1999.07.16 19:40 "RE: RGB Gradient Code, adapted from Andreas Neumann (sol'n)", by Eric B. Middlecamp
"Grissom, Ed" <firstname.lastname@example.org> on 07/16/99 02:36:10 PM
To: Eric B. Middlecamp/MIPP/Imation@Imation
Subject: RE: RGB Gradient Code, adapted from Andreas Neumann
> Eric -
> I _know_ from some discussion on comp.graphics.apps.photoshop that PS does > fiddle with your data WHEN THE DEPTH OF EACH BAND IS GREATER THAN 8bits and
> it must be converted for display. They do some random dithering that is not > really documented well, but the prepress people like.
> In your case, I know that you have 8bit bands, so the above does not come > into play. However, given the above, it seems reasonable that PS is doing
> something and they are not telling you the truth about your image. I think > the problem is in PS, not LibTIFF...
> Actually, I now seem to remember some folks talking about only having 7bits
> per band of "true"
> data in PS. Don't remember the specifics of that though.... Talk to
> ed grissom
I've found the culprit, and as it turns out it could be potentially embarrasing, so.. go easy on me. In Photoshop, under profile setup, there was a default ("dont ask") conversion in place for RGB images. Having removed this, everything is as it should be. Sigh.
Many thanks for the feedback,