Gutenprint review at www.monitor-calibration.net
Gutenprint review at www.monitor-calibration.net
- Subject: Gutenprint review at www.monitor-calibration.net
- From: Robert L Krawitz <email@hidden>
- Date: Wed, 21 Jun 2006 20:18:30 -0400
I have a few comments and questions about your Gutenprint review.
First of all, I think the review is very good, and there are a lot of
good insights there.
The performance of the driver in its default settings for printing
photographs isn't great. With a fast printer like the 2400, you
probably do need a fairly fast CPU to drive it effectively. It hasn't
been tuned for performance -- we've gone after some of the lowest
hanging fruit, but not much else. The performance problem prior to
5.0.0-rc3 was different. I found the 2400 was printing about 10% as
fast as it should (90% slowdown) on a 6" wide print, even when sending
a pre-generated print file to the printer. That turned out to be due
to an interaction with the printer firmware. That issue *should* be
completely resolved now. You didn't say how much slower it is than
you expect, and I'm curious about that.
The weave/stripe pattern you saw on the Gutenprint sample might be due
to its use of unidirectional mode above 720 DPI. In addition to being
slow, that produces inferior quality to bidirectional mode (only at
1440x2880 and above does unidirectional mode do better). That will be
fixed in the final release, and it should look smoother.
I'd like to better understand your issue with the reds. Is the
problem one of hue, saturation, or luminosity?
Highlight definition is something I've put a lot of work into. I've
seen prints from some of the lower end Epson printers where there's
practically no highlight definition -- it all but cuts off the
highlights, seemingly to reduce dither noise. I'm glad it's working
well. Shadows are an area I've been concerned about; I'm glad to hear
it's working for you.
If you're using the default settings for printing photographs, I'd
expect a somewhat reduced gamut. The High Accuracy color correction
mode (which is used by default if you specify an Image Type of
Photograph) deliberately reduces the gamut in order to get a decent
match with typical monitors -- for example, it lightens green, cyan,
blue, and in some cases magenta in order to clip colors that monitors
can't display (monitors can't display highly saturated dark greens and
cyans, which CMY inks can do very well). If you set the color
correction to Uncorrected, you get linearized output but otherwise the
RGB values are simply negated to produce the CMY values.
In addition, the density values are set up assuming that there's no
external ink limiting taking place. This likely compromises the gamut
a bit, particularly in highly saturated regions, since the paper can
take more ink if only one or two inks are in use than if three or four
inks are in use.
If you're doing your own profiling, you should specify Color
Correction as Uncorrected. If your profile has ink limiting, you
might also want to increase the density somewhat; I can't really say
how much is safe since I haven't played around with this very much.
If you increase it too much, of course, you'll have problems.
Anyway, thanks for the feedback.
--
Robert Krawitz <email@hidden>
Tall Clubs International -- http://www.tall.org/ or 1-888-IM-TALL-2
Member of the League for Programming Freedom -- mail email@hidden
Project lead for Gutenprint -- http://gimp-print.sourceforge.net
"Linux doesn't dictate how I work, I dictate how Linux works."
--Eric Crampton
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Colorsync-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden