AMD/ATI: (colored are optimized >=1.28 app values, defined by Petrion) HD 7970 ----> 1x~650, 2x~950, 4x~1,800, 5x~2,200 HD 7950 ----> 3x~1860 HD 7950 ----> 1x 1,145
HD 7950 ------> 2x 3,400, 3x 4,500
HD 7870
HD 7850 HD 7770 ----> 1x~1960, 2x~3600
HD 7750 ------> 2x~11,000 HD 5870 ------> 2x~3,105 HD 5850 ------> 1x 1,800, 2x 6,085 HD 5830 ------> 1x 2,916
HD 6970
HD 6950(1536)-> 2x 6700 HD 6950 ------> 2x 3,500
HD 6990
HD 6870
HD 5970 HD 6850 ------> 1x~2,300
HD 6790
HD 5770 ------> 1x 7,750+
HD 6770
HD 5670 ------> 1x 11,100
HD 5570 ------> 1x~15,000
HD 5450 ------> 1x~36,500!
---------------------------
My values, AMD Radeon HD6850 (2048MB), Windows 7 - 64bit Ultimate.
Comparison run over 100 v1.32 tasks, on average ~2,359 seconds.
Compared v 1.32 to v1.28
79 tasks v1.28, on average 2,332 seconds. 79 v1.28s is all I had ;-)
79 tasks v1.32, on average 2,362 seconds.
AMD/ATI: (colored are optimized >=1.28 app values, defined by Petrion) HD 7970 ----> 1x~650, 2x~950, 4x~1,800, 5x~2,200 HD 7950 ----> 3x~1860 HD 7950 ----> 1x 1,145
HD 7870
HD 7850 HD 7770 ----> 1x~1960, 2x~3600
HD 7750 ------> 2x~11,000 HD 5870 ------> 2x~3,105 HD 5850 ------> 1x 1,800, 2x 6,085 HD 5830 ------> 1x 2,916
HD 6970
HD 6950(1536)-> 2x 6700 HD 6950 ------> 2x 3,500
HD 6990
HD 6870
HD 5970 HD 6850 ------> 1x~2,300 HD 6850 ------> 1x~2,359
HD 6790
HD 5770 ------> 1x 7,750+
HD 6770
HD 5670 ------> 1x 11,100
HD 5570 ------> 1x~15,000
HD 5450 ------> 1x~36,500!
I have been doing some testing and tweaking with my new 7970 in Linux. Thanks to astrocrab for suggesting updating the driver to 12.11 beta. The driver is running at least 20% faster compared to the previous driver version for this project.
I see a bit of fluctuation in processing time when running multiple tasks. Perhaps this is due to some kind of throttling from higher GPU temps. When I opened my window temporarily to let freezing cold air in, the processing time appeared to lower a bit.
I ran a PCI-E bandwdith test using an OpenCL application called BufferBandwidth from the AMD SDK.
Here are the results without BRP4 running and the GPU idle (PCI-E 3.0 x16):
The D->H bandwidth fluctuated between 4.36 and 7.18 GB/s while running BRP4 but is generally between 5-6 GB/s. Perhaps the difference gives a rough estimate as to bandwidth usage of BRP4 with the OpenCL application.
Perhaps this is due to some kind of throttling from higher GPU temps.
if so, i'll try to set pc near an open window to see if runtime will change.
btw, what temperate does your gpus report? aticonfig --adapter=all, --odgt
some update:HD 7770 ---->
)
some update:
HD 7770 ----> 1x~1960, 2x~3600
not too bad
AMD/ATI: (colored are
)
AMD/ATI: (colored are optimized >=1.28 app values, defined by Petrion)
HD 7970 ----> 1x~650, 2x~950, 4x~1,800, 5x~2,200
HD 7950 ----> 3x~1860
HD 7950 ----> 1x 1,145
HD 7950 ------> 2x 3,400, 3x 4,500
HD 7870
HD 7850
HD 7770 ----> 1x~1960, 2x~3600
HD 7750 ------> 2x~11,000
HD 5870 ------> 2x~3,105
HD 5850 ------> 1x 1,800, 2x 6,085
HD 5830 ------> 1x 2,916
HD 6970
HD 6950(1536)-> 2x 6700
HD 6950 ------> 2x 3,500
HD 6990
HD 6870
HD 5970
HD 6850 ------> 1x~2,300
HD 6790
HD 5770 ------> 1x 7,750+
HD 6770
HD 5670 ------> 1x 11,100
HD 5570 ------> 1x~15,000
HD 5450 ------> 1x~36,500!
