On Monday 26 July 2010 17:20:34 Stefan Champailler wrote:
> On Monday 26 July 2010 16:10:52 Gerhard Jaeger wrote:
> > On Monday 26 July 2010 15:58:10 Stefan Champailler wrote:
> > > > > Maybe Gerhard knows something about this. In the end, it may just
> > > > > vibrations due to different motor settings between SANE and the
> > > > > Windows driver.
> > > >
> > > > your guess is right. While playing with different settings, the
> > > > grinding noise was my partner - and as you now observe sometimes this
> > > > ghost comes back :(
> > > > To solve this problem you might try to play with the settings in
> > > > plustek-usbdevs.c line 2044/2045. Try to reduce the dMaxMotorSpeed
> > > > and dMaxMoveSpeed until the noise dissapears. It's been a long time
> > > > since I tweaked these settings, so I can't tell you wich one to
> > > > tweak. But always change only one setting at a time.
> > >
> > > Cool, I'll look at it... However, before I start playing, could you
> > > tell me what kind of increment I may try ? According to this :
> > >
> > > 0.86, /* dMaxMotorSpeed */
> > > 0.243, /* dMaxMoveSpeed */
> > >
> > > I think a safe guess is to have increments of 0.02 for the max motor
> > > speed and 0.05 for the move speed, but that's complete guess. Do you
> > > remember the kind of increments I may use ? By any chance do you also
> > > remember the units of those values, what they represent physically ? I
> > > ask because I don't know if there's a risk of breaking the hardware
> > > with bad values... (I once broke a CRT monitor by programming a bad
> > > refresh speed :-()
> >
> > well there where indeed some problems on the EPSON 1260, where I
> > blow a fuse with the wrong settings, but I never had problems with
> > Canon devices. 0.05 should be okay. Regarding the units, I'm
> > not sure could be inches per second... you might check the formulas
> > in doc/plustek/Plustek-USB.txt but in the code there are several
> > corner cases and in the end the trial-and-error approach will be
> > the fastest here ;)
>
> Looks like these work :
>
> static HWDef Hw0x04A9_0x2206 =
> {
> 0.76, /* 0.86 dMaxMotorSpeed */
> 0.243, /* 0.243 dMaxMoveSpeed */
>
> I've just tested with scanimage defaults (they work making the noise
> before)...
>
> Now I'm pretty surprised... Why on earth would Linux use different settings
> than on windows (assuming it's just a question of settings...). Has anyone
> already looked into that ? I've seen complaints here and there about this
> scanner so I guess the problem is not entirely fixed. Moreover, I still
> wonder why it was so easy to fix it...
well I've two different N650U and they both work with the old
settings and I had nobody like you wiht a N656 to test that's all...
Anyway thanks for testing, I'll commit (is it the same wording as for
CVs?) the patch ASAP.
Regards,
Gerhard
--
To UNSUBSCRIBE, email to [email protected]
with a subject of "unsubscribe". Trouble? Contact [email protected]
