Hi Vic, > What I tried out was to invert each of the layers, then use the > 'Addition' layer mode on each layer, and invert the result back again. > Theoretically, this should be what I'm looking for, since each CMY-color > is the invert of one of the RGB-colors (isn't it?). In practice, it > looks like the right result...
You can get very approximate results using this technique, but without the use of colour profiles it will never be accurate. It can still be useful, though. What my separate plugin does (in colour-preview mode) is generate a white background layer, then solid CMYK layers with the image data in the layer masks. The mode for these layers is set to "Darken Only". > To bring this topic in relation to Scribus, I'm considering creating a > duotone print job with specific inks. A special processing of bitmaps > can restore much of the original color information, but diversifying the > image printed in both inks. Actually, my question is related to some > experiments I'm conducting in that area, of which some information can > be read here: > > http://users.pandora.be/vicmortelmans/fts/duotone Again, you might find my separate plugin interesting - it already has very limited support for this kind of duotone separation. Currently it can only use Red as the spot colour (and places the result in the Magenta channel of a CMYK image). Other hues can be pulled out just by rotating the hues before separating. If you're interested the plugin can be found, with full source, at http://www.blackfiveservices.co.uk/separate.shtml All the best, -- Alastair M. Robinson
