maemo.org - Talk

maemo.org - Talk (https://talk.maemo.org/index.php)
-   Applications (https://talk.maemo.org/forumdisplay.php?f=41)
-   -   mtPaint 3.20 Revisited (https://talk.maemo.org/showthread.php?t=17712)

unique311 2008-03-20 18:21

Re: mtPaint 3.20 Revisited
 
nice, the maintainer is here...
did you find my sourceforge post?



Quote:

Originally Posted by wjaguar (Post 157826)
This may be a more general X problem - because I get something very like this on a desktop PC running a Slackware 11.0 derivative.

After noticing one time that random strokes in mtPaint get interrupted for no apparent reason, I investigated the problem - and this is what I found out:
When some programs (Kate for one) are running in the background, and mouse is moved vigorously enough with a button pressed, X may generate spurious button-release events, and follow them by equally spurious button-press events some random time later. Any X client gets affected, and no app-level workaround seems possible - I used xev to dump X event stream, and the spurious events appearing in there are absolutely indistinguishable from normal ones.
And in mtPaint, a button release event terminates current stroke, and a button press event starts a new stroke.

-= With best regards, Dmitry Groshev, maintainer of mtPaint =-


Karel Jansens 2008-03-20 18:51

Re: mtPaint 3.20 Revisited
 
Quote:

Originally Posted by wjaguar (Post 157826)
This may be a more general X problem - because I get something very like this on a desktop PC running a Slackware 11.0 derivative.

After noticing one time that random strokes in mtPaint get interrupted for no apparent reason, I investigated the problem - and this is what I found out:
When some programs (Kate for one) are running in the background, and mouse is moved vigorously enough with a button pressed, X may generate spurious button-release events, and follow them by equally spurious button-press events some random time later. Any X client gets affected, and no app-level workaround seems possible - I used xev to dump X event stream, and the spurious events appearing in there are absolutely indistinguishable from normal ones.
And in mtPaint, a button release event terminates current stroke, and a button press event starts a new stroke.

-= With best regards, Dmitry Groshev, maintainer of mtPaint =-

So it's impossible to have a decent drawing application on a Linux/X tablet?

Nice...

unique311 2008-03-20 18:55

Re: mtPaint 3.20 Revisited
 
from what I read on the mtpaint forum on sourceforge.
The dev might be getting a n810 pretty soon.
which might make for a really nice release of mtpaint on the tablet.

I love it.
but on the other hand I am an artist.

Cheap cintiq anyone?

Quote:

Originally Posted by Karel Jansens (Post 157899)
So it's impossible to have a decent drawing application on a Linux/X tablet?

Nice...


awe215 2008-03-20 20:46

Re: mtPaint 3.20 Revisited
 
1 Attachment(s)
Here is a quick sketch made with mtpaint. I have been trying to sketch from life with this tablet. Self portraits done a with mirror seem impossible! Here is a quick self portrait... of my finger!

Bundyo 2008-03-20 21:19

Re: mtPaint 3.20 Revisited
 
Quote:

Originally Posted by wjaguar (Post 157826)
This may be a more general X problem - because I get something very like this on a desktop PC running a Slackware 11.0 derivative.

After noticing one time that random strokes in mtPaint get interrupted for no apparent reason, I investigated the problem - and this is what I found out:
When some programs (Kate for one) are running in the background, and mouse is moved vigorously enough with a button pressed, X may generate spurious button-release events, and follow them by equally spurious button-press events some random time later. Any X client gets affected, and no app-level workaround seems possible - I used xev to dump X event stream, and the spurious events appearing in there are absolutely indistinguishable from normal ones.
And in mtPaint, a button release event terminates current stroke, and a button press event starts a new stroke.

-= With best regards, Dmitry Groshev, maintainer of mtPaint =-

So that means that there should be a problematic application/process in the background and if removed mtPaint will run fine?

Strangely enough the ones i observed do not include a button press event - only a release one. The stroke stops and the mtpaint mouse pointer stays there, while the X one continues moving.

Quote:

Originally Posted by Karel Jansens (Post 157899)
So it's impossible to have a decent drawing application on a Linux/X tablet?

You judge too fast. Read first :)

Benson 2008-03-20 21:39

Re: mtPaint 3.20 Revisited
 
Quote:

Originally Posted by awe215 (Post 157928)
Here is a quick sketch made with mtpaint. I have been trying to sketch from life with this tablet. Self portraits done a with mirror seem impossible! Here is a quick self portrait... of my finger!

Lefty, eh?

Karel Jansens 2008-03-20 21:51

Re: mtPaint 3.20 Revisited
 
Quote:

Originally Posted by Bundyo (Post 157944)
You judge too fast. Read first :)

But I allways have programs running in the background. It's Linux.

Bundyo 2008-03-20 22:09

Re: mtPaint 3.20 Revisited
 
It was said "some programs" :)

Anyway, on the topic - i ran a xev on it and it seems that after time any events just stop until the stylus is released. No presses/releases. Weird.

wjaguar 2008-03-21 08:14

Re: mtPaint 3.20 Revisited
 
Quote:

Originally Posted by Bundyo (Post 157982)
Anyway, on the topic - i ran a xev on it and it seems that after time any events just stop until the stylus is released. No presses/releases. Weird.

The actual mode of breakdown may depend on device.
On further investigation, the source of spurious button events in my case seems to be the kernel itself, and not X: 'evtest /dev/input/event2' shows "Key" events (left button releases and presses) appearing from nowhere when I move mouse with left button pressed. (Actually, when evtest is attached, the problem manifests with a vengeance - nearly any mouse movement with button pressed causes a spurious button release.)

P.S.: The 'evtest' utility is from OpenSUSE 'input-utils' package.

Bundyo 2008-03-21 08:53

Re: mtPaint 3.20 Revisited
 
Okay, I'll try to test it too.

On a side note - I'm also using openSUSE - 10.3 x86_64 and mind you the scratchbox is working without any problem, the installation is rather tricky since it tests if i'm running 64bit platform and quits. If you try to install it - look here:

http://en.opensuse.org/Maemo


All times are GMT. The time now is 03:09.

vBulletin® Version 3.8.8