Changing 24 to 29.97 and sync audio..

[Hofuzz]

Hello,

Still stuck after some reading and trial & error..
I'm trying to change my 24 fps video source to 29.97 and sync the original audio track with it:

First I tried:
Video=mpeg2source("myVid.d2v",cpu=6)
Audio=WAVSource("myWav.wav")
AudioDub(Video,Audio)
assumeFps(29.97,sync_audio=true)
trim(0,400)
This works well but the audio is "sped up"

Then
Video=mpeg2source("myVid.d2v",cpu=6)
Audio=WAVSource("myWav.wav")
AudioDub(Video,Audio)
changefps(29.97)
trim(0,400)
The audio is out of sync by a second, how should I correct it to have the fps changed and audio at correct pace?

Thank you very much.

[mirkosp]

cpu=6 is pretty high, imho...
But that aside, and to answer your question, you could try to change the framerate before getting audio and video together. If that still fails, you could use a delayaudio() to manually adjust it before trimming it, I guess... I rarely work with audio inside avisynth, I prefer to avoid extra recompressions if any at all are needed...

[Hofuzz]

cpu=6 is pretty high, imho...

That's one of the things that as a n00b I dont quite get even after reading the A/V Tech guide (not that the guide isn't detailed enough)
so I just follow the examples...

But that aside, and to answer your question, you could try to change the framerate before getting audio and video together. If that still fails, you could use a delayaudio() to manually adjust it before trimming it, I guess... I rarely work with audio inside avisynth, I prefer to avoid extra recompressions if any at all are needed...

Thanks, I'll try that approach, I haven't ever tried using audio in Avisynth as well but I want to use a small part of the dialog from the Anime to lead into the song... and submission has to be in NTSC...

[mirkosp]

cpu=6 is pretty high, imho...

That's one of the things that as a n00b I dont quite get even after reading the A/V Tech guide (not that the guide isn't detailed enough)
so I just follow the examples...

cpu is a deblocking/derining parameter... the higher it is, the more it acts. Cpu=6 is really high and tends to kill detail and blur the image overall. Cpu=4 is probably as high as you'll ever want to go, but generally you would want to use cpu=0 and just deblock/dering with something else, if necessary at all.

But that aside, and to answer your question, you could try to change the framerate before getting audio and video together. If that still fails, you could use a delayaudio() to manually adjust it before trimming it, I guess... I rarely work with audio inside avisynth, I prefer to avoid extra recompressions if any at all are needed...

Thanks, I'll try that approach, I haven't ever tried using audio in Avisynth as well but I want to use a small part of the dialog from the Anime to lead into the song... and submission has to be in NTSC...

Well, in that case using avisynth for audio is probably fine, then. You'd have to re-encode anyway.

[Hofuzz]

cpu is a deblocking/derining parameter... the higher it is, the more it acts. Cpu=6 is really high and tends to kill detail and blur the image overall. Cpu=4 is probably as high as you'll ever want to go, but generally you would want to use cpu=0 and just deblock/dering with

Hmm, learnt something on the side! Thx! (changes to Cpu=0)

[Scintilla]

cpu=6 is pretty high, imho...

That's one of the things that as a n00b I dont quite get even after reading the A/V Tech guide (not that the guide isn't detailed enough)
so I just follow the examples...

cpu is a deblocking/derining parameter... the higher it is, the more it acts. Cpu=6 is really high and tends to kill detail and blur the image overall. Cpu=4 is probably as high as you'll ever want to go, but generally you would want to use cpu=0 and just deblock/dering with something else, if necessary at all.

I thought the great advantage of using the postprocessing in MPEG2Source rather than using some other deblocker was that MPEG2Source can read the quantization coefficients directly from the MPEG-2 stream, for more intelligent dynamic deblocking/deringing?

[kickass331]

if you have to output the video as ntsc, rerip the dvd and this time don't ivtc it. That way you can skip 2 re-encodes.

[kickass331]

alalso, if the reasoning for the framerate is due to you needing a dvd for a convention, or exhibition purposes, the dvd-video standard supports 24fps progressive scan. If you are doing an mep, then you can use an adaptive decimator when you ivtc, and specify a custom framerate in the parameters.

[mirkosp]

I thought the great advantage of using the postprocessing in MPEG2Source rather than using some other deblocker was that MPEG2Source can read the quantization coefficients directly from the MPEG-2 stream, for more intelligent dynamic deblocking/deringing?

Years bring up better filters so the more intelligent deblocking doesn't hold up with them. I do honestly suggest you to try Deblock_QED and comparing results. In my experience it ends up being better. Although tbh, I rarely have to do deblocking at all...

[kickass331]

I never thought I'd see the day when mirkosp would beat Scintilla. Scintilla has always had the upper hand. You may have won the battle mirk, but Scint is still leading the charge to victory.