Todd,
... I sometimes notice hash tonal transitions in my prints, usually in
the
3/4 tones. Usually it's evident in my screen view too, but it seems to be
exacerbated in print. How much of this do you think is related to gaps in
the histogram, vs just the way tones "break" along my tonal curves, or the
curves (lack of linearity) of some of our printers, and materials?
[Paul Roark]
I sounds like it's in your file. The print is just a much more critical
medium than the monitor, so defects will be much more apparent. I'm, of
course, not excluding the printing as a source of problems (none of these
systems is perfect), but if you see the problem on the monitor, you'll see
it in spades on the print.
I can't really provide more specific details, because I'm not sure I've
established a definite pattern yet. I think it occurs at times in 16-bit
mode too.
[Paul Roark]
I played with 16 bit and found no advantage in printing. I always scan at
maximum depth and/or make multiple scans with different curves applied in
the scanner software, but for the final image, 8 bit/channel seems to be
enough. (Truth is, it's hard to achieve even 256 true steps in a final file
after all the manipulations.)
I'm wondering if it's related to papers too, in the way even when
silver printing, sometimes an image is a bear to print on one brand of
paper, but it falls easily into place on another.
[Paul Roark]
I found that Media Street's papers -- Glacier and DS matte -- introduce or
exacerbate artifacts. One reason I like Epson Archival Matte is that it
seems to have a "tooth" that results in very smooth output. I've also
noticed and been told by others that glossy paper really shows the defects
and dots.
By the way, some one told me that the variable-tone/mix inks are drying
fairly well on Epson Photo Glossy. I haven't tried it yet.
Paul
http://www.PaulRoark.com
[Non-text portions of this message have been removed]Message
RE: [Digital BW] Greetings -- dots, and curves
2001-08-09 by Paul Roark
Attachments
- No local attachments were found for this message.