Overclock.net › Forums › Industry News › Hardware News › [pcper] NVIDIA Talks DX12, DX11 Efficiency Improvements
New Posts  All Forums:Forum Nav:

[pcper] NVIDIA Talks DX12, DX11 Efficiency Improvements - Page 16

post #151 of 203
Quote:
Originally Posted by Alatar View Post

Well whatever you did you ran a completely different scenario...

E: besides, how did you get a score 4 minutes after I posted the csv file...? the SS bench takes longer than that...

Because i saw your previous post when you upload your settings. I think its so boring that you always try this. We all know that they are pr graphs. First we know that follow scenario is completely random. Second they use 14.2 drivers for thief and 55~ ish fps for 290x max settings is really slow. Probably they used quiet bios. And then look the Y-axis. Do you remind you amd pr graphs? Oh yes. And then the question why now? Dx11 is from 2008. Now they find out the new super API optimizations? They only thing that they can do is bypass some dx11 fuctions for their own. Nothing else.

And if dx11 can rival mantle why microsoft even care to release dx12? A completely new API? And nvidia demonstrate these new drivers with a 3960x and a 780ti.They can do the same with low end cpus? Because we all know low level API helps more low end cpus.
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 #152 of 203
Quote:
Originally Posted by Kuivamaa View Post

Check your own test numbers.
280X DX11 avg unit count 3961
280X Mantle avg unit count 4436
Titan DX11 avg unit count 4057

It most certainly isn't a directly comparable scenario (280X under mantle had more stuff going on on average). Even with the same amount of units I doubt we could possibly have identical twin runs. Mantle is obviously nondeterministic (AI does whatever it sees fit between different runs).

Good eyes thumb.gif
Intel Killer
(14 items)
 
  
CPUMotherboardGraphicsRAM
AMD 8320 4.4ghz Gigabyte 990fx XFX R9 290 (1100/1250) 8GB DDR3 1866 
Hard DriveCoolingOSMonitor
750GB Momentus XT Corsair H80i Windows 8 3x 23" 1080p 
KeyboardPower
Logitech Mk 710 OCZ 1000 watts 
  hide details  
Reply
Intel Killer
(14 items)
 
  
CPUMotherboardGraphicsRAM
AMD 8320 4.4ghz Gigabyte 990fx XFX R9 290 (1100/1250) 8GB DDR3 1866 
Hard DriveCoolingOSMonitor
750GB Momentus XT Corsair H80i Windows 8 3x 23" 1080p 
KeyboardPower
Logitech Mk 710 OCZ 1000 watts 
  hide details  
Reply
post #153 of 203
Quote:
Originally Posted by sugarhell View Post

And if dx11 can rival mantle why microsoft even care to release dx12? A completely new API? And nvidia demonstrate these new drivers with a 3960x and a 780ti.They can do the same with low end cpus? Because we all know low level API helps more low end cpus.

Because, strangely, it's not just about squashing the other guy, as much as you seem to be hellbent on making sure whoever gets squashed isn't AMD.

There doesn't have to be a loser in order for us to be winners.
post #154 of 203
What Alatar really mean is that a 270X with Mantle enabled, is about as fast as a Titan. In this benchmark at least. tongue.gif

Oxygène
(15 items)
 
  
CPUMotherboardGraphicsRAM
Ryzen 5 1600 MSI B350 Tomahawk Arctic VTX3D R9 390 8GB G.Skill Ripjaws  
Hard DriveHard DriveHard DriveHard Drive
Kingston Hyper X Kingston HyperX Fury ADATA SP550 ADATA SX8000 
CoolingOSMonitorKeyboard
Noctua NH-D15 SE-AM4 Windows 10 Dell U2711 Microsoft WK 800 
PowerCaseMouse
Chieftec A-135 Series 1000W NZXT. Phantom 410 Microsoft WM 1000 
  hide details  
Reply
Oxygène
(15 items)
 
