On Sat, 21 Nov 1998, Bernd Schroeder wrote:
> > I just acquired a ScanMaker X6EL, and I noticed the same strange
> > behavior: the scanned image is OK after a powercycle, but the following
> > scans result in "weird" colors.
> > I have not had time to investigate the problem so far, but here is
> > some additional info:
>
> I think it is not the same problem, it is more or less the other way
> round.
Yes, sorry, it is indeed a different problem.
> Can you try the following: when the colors are wrong, do a lineart scan,
> and then again a color scan. It would be interesting to know, whether
> the colors are correct again (at least for one scan). Some models
> show this behaviour, and the current workaround is to set up the parameters
Indeed, that behaviour is shown by the X6EL as well ...
> for a B/W scan before every scan and scan one line (or zero lines
> in the next release). This is, what do_dummy_scan() does
> (see the discussion in the other thread).
Yes! Modifying do_dummy_scan() to execute for my model also (FW
1.30) fixed the problem (perhaps the test should be "<= 1.30" ?) :
--- microtek2.c.back Thu Nov 19 22:36:26 1998
+++ microtek2.c Mon Nov 30 00:53:58 1998
@@ -1989,7 +1989,9 @@
if ( ! ( md->info[MD_SOURCE_FLATBED].model_code == 0x97
&& md->revision == 1.00 )
&& ! ( md->info[MD_SOURCE_FLATBED].model_code == 0x91
- && md->revision == 1.00 ) )
+ && md->revision == 1.00 )
+ && ! ( md->info[MD_SOURCE_FLATBED].model_code == 0x98
+ && md->revision == 1.30 ) )
return SANE_STATUS_GOOD;
DBG(30, "do_dummy_scan: ms=%p\n", ms);
I guess the X6EL can be considered as a supported model by now :) .
Thanks Bernd!
-Christian
-- Source code, list archive, and docs: http://www.mostang.com/sane/ To unsubscribe: echo unsubscribe sane-devel | mail majordomo@mostang.com