jm_1990 Posted December 10, 2011 Share Posted December 10, 2011 Thanks dude. Maybe the guy who get the 420 fps must be using the new 560 Ti version with more CUDA cores. If it matters I run LuxMark too. This are the results for GPU only and CPU+GPU: The guy who got 8 GeForce GTX 580 is insane! I thought you can only get two card by SLI and four GPU if the card is Dual GPU. Who manufactures a motherboard with 8 x16 PCI Express slots? And you'll need PCI Express 4.0 to get to work with full speed. I don't know I think is fake. What do you think? Link to comment Share on other sites More sharing options...
mitch_de Posted December 11, 2011 Author Share Posted December 11, 2011 Yep, OpenCL can, if so coded use more than one GPU. But depends on the code it uses much more cpu load for that. 8 GPUs may not be get full load by today CPU+systembus speed+system pcie speed - bottleneck in this case. But shows how fast gpus can be in the future or used with very special universitary usage (mostly complex simulations). Link to comment Share on other sites More sharing options...
Regi Yassin Posted December 12, 2011 Share Posted December 12, 2011 ASUS P8P67 Deluxe i7 2600k (no OC) Gigabyte 580 SOC MacPro4,1 OpenCL_OceanWave submited, this is the screenshots Link to comment Share on other sites More sharing options...
mitch_de Posted December 12, 2011 Author Share Posted December 12, 2011 Great Luxmark value. (Info: Luxmark is also cpu dependent - faster cpu gives higher (some) luxmark value on same gpu - even on the gpu only bench). Link to comment Share on other sites More sharing options...
Hangten Posted December 16, 2011 Share Posted December 16, 2011 Here's my score, delayed by a month! LOL Link to comment Share on other sites More sharing options...
ujen Posted December 16, 2011 Share Posted December 16, 2011 Heres my run on my new X79 build with a 3930k and a GTX 570 OC: And heres my Luxmark result: Link to comment Share on other sites More sharing options...
Badeendje Posted December 17, 2011 Share Posted December 17, 2011 Shameless Crosspost; HIS 6870 22 fps difference over my old score with an old version of the benchmark, and even more with the latest version with the stock .2 kexts! Ran the benchmark a couple of times since this much difference has not been seen with any update since I started testing. Link to comment Share on other sites More sharing options...
mitch_de Posted December 20, 2011 Author Share Posted December 20, 2011 NEWS: AMD tested new AMD 7970 vs 6970 with Luxmark (win) - 2,3 times faster. AMD needs that, because GTX 580 was much faster in OpenCL. But beside faster HW the complete new Architecture (now more like Fermi) will takes time for OpenCL drivers on non Windows systems. Belongs to Apple if and when it supports (and devleops) for that new Highend GPUs. The new mobile 7xxxx GPUs are rebranded 6xxxM gpus so there is no need for much develops in drivers. Link to comment Share on other sites More sharing options...
DarkUser89 Posted December 21, 2011 Share Posted December 21, 2011 Q9400 OC2,72GHz 4GB Zotac GTX 560 MacPro 3,1 10.7.2 Link to comment Share on other sites More sharing options...
Eiszeit Posted December 22, 2011 Share Posted December 22, 2011 Q6600 HD5450 10.7.2 Link to comment Share on other sites More sharing options...
Shamsu Posted December 28, 2011 Share Posted December 28, 2011 i5-2500 8 GB GTX 560 Ti 10.7.2 OpenCL Oceanwave GPU Benchmark 393.78 FPS Link to comment Share on other sites More sharing options...
bergoulle Posted January 6, 2012 Share Posted January 6, 2012 Q9400 P5K with AMD6850 ...HD6850.tiff Link to comment Share on other sites More sharing options...
mitch_de Posted January 7, 2012 Author Share Posted January 7, 2012 Thanks , i will update the main list (and app) soon but less frequently than in the past, because we already have lots of results collected. Link to comment Share on other sites More sharing options...
Hugo_bee Posted January 8, 2012 Share Posted January 8, 2012 hy i've the gtx 460m.. I can have a semistable qe/ci.. but very slow cinebench so this bench is slow too.. I never see thos hex values in my opencl binary.. I tried to add my devid to nvda100 kext (mine is 10de 0dd1 but.. strange.. I run 1920x1080 without nvda100 ! so what can I do if I have not those hex id in opencl.. ? where the complete guide also for the dylib ? Link to comment Share on other sites More sharing options...
mitch_de Posted January 9, 2012 Author Share Posted January 9, 2012 slow Cinebench tells that also OpenGL is slow. Perhaps some GPU Powermanagement Problems (gpu runs always on slow MHz, even on load) which does effect both: OpenGL + OpenCL performance Check + ask: http://www.insanelym...howtopic=266036 For OpenCL Fermi patching check + ask: http://www.insanelym...pic=266318&st=0 Link to comment Share on other sites More sharing options...