  
CPUMotherboardGraphicsRAM
Ryzen 5 1600 MSI B350 Tomahawk Arctic VTX3D R9 390 8GB G.Skill Ripjaws  
Hard DriveHard DriveHard DriveHard Drive
Kingston Hyper X Kingston HyperX Fury ADATA SP550 ADATA SX8000 
CoolingOSMonitorKeyboard
Noctua NH-D15 SE-AM4 Windows 10 Dell U2711 Microsoft WK 800 
PowerCaseMouse
Chieftec A-135 Series 1000W NZXT. Phantom 410 Microsoft WM 1000 
  hide details  
Reply
post #155 of 203
Quote:
Originally Posted by sugarhell View Post

Because i saw your previous post when you upload your settings. I think its so boring that you always try this. We all know that they are pr graphs. First we know that follow scenario is completely random. Second they use 14.2 drivers for thief and 55~ ish fps for 290x max settings is really slow. Probably they used quiet bios. And then look the Y-axis. Do you remind you amd pr graphs? Oh yes. And then the question why now? Dx11 is from 2008. Now they find out the new super API optimizations? They only thing that they can do is bypass some dx11 fuctions for their own. Nothing else.

And if dx11 can rival mantle why microsoft even care to release dx12? A completely new API? And nvidia demonstrate these new drivers with a 3960x and a 780ti.They can do the same with low end cpus? Because we all know low level API helps more low end cpus.

Why are we talking about the Nvidia graph? I'm trying to do some non 1st party Nvidia DX11 CPU overhead testing.

Anyway, your run is still completely different. All of my runs have 33-35K batch count. And that's the number you would also get if you ran the same scenario. Instead you have like 8K which is over 4 times lower...

"My" scenario starts out like this:



And that's where the camera stays all the time.

Quote:
Originally Posted by bencher View Post

Good eyes thumb.gif

We've been through this already, if the variance between runs is sub 2fps it doesn't affect the conclusion anyway when Nvidia had a 24% advantage in DX11.
 
Benching
(17 items)
 
 
CPUMotherboardGraphicsRAM
[i7 5960X @ 4.8GHz] [Rampage V Extreme] [Titan 1400MHz (1500MHz bench)] [Various] 
Hard DriveCoolingCoolingCooling
[250GB 840EVO +2x SpinpointF3 1TB RAID0] [LD PC-V2 SS Phase Change] [XSPC X2O 750 pump/res] [Monsta 360 full copper + EK XT 360 + XT 240] 
MonitorPowerCaseAudio
[Crossover 27Q LED-P 1440p+ASUS 1200p+LG 1080p] [Corsair AX1200] [Dimastech Easy v3.0] [Sennheiser HD558s] 
CPUCPUMotherboardGraphics
FX 8320, FX 8350, Phenom II x2 555BE i7 3930K, i7 860, i7 4770K, 68x Celeron D CVF, commando, 2x RIVE, Z87X-OC Asus 4870x2, Sapphire 4870 
GraphicsGraphicsGraphicsGraphics
2x 5870, 5850, 5830, 5770 2x 3870x2, 3870 GTX Titan, GTX 480, GTX 590 GTX 285, GTX 260, 4x 9800GT, 8800GTX 
RAMHard DriveCoolingCooling
4x4GB vengeance, 2x4GB predatorX, 2x1GB OCZ DDR2 Intel X25-M 80GB LD PC-V2 SS Phase Change OCN Marksman 
CoolingCoolingOSPower
2x old tek slims (GPU) Various watercooling stuff win7, winxp AX1200 
Case
test bench / cardboard box 
  hide details  
Reply
 
Benching
(17 items)
 
