Re: touchscreen calibration again

Hello,

In entered my raw measures in OpenOffice.org Calc and displayed some graphs.
The results confirm what we already know: x coordinates are pretty
linear, but y coordinates are less linear.

However, y coordinates don't look so weird... they just look like they
went through a lens. So that should be something Lau can handle...

See
http://michaelo.free.fr/pda/h2210/ts_harvest/ts_harvest_michaelo1.pdf or
http://michaelo.free.fr/pda/h2210/ts_harvest/ts_harvest_michaelo1.sxc
(OpenOffice.org calc).

Cheers,

    Michael.

> Hello,
>
> Here's the program (at last!):
> http://michaelo.free.fr/pda/ipaq/starterkit/ts_harvest_may14.tar.bz2
>
> Just copy the directory somewhere in your rootfs, and read the README
> file. It's easy to run, and the 'run' wrapper also takes care of
> loading a shamcop_ts module without Alain's initial linearization code.
>
> You will have 120 points to input. Please be as accurate as possible,
> otherwise it won't help. Then, send your output file to Lau (feel free
> to copy the list... URL download links are preferred, but plain text
> attachments are accepted).
>
> Have fun!
>
> :-)
>
> Michael.
>
>>
>>>
>>> I'm going to fix this by checking time information from the
>>> touchscreen device. I will trash all data recorded before I draw the
>>> crosshair corresponding to the new point.
>>>
>> Unfortunately, this doesn't work. The input system gives me the time
>> the sample is delivered to me, not the time it was recorded.
>> I'll try to find something else, provided it is simple. Otherwise,
>> I'll ask users to be careful...
>>
>> Cheers,
>>
>> Michael.
>>
>
>

-- 
Michael Opdenacker
http://opdenacker.org/