mitch_de Posted February 19, 2012 Author Share Posted February 19, 2012 Great! Some AMD have problems (other than the only cosmetic glitch) , they get OpenCL init failed. I mad an new OceanWave_ML Mountain Lion ONLY!!! version. Wasnt easy, because it was not only OpenCL 10.8 SDK recompile action, also OpenGL seems to be changed much (also GLUT). Therefor i must enable multisampling to avoid that glitch (flashing + wrong OpenGL window). The results with multisampling enabled cant be compared with the older version - the slower the gpu the more fps diff between both. My 9600 GT gone down from 104 to 89 fps using multisampling. Faster gpu may not have that much diff. Diff comes because gpu has a bit more to do - beside OpenCL it must also do the opengl display - and multisampling does take some more gpu time - at least for slow gpus. OceanWave_ML Mountain Lion ONLY!!! - wont run 10.6 / 10.7 you may need to setup mail account first, if you want to submit results. OpenCL_OceanWave_ML.zip Link to comment Share on other sites More sharing options...
diskeeper Posted February 19, 2012 Share Posted February 19, 2012 With new Mountain Lion version on EVGA GTX460 SC: Link to comment Share on other sites More sharing options...
Atomic999 Posted February 19, 2012 Share Posted February 19, 2012 Results on 10.7.3 for my 560TI-448 on Q6600/8GB, GA-X38-DS5: 428,43 FPS Link to comment Share on other sites More sharing options...
iLeopod Posted February 20, 2012 Share Posted February 20, 2012 Hi mitch_de, can u also upload a 32bit version? Link to comment Share on other sites More sharing options...
mitch_de Posted February 20, 2012 Author Share Posted February 20, 2012 Here you go with 32 Bit (10.6/10.7 only, NOT 10.8 DP1) Should not be slower than 64 Bit - i tested it and it has near same FPs (even little higher )as 64 Bit Version. OpenCL_OceanWave_32Bit.zip Link to comment Share on other sites More sharing options...
iLeopod Posted February 20, 2012 Share Posted February 20, 2012 Here my results: GTX750: ION1/9400m: Link to comment Share on other sites More sharing options...
mitch_de Posted February 20, 2012 Author Share Posted February 20, 2012 Great to see that 32 Bit Version is working on your ATOM / NV 9400M Do you see any big fps diff comparing 64 Bit vs 32 Bit verson using i7 / GTX 570 ? from cmf: OpenCL 10.8 DP1: opencl works ootb on fermi , but apple seems to have removed the sm_20 code and hard setting it to 1.2 or 1.3 won't work either , means: luxmark and all other programs that want to use more complex opencl stuff (like local memory atomics) will fail) Link to comment Share on other sites More sharing options...
iLeopod Posted February 20, 2012 Share Posted February 20, 2012 Yes with standard size: x32: 489 x64:523 But the results are different from try to try. By making the window as small as possible i get: x32: 590.8 x64: 603.96 Link to comment Share on other sites More sharing options...
uphuck Posted March 4, 2012 Share Posted March 4, 2012 Great to see that 32 Bit Version is working on your ATOM / NV 9400M Do you see any big fps diff comparing 64 Bit vs 32 Bit verson using i7 / GTX 570 ? from cmf: OpenCL 10.8 DP1: opencl works ootb on fermi , but apple seems to have removed the sm_20 code and hard setting it to 1.2 or 1.3 won't work either , means: luxmark and all other programs that want to use more complex opencl stuff (like local memory atomics) will fail) Then this luxmark test on 10.8 DP1 with GTX470 (sm_20) is confusing.. Score is ~10 times higher than the test applied in lion using similar cards. http://i.imm.io/hPfv.png Link to comment Share on other sites More sharing options...
mitch_de Posted March 4, 2012 Author Share Posted March 4, 2012 Nope ! Look at the rendered window result and compare with the 10.6 / 10.7 screenshoots! 10.8 DP1 and Luxmark 2.0 didnt work really (rendered preview wrong + unsualbe results) and need to be updated in the future. Its OpenCL code itself AND also QT (the QT SDK from Nokia ) Will take much time, because actual QT SDK doesnt support 10.8. Link to comment Share on other sites More sharing options...
uphuck Posted March 5, 2012 Share Posted March 5, 2012 Nope ! Look at the rendered window result and compare with the 10.6 / 10.7 screenshoots! 10.8 DP1 and Luxmark 2.0 didnt work really (rendered preview wrong + unsualbe results) and need to be updated in the future. Its OpenCL code itself AND also QT (the QT SDK from Nokia ) Will take much time, because actual QT SDK doesnt support 10.8. Yes you are right. It also did not perform the test on GTX 560 (sm_21) which seems to be ok on 10.8 DP1. Link to comment Share on other sites More sharing options...
