[computerbase.de] DOOM + Vulkan Benchmarked. - Page 23 - Overclock.net - An Overclocking Community

Forum Jump: 

[computerbase.de] DOOM + Vulkan Benchmarked.

Reply
 
Thread Tools
post #221 of 632 (permalink) Old 07-15-2016, 07:02 PM
High Clocker
 
bmgjet's Avatar
 
Join Date: Nov 2011
Posts: 3,114
Rep: 187 (Unique: 161)
Decrease in average performance on my 980ti OC.
OpenGL averages 140fps drops to 90fps.
Vulkan averages 130fps drops to 110fps.
Crazy as coil wine on vulkan as well, opengl there was no coil wine at all and on other DX11 games very slightly noise.


SLI Voodoo 2 - > GeForce4 MX 420 -> GF 6600GT -> GF 6800GT -> GF 8800Ultra -> AMD 4870 -> AMD 4890 -> CF AMD 5820 -> CF AMD 6850 -> CF AMD 7970 -> SLI GF 680 -> SLI GF 780 -> CF AMD 290X -> GF 980ti -> SLI GF 980ti -> GF 1080ti -> SLI GF 1080ti -> <- RTX 2080ti (DOA/Refunded)
CPU
AMD 8120
CPU
FX 8350
GPU
7970 CFX
GPU
680 SLI
▲ hide details ▲


bmgjet is offline  
Sponsored Links
Advertisement
 
post #222 of 632 (permalink) Old 07-15-2016, 08:30 PM
New to Overclock.net
 
Mahigan's Avatar
 
Join Date: Aug 2015
Location: Ottawa, Canada
Posts: 1,749
Rep: 874 (Unique: 233)
Quote:
Originally Posted by Glottis View Post

but but nvidia doesn't have async compute gains, they said, DOOM not yet supporting async compute for nvidia doesn't matter, they said.


I am not disputing that Pascal can get a boost under light loads... never have. As for 3D Mark Time Fly... See concurrent vs parallel execution. All of the current games supporting Asynchronous Compute make use of parallel execution of compute and graphics tasks. 3D Mark Time Fly support concurrent. It is not the same Asynchronous Compute.

Concurrency fills in gaps which are in the execution pipeline. Parallelism executes two tasks at the same time.


Notice the context switch involved?

If 3D Mark Time fly were using Parallel executions then there would be synchronization points between the two contexts (Graphics and Compute). There would also be pipeline stalls on Maxwell GPUs. Both the pipeline stalls and the flush required for a synchronization point would add latency thus leading to Maxwell losing performance when running this variant of Asynchronous compute. We do not see Maxwell losing performance under 3D Mark Time Spy. We see a tiny performance boost. Thus 3D Mark Time Spy is not running Asynchronous Compute + graphics. You see parallel executions = Asynchronous Compute + Graphics. Concurrent execution = Asynchronous Compute. They are not the same thing.
Quote:
With DirectX 12, GPUs that support asynchronous compute can process work from multiple queues in parallel.
They can but that is not what 3D Mark is doing.
Quote:
In Time Spy, asynchronous compute is used heavily to overlap rendering passes to maximize GPU utilization. The asynchronous compute workload per frame varies between 10-20%. To observe the benefit on your own hardware, you can optionally choose to disable async compute using the Custom run settings in 3DMark Advanced and Professional Editions.

That is from 3DMark and can be found in the PC Per review. http://www.pcper.com/reviews/Graphics-Cards/3DMark-Time-Spy-Looking-DX12-Asynchronous-Compute-Performance

Yeah... even PCPer went a step further and attacked "AMD Fanboys" when in reality... PC Per do not even know the difference. Tech journalism.... rolleyes.gif

What is concurrency?
https://en.wikipedia.org/wiki/Concurrent_computing
Quote:
Concurrent computing is a form of computing in which several computations are executed during overlapping time periods —concurrently— instead of sequentially (one completing before the next starts).

