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#6641 : Speex plays slightly faster than realtime



FS#6641 - Speex plays slightly faster than realtime

Attached to Project: Rockbox
Opened by Paul Louden (darkkone) - Saturday, 10 February 2007, 02:01 GMT
Task Type Bugs
Category Codecs
Status Closed
Assigned To No-one
Operating System Gigabeat F/X
Severity Low
Priority Normal
Reported Version
Due in Version Undecided
Due Date Undecided
Percent Complete 100%
Votes 0
Private No


On my Gigabeat speex plays slightly faster than realtime (If I start a speex file on my PC and on the device, in sync, after a short time they fall out of sync with the device playing sounds earlier than the computer, as well the timer looks slightly fast).
This task depends upon

Closed by  Paul Louden (darkkone)
Saturday, 10 March 2007, 01:17 GMT
Reason for closing:  Duplicate
Comment by Paul Louden (darkkone) - Monday, 12 February 2007, 00:42 GMT
The file is 16khz Mono, 6hrs 55minues (~85mb) and created using the most recent command line encoder without parameters.

This bug can be reproduced on iHP-120 as well, so it's not ARM specific, at least. (Also happens on iPod Nano, for what little that's worth). The same is true of http://www.rockbox.org/tracker/task/6642

I also reproduced this on a 2mb file.
Comment by Paul Louden (darkkone) - Wednesday, 14 February 2007, 23:49 GMT
A little more 'interesting' behaviour: Sometimes the seconds counter will flicker between two times, like 23s, 24s, 23s, 24s, 23s, 24s, then continue counting like normal. This will usually happen withing about 1/3 or 1/2 of a second's time. A quick alternating when it should just have incremented.
Comment by Mikael Magnusson (mikaelh) - Thursday, 15 February 2007, 15:22 GMT
I think the flickering has happened to me sometimes when just pausing, so it might not be speex-specific.
Comment by Paul Louden (darkkone) - Saturday, 10 March 2007, 01:11 GMT
Actually, the problem seems to be related to resampling. This problem doesn't happen with 44.1khz files.