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 50

post #491 of 632
This can only get worse if Futuremark decides to increase compute load only when Volta gets released with full async capabilites.
post #492 of 632
Quote:
Originally Posted by Slomo4shO View Post

So even lower at 18.76%...

The bench definitely isn't compute heavy.

It does look that way compared to AOTS, and DOOM. I don't have ROTR, Hitman, or any other DX12/Vulkan titles to test this theory against. In the two other games, GPUView shows two rectangles(compute queues) stacked on top of each other. Time Spy never needs to process more than one at a time.


BTW, Forget what I said about the pre-emption not being executed earlier. I looked closer and it is definitely being executed. Again I don't know what that's there for, and I couldn't find the Device Context associated with it. I do not see Pre-Emption on AOTS, or DOOM though so there's that. Maybe something worth looking into? Who knows.

Time Spy Async off/on comparison
eOdqMzk.jpg
Edited by Doothe - 7/18/16 at 1:51pm
post #493 of 632
Quote:
Originally Posted by AmericanLoco View Post

So what's going on there? I don't what I'm looking at it, or how to interpret it. The only thing I notice different about 3D Mark, is that it seems like one of the 3D Queues stops when some compute work gets loaded in, so they're not being executed in parallel?
time spy compute queues are less than AotS, most of them are graphics, and it seems they do double fences which could be throttling AMD´s compute+graphics perf and/or parrallelism
Quote:
Minimize the use of barriers and fences
We have seen redundant barriers and associated wait for idle operations as a major performance problem for DX11 to DX12 ports
The DX11 driver is doing a great job of reducing barriers – now under DX12 you need to do it
Any barrier or fence can limit parallelism



then 3dmark could run a single path where It fits most hardware, with pre emption
Edited by PontiacGTX - 7/18/16 at 10:01am
Wanted: [WTB] GPU upgrade
$210 (USD) or best offer
  
Reply
Wanted: [WTB] GPU upgrade
$210 (USD) or best offer
  
Reply
post #494 of 632
Quote:
Originally Posted by PontiacGTX View Post

then 3dmark could run a single path where It fits most hardware, with pre emption

Incorrect, it fits Pascal architecture. THE ONLY architecture that currently is made for preemption.
post #495 of 632
Quote:
Originally Posted by Bidz View Post

Incorrect, it fits Pascal architecture. THE ONLY architecture that currently is made for preemption.
It can be used for GCN, but it wont take advantage of parrarellism and performance gains, Maxwell can do some degree of pre emption and it doesnt get negative performance(given how fences are limiting the contexts switching) and it can work in Pascal given it has improved pre emption

when people compares them Maxwell seems to have some degree of async compute(benchmark is aimed to it but it does pre emption) GCN can do pre emption but it isnt deliver same gains as async compute and Pascal shows their improved pre emption gains



Devs tell they use a single path but this only favors one side
Edited by PontiacGTX - 7/18/16 at 4:21pm
Wanted: [WTB] GPU upgrade
$210 (USD) or best offer
  
Reply
Wanted: [WTB] GPU upgrade
$210 (USD) or best offer
  
Reply
post #496 of 632
Quote:
Originally Posted by PontiacGTX View Post

...
Devs tell they use a single path but this only favors one side

With the given evidence, we can say that Time Spy benchmark, intentionally or not, by design, fits perfectly for the capabilites of Pascal, other Nvidia architectures are not capable of async computing at all, and most of the AMD architectures in theory are left with spare room to be requested of much heavier async computing loads.

It's like Tessellation loads were designed to fit the inferior AMD capabilities back in the day. There is a clear pattern with Futuremark controversies regardless of who's on the right or wrong, and it's that they always favor Nvidia.
post #497 of 632
Quote:
Originally Posted by Bidz View Post

With the given evidence, we can say that Time Spy benchmark, intentionally or not, by design, fits perfectly for the capabilites of Pascal, other Nvidia architectures are not capable of async computing at all, and most of the AMD architectures in theory are left with spare room to be requested of much heavier async computing loads.

