Hi Karl Heinz and other Epson 610 freaks,
On Tue, 06 Feb 2001, Karl Heinz wrote:
>I have not had time to check this in the source code, but I think
>I know what the problem is:
>The Perfection610 does not combine the RGB channels in the scanner, it
>sends them separately and I have to do this manually. There is a shift
>between the color channels that the backends receives (the actual amount
......
I now made a dirty fix that works reasonably : I hacked the scanimage frontend
and let it decrease the height in the PNM header now, but only in case a color
scan is made. Here is the hack :
Decrease height parameter in write_pnm_header() function in scanimage.c :
case SANE_FRAME_BLUE:
case SANE_FRAME_RGB:
printf ("P6\n# SANE data follows\n%d %d\n%d\n", width,
+ #ifdef EPSON610_COLOR_HACK
+ height - 5, // only enough for 75 and 150 dpi
+ #else
height,
+ #endif
(depth <= 8) ? 255 : 65535);
break;
See what you do with it : probably it can help somebody else as well. Note
that a decrease with 5 lines is only sufficient for 75 and 150 dpi: for 300 and
600 dpi it should be more.
This should be automatically detected and adjusted as well but that is something
to be done yet.
Note that the xscanner frontend (based upon the scanimage program) is
also updated : it is located on the website below.
bye,
-- Kees Lemmens, Email: C.W.J.Lemmens@its.tudelft.nl, WWW : http://ta.twi.tudelft.nl/DV/Staff/Lemmens/index.html Software: http://ta.twi.tudelft.nl/ftp/dv/lemmens Home : 010-4742919/4740254, Office: 015-2787224, Fax/XOIP: 020-8666099-- Source code, list archive, and docs: http://www.mostang.com/sane/ To unsubscribe: echo unsubscribe sane-devel | mail majordomo@mostang.com
This archive was generated by hypermail 2b29 : Fri Feb 16 2001 - 13:46:51 PST