Axel Dörfler wrote: > "Truls Becken" <truls.becken@xxxxxxxxx> wrote: > > Why not make the switch to full screen in itself disable the screen > > saver? That way, no individual application and/or graphics library need > > to think about this problem. > > You could use normal windows (like WonderBrush, or Firefox) in full > screen, too. No reason to turn off the screen saver, then. It could be a > window flag (or connected to a window in some way), though, so that the > screen saver is always disabled when that particular window is visible. I think we need a more generic approach, which will also have to work when we implement any power saving stuff. An application needs to be able to tell the system that it's "doing something". Obviously, all apps are doing something, but there has got to be a way for the system to determine if that is something the user cares about or not. For example, the SETI@Home client my peak the CPUs, but it could be regarded different from a long running rendering of some 3D application. Ideally, I'd like to be able to configure this, but have some useful defaults. Possibly even depending on daytime. For example, if the MediaPlayer is playing a movie, no matter if full screen or not, the screen saver would never go on and the system would never go into stand-by. When it plays audio, one may want to go into stand-by if the system has been idle for an hour and if it's past 23:00 or something (with a way to configure this). A function to temporarily disable the screen saver is only the first step of a frame work where applications can fine tune certain behavior of the OS with regards to the system being considered "idle". I think such a frame work will certainly be needed. Best regards, -Stephan