> If I understand correctly, you meet 2 issues:
> - 16 bits scanning gives "noise" picture.
well, that depends on which color depth xsane uses but read below.
> - every second scan stalls
I now know the reason for that one. It only happens if I use the standard size
settings in xscanimage and xsane since they are approx 0,5 cm too large. If I
reduce the settings to A4 size it works perfect without any problems. I guess
the lights get stuck whenever they have to travel that half centimeter too
Would be fine of course if the backend would take care of that and prevent it
> I'm wondering if data is swap. Could you post a small scan in a lossless
> format such as pnm for instance ? I have read the logs you have posted.
> Is lacks an export SANE_DEBUG_GENESYS_GL646=255, so low-level
> information is lacking. It would be great (if I'm not abusing your time too
> much) if you could post the log of 2 previews in a single xsane session.
> The second preview should stall. For the 16 bit issue, the pnm
> of a small area of text would be great.
Now it perhaps gets strange. I never did scans with xsane before, only with
scanimage, xscanimage and kooka. But since you demanded xsane, I installed it
and guess what.....with xsane I didn't encounter the distorted pic problem at
600. But since I didn't find a depth setting, I don't know wether it is 8 or
16 bits or whatever.
Despite of the fact it works with xsane, I did previews and scans with both,
xsane and xscanimage with the logging you wanted. When scanning with
xscanimage the distorted pic problem occured again, I added that pic too (it
was some text).
You can find all that in a tar.gz at the following URL:
I hope I didn't forget some of those debug variables this time :)
If you need further help, just ask. I'll do whatever helps you in making this
The test16.pnm picture you posted is perfectly fine (I'm using Imagemagick
to see it). The trouble you get is related to the image viewer you use.
After trying konqueror to display it, I saw the same artifacts you get. So
it's a kde a bug related to 16 bits pictures.
For the other bug -trying to scan outside usable area- bound checking is
done. I think it may be a bug in scan area origin detection bug. Origin may be
wrongly detected a few mm too much toward "buttons side", so the scanning
head goes to far. You may check it by scanning a triangle shaped paper, with
one corner touching the very top of the scanning area. If expect that a few mm
of it will be missing.
You may also wait ~45 s after powering before testing. I think origin
detection may fails if scanner isn't warm enough.
When all debug is activated, the backend write debug pnm files containing
data is scans. I'd like to have these: