AMD clobbers Nvidia in updated Ashes of the Singularity

The Nitrous engine can use up to 16 threads at 1 thread per core. It's the best threaded game engine I've seen. AoTS makes full use of it, it's pretty impressive. There is a thread about CPU load balancing for AoTS in the games and hardware section on [H].
 
What I'm saying is that with one GPU in DX11 I got around 27 FPS average, and with two GPUs in DX12 I got around 33 FPS. Not nothing but not really that significant either.

Didn't check CPU usage but if I'm bottlenecked with 8 threads at 4.7 GHz then that seems like an optimization issue in itself.
What did you get with 2 cards in DX11? Hard to compare the 1 card in DX11 to 2 cards in DX12. That does seem low.
 
The Nitrous engine can use up to 16 threads at 1 thread per core. It's the best threaded game engine I've seen. AoTS makes full use of it, it's pretty impressive. There is a thread about CPU load balancing for AoTS in the games and hardware section on [H].


I admit I am curious to see if they end up licensing out their Nitrous engine, and the capabilities seem very impressive.
 
I admit I am curious to see if they end up licensing out their Nitrous engine, and the capabilities seem very impressive.

AoTS could be their showpiece game for their engine, but I wonder if they will end up licencing it. It's looking like a pretty specialized engine. The optimizing possibilities look to be huge. I can only imagine how much the GPU Async scheduler helps optimize the process with 16 main independent threads working concurrently! Will be even more interesting once they add Vulkan support.
 
What did you get with 2 cards in DX11? Hard to compare the 1 card in DX11 to 2 cards in DX12. That does seem low.

Well, I mentioned in another thread but I guess I failed to here...the game apparently doesn't support SLI (yet?). In the menu there is a Mutli-GPU tick box that you can only select in DX12 mode, and in DX11 mode there is zero utilization on GPU2.
 
Well, I mentioned in another thread but I guess I failed to here...the game apparently doesn't support SLI (yet?). In the menu there is a Mutli-GPU tick box that you can only select in DX12 mode, and in DX11 mode there is zero utilization on GPU2.
So was the DX12 result with async on? There is good news then: SLI nor CF need driver profiles to work in DX12. Of course Oxide mentioned this part already so its good to see its true or seems to be.
 
Ok 2 more articles covering AotS for those that like more info like me:

Ashes Of The Singularity Beta: Async Compute, Multi-Adapter & Power - Introduction And Test System

upload_2016-2-29_22-21-41.png


Well this is one graph they have and they have many. DX12 can be touted as More CPU efficient as well.

DX12 Benched - Fury X CF vs. Titan X SLI in Ashes of the Singularity

upload_2016-2-29_22-24-54.png


Just basic graphs and not as much info as Toms but a quick and dirty look at CF and SLI. Next article in 2 days or so will have mixed adapter I think.
 
For an alpha game, AotS certainly is attracting a lot of attention.
 
Wait there is more and this is very interesting:

PC Gaming Shakeup: Ashes of the Singularity, DX12 and the Microsoft Store | PC Perspective

Earlier this week, the team behind Ashes of the Singularity released an updated version of its early access game, which updated its features and capabilities. With support for DirectX 11 and DirectX 12, and adding in multiple graphics card support, the game featured a benchmark mode that got quite a lot of attention. We saw stories based on that software posted by Anandtech, Guru3D and ExtremeTech, all of which had varying views on the advantages of one GPU or another.

That isn’t the focus of my editorial here today, though.

Shortly after the initial release, a discussion began around results from the Guru3D story that measured frame time consistency and smoothness with FCAT, a capture based testing methodology much like the Frame Rating process we have here at PC Perspective. In that post on ExtremeTech, Joel Hruska claims that the results and conclusion from Guru3D are wrong because the FCAT capture methods make assumptions on the output matching what the user experience feels like. Maybe everyone is wrong?

Just first few paragraphs but it is a good read and what I guess is going to upset a lot of people if true going forward.
 
Wait there is more and this is very interesting:

PC Gaming Shakeup: Ashes of the Singularity, DX12 and the Microsoft Store | PC Perspective



Just first few paragraphs but it is a good read and what I guess is going to upset a lot of people if true going forward.

Haven't finished reading but as soon as I got to this part:

What we are seeing is the first implications of a new pipeline for graphics and compositing. WDDM 2.0 (Windows Display Driver Model) is a very big shift from what existed previously with WDDM 1.3. The days of exclusive fullscreen gaming may be on the way out as Microsoft shifts developers and hardware vendors into a standard path through the OS compositor rather than bypassing it. Implications for this change are only beginning to be understood, but let’s see how it affects Ashes of the Singularity today.

