dev builds
themes manual
device status forums
mailing lists
IRC bugs
dev guide

Rockbox mail archive

Subject: Re: unreliable buttons? try this

Re: unreliable buttons? try this

From: John Donahue <>
Date: Sat, 3 Jul 2004 22:43:41 -0400 (EDT)

Thanks for trying to solve this problem.
I'm convinced that its only during the release of the button that the problem occurs. Looking at the flashing numbers in the debug section, when a button is released it doesnt always go from say 200 down to 000 cleanly. Sometimes a 100 will briefly show. I've always noticed the mechanism for button release feels a lot more 'gradual' than the 'click' when its pressed. I tried your fix and I indeed wasnt able to trigger a false positive buttonwise. F2 was a little sluggish for me but works. Two rapid pushes of any button didnt register the 2nd push however, so I wouldnt suggest this build for tetris ;)

Maybe instead just check that the reading goes back down to zero before allowing a lesser but non-zeroish value. like allow all the 26F's during a button hold but ignore that 1D2 upon release since it didnt go down to zero first. But, allow number changes that go higher as would result from button combinations. um, also the release of one of the buttons after a combo press should be recognized as ok i guess...

Several people reported unreliable button detection by Rockbox, sometimes there are misinterpretations. You can stop reading if you don't suffer from this.

I have added a kind of "debouncing" to the button code. Now, a button is only valid if the analog port reported several stable readings within a certain "corridor". Please test it and report if this is an improvement. The corridor may be too strict, causing dead buttons. So try the .ajz via rolo first, flash later, or you need the F1-start. ;-)

Join Excite! -
The most personalized portal on the Web!
Received on 2004-07-04

Page template was last modified "Tue Sep 7 00:00:02 2021" The Rockbox Crew -- Privacy Policy