Might it be related to premultiplied vs un-premultiplied alpha? Throw a clamp(color, 0.0, 1.0) CIkernel in there before hitting the addition kernel ?
On Dec 21, 2010, at 6:58 PM, kingLuma wrote: > On Dec 21, 2010, at 1:08 PM, Patrick Sheffield wrote: >> >> ...between how Quartz Composer processes the Screen blend mode and how Final >> Cut Pro and Motion do it? > > I actually don't have an answer, but it reminded me of a question I asked > some time ago regarding the QC Addition blend mode patch and a case where it > didn't seem to do what I'd expect (see message pasted below - perhaps > unrelated, but I never really was able to explain it in that case either) > > On Aug 4, 2009, at 6:53 PM, kingLuma wrote: >> In a compositing program (like Adobe After Effects for example), if I use >> the Add blend mode to composite the separated R, G, and B channels of an >> image as separate layers in an AE Comp, the result perfectly matches the >> original image. I'm trying to do the same thing in QC (compositing R and G >> with an Addition patch and then adding the result to the B channel with >> another Addition patch and then rendering that result using a sprite >> renderer), and I find that the result is obviously brighter than the >> original image. >> >> I've tried a jpeg brought in with an Image Importer patch, and a QT Movie >> brought in with a Movie Loader patch and I see the same problem in each >> case... Whether the "Enable Color Correction" checkbox on the importer >> patches is enabled or not seems to make no visible difference in the result. >> My method for separating the color channels involves using GLSL Shader >> patches within Render in Image patches, and for those Render In Image >> patches "Disable Color Correction" is unchecked (the image goes even >> brighter if this checkbox is enabled). >> >> Comparing my R, G, and B channels out of QC individually, they seem to match >> what I would get (for R, G, B) in AE so it seems the channel separation is >> correct and the problem is elsewhere (maybe in the Addition patch >> compositing?) >> >> Would adding R, G, and B using Addition patches in QC be expected to result >> in the original image (R+G+B = RGB) ? If so, is there something I'm missing >> that would cause what I'm seeing ? Any ideas appreciated. Thanks. >> >> ______ >> >> kingLu >> >> > > > > > > > > > >> >> This is an image screened over itself in Final Cut Pro/Motion: >> >> <fcs1.jpg> >> >> And in Quartz Composer: >> >> <quartz1.jpg> >> >> In Quartz Composer it's flatter and considerably more red than FCP. >> >> If I perform a Screen Blend (the inverse of the multiplication of the >> inverse), I get the same result as Quartz, but I need to figure out what >> FCP/Motion are doing differently. >> >> Does anyone have any insight? >> >> Thanks, >> >> Patrick Sheffield >> Sheffield Softworks >> >> _______________________________________________ >> Do not post admin requests to the list. They will be ignored. >> Quartzcomposer-dev mailing list (Quartzcomposer-dev@lists.apple.com) >> Help/Unsubscribe/Update your Subscription: >> http://lists.apple.com/mailman/options/quartzcomposer-dev/kingluma%40comcast.net >> >> This email sent to kingl...@comcast.net > > _______________________________________________ > Do not post admin requests to the list. They will be ignored. > Quartzcomposer-dev mailing list (Quartzcomposer-dev@lists.apple.com) > Help/Unsubscribe/Update your Subscription: > http://lists.apple.com/mailman/options/quartzcomposer-dev/doktorp%40mac.com > > This email sent to dokt...@mac.com _______________________________________________ Do not post admin requests to the list. They will be ignored. Quartzcomposer-dev mailing list (Quartzcomposer-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/quartzcomposer-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
