The D1 Buffer Project
Ken Holland and I ran into each other at the Omnibus auction. We got to talking about a project he had done at Compact Video to produce a noise reduction system using digital framestore memory. I mentioned to him the Ikonas framebuffers there at the auction, and he ended up getting a pair of them. Ken was working with Joe Matza of Composite Image Systems, who at the time had the only pin-registered film-to-tape transfer systems that worked. Ken helped them set up their entire facility, and had some good ideas about how to use a large buffer to hold a higher resolution frame scanned in from the rank with a 4:1 interlace. This meant you could then double the resolution of your film imagery, still use D1 signal processing and existing plant to process the footage, and print back to film.

I built a special input board for the Ikonas framebuffer and had the system working so that it could grab a frame into the even lines of the buffer, then a second frame into the odd lines of the buffer, effectively giving you a 720 x 970 image. If you wanted to carry the process a step further, you could achieve 1440 x 970 with horizontal half-pixel offsets also, and still process all the footage through your "traditional" video gear.

Although this was an interesting idea for its time, the industry had the good sense to just use plain old computers to process great big pictures, something many post houses even today are still trying to come to grips with. But it was fun to play with the hardware, write bit-slice code, burn PAL's, go searching for blips on the logic analyzer, learn about D1 VTR's and Abekases, and hang around with the techies at CIS.

Back to Dave's History Page
Please email me for more info: