Overclock.net › Forums › Industry News › Video Game News › [computerbase.de] DOOM + Vulkan Benchmarked.
New Posts  All Forums:Forum Nav:

[computerbase.de] DOOM + Vulkan Benchmarked. - Page 46

post #451 of 632
Quote:
Originally Posted by infranoia View Post


They had no problem doing this with tessellation. Now suddenly they've got morals?

I'd say there's a difference between doing the same workload (serially vs in parallel) and actively reducing the amount of workload with tessellation (geometry) is there not? Or am I not understanding this correctly?
My main PC
(8 items)
 
  
CPUMotherboardGraphicsRAM
Intel i7 6700k Asus ROG Maximus VIII Gene Nvidia GTX 1080Ti G.Skill Ripjaws 
Hard DriveOSKeyboardPower
Samsung 850 EVO  Windows 10 Razer Blackwidow Chroma EVGA Supernova 1300w 
  hide details  
Reply
My main PC
(8 items)
 
  
CPUMotherboardGraphicsRAM
Intel i7 6700k Asus ROG Maximus VIII Gene Nvidia GTX 1080Ti G.Skill Ripjaws 
Hard DriveOSKeyboardPower
Samsung 850 EVO  Windows 10 Razer Blackwidow Chroma EVGA Supernova 1300w 
  hide details  
Reply
post #452 of 632
Quote:
Originally Posted by infranoia View Post

This is exactly the dilemma I expected from them. There's no way to have a single render path in a DX12 benchmark without optimizing it for the lowest common denominator and punishing the silicon with extra features.

"Impartial" benchmarking has become an oxymoron with DX12. You have to optimize for each vendor or you're unfairly punishing one of them. It just about makes the whole concept of "benchmark" meaningless.

They had no problem doing this with tessellation. Now suddenly they've got morals?

Exactly, this is just a joke, the double standards are too much, now that they are coming directly from a Futuremark developer I call this Benchmark 100% pointless anymore, no matter how much they come close to reality, it's clear they favor Nvidia philosophy.
post #453 of 632
Quote:
Originally Posted by Remij View Post

I'd say there's a difference between doing the same workload (serially vs in parallel) and actively reducing the amount of workload with tessellation (geometry) is there not? Or am I not understanding this correctly?

I get you, but DX12 is not a one-size-fits-all API. Arguably DX11 was, but AMD suffered with high tess and had driver optimizations to keep such punishment within architectural limits. These driver optimizations became invalid within 3dmark, so they were left competing one-for-one with Nvidia.

OK 3dmark, that's fine if you want to look neutral, but now with DX12 AMD isn't allowed to shine with its parallel hardware-- it must remain on a level playing field with an NV-optimized render path. It's not an indication of game performance, unless that game is specifically NV-optimized and has very few if any AMD async shader optimizations.

See the theme here? The last 3dmark was NV-optimized with tessellation levels. The limitation was on the AMD side, and the fix was ignored / bypassed. This 3dmark is NV-optimized in its avoidance of Async Compute + Graphics, aka Async Shaders. The limitation is on the Nvidia side, and the fix is honored.

It's a valid benchmark as long as AMD knows its place.
Edited by infranoia - 7/17/16 at 5:34pm
Parasite
(18 items)
 
  
CPUMotherboardGraphicsGraphics
i7 4770K @ 4.7GHz Z87 MPOWER (MS-7818) Sapphire Radeon 290x @1100/1500 EVGA 1080Ti SC2 Hybrid 
RAMHard DriveHard DriveCooling
G.SKILL 2133 Samsung 850 Pro Caviar Black Corsair H100 
CoolingCoolingOSMonitor
Corsair HG10 Corsair H60 Windows 7 x64 Sony XBR65X850B 
KeyboardPowerCaseMouse
CMSTORM Quickfire XT Corsair AX1200i Antec P280 Logitec G700 
Mouse PadAudio
Black, came with my NeXTcube 25 years ago. Sound Blaster Recon 3D PCIe 
  hide details  
Reply
Parasite
(18 items)
 
  
CPUMotherboardGraphicsGraphics
i7 4770K @ 4.7GHz Z87 MPOWER (MS-7818) Sapphire Radeon 290x @1100/1500 EVGA 1080Ti SC2 Hybrid 
RAMHard DriveHard DriveCooling
G.SKILL 2133 Samsung 850 Pro Caviar Black Corsair H100 
CoolingCoolingOSMonitor
Corsair HG10 Corsair H60 Windows 7 x64 Sony XBR65X850B 
KeyboardPowerCaseMouse
CMSTORM Quickfire XT Corsair AX1200i Antec P280 Logitec G700 
Mouse PadAudio
Black, came with my NeXTcube 25 years ago. Sound Blaster Recon 3D PCIe 
  hide details  
Reply
post #454 of 632
Quote:
Originally Posted by infranoia View Post

I get you, but DX12 is not a one-size-fits-all API. Arguably DX11 was, but AMD suffered with high tess and had driver optimizations to keep such punishment within architectural limits. These driver optimizations became invalid within 3dmark, so they were left competing one-for-one with Nvidia.

