turk@Apple.COM (Ken "Turk" Turkowski) (09/28/89)
A couple of weeks ago, the British courts upheld Quantel's patent governing the use of alpha in paint and compositing systems. This is a ludicrous judgement, and the judge seems to have disregarded the testimony of the expert witnesses and supporting documentation because he wasn't familiar with the technology. The validated claims were: * Iterative read-modify-write into a frame store, with a continuous two- dimensional shape (multi-bit alpha) controlling the application of the brush color into the frame store. Note that this also includes air-brushing. * Subpixel positioning. * Composition of two images with a soft-edge mask. The expert witnesses were Alvy Ray Smith, Jim Blinn and Lance Williams. Perhaps the American witnesses weren't given as much credibility as the British prosecutor. Ironically enough, Alvy Ray Smith submitted a paper to SIGGRAPH in '79 on the subject of soft-edge compositing, but was dismissed as "too obvious". Well, if SIGGRAPH thought it too obvious, what the *&#*&# are the British courts looking for? Quantel's major competitor, Ampex, whimpered in its corner in America with its tail between its legs instead of participating in the trial, yet it probably would have had some of the strongest evidence to disprove Quantel's claim. Quantel is rumored to the going for exclusivity in Britain, i.e. refusing to license the technology, making itself God's keeper of the sacred Alpha. This means that no company would be allowed to sell any product in Britain that makes use of alpha, ESPECIALLY paint systems. The British patent was filed in 1980, and the US patent in 1986, however, under reciprocal agreements, the American courts will probably recognize the date of the British patent if it decides that the American version is indeed valid. This is, therefore, a call for papers which have been made public, and which prove that alpha and subpixel positioning were in [relatively] wide use prior to 1980. I personally have Steve Gabriel's 1975 "Point Lights" paper, received in 1979, which talks about splattering distributions of color along a line, for example, to draw an anti-aliased line. What about chroma-keying, which was widely used, I believe, at this time. Does anybody have any information about the availability of such machines at this time? -- Ken Turkowski (turk@apple.com) -- Ken Turkowski @ Apple Computer, Inc., Cupertino, CA Internet: turk@apple.com Applelink: TURKOWSKI1 UUCP: sun!apple!turk