Hey, it'd be nice to have an adaptative time for the DPMS stuff. This kind of stuff is explained by Matthew Garrett here: http://mjg59.livejournal.com/106581.html Basically, when the user uses the computer, the DPMS should “never” activates. But when the user isn't there, it should activate as fast as it can. And this is wether the computer is on AC or battery, and “user is there” doesn't mean “user inputs something”. For example, when one wants to read a long webpage, or reads IRC, or stuff like that, there won't be any user input, but the display shouldn't shut off. One way to manage to implement this would be to start with a short timeout for user input. If the timer timeouts and no user input, stay off. If there's user timeout short after the DPMS activation, then restart the display and double the timeout. Double again if it happens again, etc. There might be improvements on the workflow, but basically the keyword is “adaptative” :)
(In reply to comment #0) > Hey, > > it'd be nice to have an adaptative time for the DPMS stuff. This kind of stuff > is explained by Matthew Garrett here: http://mjg59.livejournal.com/106581.html > > Basically, when the user uses the computer, the DPMS should “never” activates. > But when the user isn't there, it should activate as fast as it can. And this > is wether the computer is on AC or battery, and “user is there” doesn't mean > “user inputs something”. For example, when one wants to read a long webpage, or > reads IRC, or stuff like that, there won't be any user input, but the display > shouldn't shut off. > > One way to manage to implement this would be to start with a short timeout for > user input. If the timer timeouts and no user input, stay off. If there's user > timeout short after the DPMS activation, then restart the display and double > the timeout. Double again if it happens again, etc. > > There might be improvements on the workflow, but basically the keyword is > “adaptative” :) Thanks for the report. i agree this is very nice time, but yes as you said it is a bit late now, let's focus on stabilizing 0.8. BTW: I also thought about adding some code to disactivate all sleep actions when the user is on the computer using the bluetooth protocol, no seriously this is nice, i don't remember where if found a program like that to automatically unlock the screen-saver when the user approach for the its computer.
Frankly I don't think that this is really a good idea or feasible at all. As xfpm already does enough poorly, let's not add another questionable feature but instead fix what we got.
Well, with light-locker I have to admit I would fine that really helpful, actually :) I set the screen off setting to 1 minute on battery in order to save as much as possible, but that also means I need to unlock everytime, which is somehow painful.
Right, but you can either increase the delay until it locks or just not let light-locker automatically lock your session upon timeout ;)
(In reply to Simon Steinbeiss from comment #4) > Right, but you can either increase the delay until it locks The point is to save battery as much as possible. > or just not let > light-locker automatically lock your session upon timeout ;) while still keeping a secure laptop in case it's left unattended. There's no static tradeoff, thus the interest for dynamic one :)