cocos2d - Hello there, my name is Ilias and I am working as an High Performance Computing Specialist with GPUs and Coprocessors being my area of expertise. That basically means that I compile commercial, educational and custom codes create the scripts to automate the whole process AND optimize their performance. From a very quick look on the source code it does not seem to pose any significant problem... My proposition is to create a bash script and automate the whole process. Wget could get the newest source, then unzip and cmake would make the whole operation robust and workable for many years to come. Plus you get a free optimization with flags like -O3 or -O2 depended on what is performing better and many other perks like possible investigation of CUDAblas, math kernel library (MKL) or others depending on the needs of the project. Different versions are of course possible to handle within the script, and I do have experience with low and high lvl languages with Python and C/C++ standing out of the lot. However, I would like to ask before hand what is the issue that makes it unique and you require a VS script or similar, meaning that why couldn't you just create a new folder each time from the compilation script based on your target build each time? Thank you very much for your consideration, and I hope we could have a long lasting relation perhaps even on a GPU source code optimization (OpenCL/CUDA/OpenACC). Neither price or milestones are written in stone!