Why would you try to correct a 5 ms audio delay? This seems to be a natural delay due to sound velocity which is about 3 ms each meter. So if the object is about 1.5 to 2 meters away from your camera that 5 ms is quite fine.

5 milliseconds is non-issue. One frame is about six times that. That being said, change the 'Time Format' display to "Time" or "Seconds" which should give you a millisecond read out. (00:00:00.000 or 0.000) and use the markers to map out 5 ms (.005) at the head, tail or elsewhere and advance/retard the audio event(s) accordingly. Make sure to un-group the event and "Quantize to frames' and other 'Snap to' parameters are disengaged (aside from snap to marker). You will need to have a significant the zoom-in time to accurately work at the millisecond resolution.

If you are moving audio only, there is an option where you can leave Quantize to Frames on and audio will move regardless of that setting.

Thanks David, forgot about that... I think Quantize to Frames is disabled by default for an audio only event.

Hello everyone and thank you all for the prompt and great responses.

Here's the problem:

I was shooting some test footage of me laying down a keyboard track (music). From my ASIO sound card: one output goes direct to the camera's line in (this is merely an audio reference track and because it's direct to the camera there is no physical or "air" delay) and the other (main) output goes to my DAW software for recording, mixing, and mastering. I know that, for example, depending on the ASIO sample rate and buffer setting, my latency is 5.1ms. Once the main audio track has been mixed and mastered then it is pulled into Vegas Pro and aligned visually and audibly with the camera's audio track (by zooming in greatly and matching some reference points in the .WAV file and ensuring that there are no audio artefacts). The camera's audio track is then deleted. No problem here so far but at this point, and assuming that the audio tracks were 100% aligned, there is still this known latency of 5.1ms. And believe me: it is noticable (well: it is to me anyway). But here is where it gets (got) interesting (and I only noticed this after starting this thread): it was worse (more noticeable) if I rendered to MAGIX AVC/AAC MP4 and uploaded to YouTube than if I rendered direct to YouTube ("Share Online"). At first I thought that maybe it's just a YouTube issue upon playback but it's not i.e. I then downloaded the files from YouTube and the audio delay is definitely more noticeable in the MAGIX AVC/AAC MP4 rendered and uploaded file. So that's the detail and the background to the issue.

I'm familiar with the operations / solutions posted (and thanks again). I was just wondering if there was some place to actually type in a known audio delay as opposed to moving the audio with the mouse. And here's why: it's pretty easy to do this with a single instrument like a keyboard (piano for example). But it's not so easy when it comes to guitar recording for example i.e. it would be simply if one could simply input a known delay / latency figure. Now I know one can use that little "arrow indicator" with the track offset that appears when you start moving the audio track. But there's one problem with it: if you move your audio but don't get it quite right on the first move then when you move it again that little indicator starts are zero again as opposed to indicating the actual offset. This differs from the red indicator that appears when you move audio (events) that come from the SAME source i.e. the red indicator shows the exact offset at any given time no matter how many times a track is moved. Sure would be nice if somehow, once tracks have been grouped, that the red indicator would appear if the tracks were from different sources.

Anyway: it would appear that the MAGIX AVC/AAC MP4 simply worsens the issue is all (aside from the fact that I know I have known latency in the system). And for the sake of interest: the absolute best quality all round that I get is rendering to one of the Sony XAVC / XAVCS templates ("Intra" if you can stomach the file sizes) (and note you can only use Sony XAVC S if you wish to create a Blu-ray disk in DVD Architect).

Just one other question:

Are you sure that 0.005 (assuming I've changed the ruler to seconds) is 5ms??? Is it not supposed to be 0.051??? Reason I ask: 0.005 appears to make no difference and 0.510 is way too much. Just curious.

Edit:

Hmmmnnn... Skip the last question above i.e. just checked and you are quite correct: 5.1ms = 0.0051 seconds. I guess the trick then is to offset the camera's audio track FIRST and only then align the main audio track to the camera's audio track (and given that this synch. is done visually and audibly there is always the chance that it's slightly off right from the get go).

Thanks again for all the input and help.

Regards,

Dale.

Why would you try to correct a 5 ms audio delay? This seems to be a natural delay due to sound velocity which is about 3 ms each meter. So if the object is about 1.5 to 2 meters away from your camera that 5 ms is quite fine.

The above is interesting and not something one would normally consider BUT I'm guessing that if you were shooting a live concert using an on camera mic and, let's say, the camera was 100m away from the stage then a) the camera's audio track would be useless as a reference / timing track and b) there would be a noticeable delay between the video and the camera's audio track (assuming sound is only coming direct from the stage and no additional delayed monitors / sound reinforcement was being used???

Regards,

Dale.

Hmmmnnn... "The plot thickens" (as they say in the classics)!!!

Using MediaInfo here are the different stats. for the same given test video:

MAGIX AVC/AAC MP4:

Video:

Duration: 24s 969ms, Frame Rate: 50 FPS

Audio:

Duration: 24s 981ms, Frame Rate: 46.875 FPS

Sony XAVC (any of the templates):

Video:

Duration: 24s 960ms, Frame Rate: 50 FPS

Audio:

Duration: 24.960ms, Frame Rate: 50 FPS

Clearly the MAGIX template is doing something odd with the audio stream.

Now while the difference ordinarily may not be too noticeable: in my case I have this anomaly in addition to my own known latency / delay issue.

Regards,

Dale.

But there's one problem with it: if you move your audio but don't get it quite right on the first move then when you move it again that little indicator starts are zero again as opposed to indicating the actual offset. This differs from the red indicator that appears when you move audio (events) that come from the SAME source i.e. the red indicator shows the exact offset at any given time no matter how many times a track is moved.

Surely you can add a Marker as a reference point so that you can get back to where you started from and/or use it as a measurement point? 😉