Overclock.net › Forums › Industry News › Software News › [Forbes] DirectX 12 Delivers: AMD, Nvidia, And Intel Hardware Tested With Awesome Improvements
New Posts  All Forums:Forum Nav:

[Forbes] DirectX 12 Delivers: AMD, Nvidia, And Intel Hardware Tested With Awesome Improvements - Page 11

post #101 of 265
Quote:
Originally Posted by Forceman View Post

Quote:
Originally Posted by Artistar View Post

Installed the latest API Overhead Feature test. Am running Windows Tech Preview 10 Build 10041. Tried to use it: didn't recognise DX12, just ran DX11? Any ideas as to this problem?

Drivers?
Quote:
Originally Posted by Fyrwulf View Post

Why the hell would you want Titan in six months? 290x is already equal to Titan in Dx12 and by then we should have the 390x, which should murder any and all comers from nVidia until the latter jumps onto the HBM bandwagon.

You don't honestly believe that this benchmark means that the 290X will match the Titan X in actual games, do you? It's an API test, not a GPU test.
Quote:
Originally Posted by omari79 View Post

I still don't get that..should't DX12 boost the performance of Titian X too? how come 290X = Titan X in that scenario?

People mistakenly believe this API means something about overall performance, while conveniently ignoring things like a 750Ti murdering a 290X in the DX11 test.
The API does make a difference because it will more efficiently use stuff like the CPU and ram along with the GPU itself DX12 / Mantle allow for much more objects and such to be implimented in games than DX11 also DX12 / Mantle offer proper multi threaded stuff like multi threaded rendering and much more reducing CPU limitations and such.

The API is nearly always the limiting factor in most games once the API is properly low level this gives DEV's ability to harness much better performance / efficiency out of hardware such as CPU / ram / GPU making performance increase across the board.

The issue currently is DX11 has way to many hoops for DEV's to hop thru and way to many limitations which DX12 / Mantle unlock for DEV's to harness much better efficiency / performance.

look at current Mantle VS DX11 and you will see what I am saying Mantle unlocks massive perofrmance and the implimentation of Mantle currently is literally in its infancy and DX12 is to take over the Mantle crown and will get full proper implimentation.

I can confirm AMD did directly work with microsoft to build DX12 I already went thru all this in other threads.
Edited by rickcooperjr - 3/28/15 at 7:34am
My System
(15 items)
 
  
CPUMotherboardGraphicsRAM
amd fx 9590 5.1ghz x8 asus sabertooth 990fx r2.0 Radeon R9 290X matrix platinum x3 16gb of ddr3 corsair dominator platinum1866mhz ... 
Hard DriveOptical DriveCoolingOS
2 x ocz vector 256gb raid0 128kb stripe (ocz ve... ASUS DVD rw  liquid cooled cpu and gpu's windows 7 64bit 
MonitorKeyboardPowerCase
3 x 32lcd/led 2560x1600 wolfking timberwolf 2x pc power & cooling turbo 1200w coolermaster haf 932 amd edition 
MouseMouse PadOther
steel series wow cataclysm 14button programable  wolfking sniper killer 2100 nic 
  hide details  
Reply
My System
(15 items)
 
  
CPUMotherboardGraphicsRAM
amd fx 9590 5.1ghz x8 asus sabertooth 990fx r2.0 Radeon R9 290X matrix platinum x3 16gb of ddr3 corsair dominator platinum1866mhz ... 
Hard DriveOptical DriveCoolingOS
2 x ocz vector 256gb raid0 128kb stripe (ocz ve... ASUS DVD rw  liquid cooled cpu and gpu's windows 7 64bit 
MonitorKeyboardPowerCase
3 x 32lcd/led 2560x1600 wolfking timberwolf 2x pc power & cooling turbo 1200w coolermaster haf 932 amd edition 
MouseMouse PadOther
steel series wow cataclysm 14button programable  wolfking sniper killer 2100 nic 
  hide details  
Reply
post #102 of 265
Quote:
Originally Posted by rickcooperjr View Post

The API does make a difference because it will more efficiently use stuff like the CPU and ram along with the GPU itself DX12 / Mantle allow for much more objects and such to be implimented in games than DX11 also DX12 / Mantle offer proper multi threaded stuff like multi threaded rendering and much more reducing CPU limitations and such.

The API is nearly always the limiting factor in most games once the API is properly low level this gives DEV's ability to harness much better performance / efficiency out of hardware such as CPU / ram / GPU making performance increase across the board.

The issue currently is DX11 has way to many hoops for DEV's to hop thru and way to many limitations which DX12 / Mantle unlock for DEV's to harness much better efficiency / performance.

look at current Mantle VS DX11 and you will see what I am saying Mantle unlocks massive perofrmance and the implimentation of Mantle currently is literally in its infancy and DX12 is to take over the Mantle crown and will get full proper implimentation.

