On Sunday 07 November 2010 17:13:04 Gerhard Jäger wrote:
> Hi,
>
> On Saturday 06 November 2010 22:29:38 Stefan Champailler wrote:
> [...]
>
> > > any news on that? Can you confirm the settings?
> >
> > ooops RL caught me... But thanks to your mail, I've made some tests
>
> > tonight. Excerpts of the code (plustek-usbdevs.c) :
> well, welcome back :)
>
> > /** Canon N650U/N656U */
> > static HWDef Hw0x04A9_0x2206 =
> > {
> >
> > /* 0.86 / 0.243 : original values
>
> [...]
>
> > 0.85 / 0.243 / 600dpi low grinding --> sounds almost normal
> > (although
> >
> > I don't know what normal is :-))
> >
> > 0.85 / 0.240 / 600dpi small grinding
> >
> > So what does this all mean ? I've made several tests and for some reason,
> > the 200 dpi resolution is really not a good idea on my scanner : I can't
> > get rid of the noise (it scans but it makes a very creepy noise)
> >
> > When using 600 dpi resolution, then the scanner behaves much better. I
> > have some very light grinding but the scan always completes... Except
> > when MaxMotorSpeed is set to 0.86. This seems to be a threshold value
> > (even at 200 dpi). When this is set, most of the time the scanner don't
> > even scan (the "head" doesn't move up the page)
> >
> > The maxMoveSpeed doesn't seem very important : changing it doesn't affect
> > the behaviour in a significant way in my tests.
> >
> > I wonder what I could do now ? Because what I've found is that the
> > working values for maxMotorSpeed are not the same depending on the
> > resolution... And according to the way the parameter tables are set,
> > this doesn't seem to be the way the code was meant to work.
> >
> > I'll sleep on it and start over tomorrow evening
>
> IIRC, values like 150, 300, 600 - multiples of 75 are best to
> test. I can recall the exact meaning of/difference between maxMoveSpeed
> and maxMotorSpeed. So you might have to check and understand
> the source code. You might also check the settings for the N670...
>
Managed to turn the debug on and did some more tests in 300 dpi this time
(following your advice)
0.76 / 0.243 / 300dpi a bit too noisy for my taste
0.80 / 0.243 / 300dpi a bit too noisy for my taste
so not very successful. As far as I can see, the important value is
dMaxMotorSpeed. However, it seems to be only involved in this expression :
ssize = (u_short)((double)CRYSTAL_FREQ / ( mclkdiv * 8.0 *
(double)m_bCM * hw->dMaxMotorSpeed * 4.0
* (double)hw->wMotorDpi));
(it's used in other two, but they seem to be special cases of this one). I
guess from the wording used that this is the step-size of the motor.
So we clearly have a linear equation so I'm rather surprised that the bad
sound doesn't show up on a linear basis... Well, now that I write it I realize
that there's no mention of the resolution I use in the expression. However,
the sound changes when the resolution changes. So I have the feeling I miss
something somewhere....
Stay tuned :-)
stF
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
