Overclock.net › Forums › Graphics Cards › AMD/ATI › AMD MANTLE Discussion Thread
New Posts  All Forums:Forum Nav:

AMD MANTLE Discussion Thread - Page 113

post #1121 of 2769
i knew something like this would happen.
post #1122 of 2769
Unbelievable. Another month and no Mantle as promised.

I don't even really care about Mantle so much as XFire frame pacing and Vsync fixes. These 290's have a lot of potential, but without the proper driver, it's discouraging.
post #1123 of 2769
Quote:
Originally Posted by Obi-Chad Kenobi View Post

Unbelievable. Another month and no Mantle as promised.

I don't even really care about Mantle so much as XFire frame pacing and Vsync fixes. These 290's have a lot of potential, but without the proper driver, it's discouraging.


I could not agree with you more the way current drivers handle the 290 / 290X is very pitifull and we drastically need a driver fix / update just to get them to work properly especially in crossfire little alone Manlte support at sec we get about 80% of our DirectX potential out of these cards because of the driver issues we are dealing with now. I truly hope Mantle will bring what they say it will and with what little I got to play with it it did quite a bit of a performance increase so hopefully they fix both our issues on directX side and give us good Mantle support in the next driver release for our 290 / 290x setups.
Edited by rickcooperjr - 1/31/14 at 11:06am
My System
(15 items)
 
  
CPUMotherboardGraphicsRAM
Intel I7 5820k 4.6ghz core 4.3ghz CPU cache 1.345v asus sabertooth X99 sold R9 290x's am currently running a R9 280A X... 16gb of ddr4 corsair vengeance LPX 3200mhz 1t 1... 
Hard DriveOptical DriveCoolingOS
M.2 Toshiba OCZ RD400 512gb ASUS DVD rw  liquid cooled cpu and gpu's windows 10 pro 64bit 
MonitorKeyboardPowerCase
3 x 32lcd/led 2560x1600 wolfking timberwolf EVGA supernova 1300w G2  Thermaltake core X9 
MouseMouse PadOther
steel series wow cataclysm 14button programable  wolfking sniper Intel PRO PT dual port PCIe server NIC 
  hide details  
Reply
My System
(15 items)
 
  
CPUMotherboardGraphicsRAM
Intel I7 5820k 4.6ghz core 4.3ghz CPU cache 1.345v asus sabertooth X99 sold R9 290x's am currently running a R9 280A X... 16gb of ddr4 corsair vengeance LPX 3200mhz 1t 1... 
Hard DriveOptical DriveCoolingOS
M.2 Toshiba OCZ RD400 512gb ASUS DVD rw  liquid cooled cpu and gpu's windows 10 pro 64bit 
MonitorKeyboardPowerCase
3 x 32lcd/led 2560x1600 wolfking timberwolf EVGA supernova 1300w G2  Thermaltake core X9 
MouseMouse PadOther
steel series wow cataclysm 14button programable  wolfking sniper Intel PRO PT dual port PCIe server NIC 
  hide details  
Reply
post #1124 of 2769
@CatalystMaker on Twitter a moment ago...
Quote:
We are currently hard at work trying to resolve an install issue with Catalyst 14.1 beta. Based on this I am guessing it wont be ready today
post #1125 of 2769
Quote:
Originally Posted by Civrock View Post

@CatalystMaker on Twitter a moment ago...

Jesus Christ...AMD are a bunch of teases!
post #1126 of 2769
Quote:
Originally Posted by Civrock View Post

@CatalystMaker on Twitter a moment ago...

Sounds like an excuse which is covering for a different problem.

Oh boy....
Asus ROG G751JY
(19 items)
 
  
CPUMotherboardGraphicsRAM
Intel i7 4710HQ @ 3.4 Intel H87 GTX 980m @ 1262/5410 w/PREMA BIOS v1.1.1 + Dre... 24GB 1600mhz 
Hard DriveHard DriveHard DriveOptical Drive
Samsung m.2 PCIE SSD OCZ Vertex 3 Samsung Evo 850 Bluray reader/writer 
CoolingOSMonitorKeyboard
Thermaltake Massive 14² Windows 10 x64 LG-D046C IPS @ 100hz  Corsair K70 Mechanical Gaming Keyboard 
PowerCaseMouseMouse Pad
230W Power Brick Asus ROG G751JY Logitech G502 Proteus Core RGB + G500 for travel HyperX smooth surface + Corsair hard surface fo... 
AudioAudioAudio
Sound Blaster Omni 5.1 USB w/dual array mic Ultrasone Pro 750 Sennheiser HD598 Special Edition 
  hide details  
Reply
Asus ROG G751JY
(19 items)
 
