Avisynth/TMPGEnc Xpress Audio Sync Issue

[Borflex]

Hi all,

I have a mkv that i want to convert to a playable dvd using my DVD Lab Pro and TMPGEnc Xpress. Recently, I found the guide on using fitcd to make scripts and have been tweaking it to work best with the TVs in the house and have the following script:
---------------------------
# -= AviSynth v2.5.6.0 script by FitCD v1.2.8 =-
DirectShowSource("C:file.mkv")
LanczosResize(640,432,0,3,640,474)
AddBorders(36,18,44,30)
#Trim(0,-1).FadeOut(150)
ConvertToRGB24() # For TMPGEnc or VFAPI
-----------------------------

The process I am using is:
Load mkv to virtualdubmod to extract the audio as an uncompressed wav. I then load the above avisynth script into tmpgenc and change the audio source to the wav file I extracted. TMPGEnc then asks me if I want to reset the cut-edit and key frame list (not sure what this means) and I hit yes. I set the format to NTSC/dvd compatible and the output to ES(Video+Audio), and then encode.

However, after I author it with dvd lab, the audio and video are out of sync. The desyncing is gradual, and by the end of the file the audio is ahead of the video by about 5-6 seconds.

Now for the weird part. Whenever I encode by directly importing the mkv into tmpgenc (instead of the avs) and follow the exact same steps as above, the audio is in perfect sync after authoring. I know this is a solution, but I find avisynth more versatile and really want to be able to use it. I'm not really sure why this is happening. Should i just output the .m2v by itself and convert the audio on its own later? Is something in my script off?

Thanks for all the help!

[45tripp]

is audio in sync in the cut-edit section of source properties?

try
DirectShowSource("C:file.mkv", fps=23.976)

[Borflex]

Thanks for the reply!

As far as I can tell, the audio is in sync in the cut-edit section. I will try adjusting my script as you suggested.

Thanks again

[Borflex]

Unfortunately it is still out of sync. I'm finding it really weird that the tmpgenc output from the direct file works fine, but the avisynth method isnt working. Is there something else I could add to my script?

Thanks again

[45tripp]

And if you use audio loaded from directshowsource?

Have you tried with a similiar source to exclude a source problem?

What is your source anyway?

Directshowsource isn't the most trustworthy thing.
It would probably be best to take steps to remove it from the equation.
More source info would be good.
If it's tmpg in the end. There are again other solutions.

You could try all the directshow commands.
You could also try loading audio via avisynth.

If you like you could try using mkvmerge of mkvtoolnix to split your mkv into a large enough sample in which the problem would still be evident and upload it to a file sharing site.
I'd try it out.

gl