 
CPUMotherboardGraphicsRAM
[i7 5960X @ 4.8GHz] [Rampage V Extreme] [Titan 1400MHz (1500MHz bench)] [Various] 
Hard DriveCoolingCoolingCooling
[250GB 840EVO +2x SpinpointF3 1TB RAID0] [LD PC-V2 SS Phase Change] [XSPC X2O 750 pump/res] [Monsta 360 full copper + EK XT 360 + XT 240] 
MonitorPowerCaseAudio
[Crossover 27Q LED-P 1440p+ASUS 1200p+LG 1080p] [Corsair AX1200] [Dimastech Easy v3.0] [Sennheiser HD558s] 
CPUCPUMotherboardGraphics
FX 8320, FX 8350, Phenom II x2 555BE i7 3930K, i7 860, i7 4770K, 68x Celeron D CVF, commando, 2x RIVE, Z87X-OC Asus 4870x2, Sapphire 4870 
GraphicsGraphicsGraphicsGraphics
2x 5870, 5850, 5830, 5770 2x 3870x2, 3870 GTX Titan, GTX 480, GTX 590 GTX 285, GTX 260, 4x 9800GT, 8800GTX 
RAMHard DriveCoolingCooling
4x4GB vengeance, 2x4GB predatorX, 2x1GB OCZ DDR2 Intel X25-M 80GB LD PC-V2 SS Phase Change OCN Marksman 
CoolingCoolingOSPower
2x old tek slims (GPU) Various watercooling stuff win7, winxp AX1200 
Case
test bench / cardboard box 
  hide details  
Reply
post #156 of 203
Quote:
Originally Posted by sugarhell View Post

Because i saw your previous post when you upload your settings. I think its so boring that you always try this. We all know that they are pr graphs. First we know that follow scenario is completely random. Second they use 14.2 drivers for thief and 55~ ish fps for 290x max settings is really slow. Probably they used quiet bios. And then look the Y-axis. Do you remind you amd pr graphs? Oh yes. And then the question why now? Dx11 is from 2008. Now they find out the new super API optimizations? They only thing that they can do is bypass some dx11 fuctions for their own. Nothing else.

And if dx11 can rival mantle why microsoft even care to release dx12? A completely new API? And nvidia demonstrate these new drivers with a 3960x and a 780ti.They can do the same with low end cpus? Because we all know low level API helps more low end cpus.
Almost as boring as you constantly defending anything AMD related like it was your first born child
ivybeech
(21 items)
 
4K-HTPC
(12 items)
 
 
CPUMotherboardGraphicsRAM
Intel i7 6700k@4.7ghz Gigabyte z170xp-sli Nvidia GTX 1080Ti 16gb Team Xtreem DDR4 @16/18/18/37 3333mhz 
Hard DriveHard DriveHard DriveHard Drive
256gb M4 SSD boot 4TB Seagate - Raid 0 4TB Seagate - Raid 0 2TB Western Digital Green for storage 
Hard DriveHard DriveHard DriveHard Drive
5TB Seagate 3 tb Seagate  2TB Seagate 4TB of various external storage 
CoolingOSMonitorMonitor
Corsair H100 in push/pull Windows 10 pro ASUS ROG Swift PG278Q 27" OC Catleap Q270SE IPS 2560x1440@120hz 
KeyboardPowerCaseMouse
Xarmor mechanical cherry mx-brown Hale90 1000w 80+gold EVGA DG-87 Logitech G502 
CPUMotherboardGraphicsRAM
Intel i5-2400 Intel DP67BG Nvidia GTX 950 8gb Gskill Ripjaws 1600mhz ddr3 
Hard DriveOptical DriveCoolingOS
256gb samsung SSD Asus Blu-ray Drive Model BC-12B1ST/BLK/B/AS Corsair H50 push/pull Windows 7 Professional 
MonitorPowerCase
Vizio m60-c3 - 3840x2160 OCZ ZT 750w Silverstone Grandia gd-05 
  hide details  
Reply
ivybeech
(21 items)
 
4K-HTPC
(12 items)
 
