'Twas brillig, and David Henningsson at 20/06/11 10:13 did gyre and gimble: > On 2011-06-20 10:31, Colin Guthrie wrote: >> Interesting. Smells like someone is patching something somewhere then. >> Like I say I don't have this problem here - volume keys will have their >> upper limit @100% >> >> I'd still be generally in favour of limiting the keys to 100% but >> ultimately it's up to the Dekstop environments to decide what they think >> is best. >> >> So I'd get in touch with the relevant gnome people and ask the question: >> Should the keys limit the volume to 0dB, but with the UI allowing up to >> +11dB? > > We'll just add a compressor plugin, so everything sounds louder ;-) Hehe! > Just kidding. Anyway, if I were a UI designer and got that proposal, I > would probably scratch my head and wonder why the PulseAudio folks > couldn't decide on one maximum volume only... Well it's about use case. In a typical setup, the user shouldn't *have* to go above PA_VOLUME_NORM. They may need to in some special cases and we don't want to make those special cases too tricky to enable, but we don't want people accidentally enabling them either as it leads to questions exactly like the one asked in this thread. So we need to make some kind of compromise. In my mind I see three possible solutions: 1. Just stick to 0dB for the media keys handling, but handle things gracefully if the volume is >0dB (e.g. volume up == noop, volume down = step down normal amount (i.e. don't reset to 0dB-step immediately on first press). 2. Just go up to +11dB and don't do anything different. Accept that upstream will get several death threats (or rather "why does PA do this? It's dump, you guys are idiots" emails). 3. Allow going up to 0dB via key repeats, but require a stall and second-press of the buttons to get >0dB, change the OSD to red or some other warning colour to let the user know they are overdriving and provide other UI hints in e.g. dock volume controls that this is the case and give the user enough feedback to be able to reset it if needs be. 1. is clearly the simplest, 2, clearly sucks, 3 might be the most holistic solution. Everyone will likely have a different opinion on what is best. Perhaps this is something you can add in to the usability study with jack detection and expectations thereof? Col -- Colin Guthrie gmane(at)colin.guthr.ie http://colin.guthr.ie/ Day Job: Tribalogic Limited [http://www.tribalogic.net/] Open Source: Mageia Contributor [http://www.mageia.org/] PulseAudio Hacker [http://www.pulseaudio.org/] Trac Hacker [http://trac.edgewall.org/]