Jens> gtkglue.o: In function `curve_new':
Jens> /usr/src/sane-0.61/frontend/gtkglue.c:701: undefined reference to `gtk_gamma_curve_new'
Jens> I had no problems compiling 0.57 and I did
Jens> compile and install the other gtk before trying the new
Jens> version of sane. If you know what's wrong please let me know.
Did you delete the old gtk/gdk/glib libraries before installing the
new ones? There is an unfortunate bug in the ltconfig file that makes
this necessary on Linux.
Jens> frontend/scanimage scans a nice preview, but at 300 dpi I only
Jens> get bad results (please have a look at
Jens> http://www.weh.rwth-aachen.de/~jens/ --- I'm sorry if you
Jens> can't get a connection. We have problems with the laser, which
Jens> hooks the house with the university. Please tell me where to
Jens> upload/send the pictures, if you have problems):
Jens> scanimage -x 50 -y 50 --resolution 300dpi -> ginger-01.jpg
Jens> you can see that some lines are missing because of the fast
Jens> stop and go of the head (especially at the "G" in the 3rd
Jens> row). In color it is even more bad:
Hmmh, I have had one other report of dropped lines in grayscale mode.
Unfortunately, I can't reproduce the problem, so I can't debug it.
Can you look into it? Maybe you could play with different buffer
sizes and see what's happening as the sizes change. The man-pages
describe a couple of environment variables that may be helpful in
tracking this down (e.g., set SANE_DEBUG_MUSTEK and
SANE_DEBUG_SANEI_SANE to 128). If you can, try it with a different
SCSI controller, too, since it's quite possible the bug is in the
driver for the Mustek card.
Jens> Does this happen because there's no model string returned by
Jens> the driver?
No, if you invoke scanimage with SANE_DEBUG_MUSTEK set to 3, you'll
probably see the correct scanner type (otherwise, scanimage would
refuse to scan).
--david
-- Source code, list archive, and docs: http://www.azstarnet.com/~axplinux/sane/ To unsubscribe: mail -s unsubscribe sane-devel-request@listserv.azstarnet.com