OK 3dmark, that's fine if you want to look neutral, but now with DX12 AMD isn't allowed to shine with its parallel hardware-- it must remain on a level playing field with an NV-optimized render path. It's not an indication of game performance, unless that game is specifically NV-optimized and has very few if any AMD async shader optimizations.

Well, I'd argue that AMD's hardware is still being allowed to shine. But then again I guess it's also true that Nvidia's hardware isn't being allowed to look worse than it would have under a different codepath.

I see where everyone is coming from.

I think Futuremark should make a highly parallel Async compute benchmark (separate from labeling it a DX12 benchmark that uses async compute) and really show what a highly parallel GPU architecture can do. And hold no punches back. If the hardware doesn't support it. It doesn't even run. If somethings altered in drivers, then the results are invalid. Nvidia will be heading in that direction eventually.. it's pretty well understood that it's the future, so that when things can be performed the exact same way on both IHVs hardware, then we can see which architecture handles parallel workloads better.
My main PC
(8 items)
 
  
CPUMotherboardGraphicsRAM
Intel i7 6700k Asus ROG Maximus VIII Gene Nvidia GTX 1080Ti G.Skill Ripjaws 
Hard DriveOSKeyboardPower
Samsung 850 EVO  Windows 10 Razer Blackwidow Chroma EVGA Supernova 1300w 
  hide details  
Reply
My main PC
(8 items)
 
  
CPUMotherboardGraphicsRAM
Intel i7 6700k Asus ROG Maximus VIII Gene Nvidia GTX 1080Ti G.Skill Ripjaws 
Hard DriveOSKeyboardPower
Samsung 850 EVO  Windows 10 Razer Blackwidow Chroma EVGA Supernova 1300w 
  hide details  
Reply
post #455 of 632
Quote:
Originally Posted by Remij View Post

I think Futuremark should make a highly parallel Async compute benchmark (separate from labeling it a DX12 benchmark that uses async compute) and really show what a highly parallel GPU architecture can do. And hold no punches back. Nvidia will be heading in that direction eventually.. it's pretty well understood that it's the future, so that when things can be performed the exact same way on both IHVs hardware, then we can see which architecture handles parallel workloads better.

They will surely do this, but not one moment before Volta shows up. They are a business, after all, and true vendor "neutrality" isn't possible outside of an open-source project and codebase.
Parasite
(18 items)
 
  
CPUMotherboardGraphicsGraphics
i7 4770K @ 4.7GHz Z87 MPOWER (MS-7818) Sapphire Radeon 290x @1100/1500 EVGA 1080Ti SC2 Hybrid 
RAMHard DriveHard DriveCooling
G.SKILL 2133 Samsung 850 Pro Caviar Black Corsair H100 
CoolingCoolingOSMonitor
Corsair HG10 Corsair H60 Windows 7 x64 Sony XBR65X850B 
KeyboardPowerCaseMouse
CMSTORM Quickfire XT Corsair AX1200i Antec P280 Logitec G700 
Mouse PadAudio
Black, came with my NeXTcube 25 years ago. Sound Blaster Recon 3D PCIe 
  hide details  
Reply
Parasite
(18 items)
 
  
CPUMotherboardGraphicsGraphics
i7 4770K @ 4.7GHz Z87 MPOWER (MS-7818) Sapphire Radeon 290x @1100/1500 EVGA 1080Ti SC2 Hybrid 
RAMHard DriveHard DriveCooling
G.SKILL 2133 Samsung 850 Pro Caviar Black Corsair H100 
CoolingCoolingOSMonitor
Corsair HG10 Corsair H60 Windows 7 x64 Sony XBR65X850B 
KeyboardPowerCaseMouse
CMSTORM Quickfire XT Corsair AX1200i Antec P280 Logitec G700 
Mouse PadAudio
Black, came with my NeXTcube 25 years ago. Sound Blaster Recon 3D PCIe 
  hide details  
Reply
post #456 of 632
...Car analogy taken out behind the shed...