  
CPUMotherboardGraphicsRAM
Intel i7 4710HQ @ 3.4 Intel H87 GTX 980m @ 1262/5410 w/PREMA BIOS v1.1.1 + Dre... 24GB 1600mhz 
Hard DriveHard DriveHard DriveOptical Drive
Samsung m.2 PCIE SSD OCZ Vertex 3 Samsung Evo 850 Bluray reader/writer 
CoolingOSMonitorKeyboard
Thermaltake Massive 14² Windows 10 x64 LG-D046C IPS @ 100hz  Corsair K70 Mechanical Gaming Keyboard 
PowerCaseMouseMouse Pad
230W Power Brick Asus ROG G751JY Logitech G502 Proteus Core RGB + G500 for travel HyperX smooth surface + Corsair hard surface fo... 
AudioAudioAudio
Sound Blaster Omni 5.1 USB w/dual array mic Ultrasone Pro 750 Sennheiser HD598 Special Edition 
  hide details  
Reply
post #1127 of 2769
Quote:
Originally Posted by Obi-Chad Kenobi View Post

Unbelievable. Another month and no Mantle as promised.

I don't even really care about Mantle so much as XFire frame pacing and Vsync fixes. These 290's have a lot of potential, but without the proper driver, it's discouraging.
For the Hawaii cards their frame pacing is pretty much up to the task so I don't know where you get this. Really we had Kyle claim in various comparisons that CF 290/290X gives a better experience than the SLI equivalent. Frame times confirm this no more major spikes on of the things however is that you would want to be running PCI-e 3.0 x8 of PCI-E 2.0 x16 speeds.

I ran CF 290 for a week or so very smooth but yeah I didn't get a BF4 copy as promised so I was mad and returned them. (and now I am using dual 5870's for which I would indeed like better frame pacing)
Quote:
Originally Posted by rickcooperjr View Post

I could not agree with you more the way current drivers handle the 290 / 290X is very pitifull and we drastically need a driver fix / update just to get them to work properly especially in crossfire little alone Manlte support at sec we get about 80% of our DirectX potential out of these cards because of the driver issues we are dealing with now. I truly hope Mantle will bring what they say it will and with what little I got to play with it it did quite a bit of a performance increase so hopefully they fix both our issues on directX side and give us good Mantle support in the next driver release for our 290 / 290x setups.
See the above post your statement has no place in reality
http://www.anandtech.com/show/7457/the-radeon-r9-290x-review/4

Anyhow for all non GCN 1.1 cards I completly agree with you they can use more work also the DX9 and +1600P patch is very welcome!
post #1128 of 2769
Quote:
Originally Posted by Jedson3614 View Post

I run nvidia has anyone tried it yet ?
Quote:
Originally Posted by skupples View Post

Warning: Oxide demo on 3930k & titans (Click to show)
We should conform to a standard when running the Oxide Demo...

I ran the oxide test 4 times. With the settings mentioned above by SugarHell. Which is RTS:MB on & RTS: MB off. HT on/off. While the results are almost the same, the graphs from MSi-AB are pretty interesting. I'm also including fraps min/max dumps for a better idea.

All tests ran with a 4.8ghz 3930k, & tri-sli titans(gpu seems to not matter one bit)
Warning: Test #1 - RTS: MB OFF - HT OFF.! (Click to show)
Code:
Frames, Time (ms), Min, Max, Avg
  4984,    118982,  29, 135, 41.889
Code:
== Configuration ==========================================
API:                            DirectX
Scenario:                       ScenarioRTS.csv
User Input:                     Disabled
Resolution:                     1920x1080
Fullscreen:                     True
GameCore Update:                16.6 ms
Bloom Quality:                  High
PointLight Quality:             High
ToneCurve Quality:              High
Glare Overdraw:                 16
Shading Samples:                64
Shade Quality:                  Mid
Motion Blur Frame Time:         16
Motion Blur InterFrame Time:    2
Detailed Frame Info:            C:\Users\THE BEAST\Documents\Star Swarm\FrameDump_14_01_30_2347.csv
===========================================================


