John Schamus - 2013-11-22

Version 0.5.9.6.2
- Added more flexibility to Set and get functions.
- Added FlipLR(), FlipUD(), & Rot90() functions.
- Added Round(), etc functions.
- Fixed a bug in Copy kernels that could cause pointer errors with low-end GPUs.
- Changed computeGrid1D() so that it does threads along y, not x.

Getting close to finalizing features for 0.6.0 release. Just need comparison functions, before beginning thorough testing. Depending on how much work it takes to bring it up to date, Linux setup may or may have to wait for 0.7.0. Big question is getting the CUDA Toolkit to install. There is nothing in the code that should stop it from being used on Linux, I just haven't tested it or provided an already setup Eclipse project.