 
CPUMotherboardGraphicsRAM
Intel i7 6700k@4.7ghz Gigabyte z170xp-sli Nvidia GTX 1080Ti 16gb Team Xtreem DDR4 @16/18/18/37 3333mhz 
Hard DriveHard DriveHard DriveHard Drive
256gb M4 SSD boot 4TB Seagate - Raid 0 4TB Seagate - Raid 0 2TB Western Digital Green for storage 
Hard DriveHard DriveHard DriveHard Drive
5TB Seagate 3 tb Seagate  2TB Seagate 4TB of various external storage 
CoolingOSMonitorMonitor
Corsair H100 in push/pull Windows 10 pro ASUS ROG Swift PG278Q 27" OC Catleap Q270SE IPS 2560x1440@120hz 
KeyboardPowerCaseMouse
Xarmor mechanical cherry mx-brown Hale90 1000w 80+gold EVGA DG-87 Logitech G502 
CPUMotherboardGraphicsRAM
Intel i5-2400 Intel DP67BG Nvidia GTX 950 8gb Gskill Ripjaws 1600mhz ddr3 
Hard DriveOptical DriveCoolingOS
256gb samsung SSD Asus Blu-ray Drive Model BC-12B1ST/BLK/B/AS Corsair H50 push/pull Windows 7 Professional 
MonitorPowerCase
Vizio m60-c3 - 3840x2160 OCZ ZT 750w Silverstone Grandia gd-05 
  hide details  
Reply
post #157 of 203
Quote:
Originally Posted by Alatar View Post

Why are we talking about the Nvidia graph? I'm trying to do some non 1st party Nvidia DX11 CPU overhead testing.

Anyway, your run is still completely different. All of my runs have 33-35K batch count. And that's the number you would also get if you ran the same scenario. Instead you have like 8K which is over 4 times lower...

"My" scenario starts out like this:



And that's where the camera stays all the time.
We've been through this already, if the variance between runs is sub 2fps it doesn't affect the conclusion anyway when Nvidia had a 24% advantage in DX11.

Test Duration: 360 Seconds
Total Frames: 8970

Average FPS: 31.92
Average Unit Count: 4024
Maximum Unit Count: 5430
Average Batches/MS: 910.24
Maximum Batches/MS: 1394.20
Average Batch Count: 39456
Maximum Batch Count: 92105

Probably it was a bug
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 #158 of 203
Quote:
Originally Posted by y2kcamaross View Post

Almost as boring as you constantly defending anything AMD related like it was your first born child

So nice of you redface.gif
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 #159 of 203
That is exactly correct.

Both the GPU producers will only release anything with large improvements if they need to so they can keep there profits up.
Neither one will release anything to dramatic as it will make it harder for them to do smaller upgrades later i guess.
post #160 of 203
Quote:
Originally Posted by Alatar View Post

Why are we talking about the Nvidia graph? I'm trying to do some non 1st party Nvidia DX11 CPU overhead testing.

Anyway, your run is still completely different. All of my runs have 33-35K batch count. And that's the number you would also get if you ran the same scenario. Instead you have like 8K which is over 4 times lower...

"My" scenario starts out like this:



And that's where the camera stays all the time.
We've been through this already, if the variance between runs is sub 2fps it doesn't affect the conclusion anyway when Nvidia had a 24% advantage in DX11.

The Titan is suppose to perform better than a 280x it is a way stronger card.
Intel Killer
(14 items)
 
  
CPUMotherboardGraphicsRAM
AMD 8320 4.4ghz Gigabyte 990fx XFX R9 290 (1100/1250) 8GB DDR3 1866 
Hard DriveCoolingOSMonitor
750GB Momentus XT Corsair H80i Windows 8 3x 23" 1080p 
KeyboardPower
Logitech Mk 710 OCZ 1000 watts 
  hide details  
Reply
Intel Killer
(14 items)
 
  
CPUMotherboardGraphicsRAM
AMD 8320 4.4ghz Gigabyte 990fx XFX R9 290 (1100/1250) 8GB DDR3 1866 
Hard DriveCoolingOSMonitor
750GB Momentus XT Corsair H80i Windows 8 3x 23" 1080p 
KeyboardPower
Logitech Mk 710 OCZ 1000 watts 
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Hardware News
Overclock.net › Forums › Industry News › Hardware News › [pcper] NVIDIA Talks DX12, DX11 Efficiency Improvements