Also happens on my Benchmark DAC-1 connected via digital coax; actually the red error light briefly lights up when skipping to the next 24/96 track, indicating loss of lock.

To add some extra meat: I suspect this is a Touch firmware issue. When skipping
from one 24/96 track to the next, whatever is sent in the small gap between one
track and the next is dropping back to 44.1k - this can be seen on the display
of DACs with that capability. I think that the digital silence that is sent
should stay at the rate of the previously playing track.
The impact of the bug varies from mildly irritating to devastating, depending
on the users audio system downstream of the Touch when using an external DAC.
Those who are using a DAC straight into a power amp can get alarming pops and
clicks as the DAC loses then regains lock. This happens on a variety of DACs.

Other players do not seem to exhibit this problem @ 24/48 for example (or the
TP at 24/96).

*** This bug has been confirmed by popular vote. ***

I would like to add that a short dropout occurs also when the touch goes from the end of one track to the next track, inserting a tiny gap in the sound. I wanted to distinguish this from the issue where pressing the >> key (skipping ahead) causes the Touch to briefly drop to 44.1. When I get the tiny gap when it should segue seamlessly, I do not see the light change on my Cambridge DacMagic, whereas I do see the light briefly jump to 44.1 when I jump to the next track using the >> key on the remote. So I can't be absolutely sure that the dropout I get during the segue from the end of one track to the beginning of the next is symptomatic of the jumping ahead to the next track with the remote problem (but I suspect it is and don't think it's proper to file a separate bug report but will if asked).

Also happening on a PS Audio DLIII using coax port. Loss of lock indicated by lock LED briefly extinguishing when skipping between tracks. No audible pop or click that I can detect. DAC is not directy connected to a power amp.

Just to add my TACT displays the intermittant loss of lock/dropping back to 44.1 problem as well.

I think I might have found where this is happening. In decode_alsa.c the decode_alsa_stop function sets the sample rate to 44100. If I'm not mistaken this gets called when a song finishes. 

I'm going to try commenting this out tonight and see if that fixes the problem. 

The big question is, why is the sample rate getting set to 44.1? If the sample rate is left alone is it going to break something else?

BTW exactly the same thing happens in decode_portaudio.c.

John S.

Richard: maybe you can comment on why decode_alsa_stop sets the sample rate to 44.1?

(In reply to comment #4)
> I would like to add that a short dropout occurs also when the touch goes from
> the end of one track to the next track, inserting a tiny gap in the sound. I
> wanted to distinguish this from the issue where pressing the >> key (skipping
> ahead) causes the Touch to briefly drop to 44.1. When I get the tiny gap when
> it should segue seamlessly, I do not see the light change on my Cambridge
> DacMagic, whereas I do see the light briefly jump to 44.1 when I jump to the
> next track using the >> key on the remote. So I can't be absolutely sure that
> the dropout I get during the segue from the end of one track to the beginning
> of the next is symptomatic of the jumping ahead to the next track with the
> remote problem (but I suspect it is and don't think it's proper to file a
> separate bug report but will if asked).

I retract my original comments, above, in comment #4. It turns out the gaps were actually at the beginning of the source material I was using. There is no gapless problem with hirez. I tested this again with several difference sources, some of which I created... sorry for muddying the waters. If a moderator wants to nuke the comment go ahead.

I tried commenting out the offending line and made a new image for the touch and tried it out and it does fix the problem. The S/PDIF out stays at 96 between two 96 tracks. 

I tried letting one track go to the next normally, skipping to the next track with ">>" and moving around within a track using the slider and they all worked flawlessly.

I didn't find any problems due to this change. I'll do some more testing but so far there doesn't seem to be a downside.

(In reply to comment #10)
> I tried commenting out the offending line and made a new image for the touch
> and tried it out and it does fix the problem. The S/PDIF out stays at 96
> between two 96 tracks. 
> I tried letting one track go to the next normally, skipping to the next track
> with ">>" and moving around within a track using the slider and they all worked
> flawlessly.
> I didn't find any problems due to this change. I'll do some more testing but so
> far there doesn't seem to be a downside.

