[sane-devel] Patch for sane-fujitsu: Fixes fi-5110eox ignoring page-width]

[m. allan noah]

On 7/13/08, m. allan noah <kitno455 at gmail.com> wrote:
> On 7/12/08, spammed <spammed at get2net.dk> wrote:
>  > > then --overscan is the only option
>  > > which is exposed by the backend for this scanner, but which does not
>  > > actually work. Maybe the option could be hidden when the scanner returns
>  > > (os_x_basic, os_y_basic) = (0, 0), as this scanner does?
>  >
>  > yes- i dont have a good way to detect the overscan capability, other
>  > than the existence of the page code that is used to enable it, which
>  > also enables other things. do you get an error when you set overscan?
>  > if not, then the problem might be that your os_x/y_basic need to be
>  > overridden, like the 4120c?
>  >
>  >
>  >
>  >  There is no error. I will try overriding the settings.
>  >
>  >  I copied the code for the 4120C and tried several values around 2 mm and
>  > 2.5 mm (in 600dpi units) with no change. Do you think there is a slim chance
>  > that the values need to exactly match what the scanner expects (the missing
>  > vpd-values) to activate overscan, and otherwise they are just ignored? How
>  > were the values for the 4120C established?
>
>
> those values came from trial and error. if you dont see a bit of the
>  background before the paper with overscan enabled, then it must not
>  work on this machine. i'll just disable it.

i've just committed a patch which disables overscan, and centers the
scan area in software for your scanner. i used some different values
for the math, so please check that it works.

i also added workarounds for firmware errors in the fi-5330 and
fi-5650 (brightness/contrast), and fi-5900 (background color).

allan
-- 
"The truth is an offense, but not a sin"