Even though Ashes of the Singularity is not a Windows Store application, the behavior we are seeing is part of the push that Microsoft is making to sell games through that store with a unified platform. The debate of app store based games versus free standing and open gaming has been a debate in the community since MS first starting discussing it – we just happen to have a real-world implication of it in front of us today.

Back on topic to our specific testing scenario, the AMD platform is more closely emulating what Microsoft would like to see done as DX12 gaming progresses. They never want applications to enter into anything that resembles a “Vsync off” state, which front buffer flips that lead to horizontal tearing.

I was like "alright, I am NEVER upgrading to Win10 if I can help it". Maybe my opinion will change when I actually finish reading the piece (I know I know...), but yeah, fuck M$ and its Win10 forceware bullshit.

Edit: Finished reading and welp, if M$ gets its way, PC gaming is done for. Time for PCMR to become the Personal Console Master Race lol...
 
Last edited:
Here's my screenshot. I had to back my CPU to stock as it was having issues. See my specs below. DX11 R390X Quick Edit: I don't know how to make the screenshot larger. 43FPS Average.

2K
4XAA
High & Medium
 

Attachments

  • 2016-03-01_00001.jpg
    2016-03-01_00001.jpg
    192.5 KB · Views: 11
Edit: Finished reading and welp, if M$ gets its way, PC gaming is done for..
Indeed. Everyone read this quote from the article

This post over at the PC Master Race subreddit gives even more examples of things that are going to change for games that are released through Microsoft’s Store implementation. No modding, no custom mouse bindings, no controller support outside of Xbox controllers; clearly this is going to shake up our lives as PC gamers. I also don’t think that you’ll be able to live stream out your games through XSplit and OBS either.

No modding? No custom mouse binding?

WTF?
 
If all of this comes to pass with MS's bullshit store, then people and publishers will finally have a solid reason to ditch the MS ecosystem once and for all and go the Vulkan + Linux route.
 
I bought AotS when it was at a reduced price on Steam just to compare performance of my two rigs below. I have a single EVGA GTX980TI SC (core1102) Vram 6G in the 5960x rig and 2 Sapphire Tri-X OC 290s (core1000) Vram 4G in CF in the 4790k rig. Both rigs are custom WC'd - EK Supremacy EVO cpu blocks and EK TitanX gpu block for the 980TI and EK 290/290X gpu blocks for both of the 290s. I have the 5960x OC'd to 4.4Ghz stable and the 4790k to 4.7 GHZ stable.

The 5960x has a DellU3415W monitor- native resolution 3440x1440 and the 4790k has the BenQ BL3200PT monitor - 2560x1440 native resolution.

Until the Newest Beta 2 update of AotS, multi-gpu support was not available and obviously the 980TI was way head of the single 290. Once the newest beta was released I ran tests on both rigs setting resolution to 2560x1440 to make them somewhat equal. I know that the cpus are different and clocked different but since they are custom WC'd switching gpus, though possible, was too much work - drain system, fill and purge system, etc., etc.!

Here are the results on the machines. All AotS settings on High at 2560x1440
5960x - Async enabled
Average=38.3 fps
Normal=43.1
Medium=38.3
Heavy=34.4

5960x - Async disabled
Average=40.1
Normal=45.4
Medium=40.2
Heavy=35.8

4790k single 290
Async enabled
Average=28
Normal=32.8
Medium=28
Heavy=24.3

Async disabled
Average=26.8
Normal=30.8
Medium=27.1
Heavy=23.5

4790k 290s in CF
Async enabled
Average=48.3
Normal=55.4
Medium=48.1
Heavy=42.8

Async disabled
Average=48
Normal=55.7
Medium=47.6
Heavy=42.5


I hope this info helps the discussion. I'll be willing to run other settings if it will help you.
 
Haven't finished reading but as soon as I got to this part:



I was like "alright, I am NEVER upgrading to Win10 if I can help it". Maybe my opinion will change when I actually finish reading the piece (I know I know...), but yeah, fuck M$ and its Win10 forceware bullshit.

Edit: Finished reading and welp, if M$ gets its way, PC gaming is done for. Time for PCMR to become the Personal Console Master Race lol...
In that article they mention interviews today or tomorrow so hopefully we will have definitive answers. Problem I had with the article/editorial was it was mostly speculation because the data didn't give clear evidence to what was going on. He gave a reasonable hypothesis, but not a definitive one. So give a few more days and then we can see what the actual facts are, at least I hope we get them.
 
Back
Top