I've updated the first post with the latest release.

This includes some make file fixes and speed improvements for the OpenCL miner.  Tweaking the compiler options have also tremendously increased the speed of the 4way hash in the Windows binaries for processors capable of SSE2.

Remember, to get the most out of the CUDA and OpenCL miner, you must tweak the options.  The defaults are just there to give you a baseline.

Use CMake.