2020-02-05 14:45:46 -0600 | received badge | ● Necromancer (source) |
2020-02-05 04:24:34 -0600 | received badge | ● Supporter (source) |
2020-02-05 04:13:51 -0600 | received badge | ● Editor (source) |
2020-02-05 04:13:51 -0600 | edited answer | No effect from using cuda::Stream? I think the real issue is that CUDA needs pinned / pagelocked host-memory to do asynchronous transfers to the GPU. If yo |
2020-02-05 04:02:50 -0600 | answered a question | No effect from using cuda::Stream? I think the real issue is that CUDA needs pinned / pagelocked host-memory to do asynchronous transfers to the GPU. If yo |
2020-01-31 08:15:02 -0600 | commented question | i need to detect human body (dead,alive,injured),what method do i use? A stethoscope maybe? |
2017-01-25 06:59:44 -0600 | commented question | OpenCL TAPI mixed performance The operations mean and sum should be almost identical (see CPU results for reference). The "backend" is the NVidia GPU driver, which provides almost identical performance on Windows and Linux (see popular benchmarks). I doubt that the GPU "is somehow configured incorrectly", but I'm open to suggestions. |
2017-01-25 02:19:10 -0600 | received badge | ● Student (source) |
2017-01-25 02:18:38 -0600 | asked a question | OpenCL TAPI mixed performance I'm getting mixed results when using the OpenCL transparent API in terms of performance, so I wrote a simple test application for measuring the execution time of a few OpenCV methods. I'm testing the methods The following results were obtained on the same machine, using a GTX 980 GPU with the latest drivers on both Linux and Windows, built with OpenCV 3.2: The CPU results are perfectly comparable. On Linux the GPU results from Also note that on both systems the performance of |