What is quite funny as that the guy's "optimized" version of code originally written for a 7Mhz amiga runs about 32 fps on my quad-core 2.5Ghz machine. From 7mhz to 10,000mhz and no performance improvement.
Modern PC's and Macs are so slow compared to what they should be if the OS's were up to using the hardware properly. Just comparing MorphOS and OS X Leopard running on the same Mac shows how fast they could be.