Hey guys, about a week ago I began looking into CUDA programming and so far it has been a pretty smooth ride. However, I recently found out that CUDA does not support double variables by default, which is problematic to me as I need double precision for my programs. From what I have found, I need to add the flag “-arch sm_13” but I have absolutely no idea how to do this with Visual Studio 2012. Anybody mind pointing me in the right direction?
Related topics
| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
| double doesnt work in kernel | 7 | 3685 | October 23, 2008 | |
| enable doubles in kernel | 2 | 1993 | October 12, 2008 | |
| set a flag under xp/visual studio 2008 | 0 | 10314 | April 19, 2011 | |
| Adding Compiler Flag in Visual Studio 2010 | 2 | 1576 | December 12, 2012 | |
| Double not supported; demoting to float Compiling on a comp: 2.1 device and getting precision errors | 5 | 2569 | February 28, 2012 | |
| How To Define Architecture in Makefile | 1 | 22079 | April 12, 2011 | |
| support double precision in cuda | 12 | 1849 | January 26, 2014 | |
| How to support Double Precision on CUDA2.1? Double precision works well on CUDA2.0 but can not work | 2 | 2712 | February 19, 2009 | |
| Problem about double configuration and GTX 690 | 9 | 2340 | March 27, 2013 | |
| Does CUDA 4.0 keep double precision on by default on GPUs with Compute Capability 2.0? | 2 | 2194 | August 6, 2011 |