Session G358 - Making Movies: Animation of Large Scale Computations

August 23-28, 1987

A chemist from IBM talked about a graphics system they built to display models of large molecules. This probably doesn't have much applicability to my employer, but I try to stay current in new graphics development. Once upon a time I wrote a program in Apple Pascal to rotate wire-frame models in three-space, and I wanted to see what IBM could do with a 3090 that I couldn't do with an Apple II.

Wow! It helps to be able to crunch many numbers simultaneously. They showed a videotape in which they hydrated a large double-helixed DNA molecule. They were doing hidden-surface processing, reflections, refractions, multiple light sources and LOTS of hairy mathematics.

For the record, they were building images on a 3090 under VM, then sending them to a PC-AT through a 3270 emulator card running a file-server program. The AT contained a high-resolution RGB display card, which in turn was connected to an RF modulator, which fed a stock VCR, which was under program control. It wasn't very pretty, but it got the job done, and the results were spectacular.

Back to session index
Back to index of SHARE meetings
Read the disclaimer