Warning: mysqli_real_connect(): Headers and client library minor version mismatch. Headers:50625 Library:50543 in /sites/rockbox.org/flyspray/adodb/drivers/adodb-mysqli.inc.php on line 108 FS#12883 : Opus becomes unresponsive at lowest CPU frequency



FS#12883 - Opus becomes unresponsive at lowest CPU frequency

Attached to Project: Rockbox
Opened by Mark Mitchinson (xanikseo) - Saturday, 20 July 2013, 15:07 GMT
Task Type Bugs
Category Codecs
Status Assigned
Assigned To amaury pouly (pamaury)
Operating System Sansa Fuze+
Severity Low
Priority Normal
Reported Version Release 3.12
Due in Version Undecided
Due Date Undecided
Percent Complete 0%
Votes 0
Private No


After playing a few opus files (~128 kbps) the system becomes so unresponsive that it cannot decode the file in real time, and the device cannot respond to any input within a reasonable timeframe (have to hard power off).

This does not occur when you set the cpu frequency beyond the highest setting in the debug menu!

To reproduce:
play opus files in a playlist with normal cpu scaling
on beginning of 2nd or 3rd file, system will become unresponsive.
This task depends upon

Comment by Mark Mitchinson (xanikseo) - Saturday, 20 July 2013, 15:12 GMT
I am not certain, but I have a gut feeling that something during the frequency change is botched, rather than the low frequency itself being a problem.
Comment by Mark Mitchinson (xanikseo) - Sunday, 08 September 2013, 11:44 GMT
this seems to be fixed with the recent updates
Comment by Mark Mitchinson (xanikseo) - Tuesday, 10 September 2013, 09:25 GMT
scratch that. I must have just been lucky. I charged the thing up and now the problem is as it was before
Comment by amaury pouly (pamaury) - Sunday, 12 January 2014, 21:00 GMT
Does the problem still exist ? Can you link to a file which exhibit the problem so I can test it ?
Comment by Mark Mitchinson (xanikseo) - Sunday, 12 January 2014, 21:09 GMT
I haven't had access to the Internet for a while, but my version is still fairly up to date (within three weeks old I would say). the issue still occurs, but I will update the version the next opportunity I get to say for sure that it occurs with the current version. The issue occurs in the transition between any two tracks as far as I can tell, always with the cpu frequency down. It is easier to reproduce when crossfade is used. The majority 90% of my library is Opus 128kbps, mostly on SD. I should mention, I'm playing back with 48khz sample rate.
Comment by Mark Mitchinson (xanikseo) - Monday, 13 January 2014, 22:20 GMT
I can confirm that the bug is still there with the latest git, also when using the corresponding bootloader. The first track change usually goes smoothly, however the second usually results in unresponsiveness