News:

--

Main Menu

Do I have an audio desync after a split?

Started by Bruno Augusto, May 14, 2018, 04:24:03 PM

Previous topic - Next topic

Bruno Augusto

I never used AviDemux before and recently, when looking to split an MKV into two parts, I read it could be done with it.

Despite the slight sluggishness I had with the seek bar, I could split without any quality loss and very fast, however when I checked both the original file and the split one with MediaInfo, the one I made reported Delay relative to video: -110 ms

I played the video, at different moments, and I couldn't notice the audio delayed as indicated. I also tried to compensate these -100 with VLC controls and, once again, apparently no differences.

So, does this entry means I have a real delay or not?

In the event you need a sample of the file, any of these, originally found in a Reddit post will do.

Thank you for your time

eumagga0x2a

Presumably there is no desync, but please use the latest nightly, some issues resulting in excessive video delay have been fixed since the last release (2.7.0).

http://avidemux.org/nightly/


Bruno Augusto

Alright, I've downloaded the latest version (May, 4th) and tried again. One of the videos reporting the mentioned delay still are but instead of -100ms I've got -65ms which, in a different scene that I finally could notice if it was delayed or not, now became unnoticeable.

Something that peaked my attention while testing this nightly build was that the file with the... "left" portion of the video, I mean, from the start and up to the length of the first episode, did not have this so-called delay, only the "right" one, something around the middle of the video and up to the end of the file.

Because I'll need to edit the resulting files through Premiere for some linear adjustments, like adding missing vignettes, I always have to work with at least MP4 instead of MKV (because Adobe is picky :p) and when using the MP4 Muxer, none of the resulting files had any delay. \o/

Anyway, it seems the issue has been solved but I thought this bit of information could be useful for the future, be it as a reference or any possible update.