Disclosure: I suffer from incurable biases on the topics of CUDA, NVIDIA, and NVIDIA’s history, having not only worked on CUDA from early 2005 to late 2010 (through v4.0) but also having worked with NVIDIA while at Microsoft in the 1990s (I was the development lead of Direct3D for DX5 and DX6 [c. 1996-1998])
The rest of this post is for members.
A CUDA Handbook membership is ad-free reading of the whole book plus full access to paid posts, from $10/month. Become a member. Already a member? Log in to read it here.