From: Dave Jones <dave_at_hfpacoustical.com>
Date: Fri, 17 Oct 2003 14:05:05 -0500

I just stopped a AJBR unit that has been recording now for almost 9
days. That 9 days of continuous recording! Amazing. Awesome job, Rockbox
team! That's some stable software.

I've posted a "Panic* rec upd: -6 during split time recording" to the
tracker system along with a simple patch to mpeg.c that adds a
if(!global_settings.rec_timesplit) in front of the VBR header creation
code. This seems to clear up the bug, with the loss of record-time VBR
header creation.

I've gone through and listened to the beginning and ending few minutes of
each file for all of the files in the 9 day recordings above. All of the
files are there (each day has 12 two hour recordings) and they all have a
length that looks about right. Of course, there's no VBR header but
otherwise...

Linus, are there any tests you'd like me to run on these files?




Dave





At 11:02 AM 10/15/2003 +0200, you wrote:
>Dave Jones wrote:
>>Should I post this as a bug to the bug tracker system? I'm not sure
>> it's worth much development time if someone is really working on a
>>frame perfect recording split feature.
>
>Yes it is worth it. This bug will still be there even when I commit the
>frame-accurate split.
>
>
>>I could also submit a patch that would turn off the VBR header
>>writing code if the record split feature is enabled. That would be
>>the easiest short-term work-around. Any comments?
>
>I would like you to confirm that all the recorded files in your 6-day
>session are there, intact and correctly named. The -6 panic means that the
>file to be Xing-updated couldn't be found, something that definitely
>shouldn't happen.
>
>Linus

_______________________
Dave Jones P.E.
Senior Project Engineer
HFP Acoustical Consultants Inc.
6001 Savoy Dr., Suite 115
Houston, Texas 77036
713.789.9400 (phone)
713.789.5493 (fax)
www.hfpacoustical.com
_______________________
Received on 2003-10-17