2013-07-19 14:50:30 -0600 | asked a question | NVCC cannot find a supported cl version. Only MSVC 9.0 and MSVC 10.0 are supported I did build openCV2.4.6 with Cmake2.8.11.2 and configured it. As I have VS2012 so I chose "Visual Studio 11" (32 bit). After a successful configuration that I chose "Automatic" for "CUDA_GENERATION" and checked the CUDA_FAST_MATH. The solution was generated successfully. But, when I tried to build the solution in VS2012, I got this error: "nvcc cannot find a supported cl version. Only MSVC 9.0 and MSVC 10.0 are supported". Is there any way around it? |
2013-07-19 12:42:17 -0600 | received badge | ● Supporter (source) |
2013-07-19 12:41:40 -0600 | commented answer | getCudaEnabledDeviceCount() returns 0 ?!! Hmm... Thanks for the info @StevenPuttemans |
2013-07-19 12:39:55 -0600 | received badge | ● Scholar (source) |
2013-07-17 17:41:40 -0600 | commented answer | getCudaEnabledDeviceCount() returns 0 ?!! Thanks guys, this sounds strange, why they did such a thing?!! |
2013-07-16 09:45:04 -0600 | asked a question | getCudaEnabledDeviceCount() returns 0 ?!! Hi, I have installed the pre-compiled OpenCV 2.4.6 and CUDA 5.0 with latest graphics drivers. While I call the getCudaEnabledDeviceCount() it returns 0 and I can not use its functions. However, when I ran the deviceQuery test I saw that I have a compatible graphics card and every thing is OK! (The code successfully runs with CPU functions) I wonder why openCV can not access my Graphic Card? |