So yeah... 3D Mark does not use the same type of Asynchronous compute found in all of the recent game titles. Instead.. 3D Mark appears to be specifically tailored so as to show nVIDIA GPUs in the best light possible. It makes use of Context Switches (good because Pascal has that improved pre-emption) as well as the Dynamic Load Balancing on Maxwell through the use of concurrent rather than parallel Asynchronous compute tasks. If parallelism was used then we would see Maxwell taking a performance hit under Time Fly as admitted by nVIDIA in their GTX 1080 white paper and as we have seen from AotS.



GCN can handle these tasks but performs even better when Parallelism is thrown in as seen in the Doom Vulkan results. How? By reducing the per Frame latency through the parallel executions of Graphics and Compute Tasks. A reduction in the per-frame latency means that each frame takes less time to execute and process. The net result is a higher frame rate. 3DMark lacks this. AotS makes use of both parallelism and concurrency... as does Doom with the new Vulkan patch. See below...


If 3D Mark Time Fly had implemented a separate path and enabled both concurrency and parallelism for the FuryX... it would have caught up to the GTX 1070. No joke.

If both AMD and nVIDIA are running the same code then Pascal would either gain a tiny bit or even lose performance. This is why Bethesda did not enable the Asynchronous Compute + Graphics from the AMD path for Pascal. Instead... Pascal will get its own optimized path. They will also call it Asynchronous Compute... people will think it is the same thing when in reality... two completely different things are happening behind the scene.

See why understanding what is actually happening behind the scenes is important rather than just looking at numbers? Not all Asynchronous Compute implementations are equal. You would do well to take note of this.

Where are the tech journalists these days?

"Once you eliminate the impossible, whatever remains, no matter how improbable, must be the truth." - Arthur Conan Doyle (Sherlock Holmes)
Mahigan is offline  
post #223 of 632 (permalink) Old 07-16-2016, 01:25 AM
 
Join Date: Mar 2011
Location: New York
Posts: 3,914
NvidiaMark once again.
FLCLimax is offline  
Sponsored Links
Advertisement
 
post #224 of 632 (permalink) Old 07-16-2016, 02:12 AM
Linux Lobbyist
 
EightDee8D's Avatar
 
Join Date: Jan 2015
Posts: 1,864
Rep: 184 (Unique: 64)
Quote:
Originally Posted by Mahigan View Post



Where are the tech journalists these days?
Bought out by big bad boyz (intl/nvda) to slack, so they can con many peoples.

Quote:
"if (c->x86_vendor != X86_VENDOR_AMD)
setup_force_cpu_bug(X86_BUG_CPU_INSECURE);"
EightDee8D is offline  
post #225 of 632 (permalink) Old 07-16-2016, 02:17 AM - Thread Starter
New to Overclock.net
 
Randomdude's Avatar
 
Join Date: Aug 2011
Posts: 486
Rep: 45 (Unique: 28)
I am very underwhelmed by this GPU cycle. Honestly, new memory,new node,new architecture - these things had my hopes up for something different. What it turns out is that nVidia is yet again staying in the way of innovation (to make more money, who can blame them) with their Pascal shenanigans, especially after seeing Mahigan's Time Spy posts. And on the other hand is AMD that might have the better hardware, but you'd be a fool to buy them when the market is controlled by nV. I really want to buy a Vega, but I know that the Titan will be likely faster. This is like an exact repeat of the Maxwell/Fiji generation. I am not going to buy a Pascal (Maxwell v3) regardless of its performance when I know that it will be times more obsolete than Maxwell is now when Volta hits. If I go with a Vega then I'll have to deal with nVidia strong arming the market and it would be just as obsolete (having features it can't make use of) as Fiji was... Really sad, imo.

Randomdude is offline  
post #226 of 632 (permalink) Old 07-16-2016, 02:53 AM
New to Overclock.net
 
scorch062's Avatar
 
Join Date: Jun 2016
Posts: 160
Rep: 13 (Unique: 9)
DF did another video, now mostly about Fury X at 4k with some discussion:
Pretty impressive to sustain 50 to 60 FPS at 4K for Fury X on Ultra. 1070, as little has been in the video, gains nothing so far and under Vulkan the V-sync cannot be disabled for it.

