Getting echo compressing for DVD using TMPGEnc 2.58 Plus

[hhn]

Hi All,

I know this subject have been covered a number of times before, but until now I still haven't found any usable solution to my problem. So here it is.....

I've been doing some MPG capturing from TV (PAL), using my ATI AIW Radeon. This have given me some 6-8GB MPG files, that I now want to encode/compress for burning on my HP DVDWriter 200j.

To compress the MPG stream, I using TMPGEnc 2.58 PLUS. But this produces a very crappy sound, since I have this echo all the time.
So I looked around, and mostly people have stated, that this could be avoided if I extract the sound to WAV, and then tell TMPGEnc to use the WAV as the audio source.

But when asking TMPGEnc to use my MPG as souce for the video stream and WAV as source for the audio stream, I get an even worse echo!! It seems to me, that the sound from the MPG stream and the sound from the WAV is now combined.

So the questions....
1. Am I taking the right path when extracting the sound from the MPG stream to solve this issue?
2. Is there a way to tell TMPGEnc to use only the video from the MPG stream, and not both video & audio?
3. Does anyone have any suggestions for a better path than the one i'm traveling on?


Kind regards,
HHN

[bugster]

Lets assume the audio in the original captures sounds fine (no echo).

First off, if you are capturing with the intention of burning to DVD format, make sure you capture the audio at 48Khz sampling rate, as this is required for DVD compliance.

Secondly, you could try using the toolame mp2 audio encoder with Tmpgenc to improve the audio quality. Also if you already have captures that are NOT at 48khz, I would suggest using ssrc.exe as your sample rate converter. Again it can be hooked into Tmpgenc to make the process automatic.


(Tmpgenc's audio encode/convert routines are not the worlds best and the authors recognise this which is why they provide the facility for external tools).

Hope this helps.

[hhn]

Bugster,

Thanx for taking time to help me on this one.
First - your assumption about the source video being fine, is correct.

But I'm kinda having a problem with capturing the sound in 48Khz, since the capturing software does not support this. I'm using the ATI MMC 7.1...

But having used Virtualdub, I extracted and converted the sound in 48Khz. I'm now trying to convert again (as I write), using TMPGEnc, and will let you know how it turns out when it's done in 16 hours.

Also - thanx for the tips about ssrc & toolame - I'll check those out asap.


Many thanx,
HHN

[hhn]

Bugster,

It turned out that my problem was not having converted the sound to 48Khz. Having done that using SSRC, the actual encoding (using TMPGEnc) produced a fine result.

Many many thanx for pointing me in the right direction here!


Regards,
HHN