2016 at 09:59

kaap - I think the presets of high>extreme >crazy are the ones used

Mauller · 26 Feb 2016 at 10:00

fs123 said:
We need a new AMD vs Nvidia video comparison. This older comparison video shows significant lighting differences between a 390X running the latest AoTS benchmark vs a 980 running (allegedly) an older version. Will be interesting to know if that difference still exists in the latest version.

Greg has made a video with his Titan X, i am sure it is missing some lighting effects compared to any of the AMD videos i have seen. But it would be good for someone with an AMD card and the bench to record and upload a run.

Kaapstad · 26 Feb 2016 at 10:01

Harlequin said:
kaap - I think the presets of high>extreme >crazy are the ones used

I always use max settings when I bench something as it reduces any CPU bottlenecks.

Kaapstad · 26 Feb 2016 at 10:03

Mauller said:
Greg has made a video with his Titan X, i am sure it is missing some lighting effects compared to any of the AMD videos i have seen. But it would be good for someone with an AMD card and the bench to record and upload a run.

I ran my Fury Xs earlier and the effects looked the same as on NVidia. Having said that I was looking at other things as well so I could be wrong.

Harlequin · 26 Feb 2016 at 10:06

tjhats my `cheap` rig at 1080p on extreme

fs123 · 26 Feb 2016 at 10:07

Kaapstad said:
I ran my Fury Xs earlier and the effects looked the same as on NVidia. Having said that I was looking at other things as well so I could be wrong.

Even a side by side screenshot should show if there is a difference. The video above shows the flying craft exhaust flames are much brighter than the 980 versions.

Gregster · 26 Feb 2016 at 10:12

Mauller said:
Greg has made a video with his Titan X, i am sure it is missing some lighting effects compared to any of the AMD videos i have seen. But it would be good for someone with an AMD card and the bench to record and upload a run.

Here is a bench with a Nano on the press release (which looks the same)

And here is mine

Looks the same to me, albeit mine isn't a great recording but you might be seeing something I am not.

fs123 · 26 Feb 2016 at 10:27

Not sure what is going on but the video I posted clearly shows a difference and is most noticeable on the exhaust flames in the following screenshot:

The new benchmark doesnt have the same flames coming out of ships so maybe they turned down the bloom effect of the lighting for performance reasons. Is there an option for such a feature? The old comparison is using a older version so anythings possible I guess. I remember the old Oxide Starswarm demo had similar bloom effects though. I prefer the old style lighting even if it was over the top.

Dygaza · 26 Feb 2016 at 10:31

Kaapstad said:
Am I the only one using 8xMSAA ?

Yes you are.

Dygaza · 26 Feb 2016 at 10:32

fs123 said:
Not sure what is going on but the video I posted clearly shows a difference and is most noticeable on the exhaust flames in the following screenshot:

The new benchmark doesnt have the same flames coming out of ships so maybe they turned down the bloom effect of the lighting for performance reasons. Is there an option for such a feature? The old comparison is using a older version so anythings possible I guess. I remember the old Oxide Starswarm demo had similar bloom effects though.

Again this video. It has never been comparable because 390x and 980 ain't running same game version.

Gregster · 26 Feb 2016 at 10:35

Just been reading this article from Guru3D when looking for comparison vids to check IQ and some interesting facts have cropped up and I don't think the AMD boys will like it

http://www.guru3d.com/articles_pages/ashes_of_singularity_directx_12_benchmark_ii_review,11.html

What Do These Measurements Show?

Basically, what these measurements show are anomalies like small glitches and stutters that you can sometimes (and please do read that well, sometimes) see on screen. Below I'd like to run through a couple of titles with you. Bear in mind that Average FPS matters more than frametime measurements.

Right, we are now focusing on the GeForce cards sole on this page. The second we add a Radeon card, VSYNC kicks in making the recording VSYNC enabled and thus useless.

As the FPS benchmarks have shown, combining a GeForce GTX 980 and 980 Ti actually works and shows an increase in performance. But learning experiences in the past have teaches us that RAW and brute Multi GPU implementations often can result in a phenomenon called micro-stutter. Basically multi GPU solutions apply AFR (alternate frame rendering), two GPUs alternate in processing frames. These days with Crossfire and SLI both AMD and Nvidia keeps things smooth by syncing and even delaying frames. By syncing both vendors are delaying one frame to compensate the offset that you normally get when rendering as fast as possible.

Let me first grab the result from the previous page, a single card:

A single card with just one GPU doesn't have to deal with stuff like frame pacing. Aside from some small anomalies we see a reasonably okay rendering here with no on-going micro-stuttering. Now let's load up the GTX 980 and GTX 980 Ti in the Explicit Multi-GPU mode:

And boom ... there is is again, frames bouncing up and down with a latency up-to even 7ms per frame. Above what you see is a classic pattern for micro-stuttering. A problem that took us years to get rid of is now back.

Update: hours before the release of this article we got word back from AMD. They have confirmed what we are seeing. Radeon Software 16.1 / 16.2 does not support a DirectFlip in DX12, which is mandatory to solve to this specific situation/measurement. AMD intends to resolve this issue in a future driver update. Once that happens we'll revisit FCAT.

Correct see the frames bouncing into sync with that middle straight line at roughly 16~17ms ? Do the math with me: 1000/16,5 = ~60 FPS. This pattern is 100% VSYNC. All Radeon cards (and yes seriously we tested ALL cards) render with VSYNC on. Now we contacted the Ashes devs about this one. Unfortunately it has become a rather sizable discussion with the developer who claims this to be a DX12 issue. Here's the thing, if your GPU is fast enough the numbers the benchmark produces exceed 60 FPS / 60 Hz. The game measures at the rendering port (and not the display end), then is post processed, passed to DX12 and the output rendered. So if at that stage your score can be like 92 FPS. However after that segment in the pipeline a lot of other stuff happens. In our case a 60Hz / FPS framerate . Where we measure with FCAT is definitive though, it's what your eyes will see and observe. As stated we cannot turn this VSYNC like behaviour off, the feedback from Stardock is that Vsync on/off works differently in D3D12 then D3D11 due to changes in Windows, so they will not be apples to apples. In D3D12 you can only flip the backbuffer, not blit into like D3D11. Thus, you may not see the same type of screen tearing or present behavior that you will in D3D11.

Stutter city for AMD

Harlequin · 26 Feb 2016 at 10:38

FCAT is near useless under DX12 as its assuming DIRECTFLIP , which isn't in use by AMD and some of the results are based on that assumption by the software .ergo the results for AMD can`t be used

muziqaz · 26 Feb 2016 at 10:39

Gregster said:
Just been reading this article from Guru3D when looking for comparison vids to check IQ and some interesting facts have cropped up and I don't think the AMD boys will like it

http://www.guru3d.com/articles_pages/ashes_of_singularity_directx_12_benchmark_ii_review,11.html

Stutter city for AMD

Greg, next time try reading this thread more carefully

guru3D has no clue that FCAT does not support dx12 correctly.
There is an article on extremetech explaining this. There is no stutter on single or dual GPUs from AMD. period