Async compute is being used in lots of console games and is being added to the major engines.
There will be widespread use of it as DK12 adoption continues from developers.
Actually AMD fanboys are the ones who at least have no memory...
Oxide is a AMD fanclub, remember how they created the Star Swarm as a Mantle PR and it was biased vs DX11 and intencionally did not supported Context Lists just because AMD did not supported it? Remember how everything was about how awesome AMD was with Mantle and lots of useless raw draw calls? ring a bell?
Now they are taking Oxide devs words as the only truth, and now everything is about how awesome AC is? im sorry but this is a huge deja vu.
No no, only AMD fanboys can take anything from Oxide as a parameter of anything, anyone who is not biased and has memory can tell you, they already did this crap with Star Swarm as a AMD PR...
If you want to call Nvidia fanboy to someone that is showing hard evidence in your face go ahead.
BTW i have no doub that AC is a nice feature, i doubt it will become relevant when is only working in like 20% of the market... AND NO, dont come out with the "AMD is in consoles" BS, the consoles has been around for more than 2 years and things just work worse and worse on AMD hardware.
We need to wait to see in anyone implement it beyond the ones that AMD pay them for.
No I dont. Because its a demo made for size, not for optimization. You can most likely find other demos from the same event with the exact opposite outcome. You are down to what card support what instruction you may need for that single demo. It could for example be the same as mining on an NVidia card vs AMD. All down to a couple of instructions to make all the difference there.
Does Ashes support multithreaded AI? Supreme Commander gets sluggish on a 40KM+ map with maxed unit count. The game loads all the AI on one core. Near the end (or middle) of the battle, the game becomes choked by that core.
If Ashes can be GPU limited all the time, I will get it for that reason alone. Even Stardock's Sins of a Solar Empire was bottlenecked the SAME way. Single threaded performance is king in those older RTS games. Even now, a 5 GHz Skylake core is nowhere near enough for those games.
Does Ashes support multithreaded AI? Supreme Commander gets sluggish on a 40KM+ map with maxed unit count. The game loads all the AI on one core. Near the end (or middle) of the battle, the game becomes choked by that core.
If Ashes can be GPU limited all the time, I will get it for that reason alone. Even Stardock's Sins of a Solar Empire was bottlenecked the SAME way. Single threaded performance is king in those older RTS games. Even now, a 5 GHz Skylake core is nowhere near enough for those games.
...in which light sources are imitated by sprites. You do realize that?
Absolutely agree on this one.I really hope that Stardock releases Sins of a Solar Empire 2 using the Ashes engine. I love Sins but even on my 4.3ghz 3770k the late game slows down big time because one thread is getting hammered to death.
Supreme Commander has large maps, too...
"Sprites"? We are more than a decade after sprites...
The graphic in Ashes is just outdated. Here is a screen from Supreme Commander 2:
Released 6 years ago, using DX9 and has better texture quality and the shadows are leagues better.
And DX9 hasnt supported something like DirectCompute at this time...
Actually AMD fanboys are the ones who at least have no memory...
Oxide is a AMD fanclub, remember how they created the Star Swarm as a Mantle PR and it was biased vs DX11 and intencionally did not supported Context Lists just because AMD did not supported it? Remember how everything was about how awesome AMD was with Mantle and lots of useless raw draw calls? ring a bell?
Now they are taking Oxide devs words as the only truth, and now everything is about how awesome AC is? im sorry but this is a huge deja vu.
No no, only AMD fanboys can take anything from Oxide as a parameter of anything, anyone who is not biased and has memory can tell you, they already did this crap with Star Swarm as a AMD PR...
If you want to call Nvidia fanboy to someone that is showing hard evidence in your face go ahead.
BTW i have no doub that AC is a nice feature, i doubt it will become relevant when is only working in like 20% of the market... AND NO, dont come out with the "AMD is in consoles" BS, the consoles has been around for more than 2 years and things just work worse and worse on AMD hardware.
We need to wait to see in anyone implement it beyond the ones that AMD pay them for.