Carry on. biggrin.gif
Edited by infranoia - 7/17/16 at 6:13pm
Parasite
(18 items)
 
  
CPUMotherboardGraphicsGraphics
i7 4770K @ 4.7GHz Z87 MPOWER (MS-7818) Sapphire Radeon 290x @1100/1500 EVGA 1080Ti SC2 Hybrid 
RAMHard DriveHard DriveCooling
G.SKILL 2133 Samsung 850 Pro Caviar Black Corsair H100 
CoolingCoolingOSMonitor
Corsair HG10 Corsair H60 Windows 7 x64 Sony XBR65X850B 
KeyboardPowerCaseMouse
CMSTORM Quickfire XT Corsair AX1200i Antec P280 Logitec G700 
Mouse PadAudio
Black, came with my NeXTcube 25 years ago. Sound Blaster Recon 3D PCIe 
  hide details  
Reply
Parasite
(18 items)
 
  
CPUMotherboardGraphicsGraphics
i7 4770K @ 4.7GHz Z87 MPOWER (MS-7818) Sapphire Radeon 290x @1100/1500 EVGA 1080Ti SC2 Hybrid 
RAMHard DriveHard DriveCooling
G.SKILL 2133 Samsung 850 Pro Caviar Black Corsair H100 
CoolingCoolingOSMonitor
Corsair HG10 Corsair H60 Windows 7 x64 Sony XBR65X850B 
KeyboardPowerCaseMouse
CMSTORM Quickfire XT Corsair AX1200i Antec P280 Logitec G700 
Mouse PadAudio
Black, came with my NeXTcube 25 years ago. Sound Blaster Recon 3D PCIe 
  hide details  
Reply
post #457 of 632
Thats 2% difference, actually within margin of error. And thats not a drastic loss in in performance either. With the disparity between runs in AotS, I have async runs that show higher fps than my non async.
post #458 of 632
Quote:
Originally Posted by Remij View Post

Well, I'd argue that AMD's hardware is still being allowed to shine. But then again I guess it's also true that Nvidia's hardware isn't being allowed to look worse than it would have under a different codepath.

I see where everyone is coming from.

I think Futuremark should make a highly parallel Async compute benchmark (separate from labeling it a DX12 benchmark that uses async compute) and really show what a highly parallel GPU architecture can do. And hold no punches back. If the hardware doesn't support it. It doesn't even run. If somethings altered in drivers, then the results are invalid. Nvidia will be heading in that direction eventually.. it's pretty well understood that it's the future, so that when things can be performed the exact same way on both IHVs hardware, then we can see which architecture handles parallel workloads better.

If async compute was hold back because it gave AMD too much of a lead (and NOT by punishing nVIDIA, which could run the test just as well with it off), then the test is not vaid. Async compute is a feature that brings performance, just like hyperthreading does, when it's properly implemented ( or it can bring higher image quality if you want to keep the same performance, but add some effects). For AMD is the way the hardware was designed to work in the first place in order to achieve maximum performance.

At the moment, the only logical explanation would be that 3DMark is trying not to upset nVIDIA too much. If you take out 1070 and 1080, their older products are being left behind by the even older products from AMD, such as R290/x
post #459 of 632
Funny that people cry fault when 3DMark maybe is not using more Async to make AMD shine more, but no one cares that DOOM Vulcan doesn't even support Async for Nvidia, and that Pascal is losing about 6% in Vulcan benchmarks because of this.
Quote:
Does DOOM support asynchronous compute when running on the Vulkan API?

Asynchronous compute is a feature that provides additional performance gains on top of the baseline id Tech 6 Vulkan feature set.

Currently asynchronous compute is only supported on AMD GPUs and requires DOOM Vulkan supported drivers to run. We are working with NVIDIA to enable asynchronous compute in Vulkan on NVIDIA GPUs. We hope to have an update soon.
https://community.bethesda.net/thread/54585?tstart=0
post #460 of 632
Quote:
Originally Posted by Glottis View Post

Funny that people cry fault when 3DMark maybe is not using more Async to make AMD shine more, but no one cares that DOOM Vulcan doesn't even support Async for Nvidia, and that Pascal is losing about 6% in Vulcan benchmarks because of this.
https://community.bethesda.net/thread/54585?tstart=0
Because what you believe is async after seeing on time spy is not actually async. read what mahigan explained before crying.

why does nvidia's own sponsored game doesn't have async ? hint - they don't support it. means pascal doesn't support graphics + compute at the same time.
Edited by EightDee8D - 7/18/16 at 1:16am
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Video Game News
Overclock.net › Forums › Industry News › Video Game News › [computerbase.de] DOOM + Vulkan Benchmarked.