Well done John - seriously good work. Hope fully we can get a firmware fix from Richard et al soon...?

Don't want to hassle anyone, but when is the bug meeting going to assign this ?
:-)

John, I cannot see how decode_alsa_stop() will get called as part of normal track-to-track transition. It gets called by decode_output_end() and decode_output_flush(). decode_output_end() gets called only by decode_stop()/decode_stop_handler() and neither decode:stop nor decode:flush should get called as part of normal track-to-track transition. Perhaps I am missing something.

Alan notes he can now see how this might happen, but recalls some other reason why this fix might have other problems.  He will review it further before committing it.

I'm also using a Benchmark DAC1, and I'm getting clicks when pausing, skipping, or ff/rewinding in the middle of a track, even though all my files are 44.1khz FLAC's. I'm on an SB Touch and I've had the same issue since SB3, in all versions of SBS that I can remember. I'm on 7.5.1 now.

Alan - any movement on this? It's been a while...
Touch users are increasing testing with 24/96 files and some DAC's seem to not like the current firmware behaviour.
many thanks
Phil

 == Auto-comment from SVN commit #9061 to the jive repo by ayoung ==
 == http://svn.slimdevices.com/jive?view=revision&revision=9061 ==

bug 16146: During the inter-track skip, the Touch seems to revert to 44.1 
Do not revert to 44100 when stopping output (such as happens with a skip)
but remember to force 44100 for loopback use.

I add my vote: if playing 96k tracks, and I skip between tracks, the Touch momentarily switches to 44.1K and then switches back to 96K. Causes an audible "pop" to be produced.

(In reply to comment #18)
> I add my vote: if playing 96k tracks, and I skip between tracks, the Touch
> momentarily switches to 44.1K and then switches back to 96K. Causes an audible
> "pop" to be produced.

look at comment #17 it's says that a fix is provided in ver 9061 .
So it is the latest 7.6 beta fw  .

I just downloaded lateset 7.6 and it is still a problem for me--not fixed.

(In reply to comment #20)
> I just downloaded latest 7.6 and it is still a problem for me--not fixed.

Did you *really* wait for the Touch to upgrade it's firmware ?

After you install 7.6 it will download new fw for the Touch this can take up to 12hr, then you will be prompted for upgrade.
Maybe you get an upgrade next time you boot the computer running squeezeboxserver ?

On your Touch go to settings > advanced > about what firmware version do you see ?
Since todays upgrade it should be 7.6.0 r9103

Hi,

I have just purchased a Touch and have noted the same audio popping when skipping tracks, naturally migrating from one track to the next or fast forwarding/rewinding but only with 24bit/96KHz FLAC files.  (To be clear this does not happen with WAVE nor MP3 files).

It also does it when the DAC changes to or from a high resolution 'format' e.g., when transitioning from a WAVE/MP3 to 24/96 FLAC file (or the reverse).  For clarity, the audio popping is not heard when transitioning between MP3 and WAVE files.

I tried the latest Touch firmware (from the Squeezebox Server nightly build SqueezeboxServer-7.6.0-31483.exe) and can confirm that the audio popping when skipping, naturally migrating or ffwd/rrwd has been elimited, however, the problem is still there when transitioning from a WAVE (or MP3) file to 24/96 FLAC or the reverse.

I do not use an external DAC, this audio popping is present on the Touch's analogue audio outputs.

Hmm this is not backported to 7.5.2 or 7.5.3 ?

I know that choff001@centurytel.net  said that this is not fixed for him. Please can we have reports from any others for whom this was an issue and who have tried 7.6 r9061 or better.

(In reply to comment #24)
> I know that choff001@centurytel.net  said that this is not fixed for him.
> Please can we have reports from any others for whom this was an issue and who
> have tried 7.6 r9061 or better.

This issue have been solved for me since r9061. I am using flac files. 
Still I have an issue with a small "click" when actually changing from say 44.1 to 48kHz or 96kHz. (My Duet is absolutely silent in this regard.)

best,
Paal

I'm using a Touch with the latest official updates, and don't have the issue anymore.

Closing due to comments.