RobertX Posted April 19, 2012 Share Posted April 19, 2012 Zotac GT 520 GeForce 1GB on Lion(10.7.2) ...the app doesn't show my GPU clock speed...but it does run at least...here's my results Link to comment Share on other sites More sharing options...
bkdwt Posted May 1, 2012 Share Posted May 1, 2012 But darn, here just picked up 19 FPS on a GTX 460 in SL 10.6.8, and have the table scoring, GTX 460 making 271 FPS!. Ahhh that anger! :hammer: Link to comment Share on other sites More sharing options...
mitch_de Posted May 7, 2012 Author Share Posted May 7, 2012 Luxmark 2.0 with ML 10.8 Dp3+ update starts working again (preview + results wrong in ML Dp1+DP2) OpenCL Oceanwave stays working. Link to comment Share on other sites More sharing options...
mitch_de Posted May 27, 2012 Author Share Posted May 27, 2012 UPDATE to Mai 2012 Version Updated Apple changes in my Benchmark, " Version 1.3, 2012-05-22 Fix warnings pertaining to use of deprecated APIs Fix a bug related to redundant call to glutSwapBuffer" added fullscreen mode (look first post), DL first post Link to comment Share on other sites More sharing options...
RobertX Posted May 28, 2012 Share Posted May 28, 2012 works here...gt 520...results from fullscreen... ...windowed mode around 44 fps...but it does work... ...a few fps slower than the older build... Link to comment Share on other sites More sharing options...
T a D Posted June 18, 2012 Share Posted June 18, 2012 GA-P35-DS3P (rev 1.1) Q6700 @2.3 GHz 4GB RAM GeForce GTX460 @ 1430 MHz - Netkas OpenCL patch applied to GeForceGLDriver.bundle and NVDAGF100Hal.kext in both 10.7.4 and 10.8DP4 Lion (10.7.4) Windowed: 215.2 fps Full Screen: 149.9 fps Mountain Lion (10.8DP4) Windowed: 241.1 fps Full Screen: 164.4 fps 1 Link to comment Share on other sites More sharing options...
dddirtman Posted July 21, 2012 Share Posted July 21, 2012 evga gts 460 oc i7 950(stock) lion 10.7.4 244.5 fps Link to comment Share on other sites More sharing options...
DarkUser89 Posted July 21, 2012 Share Posted July 21, 2012 i5 2500K @ 4,0GHz GA-Z68X-UD3H-B3 rev 1.3 @ 16GB RAM GTX 560 OC (PCIe 8x) 224.6 fps Link to comment Share on other sites More sharing options...
TH3L4UGH1NGM4N Posted July 21, 2012 Share Posted July 21, 2012 AMD Radeon 6870 Link to comment Share on other sites More sharing options...
Wayang-NT Posted July 22, 2012 Share Posted July 22, 2012 2500K @ 4.4GHZ - AMD HD6870 - OS X build 12A269 .... Link to comment Share on other sites More sharing options...
jaymz60 Posted August 5, 2012 Share Posted August 5, 2012 i73960X GTX580 1 Link to comment Share on other sites More sharing options...
mitch_de Posted August 6, 2012 Author Share Posted August 6, 2012 Wow , GTX 580 performs really fast. Much faster than the GTX 4xxx cards. Link to comment Share on other sites More sharing options...