scorch062 is offline  
post #227 of 632 (permalink) Old 07-16-2016, 02:58 AM
 
Join Date: Mar 2011
Location: New York
Posts: 3,914
No need to upload my video then.
FLCLimax is offline  
post #228 of 632 (permalink) Old 07-16-2016, 03:00 AM
New to Overclock.net
 
Remij's Avatar
 
Join Date: Apr 2012
Posts: 574
Rep: 44 (Unique: 27)
Quote:
Originally Posted by Randomdude View Post

I am very underwhelmed by this GPU cycle. Honestly, new memory,new node,new architecture - these things had my hopes up for something different. What it turns out is that nVidia is yet again staying in the way of innovation (to make more money, who can blame them) with their Pascal shenanigans, especially after seeing Mahigan's Time Spy posts. And on the other hand is AMD that might have the better hardware, but you'd be a fool to buy them when the market is controlled by nV. I really want to buy a Vega, but I know that the Titan will be likely faster. This is like an exact repeat of the Maxwell/Fiji generation. I am not going to buy a Pascal (Maxwell v3) regardless of its performance when I know that it will be times more obsolete than Maxwell is now when Volta hits. If I go with a Vega then I'll have to deal with nVidia strong arming the market and it would be just as obsolete (having features it can't make use of) as Fiji was... Really sad, imo.

Oh lord... if you like AMD and what they are doing.. support them. They need it.

But it's funny to me that people like you claim that Nvidia are holding technology back, having to debate which GPU vendor to buy from because on one hand.. AMD has innovative hardware, yet on the other hand Nvidia has the best performance.. It doesn't sound like they are holding technology back to me...

Blame AMD for not having an ultra high-end enthusiast card out there to battle the 1080 yet

If you like that AMD cards hold performance due to more forward thinking technologies, then support AMD for doing that. But I think it's time we stop blaming Nvidia for holding things back when right now AMD isn't even giving them any challenge on the high end.

Mid-range, I think the choice is fairly clear which company stands to provide the best performance/value ratio. AMD has always been good in that regard, and always been a more value oriented purchase for consumers in that range.

But all this strong arming that people say Nvidia is doing with their marketshare influence is no worse than the advantage AMD has with the GCN architecture being in all 3 current gen consoles and those coming in the foreseeable future. Nvidia has the right and duty to push it's platform and make it THE hardware to want to play games on. AMD has the same duty. They have the hardware, it's their job to make sure developers use it to it's fullest extent. There was a wait, but now things are starting to come together for them. They've got much better APIs for their hardware, they have better drivers (seemingly) than before. They have their hardware in all game consoles. And now they have a newer architecture on the horizon.

However, after all that if you're still not convinced to buy from them (especially because the competition still has a faster GPU out) then you need to just drop the moral dilemma from the equation and buy what suits your needs and be happy with it.

Remij is offline  
post #229 of 632 (permalink) Old 07-16-2016, 03:05 AM
 
Join Date: Jul 2014
Posts: 2,899
Quote:
1070, as little has been in the video, gains nothing so far and under Vulkan the V-sync cannot be disabled for it.
dunno about 1070, bur neither of those is true for my G1 1080

there is a slight increase in avg fps, and a decent one in min fps

Vsync can absolutely be disabled with Vulkan, Vsync line in Options says Off and my fps counter isnt 60, but up to 200 fps, as it should be
ChevChelios is offline  
post #230 of 632 (permalink) Old 07-16-2016, 04:13 AM
New to Overclock.net
 
daviejams's Avatar
 
Join Date: Feb 2014
Posts: 1,431
Rep: 33 (Unique: 24)
Fury x 27% faster than the 1070 in the first scene in the Eurogamer video above using vulcan

Impressive
daviejams is offline  
Reply

Quick Reply
Message:
Options

Register Now

In order to be able to post messages on the Overclock.net - An Overclocking Community forums, you must first register.
Please enter your desired user name, your email address and other required details in the form below.
User Name:
If you do not want to register, fill this field only and the name will be used as user name for your post.
Password
Please enter a password for your user account. Note that passwords are case-sensitive.
Password:
Confirm Password:
Email Address
Please enter a valid email address for yourself.
Email Address:

Log-in



Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 
Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may post new threads
You may post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off