It's like Tessellation loads were designed to fit the inferior AMD capabilities back in the day. There is a clear pattern with Futuremark controversies regardless of who's on the right or wrong, and it's that they always favor Nvidia.


are you going to sell a benchmark that favours the minority?

people who buy this are users like you and i and if we know from forums and what not that it won't work well on our cards we won't buy it.
Gaming Rig
(15 items)
 
Sons Rig
(11 items)
 
File Server
(13 items)
 
CPUMotherboardGraphicsGraphics
AMD Ryzen 1700x ASUS Prime X370 Pro Gigabyte GeForce GTX 980 Gigabyte GeForce GTX 980 
RAMHard DriveCoolingOS
16gb's Team Select 3200Mhz RGB  Crucial M500 SSD 240gb Corsair H55 Windows 7 Ultimate 
MonitorMonitorPowerCase
ASUS PB287Q 4K Monitor HTC Vive Coolermaster Silent Pro M 1000W Rosewill Blackhawk Ultra 
MouseMouse PadAudio
Razer Deathadder 2013 Steelseries Qck Mass Super Thick Cloth Mouse Pad Genius SW-G2.1 1250 4PC Gaming Speakers 
CPUMotherboardGraphicsGraphics
Core i7 3770k Asus P8Z68 Deluxe Galaxy Geforce GTX 780 Galaxy Geforce GTX 780 
RAMHard DriveCoolingOS
16 Gb's G.Skill DDR3 1866 Crucial M500 SSD 240Gb Coolit ECO C240 Windows 7 Ultimate 
MonitorPowerCase
Benq G2420HD Coolermaster Silent Pro 1000 Watt Deepcool Kendomen 
CPUMotherboardRAMHard Drive
Core i3 2120 Supermicro X9SCM-F 4 Gb Kingston 1600mhz DDR3 ECC 12 Segate 2tb drives (RAID 6) 
Hard DriveHard DriveOSPower
ADATA SP600 SSD 24 Toshiba DT01ACA200 drives (2 RAID 6's) Windows 7 Ultimate 64 bit. OCZ ZT750 Supermicro PWS-665-PQ 
CaseOtherOtherOther
Norco 4020 and Norco 4224 LSI 9260-4i Raid Card Intel RES2SV240 20 port Expander. HP SAS 24 por... Voltaire 410Ex Hca Infiniband HBA 
  hide details  
Reply
Gaming Rig
(15 items)
 
Sons Rig
(11 items)
 
File Server
(13 items)
 
CPUMotherboardGraphicsGraphics
AMD Ryzen 1700x ASUS Prime X370 Pro Gigabyte GeForce GTX 980 Gigabyte GeForce GTX 980 
RAMHard DriveCoolingOS
16gb's Team Select 3200Mhz RGB  Crucial M500 SSD 240gb Corsair H55 Windows 7 Ultimate 
MonitorMonitorPowerCase
ASUS PB287Q 4K Monitor HTC Vive Coolermaster Silent Pro M 1000W Rosewill Blackhawk Ultra 
MouseMouse PadAudio
Razer Deathadder 2013 Steelseries Qck Mass Super Thick Cloth Mouse Pad Genius SW-G2.1 1250 4PC Gaming Speakers 
CPUMotherboardGraphicsGraphics
Core i7 3770k Asus P8Z68 Deluxe Galaxy Geforce GTX 780 Galaxy Geforce GTX 780 
RAMHard DriveCoolingOS
16 Gb's G.Skill DDR3 1866 Crucial M500 SSD 240Gb Coolit ECO C240 Windows 7 Ultimate 
MonitorPowerCase
Benq G2420HD Coolermaster Silent Pro 1000 Watt Deepcool Kendomen 
CPUMotherboardRAMHard Drive
Core i3 2120 Supermicro X9SCM-F 4 Gb Kingston 1600mhz DDR3 ECC 12 Segate 2tb drives (RAID 6) 
Hard DriveHard DriveOSPower
ADATA SP600 SSD 24 Toshiba DT01ACA200 drives (2 RAID 6's) Windows 7 Ultimate 64 bit. OCZ ZT750 Supermicro PWS-665-PQ 
CaseOtherOtherOther
Norco 4020 and Norco 4224 LSI 9260-4i Raid Card Intel RES2SV240 20 port Expander. HP SAS 24 por... Voltaire 410Ex Hca Infiniband HBA 
  hide details  
Reply
post #498 of 632
They clearly said that they have only a neutral optimization path. They support only FL_11.0 and the only way to achieve async compute on both architectures is by supporting the method that both GCN and Paxwell can support.