I can confirm AMD did directly work with microsoft to build DX12 I already went thru all this in other threads.

ok..so back to my original question i guess..why is the 290X as fast as the TitanX if both are using the same DX12 API?
Underdog
(17 items)
 
  
CPUMotherboardGraphicsRAM
Intel C2Q6600 G0 @ 3.40Ghz 8 x 425Mhz 1.26v Gigabyte GA-EP45-UD3R MSI N460GTX Cyclone 1GD5/OC @ 850Mhz/4200Mhz 1.08v Kingston 4x2GB PC2-6400 5-5-5-15 1.90v 1:1 
Hard DriveHard DriveOptical DriveCooling
Samsung Spinpoint 250GB  WesternDigital Blue 160GB Samsung SATA DVD-RW +/- Cooler Master Hyper T4  
OSOSMonitorKeyboard
Windows 7 Ultimate 64Bit Windows 8 Pro 64Bit ASUS VS228-D A4TECH GR-152 
PowerCaseMouseMouse Pad
Corsair VS450W Everest 707BR A4TECH G9-730FX Black 
Audio
Realtek ALC889A 
  hide details  
Reply
Underdog
(17 items)
 
  
CPUMotherboardGraphicsRAM
Intel C2Q6600 G0 @ 3.40Ghz 8 x 425Mhz 1.26v Gigabyte GA-EP45-UD3R MSI N460GTX Cyclone 1GD5/OC @ 850Mhz/4200Mhz 1.08v Kingston 4x2GB PC2-6400 5-5-5-15 1.90v 1:1 
Hard DriveHard DriveOptical DriveCooling
Samsung Spinpoint 250GB  WesternDigital Blue 160GB Samsung SATA DVD-RW +/- Cooler Master Hyper T4  
OSOSMonitorKeyboard
Windows 7 Ultimate 64Bit Windows 8 Pro 64Bit ASUS VS228-D A4TECH GR-152 
PowerCaseMouseMouse Pad
Corsair VS450W Everest 707BR A4TECH G9-730FX Black 
Audio
Realtek ALC889A 
  hide details  
Reply
post #103 of 265
Quote:
Originally Posted by omari79 View Post

ok..so back to my original question i guess..why is the 290X as fast as the TitanX if both are using the same DX12 API?

My guess would be 512-bit memory bus vs 384-bit memory bus,and also maybe AMD being slightly ahead of the curve in implementing the new API's.
RCPC#1
(17 items)
 
Professional
(13 items)
 
 
CPUMotherboardGraphicsRAM
AMD Phenom II X6 960T Asus M4A88T-VEVO Asus Strix R7 370 SuperTalent Perfomance 
RAMHard DriveHard DriveOptical Drive
GSkill Snipers Monster Daytona Seagate Barracuda 500GB 7,200 RPM 16Mb cache Memorex DVD/RW 
CoolingOSMonitorKeyboard
Corsair H60 Windows 8N IBM 9494 19" LCD IBM 
PowerCaseMouseMouse Pad
Corsair GS500 In Win H-Frame Wolfking OCZ Behemoth 
Audio
JBL Creature 
CPUMotherboardGraphicsRAM
Phenom II X6 1100t MSI 890FX GD65 MSI Radeon HD5670 GSkill RipjawsX DDR3 PC3 12800 2x4GB CL8 
Hard DriveOptical DriveCoolingOS
WD Black 1TB SATA III Samsung BD Zalman 9900MAX Windows 7 64 Professional 
MonitorKeyboardPowerCase
AOC 22" LED Logitech Kingwin Lazer Platinum 500w Fractal Design R3 
Other
Samsung 470 SSD 128GB 
  hide details  
Reply
RCPC#1
(17 items)
 
Professional
(13 items)
 
 
CPUMotherboardGraphicsRAM
AMD Phenom II X6 960T Asus M4A88T-VEVO Asus Strix R7 370 SuperTalent Perfomance 
RAMHard DriveHard DriveOptical Drive
GSkill Snipers Monster Daytona Seagate Barracuda 500GB 7,200 RPM 16Mb cache Memorex DVD/RW 
CoolingOSMonitorKeyboard
Corsair H60 Windows 8N IBM 9494 19" LCD IBM 
PowerCaseMouseMouse Pad
Corsair GS500 In Win H-Frame Wolfking OCZ Behemoth 
Audio
JBL Creature 
CPUMotherboardGraphicsRAM
Phenom II X6 1100t MSI 890FX GD65 MSI Radeon HD5670 GSkill RipjawsX DDR3 PC3 12800 2x4GB CL8 
Hard DriveOptical DriveCoolingOS
WD Black 1TB SATA III Samsung BD Zalman 9900MAX Windows 7 64 Professional 
MonitorKeyboardPowerCase
AOC 22" LED Logitech Kingwin Lazer Platinum 500w Fractal Design R3 
Other
Samsung 470 SSD 128GB 
  hide details  
Reply
post #104 of 265
This bench does not touch shader performance where Titan X is better.
Mastodon
(11 items)
 
