Well, I'm encountering this nagging issue. I have hundreds of gigabytes of DV video, mostly recorded years ago, which I am now undertaking to edit and process. Most of the captures are roughly two hours in length. Each of them has an audio sync issue which worsens the longer the video plays. Premiere Pro does not adjust for this, so I am forced to edit out pieces of the audio myself, about once every half hour, to keep things roughly in sync (as close as a 30th of a second, in any event).
It seems that video captured as DV type 1 would have avoided this issue, because of a difference in how the video and audio are muxed. Fine and dandy. That's unfortunately not the way my video was captured, and I'm stuck with it. However! Because the DV type 2 sync issue is very well established, and the sync inaccuracy over time is a known quantity, it stands to reason that someone may have written a simple little app which converts DV type, or resamples the audio, or does whatever needs doing to force the video to behave.
So what are my options? ;p
+ Reply to Thread
Results 1 to 7 of 7
You're getting 2 different techniques mixed up: DV type conversion, & (locked vs unlocked) or (drop vs non-drop) audio sync capturing differences.
In (not quite) a nutshell,
On a camera, it's all one raw interleaved dv stream.
On a computer the dv stream usually needs to be wrapped in a multimedia container file so that regular apps can recognize it and know how to correcly decode it. So it gets put into an AVI stream (I'll ignore QT at the moment).
DV Type 1 AVI = 1 stream only of "IVAS" (the raw, interleaved video+audio dv stream), no other streams
DV Type 2 AVI = 1 stream of the same "IVAS" stream--NOW renamed to "VIDS" stream (standard video, but still including the audio in there) + 1 stream of the audio copied from the original IVAS into its own "AUDS" stream.
To go from Type 1 to Type 2, an app will read the IVAS and copy the audio and rename/remux.
To go from Type 2 to Type 1, an app will strip off the audio and rename/remux.
None of this has to do with SYNC.
Sync could be made wrong because you captured audio via a separate card, or had overhead/bottleneck issues in your system and frames were dropped (In capturing, Type 1 has lower overhead, Type 2 has more overhead and therefore more prone to dropping, but is generally much more compatible). Or a setting for audio did not match the video setting. You'll need to demux the audio and stretch/shrink the audio to match (in a good audio editor).
If the sync mismatch varies (sometimes rushing, sometimes dragging), you'll have to manually make many small adjustments throughout the piece, trying to sync by common A + V occurrences (like clapper, drum hits, door shut, etc). This is the hardest to fix.
If the sync mismatch gets progressively worse, all you have to do is figure out the time difference between the audio and video and divide by the total time, and that's the percentage difference you need to change the audio.
Example would be difference between drop and non-drop frame (don't ask me why audio would even be referenced by a "framerate", it's too complicated). The difference would be a total of 3.6 seconds over the course of an hour.
This = .05%
For apps, I would use Hypercube Time Stretcher or SyncView.
You probably don't need to change DV type (and shouldn't) unless you're going from one app that only uses one type (like WMM) to another app that only uses the other.
Your primary problem is Premiere Pro. It has always handled NTSC audio poorly. It also doesn't care very much for Type-1 DV AVIs that are NTSC. Seems OK with PAL. (This is my painful experience with PPro 1.5 and every preceding version).
Your secondary problem is that nearly every Type-2 to Type-1 converter will take the separate video and audio streams and mux them (as Cornucopia states). What you really want to do is just get at the original DV stream and save that to a new file - completely ignore the second audio stream.
As it happens, last week a wrote a small utility program that will let you do that. It is actually designed to repair DV AVI files with corrupt/missing headers but serves as a Type-1/Type-2 converter. It has a very simple interface and can be run from the command line for batch processing. Since you have 100's of GB, you'll probably want to use batch processing.
You can get it here:
Enosoft AVI Repair Tool
This utility doesn't care about the separate audio stream - it creates the new file only from the original DV audio and video.
I wrote this to repair corrupt DV AVIs but I didn't think to use it as a Type-1/Type-2 converter!
Thanks for the recommendations.
I've given the Enosoft repair app a go. Unfortunately, it generated a video filled with corrupt superblocks. On the other hand, after that attempt, I went ahead and recaptured one of my miniDV tapes as type 1 (my first ever capture in said type) and it worked great in PPro, with no sync issues.
The only other app I've seen which is reported to be capable of converting type 2 to type 1 is DVDate, so I'll give that one a shot soon.
(Update: Looks like DVDate works like a charm.)
Thanks for the feedback - I don't why the repair tool would have done that - it was thoroughly tested on a variety of files.
Glad you have found a solution, though