My DVD player plays pretty much anything I chuck into it with bitrates under 2400 Kbits/sec. I have MPEG-2 files with bitrates of approximately 6 Mbits/sec that I'm trying to re-encode down so the player will play them and still be able to get a reasonable quality in a reasonable space (CD for now).

I started by using TMPGEnc with KVCD templates to get very good quality down-converts in very reasonable sizes but it takes a looonnnnnnnnng time. I then tried DVD2SVCD using the CCE encoder and was very, very impressed with how incredibly good the conversion from DVD chapters was in a relatively reasonable time. This encouraged me to begin looking at the whole manual process of frameserving and the customize-ability it offers.

What I'm finding is that I'm having terrible audio-synching problems that I didn't have when I was using TMPGEnc with my files or DVD2SVCD with DVD source files. The audio on a 1-hour video can be in synch at the start and be lagging by over a second at the end.

The general steps I've taken are:
1) Open MPEG2 in DVD2AVI and save project (.d2v and .mpa files)
2) Use either AviSynth to frameserve a .d2v or VFAPI to frameserve a .avi to CCE
3) Re-encode with CCE (I've tried 1-pass, 3-pass, and 4-pass VBR)
4) Multiplex the CCE .mpv output with the DVD2AVI project .mpa file in TMPGEnc MPEG Tools.

I've also used DVD2SVCD as AVI2SVCD and frameserved it a VFAPI .avi via the DVD2AVI .d2v file, but I haven't really figured out how to give it the audio file. I would then find the .mpv file and re-mux it with the .mpa file.

All the ways I've tried so far where CCE was utilized (except with DVD .VOB source in DVD2SVCD) have resulted in this audio-synch problem. I haven't had this problem when simply using a KVCD tempate in TMPGEnc, but I'm wanting more control and better speed.

By the way, when I load the original source file into VirtualDub, even one with the audio apparently synched, it will show me a number other than the source framerate that would need to be used "so video and audio durations match". This seems to typically be a ratio of 29.970 to 29.961 or so. I figure there has to be something to this, but I have no idea what.

I tried to use BeSweet to transcode the audio file into a different "framerate", but all I get is output with severe chatter.

What am I doing wrong? What should I be doing?

Thanks!

Tim