To add to Tyler's response regarding banding - make sure you're viewing the image at 100% when you're evaluating banding. Anything less than 100% is prone to showing banding that doesn't exist when viewed at full size. Also, as Tyler suggested, adding noise to the mask works wonders for improving the "smoothness" of the gradation. The alpha channels are still 8 bit. Brian --- In DigitalBlackandWhiteThePrint@yahoogroups.com, "Tyler Boley" <tyler@...> wrote: > > --- In DigitalBlackandWhiteThePrint@yahoogroups.com, "Edgar" > <bakersfieldabalone@> wrote: > > > > My experience with the masks on adjustment layers has been mixed with > > regard to the bit depth. I find that I am able to see on my monitor > > (although not in my QTR/piezotone prints) the gradations between > > shades when generated from a soft brush or from the gradation tool > > showing through the adjustment layer especially in flat sky areas. > > This happens in both 8 and 16 bit modes for me. Are the layer masks 8 > > bit deep even when working in a 16 bit mode? > > I try to stay away from that kind of editing for the reasons you > state. Using a huge blur, or the grad tool, etc. can create these > problems if in certain kinds of image areas like skies. I haven't seen > the problem for a while, so not sure if latest PS versions have hi bit > masks or I've just learned to stay way from these kinds of edits. > Given the rapid escalation of hi bit file size when adding layers in > recent versions, I've wondered if the alpha channels are still just 8 bit. > I tend to use curve adj layers a lot, with masks. If I were trying to > darken a sky that way with gradation in the mask, I'd add noise to the > mask if there is a problem. > Definitely do not run tonal editing on such a layer mask after > creating it, I'll band it for sure, better to rebuild the mask as desired. > > > More to the point at hand, Tyler, do you see any affect of the 8bit > > masks when applied to your 16bit file either on screen or in print? > > I hope the above answers this, no I don't. However, as you are > implying, there would be nothing gained from the application since the > source (8 bit alpha channel) was created in 8. Again, perhaps I'm > staying away from the problem anyway, so not seeing it for that reason. > Alternately, I wonder what options would exist that don't result in > the same problem. You have to make that gradated edit somehow. Even a > highly feathered selection or gradated quick mask, then tonal edit on > a high bit file, the selection is a temporary 8 bit alpha channel, right? > By the way, monitors are notorious for showing banding that doesn't > exist in the file, particularly monochrome or pastels. > Tyler >
Message
Re: [Digital BW] Tyler's Memory Work-a-round
2007-04-15 by Brian Chapman
Attachments
- No local attachments were found for this message.