Are there any other posters on this thread who actually own AOTS and have benched it with their rig/rigs?
Are there any other posters on this thread who actually own AOTS and have benched it with their rig/rigs?
Interesting context: Compute/graphics benchmark
http://www.pcgameshardware.de/Grafi...PC-4K-Demo-Radeon-Geforce-Future-GPU-1187323/
That is a bit... scary, to be honest.
At 1.4 GHz GTX 970 scores less than R9 380X on 1020 MHz.
Yeah, but don't wanna spam bench pictures to this forum.
You can find my results here (those are all 8xmsaa results as ... rules are rules ) Ultra + 8xMSAA
https://forums.overclockers.co.uk/showpost.php?p=29220869&postcount=19
Without knowing more about it, it's just like any synthetic. I mean if they push Compute fully like what I am getting from reading the description, ie. it's actually compute-based rendering, then yeah, GPUs with parallel compute engines will excel.
Games don't use that ratio % of compute vs graphics though. It still is mostly graphics and compute are added on for effects.
Thank you Dygaza!
Mr. Teal would love to see your results especially with that 5930k!:thumbsup:
Total Time: 60.009888 ms per frame
Avg Framerate: 32.326557 FPS (30.934319 ms)
Weighted Framerate: 31.718487 FPS (31.527355 ms)
CPU frame rate (estimated if not GPU bound): 98.768288 FPS (10.124707 ms)
Percent GPU Bound: 35.976933 %
Driver throughput (Batches per ms): 6846.007813 Batches
Average Batches per frame: 18543.445313 BatchesTotal Time: 60.006493 ms per frame
Avg Framerate: 40.025700 FPS (24.983948 ms)
Weighted Framerate: 39.242874 FPS (25.482332 ms)
CPU frame rate (estimated if not GPU bound): 107.213821 FPS (9.327155 ms)
Percent GPU Bound: 44.509495 %
Driver throughput (Batches per ms): 6322.104980 Batches
Average Batches per frame: 14600.041992 BatchesSure. For some reason I still can't get multi-CPU working properly, but this is the single-GPU scores.
Extreme 4k
High 4kCode:Total Time: 60.009888 ms per frame Avg Framerate: 32.326557 FPS (30.934319 ms) Weighted Framerate: 31.718487 FPS (31.527355 ms) CPU frame rate (estimated if not GPU bound): 98.768288 FPS (10.124707 ms) Percent GPU Bound: 35.976933 % Driver throughput (Batches per ms): 6846.007813 Batches Average Batches per frame: 18543.445313 Batches
[/QUOTECode:Total Time: 60.006493 ms per frame Avg Framerate: 40.025700 FPS (24.983948 ms) Weighted Framerate: 39.242874 FPS (25.482332 ms) CPU frame rate (estimated if not GPU bound): 107.213821 FPS (9.327155 ms) Percent GPU Bound: 44.509495 % Driver throughput (Batches per ms): 6322.104980 Batches Average Batches per frame: 14600.041992 Batches
Did you download and install Radeon Hotfix 16.2?
Async compute is being used in lots of console games and is being added to the major engines.
There will be widespread use of it as DK12 adoption continues from developers.
Actually AMD fanboys are the ones who at least have no memory...
Oxide is a AMD fanclub, remember how they created the Star Swarm as a Mantle PR and it was biased vs DX11 and intencionally did not supported Context Lists just because AMD did not supported it? Remember how everything was about how awesome AMD was with Mantle and lots of useless raw draw calls? ring a bell?
Now they are taking Oxide devs words as the only truth, and now everything is about how awesome AC is? im sorry but this is a huge deja vu.
No no, only AMD fanboys can take anything from Oxide as a parameter of anything, anyone who is not biased and has memory can tell you, they already did this crap with Star Swarm as a AMD PR...
If you want to call Nvidia fanboy to someone that is showing hard evidence in your face go ahead.
BTW i have no doub that AC is a nice feature, i doubt it will become relevant when is only working in like 20% of the market... AND NO, dont come out with the "AMD is in consoles" BS, the consoles has been around for more than 2 years and things just work worse and worse on AMD hardware.
