Imaging and computer graphics being programmable was a great breakthrough, and we invented that. On top of that, we searched for ways to express more sophisticated algorithms that could be computed on our processor, which is very different from a CPU.
We created something called CG—I think it was around 2003—C for GPUs. It predated CUDA by about three years. The same person who wrote the textbook that saved the company, Mark Kilgard, wrote that textbook.
CG was super cool. We wrote textbooks about it, started teaching people how to use it, and developed tools and such. Several researchers discovered it. Many of the researchers, and students here at Stanford, were using it.
Many engineers who later joined NVIDIA were playing with it. A couple of doctors at Mass General picked it up and used it for CT reconstruction. I flew out and saw them and asked, what are you doing with this thing? They told me about their work.
Then, a quantum chemist also used it to...