AMD A8 3870 -> 1x 6,489
NVIDIA: (colored are optimized >=1.28 app values, defined by Petrion)
GTX 690
GTX 590
GTX 680 ------> 1x~750
GTX 680 ------> 3x 3,100(Win7)
GTX 680 -----> 2x 1,945(Linux)
GTX 580 ------> 1x 834, 3x~2,500
GTX 580 ------> 3x 3,350(Windows)
GTX 580 -----> 3x 3,050(Linux)
GTX 670 ------> 3x~4,300(vista)
GTX 660Ti ----> 1x~1,180, 2x~2,170
GTX 660Ti ----> 1x~1,700, 2x~2,900, 3x~4,500, 4x~6,030, 5x~8,660, 6x~12,760
gtx650 ----> 1x2630 sec, 2x4340 sec
GTX 570
GTX 670
GTX 480 ------> 2x~2,200
GTX 470 ------> 2x~3,000, 3x 3,800
GTX 560 [448] -> 1x 1,550, 2x 2,500
gtx 560 TI ----> 2x2030
GTX 560 Ti ----> 1x~1,100, 2x 2,654, 6x 6,400
GTX 560 Ti ----> 1x~1,100, 2x 2,000, 4x 4,100, 5x 5,200
GTX 560 Ti ---> 1x 1,583 (OC'd)
GTX 560 ------> 2x 2,300
GTX 560 ------> 1x 3,300, 2x 4800
GTX 460 -> 1x3000, 2x4800
GTX 465
GTX 460 SE
GTX 550 Ti ---> 1x 1,793, 2x 2,961
GTX 550 Ti ---> 1x 3,065, 2x 5,600
GT 640 -------> 1x~5,700
GT 440
GTS 450 ----> 1x~2,200, 2x 4,200
GF 610M ------> 1x~7,800
GT 430 -------> 2x 9,100
GT 430 -------> 1* 4860
GT 520 -------> 1x~9,600(Linux)
FirePro V4800-> 1x 10,620
Older cards (not openCL v1.1 capable) but still interesting comparison:
GT 295 -------> 1x 2,000(Linux)
GTX 285 ----> 2*3000
GTX 260 ----> 1*2200
8800GT G92 ---> 1x 2,940(Linux)
8800GT G92 ---> 1x 3,600(Linux)
8800GTS G80 --> 1x 4,020(Linux)
GTS 250 ------> 2x~5,484
GT 240 ------> 1x 4,035(OC'd)
GT 240 -------> 1x~4,500
GT 240 ----> 1x~5,400, 2x 10,500
GT 220 -------> 2x 19,400[/b]
DSKAG Austria Research Team: [LINK]http://www.research.dskag.at[/LINK]
Great job, THX!
)
Great job, THX!
One more from the vintage
)
One more from the vintage card department, setting the record straight for the GT 240:
GT 240 ----> 1x~3460 (Linux)
e.g. http://einsteinathome.org/task/319290633
And
GTX 650 Ti ----> 3x ~ 5900 (Linux ,PCIe 2)
http://einsteinathome.org/task/318765183
HBE
RE: AMD/ATI: (colored are
)
New round, new applications, new values.
Linux:
v1.31 (BRP4cuda32nv270)
v1.31 (opencl-ati)
Mac OSX:
v1.31 (BRP4cuda32OSX)
v1.31 (opencl-ati-lion)
Windows:
v1.32 (BRP4cuda32)
v1.32 (BRP4cuda32nv301)
v1.32 (opencl-ati)
---------------------------
My values, AMD Radeon HD6850 (2048MB), Windows 7 - 64bit Ultimate.
Comparison run over 100 v1.32 tasks, on average ~2,359 seconds.
Compared v 1.32 to v1.28
79 tasks v1.28, on average 2,332 seconds. 79 v1.28s is all I had ;-)
79 tasks v1.32, on average 2,362 seconds.
RE: AMD/ATI: (colored are
)
DSKAG Austria Research Team: [LINK]http://www.research.dskag.at[/LINK]
I have been doing some
)
I have been doing some testing and tweaking with my new 7970 in Linux. Thanks to astrocrab for suggesting updating the driver to 12.11 beta. The driver is running at least 20% faster compared to the previous driver version for this project.
BRP4 v1.31 64-bit
Linux: 1x 7970 @ PCI-E 3.0 x16 - 1x~623 , 2x 927-969 , 4x 1691-1792 (avg 1758)
I see a bit of fluctuation in processing time when running multiple tasks. Perhaps this is due to some kind of throttling from higher GPU temps. When I opened my window temporarily to let freezing cold air in, the processing time appeared to lower a bit.
I ran a PCI-E bandwdith test using an OpenCL application called BufferBandwidth from the AMD SDK.
Here are the results without BRP4 running and the GPU idle (PCI-E 3.0 x16):
Host->Device: 13,189.0 MB/s
Device->Host: 12,267.5 MB/s
With BRP4 running:
Host->Device: 10,537.0 MB/s
Device->Host: 4,464.6 MB/s
The D->H bandwidth fluctuated between 4.36 and 7.18 GB/s while running BRP4 but is generally between 5-6 GB/s. Perhaps the difference gives a rough estimate as to bandwidth usage of BRP4 with the OpenCL application.
RE: Perhaps this is due to
)
if so, i'll try to set pc near an open window to see if runtime will change.
btw, what temperate does your gpus report? aticonfig --adapter=all, --odgt
first 7970@54C second
)
first 7970@54C
second @52C
no performance change
I will insert all values
)
I will insert all values after some posts with new values, because i want prevent this thread to be too long because scrolling ;)
DSKAG Austria Research Team: [LINK]http://www.research.dskag.at[/LINK]