== Results ================================================
Test Duration:                  120 Seconds
Total Frames:                   5103

Average FPS:                    42.52
Average Unit Count:             3486
Maximum Unit Count:             5445
Average Batches/MS:             840.13
Maximum Batches/MS:             1397.63
Average Batch Count:            19026
Maximum Batch Count:            36637
===========================================================


Ok, so as you can see here. GPU usage is low, & CPU usage is also kinda low, fluctuating greatly, rarely hitting 99%, running on all 6 cores.

Warning: TEST #2 - RTS : MB ON - HT OFF (Click to show)
Code:
Frames, Time (ms), Min, Max, Avg
  1607,    119091,   6,  63, 13.494
Code:
== Configuration ==========================================
API:                            DirectX
Scenario:                       ScenarioRTS.csv
User Input:                     Disabled
Resolution:                     1920x1080
Fullscreen:                     True
GameCore Update:                16.6 ms
Bloom Quality:                  High
PointLight Quality:             High
ToneCurve Quality:              High
Glare Overdraw:                 16
Shading Samples:                64
Shade Quality:                  Mid
Motion Blur Frame Time:         16
Motion Blur InterFrame Time:    2
Detailed Frame Info:            C:\Users\THE BEAST\Documents\Star Swarm\FrameDump_14_01_30_2356.csv
===========================================================


== Results ================================================
Test Duration:                  120 Seconds
Total Frames:                   1727

Average FPS:                    14.38
Average Unit Count:             3401
Maximum Unit Count:             5542
Average Batches/MS:             1336.74
Maximum Batches/MS:             1682.49
Average Batch Count:            91791
Maximum Batch Count:            175239
===========================================================



Usage is pretty much the same, results with the special motion blur ON are staggeringly low.

Warning: TEST #3 - RTS: MB OFF - HT ON (Click to show)
Code:
Frames, Time (ms), Min, Max, Avg
  4686,    119372,  28, 133, 39.255
Code:
== Configuration ==========================================
API:                            DirectX
Scenario:                       ScenarioRTS.csv
User Input:                     Disabled
Resolution:                     1920x1080
Fullscreen:                     True
GameCore Update:                16.6 ms
Bloom Quality:                  High
PointLight Quality:             High
ToneCurve Quality:              High
Glare Overdraw:                 16
Shading Samples:                64
Shade Quality:                  Mid
Motion Blur Frame Time:         16
Motion Blur InterFrame Time:    2
Detailed Frame Info:            C:\Users\THE BEAST\Documents\Star Swarm\FrameDump_14_01_31_0011.csv
===========================================================


== Results ================================================
Test Duration:                  120 Seconds
Total Frames:                   4801

Average FPS:                    40.01
Average Unit Count:             3892
Maximum Unit Count:             5516
Average Batches/MS:             871.94
Maximum Batches/MS:             1397.60
Average Batch Count:            20868
Maximum Batch Count:            34877
===========================================================




So, as you can see the min/max are within margin of error when compared to test #1. The interesting thing here is that the bench is now only using three cores, & with much lower usage. I unfortunately forgot to flip on the threads in the hardware monitor, maybe some one with better understanding can explain why the engine would work this way, & the possible benefit. so sorry about that. Major derp. I will run them again tomorrow when it's not 1:00 AM. It may have been placebo effect, but I swear the visuals looked different with HT on. While it's a benchmark, it's not 100% canned. The runs were different looking every time. They start the same, but that's about it.

