clicking noise upon "pause" and "continue"

Bug #694259 reported by frizzle21
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
Banshee
New
Undecided
Unassigned
Rhythmbox
New
Undecided
Unassigned

Bug Description

when pausing the playback via sound menu, rhythmbox often produces a clicking noise, same when continuing playback.
this is a well known problem in digital audio, it is circumvented by either emulating tape-stop and tape-start, or by applying brief, virtually inaudible fades to the ending and beginning of every playback period / sample.

i would find tape-stops upon pause and tape starts upon continue quite mean, that would give ubuntu some face, but that would be a little outside the box..

Revision history for this message
frizzle21 (frederik-nnaji) wrote :

Workaround:
enable crossfade engine in Rhythmbox.

Note:
the crossfade engine configuration needs some love..
i have only a few millimeters slider to set the crossfade time usefully to something between 0 and 2 seconds

i think the crossfade engine needs to focus on everything below 15sec, instead of offering a 60sec crossfade.
what are the use cases of such long crossfades?

Revision history for this message
Conor Curran (cjcurran) wrote :

Considering MPRIS does support volume change this is something that could be working on easily.
thx Frederick.

Revision history for this message
frizzle21 (frederik-nnaji) wrote :

Anytime, Conor!
let me know in case you need any more info to close this bug..
happy new year ;)

Conor Curran (cjcurran)
Changed in indicator-sound:
importance: Undecided → Wishlist
status: New → Opinion
assignee: nobody → Conor Curran (cjcurran)
Revision history for this message
frizzle21 (frederik-nnaji) wrote : Re: [Bug 694259] Re: clicking noise upon "pause" and "continue"

Opinion:
EMU Samplers used to autocreate microfades at every start|stop playback
event.. in the millisecond range.
Just to make sure that non-continuous wave-information doesn't make the
hardware "crackle", "pop", "plop" or "click" ;)

What do you need the opinion on? Whether to implement microfades at all,
whether to implement them in MPRIS, Rhythmbox or PA or whether this bug can
be confirmed in the first place?

I stand to my suggestion to use Tape-stops in Rhythmbox's case, but that's
not a generic solution i know.

Revision history for this message
Conor Curran (cjcurran) wrote :

Okay so I would prefer for the players to handle this internally. Spamming the dbus with exponential fades everytime a direct user transport action occurs is not ideal. The second part to this would be to build a pulse module to do an automatic fade on mute.

Changed in indicator-sound:
status: Opinion → Confirmed
Revision history for this message
frizzle21 (frederik-nnaji) wrote :

yeah, a sane setting such as X-Fade engine set to something < .5 sec in Rhythmbox should fix this for Maverick users.
Would it be ok to ship that sort of setting per default?
The fade on mute idea sounds mature and very elegant..

Conor Curran (cjcurran)
tags: added: design
Conor Curran (cjcurran)
Changed in indicator-sound:
status: Confirmed → In Progress
Conor Curran (cjcurran)
tags: removed: design
Conor Curran (cjcurran)
Changed in indicator-sound:
status: In Progress → Triaged
Revision history for this message
Lars Karlitski (larsu) wrote :

Yeah, this is something for the players to handle.

no longer affects: indicator-sound
Revision history for this message
Chow Loong Jin (hyperair) wrote :

Odd, I thought the clicking problems were purely hardware-related, and fixed with the whole bunch of hda_power_save bugfixes.

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.