Odd audio issue when re-joining split clips

[Adam Gold]

So this is interesting...

When using the Sony ru_util Utility to join and import clips that have been split due to the 4GB FAT32 file size limitation, twice in recent days I've found an odd audio issue: the audio from one subclip gets attached to a different subclip. That is, the audio from part one will suddenly begin replaying where part two starts (or would have started had they not been rejoined). This is only a sporadic issue and is not present on the original subclips, which play fine when imported into Premiere and laid sequentially on the timeline. I assume this has something to do with that tiny 1 or 2 KB file they tell you not to remove which apparently serves as a sort of index for the split clips. I assume it has been somehow corrupted and thus does not reassemble the files correctly.

It's happened on two different MRC1ks on two different cards on two different cams on two different days.

I also found that while it takes about 20 minutes for the Sony SW to join and import the clips as one large 1-hour file, to transfer the individual files using a straight Windows file copy just takes a couple of minutes.

Anyone else heard about or had this experience? Opinions or thoughts?

[Steve Renouf]

Originally Posted by Adam Gold

So this is interesting...

When using the Sony ru_util Utility to join and import clips that have been split due to the 4GB FAT32 file size limitation, twice in recent days I've found an odd audio issue: the audio from one subclip gets attached to a different subclip. That is, the audio from part one will suddenly begin replaying where part two starts (or would have started had they not been rejoined). This is only a sporadic issue and is not present on the original subclips, which play fine when imported into Premiere and laid sequentially on the timeline. I assume this has something to do with that tiny 1 or 2 KB file they tell you not to remove which apparently serves as a sort of index for the split clips. I assume it has been somehow corrupted and thus does not reassemble the files correctly.

It's happened on two different MRC1ks on two different cards on two different cams on two different days.

I also found that while it takes about 20 minutes for the Sony SW to join and import the clips as one large 1-hour file, to transfer the individual files using a straight Windows file copy just takes a couple of minutes.

Anyone else heard about or had this experience? Opinions or thoughts?

Was the transfer via a card reader or via the MRC1K firewire? Are the two cards that this happened on the same type of card? Have you used other (different) cards which have been OK? I've not come across this problem but I always transfer via a cardreader. Do you delete the files on the MRC1K or on the PC? I suspect, like you, that maybe the index file has got messed up somehow.

I don't have a ready solution but it may help to go through an elimination process.

[Adam Gold]

Same brand of card -- Sandisk Extreme IIIs -- but two different cards, transferred via external reader. Other cards -- both this brand and others -- were fine, as were these cards on other projects. When I clear the cards for the next shoot, I do it on the MRC, but only as a DELETE ALL, not a re-format.

As this is multicam and I can afford to have a momentary glitch on one cam at a time, I'm just using the individual clips on the timeline and everything's fine.

[Steve Renouf]

Originally Posted by Adam Gold

Same brand of card -- Sandisk Extreme IIIs -- but two different cards, transferred via external reader. Other cards -- both this brand and others -- were fine, as were these cards on other projects. When I clear the cards for the next shoot, I do it on the MRC, but only as a DELETE ALL, not a re-format.

As this is multicam and I can afford to have a momentary glitch on one cam at a time, I'm just using the individual clips on the timeline and everything's fine.

Those are the same cards I use - and it sounds like the same workflow too. It certainly sounds like an odd glitch. Could the recording have been stopped and re-started too quick on those particular points and causing the out of sync glitch?

[Adam Gold]

No, it was all continuous for an hour. No stopping at all. As I mentioned, the individual subclips are fine; it's only when they're re-joined that the audio issue occurs.

This is why I still roll tape as a backup.

I should probably just talk to Sony about this, but my past experience with their support people had left me feeling I know more about their gear than they do....

[Steve Renouf]

Originally Posted by Adam Gold

No, it was all continuous for an hour. No stopping at all. As I mentioned, the individual subclips are fine; it's only when they're re-joined that the audio issue occurs.

This is why I still roll tape as a backup.

I should probably just talk to Sony about this, but my past experience with their support people had left me feeling I know more about their gear than they do....

Adam, I feel for you on that one - I've had my share of support people knowing next to nothing apart from what's on their script.

Did you try re-doing the transfer and did it happen exactly the same, or did it mysteriously clear itself? I've had strange things happen on the PC on occasions (possibly through spikes in the electricity supply etc. or windows housekeeping suddenly deciding to do something) which have then cleared on the second attempt.

I do mainly live performances where the cam is running non-stop for up to a couple of hours with Sandisk III 32GB cards which I then transfer to a portable drive via my HP laptop
so I can clear the cards for re-use for the next performance and I've not had any issues like you describe. My workflow also includes using Cineform intermediates, so I don't wipe the cards until I've done the intermediates. That way I effectively have a backup of the file as .mts and Cineform .avi on the drive before clearing the original off the card.

[Adam Gold]

Yeah, I did the transfer using the Sony utility three times and it happened each time with this particular set of clips. But overall we've got about a 95% success rate doing this and it only happened with these two sets of clips, as I mentioned above.

Curious.