plustek@linuxhacker.org

On Mittwoch, 7. Januar 2004 19:33, Alain Wenmaekers wrote:
> < can you please change the 110 to 0 and do a scan of the first few cm
> < a 300dpi and then count the lines of the black and the white calibration
> < strip?
>
> OK, I just did that.

Thanks.

>
> What I can calculate:
> - At the left side, the black strip ends at position 30. At the right, it
> ends on 25. In one of my scans the left side started on position 15 (but
> that only occured once. calibration also went wrong with this scan), all
> other tries are consistent with position 25.
> http://users.skynet.be/bluedigit/calibration.png
>
> Is it normal that the strip isn't symmetric?

No, that's unusual. Probably the strip moved a little bit, as it is in the
plastic "band" where the paper sizes are printed on. This one is not
fixed very good (at least I managed to open a defective LiDE30)...

Anyway, what makes me wonder is the fact, that you get better results,
when you use 25 instead of 35. Can you do me again a favour and
do two sample scans, one with 25 and one with 35. Please enable in
both cases the debug-messages:
export SANE_DEBUG_PLUSTEK=20 ; xsane

Can you please send me the pictures and the log messages!
(Send it to my private mail address 'cause the mailing list is limited
to 30k)

Back to these 35/25 values. If the calibration here on my scanner is
done with the 25, the sensor is over the black strip, and therefore, tha
picture is completely overexposed and this should also be the same
in your case. It also makes me wonder, that vuescan will wokr correctly
here, probably it has some kind of "white-strip-detection" - I'll ask Ed
Hamrick...

Thanks for your patience,
  Gerhard