Getting started with CUDA
I've recently gotten interested in parallel programming using graphics cards. A graduate student of mine has been working on the development of the code, but recently I've decided to try some coding myself. My plan with these posts is to outline the steps required for getting up and running with CUDA development on a Win7 machine using Visual Studio 2008. Yes this has probably been done before, but isn't this what the internet is all about...massively repeated information. So here goes my approach and I hope it is helpful to someone.
The very first step is to download the Getting started guide for windows. This document will step you through the process to ensure that you have the correct hardware and software on your computer to enable CUDA development and execution. The only difference from the instructions that I found were the directory paths even though I installed into the default directory. I think this is a Win7 (possibly also Vista) discrepancy from WinXP. So instead of
C:\Documents and Settings\All Users\Application Data\NVIDIA Corporation\NVIDIA GPU Computing SDK
I had to use
C:\ProgramData\NVIDIA Corporation\NVIDIA GPU Computing SDK 3.2
Now, I realize that I had previously installed a CUDA SDK and have already installed version 3.2 despite the documentation suggesting that you uninstall previous version of the SDK. I am now uninstalling the previous version (post installation of version 3.2). If I have problems in the future, I will uninstall/reinstall version 3.2.
blog comments powered by Disqus
blog comments powered by Disqus