We need to wait to see in anyone implement it beyond the ones that AMD pay them for.
Q. What about deferred contexts? Dont those help multithreaded applications in DirectX?
A. Until the very latest drivers, we saw equal or even worse performance using deferred contexts in Nitrous. The newest drivers, however, have seen performance gains and so the latest build of Star Swarm can be set to use deferred contexts. Users can enable deferred contexts by modifying the lineEnableDeferredContexts=FALSE*to*=TRUE*in their settings files (for more information on customizing settings, see the link below). We have seen some stability issues with deferred contexts enabled, so change the setting at your own risk.
Q. Is Star Swarm unfairly targeting DirectX?
A. No. In fact our performance under DirectX is quite good. With a high-end CPU with excellent single-core performance and a high-end GPU, Nitrous can push up to 30,000 batches at 30 frames per second. We believe this number to be quite competitive with any other engine out there. We have also made additional optimizations specific to the DirectX version which are not required on other platforms.
Q. Are games using Nitrous going to have the same DirectX performance as Star Swarm?
A. No. We are working closely with all partners involved with DirectX to improve its performance going forward. Star Swarm is intended as a stress test, to ensure that games built on Nitrous will have an opportunity to efficiently use all the hardware available and not be artificially limited by the platform it is running on. In addition, all games will tailor their graphics settings to allow for optimal performance on each platform. Since Star Swarm is a stress test, it makes no distinction between DirectX and other platforms.
Meanwhile DirectX 11 deferred context support is currently broken in the combination of Star Swarm and NVIDIA's drivers, causing Star Swarm to immediately crash, so these results are with D3D 11 deferred contexts disabled.
Interesting context: Compute/graphics benchmark
http://www.pcgameshardware.de/Grafi...PC-4K-Demo-Radeon-Geforce-Future-GPU-1187323/
That is a bit... scary, to be honest.
At 1.4 GHz GTX 970 scores less than R9 380X on 1020 MHz.
That's not a benchmark. Not at all. It's a challenge, of how much graphic effects you can press into 4kB program code. Not 4k resolution, but 4kB binary size.Interesting. Google translate makes it difficult for me to decipher exactly what it all means, but AMD certainly likes this benchmark.
We could go back to the Mantle thread 2 years ago and i could read the same damn thing, yet, Dice the ones who "asked AMD to do Mantle", abandoned Mantle after BF4...
Its the same reason of why DX11 contexts lists(Nvidia Only) and Mantle (AMD Only) where never widely used, companies does not like to invest their own money to support some feature that only works on a portion of the market, because they target the whole market.
What they do on consoles does not matter,
What they do on consoles does not mean its good for us.
Could you provide some context? You aren't making any sense.i still remember how everyone said that consoles where gona help AMD processors, it never happen.
There are a lot of ports from console to PC. The nextgen PC APIs are a lot closer to the ones used on consoles. Console devs are also making heavy use of AC. They have to to squeeze every bit out of the hardware. It stands to reason that it does matter.
DirectX 12 allows us much better control over the CPU load with heavily reduced driver overhead. Some of the overhead has been moved to the game where we can have control over it. Our main effort is in parallelising the rendering system to take advantage of multiple CPU cores. Command list creation and D3D resource creation are the big focus here. We're also pulling in optimisations from UE4 where possible, such as pipeline state object caching. On the GPU side, we've converted SSAO to make use of async compute and are exploring the same for other features, like MSAA.
They say similar things for Hitman, Quantum Break, Deus Ex, Mirror's Edge...
And even Rise of the Tomb Raider, despite NV sponsorship, the game's original code is DX12 (with Async Compute usage!), they had to port it to DX11 for the PC and removed DX12 in the release, despite having DX12 files in the Steam folder (because reasons...)!
In the DX11 era, many modern AAA games are very well threaded and performance is good on AMD's weak FX CPUs, unlike older titles where it's single thread. It's not a coincidence, when consoles have lots of weaker cores.