HP Z220
(8 items)
 
 
CPUMotherboardGraphicsRAM
FX 8320 M5A99FX Pro 2.0 Asus R9 290X DirectCU II Crucial Ballistix 
Hard DriveCoolingOSMonitor
Pny SSD 240GB CM Nepton 280L Win 8.1 Acer Predator XG270HU Freesync 
PowerCaseMouse
XFX 750W Pro HAF XM Logitech G300 
CPUMotherboardGraphicsCooling
i7 3770 HP Quadro K2000 HP 
OSPowerCaseMouse
Win 7  HP 400W HP CMT RAT 7 
  hide details  
Reply
Mastodon
(11 items)
 
HP Z220
(8 items)
 
 
CPUMotherboardGraphicsRAM
FX 8320 M5A99FX Pro 2.0 Asus R9 290X DirectCU II Crucial Ballistix 
Hard DriveCoolingOSMonitor
Pny SSD 240GB CM Nepton 280L Win 8.1 Acer Predator XG270HU Freesync 
PowerCaseMouse
XFX 750W Pro HAF XM Logitech G300 
CPUMotherboardGraphicsCooling
i7 3770 HP Quadro K2000 HP 
OSPowerCaseMouse
Win 7  HP 400W HP CMT RAT 7 
  hide details  
Reply
post #105 of 265
Quote:
Originally Posted by Redwoodz View Post

My guess would be 512-bit memory bus vs 384-bit memory bus,and also maybe AMD being slightly ahead of the curve in implementing the new API's.

I thought that as well, however do you think it's possible the architecture fairs better on the new API just based off of GCN design?
post #106 of 265
Quote:
Originally Posted by omari79 View Post

so what does this API test represent anyway and why should we take it seriously?

It represents the gains DX12 and Mantle make in reducing CPU overhead and dependency, which will translate into increased performance, just not in the scale this test shows. Mantle scores 10x DX11 in this rest, but real world Mantle gains are obviously much smaller. So it shouldn't be taken seriously, or at least not as serious as some people want to believe. At this point it probably shows more about the relative state of the drivers than anything else.
Quote:
Originally Posted by rickcooperjr View Post

The API does make a difference because it will more efficiently use stuff like the CPU and ram along with the GPU itself DX12 / Mantle allow for much more objects and such to be implimented in games than DX11 also DX12 / Mantle offer proper multi threaded stuff like multi threaded rendering and much more reducing CPU limitations and such.

Yes, the API certainly matters. I accidentally deleted the word test from my post when I was editing it, I meant the "API test" means little.
post #107 of 265
Quote:
Originally Posted by omari79 View Post

ok..so back to my original question i guess..why is the 290X as fast as the TitanX if both are using the same DX12 API?
Because AMD Worked with Microsoft to do their part of DX12,thing that nvidia didnt do aswell because was improving an old and limited standard
post #108 of 265
Quote:
Originally Posted by Forceman View Post

It represents the gains DX12 and Mantle make in reducing CPU overhead and dependency, which will translate into increased performance, just not in the scale this test shows. Mantle scores 10x DX11 in this rest, but real world Mantle gains are obviously much smaller. So it shouldn't be taken seriously, or at least not as serious as some people want to believe. At this point it probably shows more about the relative state of the drivers than anything else.
Yes, the API certainly matters. I accidentally deleted the word test from my post when I was editing it, I meant the "API test" means little.
No game has been written according to specific Mantle hlsl shaders, they have been the same as dx11 calls afaik. Cannot provide reference, though.
The Machine
(14 items)
 
Nexus 7 2013
(11 items)
 
 
CPUMotherboardGraphicsRAM
A10 6800K Asus F2A85-V Msi 6870 Hawx, Vtx3d 5770, Sapphire 4890oc(RIP) 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, Sapphire 4890oc(RIP) 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 #109 of 265
Why all tests focus on gpu in a cpu overhead test? Wheres the in depth reviews on cpu benches using cheap and old cpus? If this is all its made out to be can an old cpu or i3/athlon drive the best cards?
post #110 of 265
Quote:
Originally Posted by delboy67 View Post

Why all tests focus on gpu in a cpu overhead test? Wheres the in depth reviews on cpu benches using cheap and old cpus? If this is all its made out to be can an old cpu or i3/athlon drive the best cards?

The other benchmark did test a 7850k which is the closest to what you described.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Software News
Overclock.net › Forums › Industry News › Software News › [Forbes] DirectX 12 Delivers: AMD, Nvidia, And Intel Hardware Tested With Awesome Improvements