That means preemption.

Oh well it doesn't matter, it is just a benchmark.
Workstation
(4 items)
 
  
CPUMotherboardGraphicsMonitor
Xeon E5-2690 Supermicro 2011 Nvidia GP100/ Vega FE Dell ultrasharp 4k 
  hide details  
Reply
Workstation
(4 items)
 
  
CPUMotherboardGraphicsMonitor
Xeon E5-2690 Supermicro 2011 Nvidia GP100/ Vega FE Dell ultrasharp 4k 
  hide details  
Reply
post #499 of 632
Quote:
Originally Posted by Master__Shake View Post

[/B]

are you going to sell a benchmark that favours the minority?

people who buy this are users like you and i and if we know from forums and what not that it won't work well on our cards we won't buy it.
Wait! They were losing money when people get to test it for free, right?rolleyes.gif
The Machine
(14 items)
 
Nexus 7 2013
(11 items)
 
 
CPUMotherboardGraphicsRAM
A10 6800K Asus F2A85-V MSI 6870 Hawx, VTX3D 5770, AMD HD6950(RIP), Sap... G.skill Ripjaws PC12800 6-8-6-24 
Hard DriveOptical DriveOSMonitor
Seagate 7200.5 1TB NEC 3540 Dvd-Rom Windows 7 x32 Ultimate Samsung P2350 23" 1080p 
PowerCaseMouseAudio
Seasonic s12-600w CoolerMaster Centurion 5 Logitech G600 Auzen X-Fi Raider 
CPUMotherboardGraphicsRAM
Quad Krait 300 at 1.5Ghz Qualcomm APQ8064-1AA SOC Adreno 320 at 400mhz 2GB DDR3L-1600 
Hard DriveOSMonitorKeyboard
32GB Internal NAND Android 5.0 7" 1920X1200 103% sRGB & 572 cd/m2 LTPS IPS Microsoft Wedge Mobile Keyboard 
PowerAudio
3950mAh/15.01mAh Battery Stereo Speakers 
  hide details  
Reply
The Machine
(14 items)
 
Nexus 7 2013
(11 items)
 
 
CPUMotherboardGraphicsRAM
A10 6800K Asus F2A85-V MSI 6870 Hawx, VTX3D 5770, AMD HD6950(RIP), Sap... G.skill Ripjaws PC12800 6-8-6-24 
Hard DriveOptical DriveOSMonitor
Seagate 7200.5 1TB NEC 3540 Dvd-Rom Windows 7 x32 Ultimate Samsung P2350 23" 1080p 
PowerCaseMouseAudio
Seasonic s12-600w CoolerMaster Centurion 5 Logitech G600 Auzen X-Fi Raider 
CPUMotherboardGraphicsRAM
Quad Krait 300 at 1.5Ghz Qualcomm APQ8064-1AA SOC Adreno 320 at 400mhz 2GB DDR3L-1600 
Hard DriveOSMonitorKeyboard
32GB Internal NAND Android 5.0 7" 1920X1200 103% sRGB & 572 cd/m2 LTPS IPS Microsoft Wedge Mobile Keyboard 
PowerAudio
3950mAh/15.01mAh Battery Stereo Speakers 
  hide details  
Reply
post #500 of 632
Quote:
Originally Posted by Master__Shake View Post

[/B]

are you going to sell a benchmark that favours the minority?

people who buy this are users like you and i and if we know from forums and what not that it won't work well on our cards we won't buy it.

So now it's ok to drop objectivity in favor of sales?

You know, in cases of highly important stuff like measuring safety, contamination, etc, dropping objectivity in favor of "sales" can send you to prison.
Quote:
Originally Posted by sugarhell View Post

They clearly said that they have only a neutral optimization path. They support only FL_11.0 and the only way to achieve async compute on both architectures is by supporting the method that both GCN and Paxwell can support.

That means preemption.

Oh well it doesn't matter, it is just a benchmark.

It's not neutral if it's not really measuring the full extent of async capabilities by limiting em to fit one side capabilities.
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.