RobertX Posted August 8, 2012 Share Posted August 8, 2012 (edited) moved over to 10.8...can't run the benchmark now...tried the opencl fix from before...but still errors... Edit: actually installing the bundle after patching helps here's my results INITIALISING: 1 OpenCL platform found! [Platform 0] Name: Apple Vendor: Apple Version: OpenCL 1.2 (Jun 20 2012 14:18:19) Profile: FULL_PROFILE Extensions: cl_APPLE_SetMemObjectDestructor cl_APPLE_ContextLoggingFunctions cl_APPLE_clut cl_APPLE_query_kernel_names cl_APPLE_gl_sharing cl_khr_gl_event [OpenCL-only Context] 2 OpenCL devices found! [Device 0] Name: Intel(R) Core(TM)2 Duo CPU E6550 @ 2.33GHz Vendor: Intel Type: CPU Device Version: OpenCL 1.2 Driver Version: 1.1 Compute Units: 2 Work Group Size: 1024 Clock: 2327 MHz Global Memory (Total): 8192 MB Global Memory (Host): 8192 MB Global Memory (PCIe): 0 MB Local Memory: 32 KB Cache Size: 0.0625 KB Cache Line Size: 4194304 Bytes Available: Yes Double-Precision: Yes Extensions: cl_APPLE_SetMemObjectDestructor cl_APPLE_ContextLoggingFunctions cl_APPLE_clut cl_APPLE_query_kernel_names cl_APPLE_gl_sharing cl_khr_gl_event cl_khr_fp64 cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_byte_addressable_store cl_khr_int64_base_atomics cl_khr_int64_extended_atomics cl_khr_3d_image_writes cl_APPLE_fp64_basic_ops cl_APPLE_fixed_alpha_channel_orders cl_APPLE_biased_fixed_point_image_formats [Device 1] Name: GeForce GT 520 Vendor: NVIDIA Type: GPU Device Version: OpenCL 1.1 Driver Version: CLH 1.0 Compute Units: 1 Work Group Size: 1024 Clock: 1620 MHz Global Memory: 1024 MB Local Memory: 48 KB Cache Size: 0 KB Cache Line Size: 0 Bytes Available: Yes Double-Precision: No Extensions: cl_APPLE_SetMemObjectDestructor cl_APPLE_ContextLoggingFunctions cl_APPLE_clut cl_APPLE_query_kernel_names cl_APPLE_gl_sharing cl_khr_gl_event cl_khr_byte_addressable_store cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_APPLE_fp64_basic_ops [shared OpenCL+OpenGL Context] 2 OpenCL devices found! [Device 0] Name: GeForce GT 520 Vendor: NVIDIA Type: GPU Device Version: OpenCL 1.1 Driver Version: CLH 1.0 Compute Units: 1 Work Group Size: 1024 Clock: 1620 MHz Global Memory: 1024 MB Local Memory: 48 KB Cache Size: 0 KB Cache Line Size: 0 Bytes Available: Yes Double-Precision: No Extensions: cl_APPLE_SetMemObjectDestructor cl_APPLE_ContextLoggingFunctions cl_APPLE_clut cl_APPLE_query_kernel_names cl_APPLE_gl_sharing cl_khr_gl_event cl_khr_byte_addressable_store cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_APPLE_fp64_basic_ops [Device 1] Name: Intel(R) Core(TM)2 Duo CPU E6550 @ 2.33GHz Vendor: Intel Type: CPU Device Version: OpenCL 1.2 Driver Version: 1.1 Compute Units: 2 Work Group Size: 1024 Clock: 2327 MHz Global Memory (Total): 8192 MB Global Memory (Host): 8192 MB Global Memory (PCIe): 0 MB Local Memory: 32 KB Cache Size: 0.0625 KB Cache Line Size: 4194304 Bytes Available: Yes Double-Precision: Yes Extensions: cl_APPLE_SetMemObjectDestructor cl_APPLE_ContextLoggingFunctions cl_APPLE_clut cl_APPLE_query_kernel_names cl_APPLE_gl_sharing cl_khr_gl_event cl_khr_fp64 cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_byte_addressable_store cl_khr_int64_base_atomics cl_khr_int64_extended_atomics cl_khr_3d_image_writes cl_APPLE_fp64_basic_ops cl_APPLE_fixed_alpha_channel_orders cl_APPLE_biased_fixed_point_image_formats Edited August 9, 2012 by robertx Link to comment Share on other sites More sharing options...
RobertX Posted September 12, 2012 Share Posted September 12, 2012 using the newest retail nvidia drivers for 10.8.1 i get this error again..."ERROR: Failed to initialize OpenCL= mostly by OpenCL driver didnt know your gpu, like Fermi GTX.... which need to be patched" Link to comment Share on other sites More sharing options...
Recommended Posts