Warning: TEST #4 - RTS: MB ON - HT ON (Click to show)
Code:
Frames, Time (ms), Min, Max, Avg
  1676,    119746,   7,  85, 13.996
Code:
== Configuration ==========================================
API:                            DirectX
Scenario:                       ScenarioRTS.csv
User Input:                     Disabled
Resolution:                     1920x1080
Fullscreen:                     True
GameCore Update:                16.6 ms
Bloom Quality:                  High
PointLight Quality:             High
ToneCurve Quality:              High
Glare Overdraw:                 16
Shading Samples:                64
Shade Quality:                  Mid
Motion Blur Frame Time:         16
Motion Blur InterFrame Time:    2
Detailed Frame Info:            C:\Users\THE BEAST\Documents\Star Swarm\FrameDump_14_01_31_0016.csv
===========================================================


== Results ================================================
Test Duration:                  120 Seconds
Total Frames:                   1734

Average FPS:                    14.44
Average Unit Count:             3394
Maximum Unit Count:             5613
Average Batches/MS:             1327.49
Maximum Batches/MS:             1659.81
Average Batch Count:            91196
Maximum Batch Count:            171986
===========================================================




low GPU usage, & even lower CPU usage than running it with MB off, yet FPS is once again through the floor.

So, running these tests have left me with even more questions. Maybe some one can explain. Low FPS + Low GPU usage + low CPU usage = ???... Is this the limitations of d3d in the works? Iv'e never seen anything like this before. One thing is for sure, that motion blur is a serious ass kicker.

Also, before the "waaahhh qq off topic, go home nv fanboy" posts rain from the sky... I did this so we could have a direct comparison for before/after, & to also compare to the results shown above.
IT'S WHITE
(21 items)
 
 
CPUMotherboardGraphicsGraphics
4930k Rampage iv extreme Black Edition GTX Titan  GTX Titan 
GraphicsRAMHard DriveHard Drive
GTX Titan GskillZ Trident X Samsung EVO samsung 840 evo 
Optical DriveCoolingCoolingOS
Asus External Blue Ray drive 4x 480x60mm Swiftech MCP35x2 Win 8.1.2 
MonitorMonitorMonitorPower
AOC AOC AOC EVGA G2 1300w 
CaseMouseMouse PadAudio
STH10 RAT7 what is this?  Vali / Modi :(  
Other
Corsair HX850 
  hide details  
Reply
IT'S WHITE
(21 items)
 
 
CPUMotherboardGraphicsGraphics
4930k Rampage iv extreme Black Edition GTX Titan  GTX Titan 
GraphicsRAMHard DriveHard Drive
GTX Titan GskillZ Trident X Samsung EVO samsung 840 evo 
Optical DriveCoolingCoolingOS
Asus External Blue Ray drive 4x 480x60mm Swiftech MCP35x2 Win 8.1.2 
MonitorMonitorMonitorPower
AOC AOC AOC EVGA G2 1300w 
CaseMouseMouse PadAudio
STH10 RAT7 what is this?  Vali / Modi :(  
Other
Corsair HX850 
  hide details  
Reply
post #1129 of 2769
Quote:
Originally Posted by Exostenza View Post

Sounds like an excuse which is covering for a different problem.

Oh boy....

Covering for the 24 hour media exclusivity is more like it. They know they can't back out of it and they know it will piss us off, so easiest solution is to lie.
post #1130 of 2769
Quote:
Originally Posted by maarten12100 View Post

For the Hawaii cards their frame pacing is pretty much up to the task so I don't know where you get this. Really we had Kyle claim in various comparisons that CF 290/290X gives a better experience than the SLI equivalent. Frame times confirm this no more major spikes on of the things however is that you would want to be running PCI-e 3.0 x8 of PCI-E 2.0 x16 speeds.

I ran CF 290 for a week or so very smooth but yeah I didn't get a BF4 copy as promised so I was mad and returned them. (and now I am using dual 5870's for which I would indeed like better frame pacing)
See the above post your statement has no place in reality
http://www.anandtech.com/show/7457/the-radeon-r9-290x-review/4

Anyhow for all non GCN 1.1 cards I completly agree with you they can use more work also the DX9 and +1600P patch is very welcome!

If I enable Frame Pacing or Vsync in 13.12, I get terrible performance, sound crackling, stuttering. If I turn both of these off, these problems go away. We'll see if 14.2 Beta drivers fix this or not.

Running these on a i7 930 (LGA 1366) at 4Ghz in two PCI-e 2.0 x16 slots on 2560 x 1440 @ 60Hz.
Edited by Obi-Chad Kenobi - 1/31/14 at 12:24pm
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: AMD/ATI
Overclock.net › Forums › Graphics Cards › AMD/ATI › AMD MANTLE Discussion Thread