VideoGrabber: dropping last second of sound

[Ben Rubinstein]

Thanks Stephen.

Your suggestion is indeed what we expect we'll have to implement - but before 
I report this in RQCC, and before I roll out that fix, I'd be interested to 
know if anyone else has encountered this.

Your response suggests that there is a known issue with the VG not flushing 
the audio buffer on close, is that correct?  Is there an RQCC number for this? 
  (I've looked through without spotting anything, but I might have missed it.) 
  If this is known, is there lore about the factors - why it's intermittent? 
You mentioned on close, is that on revStopRecordingVideo, or on 
revCloseVideoGrabber - if I could stop the recording on time, but let the VG 
keep rolling longer, that would provide a neater solution.

TIA,

Ben


On 22/10/2010 20:52, stephen barncard wrote:
> yes, I'm guessing that it's not flushing the audio buffer on close. Another
> coding error that's been there for ages.
>
> I'd suggest that your stop sequence include some kind of delay before
> actually stopping the recording, so the talent stops speaking earlier than
> the actual time it stops, and it should all work out. *Send in time* would
> work well for this...
>
> Colin, check out videograbber, it's fairly useful for some limited video
> chores. It's inside every installation (in the mac package
> Contents:Tools:Resources:Sample Projects: folder as Video Capture.rev)  --
> they actually fixed most of it in the last updates.
>
> On 22 October 2010 10:44, Ben Rubinstein<benr_mc at cogapp.com>  wrote:
>
>> We've used LiveCode to create an unattended system in a public environment
>> which is recording several hundred short video clips a day.  The clips are
>> variable length - controlled by the person who is recording a message -
>> typically 10-30 second each.
>>
>> In a small number of cases, the last second of audio is lost.  The actual
>> sound track is shorter than the video track; and in the very small sample of
>> cases I've so far being able to inspect (2 clips) it was almost exactly 1
>> second shorter.
>>
>> Has anyone else encountered this or a similar problem?
>>
>> TIA,
>>
>> Ben