Hugo_bee Posted January 9, 2012 Share Posted January 9, 2012 ok. thanks i'll discover your posts apart from it.. I know I cannot 'force the hand' for this driver because everywhere in the net I see that the 460m gtx got frozen after a while.. but only when opengl is invoked and I tried to replace dp4-->10.7.2.. injecting the 100nvda drv.. injecting the AGPD and tring to replace with some original invidia driver or cuda driver. but in 3 days I got (except for black screens or no good starts) only 1 running benchmarks in opengl and that's ..for me.. a great great goal! (better than anything ok) so .. i'll try yours thanks Link to comment Share on other sites More sharing options...
mitch_de Posted January 12, 2012 Author Share Posted January 12, 2012 NEW OpenCL Luxmark Bench 2.0 Beta (now with submit result + new SCala scene). Very good (comlex gpu computing) OpenCL Benchmark. I made an new thread for your results. You can look after the submitted results (fpr 2.0x Version only) but there are only the Top 20 listed. I got 77 - to slow for the Top 20 Luxmark 2.0 OS X86 Thread: http://www.insanelymac.com/forum/index.php?showtopic=274549 Link to comment Share on other sites More sharing options...
tilllt Posted January 21, 2012 Share Posted January 21, 2012 i have two CUDA / OpenCl capable cards in my machine. How do i specifically run the benchmark on either one of them? Link to comment Share on other sites More sharing options...
mitch_de Posted January 21, 2012 Author Share Posted January 21, 2012 You can not select the gpu used. Only if the benches have some code to select the gpu you would so this. Luxmark for example will use automatic all known OpenCL gpus found - at least it will work if they (all) are from same vendor. EDIT: In Luxmark V2.0B2 you can select the gpu(s) which you want to use or not. Try that. YOu must set LUXMARK (after starting) in pause mode (Menue) and then select the gpus on the right side then use in menue "use selected CPU/GPU) and unpause the benchmark again. Link to comment Share on other sites More sharing options...
Ritz Carltn Posted February 3, 2012 Share Posted February 3, 2012 My results is it good? Or can i do something to improve my fps? Thanks!! How did you configure GTX 550 Ti in 10.7.2? do you share me the way? Link to comment Share on other sites More sharing options...
mitch_de Posted February 18, 2012 Author Share Posted February 18, 2012 Mac OS X 10.8 DP1: My 9600 GT works with OpenCL but shows not correct result. If you also have DP1 for test please check OceanWave with it. (Screenshoot mad with the command line OceanWave, but App shows same result - wrong) Link to comment Share on other sites More sharing options...
stefano.85 Posted February 18, 2012 Share Posted February 18, 2012 Mac OS X 10.8 DP1: My 9600 GT works with OpenCL but shows not correct result. If you also have DP1 for test please check OceanWave with it. (Screenshoot mad with the command line OceanWave, but App shows same result - wrong) Same here Link to comment Share on other sites More sharing options...
diskeeper Posted February 18, 2012 Share Posted February 18, 2012 The GeForceGLDriver "78 E8 83 F8 02 7C 11" hex for Fermi cards is not here anymore in 10.8. Link to comment Share on other sites More sharing options...
mitch_de Posted February 18, 2012 Author Share Posted February 18, 2012 I researched a bit the problem with 10.8 and OceanWave. I recomplied it with Xcode 4.4 and OS X 10.8 Frameworks and removed some optimizations (more GLflush than than before). Now it works again, but results with that version cant be compared with old results because they are little (10% in my case) slower. I will upload an new 10.8 Version (only will run on 10.8!) soon and attatch it at 1. posting. Link to comment Share on other sites More sharing options...
diskeeper Posted February 19, 2012 Share Posted February 19, 2012 The GeForceGLDriver "78 E8 83 F8 02 7C 11" hex for Fermi cards is not here anymore in 10.8. I checked with oclinfo, and says OpenCL works on my GTX460 (without any modifications!). 1 OpenCL platform found! ... 2 OpenCL devices found! [Device 0] Name: Intel(R) Core(TM)2 Quad CPU Q8300 @ 2.50GHz Vendor: Intel Type: CPU Device Version: OpenCL 1.1 Driver Version: 1.1 Compute Units: 4 Work Group Size: 1024 Clock: 2499 MHz Global Memory (Total): 4096 MB Global Memory (Host): 4096 MB Global Memory (PCIe): 0 MB Local Memory: 32 KB Cache Size: 0.0625 KB Cache Line Size: 2097152 Bytes Available: Yes Double-Precision: Yes Extensions: ... [Device 1] Name: GeForce GTX 460 Vendor: NVIDIA Type: GPU Device Version: OpenCL 1.1 Driver Version: CLH 1.0 Compute Units: 14 Work Group Size: 1024 Clock: 1526 MHz Global Memory: 1024 MB Local Memory: 48 KB Cache Size: 0 KB Cache Line Size: 0 Bytes Available: Yes Double-Precision: No Extensions: ... [shared OpenCL+OpenGL Context] 2 OpenCL devices found! [Device 0] Name: GeForce GTX 460 Vendor: NVIDIA Type: GPU Device Version: OpenCL 1.1 Driver Version: CLH 1.0 Compute Units: 14 Work Group Size: 1024 Clock: 1526 MHz Global Memory: 1024 MB Local Memory: 48 KB Cache Size: 0 KB Cache Line Size: 0 Bytes Available: Yes Double-Precision: No Extensions: ... [Device 1] Name: Intel(R) Core(TM)2 Quad CPU Q8300 @ 2.50GHz Vendor: Intel Type: CPU Device Version: OpenCL 1.1 Driver Version: 1.1 Compute Units: 4 Work Group Size: 1024 Clock: 2499 MHz Global Memory (Total): 4096 MB Global Memory (Host): 4096 MB Global Memory (PCIe): 0 MB Local Memory: 32 KB Cache Size: 0.0625 KB Cache Line Size: 2097152 Bytes Available: Yes Double-Precision: Yes Extensions: ... Link to comment Share on other sites More sharing options...
Recommended Posts