A realtime Mandelbrot zoomer in SSE assembly and CUDA
Back in 2002, Thanassis Tsiodras decided to begin investigating the power of SSE on the new Athlon CPU’s and wrote a rather complex real-time Mandlebrot zoomer program. Now, 7 years later, he revisits his code and rewrites using NVidia’s CUDA to startling effect:
In less than one hour, I went from my rather complex SSE inline assembly, to a simple, clear Mandelbrot implementation… that run… 15 times faster!
Let me say this again: 1500% faster. Jaw dropping. Or put a different way: I went from 147fps at 320×240… to 210fps… at 1024×768!