Sound level jumps to max at the end of the song

[hfujita]

Reported by Nicolas VEYSSIERE

there is a strange thing : at the end of a song the sound level goes to where it is currently to max in less of 2 seconds, then the music stop. When playing a song after that the sound level is normal.

• iHome IW2 with mpd

[iXo]

I would like to give more informations : the sound level problem doen't occurs after earch song played, but when I tell MPD to stop song playback (stop button), this is a this moment that the sound level goes to the max (then no more sound as the music is stopped). After that if restarting playback, the sound level is normal.

[josla972]

Same issue here. Using lightspark on youtube.

[Toby-Haynes]

Looking at the logs, there is actually a SET VOLUME <100%> sent when the device is switched over.

Music is playing at this point... ( 332.683| 0.000) D: [raop-sink] raop_client.c: Resending 1 packets starting at 8384 ( 332.685| 0.001) D: [raop-sink] module-raop-sink.c: Received control packet. ( 332.685| 0.000) D: [raop-sink] raop_client.c: Resending 1 packets starting at 8385 ( 332.685| 0.000) D: [raop-sink] module-raop-sink.c: Received control packet. ( 332.685| 0.000) D: [raop-sink] raop_client.c: Resending 1 packets starting at 8386 ( 335.107| 2.422) D: [raop-sink] module-raop-sink.c: Received timing packet. ( 337.207| 2.099) D: [raop-sink] module-raop-sink.c: Received timing packet. ( 339.534| 2.327) D: [raop-sink] module-raop-sink.c: Received timing packet.

Move the music stream to the internal card...

( 342.034| 2.500) D: [raop-sink] module-raop-sink.c: Received timing packet. ( 342.549| 0.514) D: [raop-sink] module-raop-sink.c: Received control packet. ( 342.549| 0.000) D: [raop-sink] raop_client.c: Resending 1 packets starting at 9638 ( 343.694| 1.145) I: [lt-pulseaudio] source-output.c: Freeing output 4 "Peak detect" ( 343.694| 0.000) D: [lt-pulseaudio] module-raop-sink.c: Volume adjusted: orig=62965 adjusted=63778 ( 343.694| 0.000) D: [lt-pulseaudio] module-raop-sink.c: Requested volume: front-left: 62965 / 96%, front-right: 62965 / 96% ( 343.694| 0.000) D: [lt-pulseaudio] module-raop-sink.c: Got hardware volume: front-left: 63778 / 97%, front-right: 63778 / 97% ( 343.695| 0.000) D: [lt-pulseaudio] module-raop-sink.c: Calculated software volume: front-left: 65536 / 100% / 0.00 dB, front-right: 65536 / 100% / 0.00 dB ( 343.695| 0.000) D: [lt-pulseaudio] raop_client.c: volume=63778 db=-0.708542 ( 343.695| 0.000) D: [lt-pulseaudio] rtsp_client.c: Sending command: SET_PARAMETER ( 343.695| 0.000) D: [raop-sink] sink.c: Requesting rewind due to started move

[sybrenstuvel]

I have this same issue, with BOSE SoundTouch 20 speakers, except that the volume change happens just before the sound stops, blasting the last half-second at full volume. That, and the volume at the device stays at 100%, so everything you play afterwards is loud too.

[lawabb]

My Yamaha Rx-S600D is working well apart from this bug. Anyone know if any progress on it? From Bug 42804 below. Is this related?

Cheers, Lawrie

While testing the packet buffer I noticed that the volume level set via PA (I use the gnome applet shipped with ubuntu 12.04) is not linear. Looking at the code and comparing it to 2, I think that the volume calculation is wrong. In the code, volume is calculated between VOLUME_MIN (-144) and VOLUME_MAX (0). I think it should be calculated between (-30.0) and (0.0), as seen in 2. Fixing this should be easy.

Greetings,

Matthias

[josla972]

I have not had this issue recently. I stopped using both gnash and lightspark for youtube though (HTML5 seems to work fine). Please try and see what result you get. You compile it by yourself right? Then it should be easy to try it.

[Toby-Haynes]

I built the v5.99.2 from git and have not seen this issue with this level. Tried quodlibet, mpd, totem and firefox (soundcloud) without any problems.