[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: Re[3]: [cgmo-webcgm] Drawing model (take 2)
At 02:32 PM 4/20/2006 -0400, Benoit Bezaire wrote: >[...] >Note, if it were up to me, I wouldn't put any examples at all (it's >too much). I would support this -- remove the examples. Thanks for the history (Porter-Duff). Interesting. Correct me if I'm wrong, but it (your equations presented with pre-mult input assumption) doesn't imply what functionality is present or how it is parameterized in the metafile, but rather just how it's effect is described for viewers. Right? I.e., when we say ... >Yes, Esc.45 is not premultiplied. what we're really saying is that the equation takes as input non-pre-mult RGB coordinates (regardless of how they are stored). But we don't mean your equations to imply, in a RGBA metafile for example, that the R, G, B are premultiplied. I.e., R,G,B are NOT premultiplied in the metafile content. But a processor would premultiply them before applying the compositing equations. Right? (I just re-read the RGBA Registry item -- no hint of pre-mult.) -Lofton.
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]