Message boards : GPUs : CUDA MBPro GeForce GT 650M
Message board moderation
Author | Message |
---|---|
Send message Joined: 20 Mar 13 Posts: 1 ![]() |
Hello all, Could someone please help me get CUDA working on my MBPro. I have OS X 10.8.3, NVIDIA GeForce GT 650M and CUDA 5 w all toolkits etc installed and a cc_config in my data folder. Here is my log. Thanks in advance. Blake Wed Mar 20 18:10:00 2013 | | Starting BOINC client version 7.0.56 for x86_64-apple-darwin Wed Mar 20 18:10:00 2013 | | log flags: file_xfer, sched_ops, task Wed Mar 20 18:10:00 2013 | | Libraries: libcurl/7.26.0 OpenSSL/1.0.1c zlib/1.2.5 c-ares/1.9.1 Wed Mar 20 18:10:00 2013 | | Data directory: /Library/Application Support/BOINC Data Wed Mar 20 18:10:00 2013 | | Processor: 8 GenuineIntel Intel(R) Core(TM) i7-3615QM CPU @ 2.30GHz [x86 Family 6 Model 58 Stepping 9] Wed Mar 20 18:10:00 2013 | | Processor features: FPU VME DE PSE TSC MSR PAE MCE CX8 APIC SEP MTRR PGE MCA CMOV PAT PSE36 CLFSH DS ACPI MMX FXSR SSE SSE2 SS HTT TM PBE SSE3 PCLMULQDQ DTES64 MON DSCPL VMX EST TM2 SSSE3 CX16 TPR PDCM SSE4.1 SSE4.2 xAPIC POPCNT AES PCID XSAVE OSXSAVE TSCTMR AVX1.0 RDRAND F16C Wed Mar 20 18:10:00 2013 | | OS: Mac OS X 10.8.3 (Darwin 12.3.0) Wed Mar 20 18:10:00 2013 | | Memory: 8.00 GB physical, 357.07 GB virtual Wed Mar 20 18:10:00 2013 | | Disk: 464.96 GB total, 356.83 GB free Wed Mar 20 18:10:00 2013 | | Local time is UTC -4 hours Wed Mar 20 18:10:00 2013 | | BOINC does not support CUDA on this computer. Wed Mar 20 18:10:00 2013 | | OpenCL: NVIDIA GPU 0: GeForce GT 650M (driver version CLH 1.0, device version OpenCL 1.1, 512MB, 512MB available, 25 GFLOPS peak) Wed Mar 20 18:10:00 2013 | | Config: use all coprocessors |
![]() Send message Joined: 29 Aug 05 Posts: 15599 ![]() |
If this is a dual-GPU MacBook Pro, then as far as I know, you can't. There is a problem with the CUDA driver, which the BOINC developers have asked Nvidia to fix, but so far they haven't come around to it. So since BOINC 7.0.47, BOINC stops trying to detect CUDA on dual-GPU Mac's. The simplest way there is to remove CUDA from the system. Most projects that give out GPU applications now have a choice for OpenCL applications as well. |
Copyright © 2025 University of California.
Permission is granted to copy, distribute and/or modify this document
under the terms of the GNU Free Documentation License,
Version 1.2 or any later version published by the Free Software Foundation.