Followed FulciLives PAL to NTSC guide too the tee, big muxing sync problem.

[Xplic1T]

Sup guys,

Well im actually converting some material for my Russian teacher. She has some pal dvds that wont play and offered to convert some of her dvds for her. This has been by far the biggest headache of my life.

I followed FulciLives wonderful guide to the tee expect theres one problem, i cannot seem to find the right way to sync up everything again. I have tried it on this same movie atleast 2 or 3 times with the same results, a movie without synced audio at all.

I've obtained adobe premiere elements 2.0 from a friend and im trying to time stretch the audio to the video and vise versa just to see if i can sync things back up.

At the end i have my converted 480 height movie with 25 fps, when i use dgpulldown its converted to 29 fps and is roughly 3 minutes longer. (I dunno how that happeneds but thats the freakn output).

The pulldown file is 1 hour 23 min. and 44 seconds

The original 25 fps video is 1 hour 20 min. and 16 seconds and the audio matches with the original 25fps video.

When i mux the pulldown and the audio together, no syncage lol. If i stretch out the audio using besweet i get chipmunk voices so the only thing i can think of is syncing up the audio in Adobe Premiere to the runtime of the pull down file.

It seems like this part of your guide has been left out FulciLives because it just says join the origional ac3 file with ur pulldown file and its supposed to sync. It doesn't.

I am just converting the movie, not the menu, no subs, just the movie itself.

Can anybody help ?

[manono]

Hi-

At the end i have my converted 480 height movie with 25 fps, when i use dgpulldown its converted to 29 fps and is roughly 3 minutes longer. (I dunno how that happeneds but thats the freakn output).

As you should know, the reencoded and pulldowned length will be the exact same as the original length, if you do it right. Forget Adobe. The audio doesn't need stretching if you do this right.

First, although what you're saying sounds about right, how are you determining the length of the video? Second, did you run the MPV through DGPulldown with it set for 25->29.97fps? And not the default 23.976->29.97fps? The reason I ask is that it sounds to me like you converted the original 25fps film to 23.976fps. That will explain the longer length.

If i stretch out the audio using besweet i get chipmunk voices

You don't convert from 25 to 29.97fps. You use the preset for 25 to 23.976fps. That's as a last resort. My suggestion is to run the M2V through DGPulldown again, set for 25->29.97fps.

[Xplic1T]

manono

ive been using 25->29.97 in dgpulldown

i open it up, load the m2v video file that dgindex demuxed for me and then i click the 25->29.97 option and convert and i swear to you they are the same exact size but when i open them up separatly one is longer then the other.

Hmm i just looked at both the pulldown file and the original file (25 fps) in TMPg source range and they are both the same size but when i loaded them up in elements they considerably werent.

Maybe the problem lies with the muxing process. Ive tryed using ifoedit to mux the movie, still desync, and then muxman and rejig and still it desyncs. I tryed ripping the audio using PCGdemux and still the same problem.

I still dunno what im doing wrong

[Xplic1T]

Update:

Ive tryed muxing the original m2v without the pulldown with the origional audio, again DESYNC so im guessing the problem lies within the TMPG conversion ?

[manono]

Hi-

i swear to you they are the same exact size but when i open them up separatly one is longer then the other.

The reencoded MPV is of a different length than is the DGPulldowned M2V? Again I ask, how are you determing the lengths? I'm not saying you're wrong. You're probably right as different lengths between the original movie and the reencoded and pulldowned one explains your problem. I sure as heck wouldn't trust Adobe to tell me though, if that's what you're using.

In any event, the lengths shouldn't be different. Check something. Open the MPV (fresh out of the encoder but before the DGPulldown step) in BitRate Viewer and let me know the framerate (the FPS). It should say 25.

I can't help you with the specifics of your encoding problem (as that's where I suspect you went wrong), because I use neither TMPGEnc nor FulciLives' excellent guide. I use AviSynth to frameserve and CCE to encode. You'll have to wait until he comes along to help out. But, I've used DGPulldown dozens, if not hundreds, of times to apply pulldown for framerates other than 23.976fps, and not once - not once - have I ever had a synch problem. It can be done.

And as suggested before, if you're getting tired of messing with it, use my BeSweet tip from the previous post, author using Muxman, and be done with it.

And next time don't be so quick to volunteer to help out your teacher when you know nothing about what's involved. I hope she's worth it. Or buy her a cheap DVD player that can play PAL.

Edit: You updated while I was writing. As mentioned above, I suspect the same thing. Something went terribly wrong during the encoding.

[Xplic1T]

Can anybody offer any guides on here that do this. I am convinced the problem is within TMPG so i wanna try CCE and AViSYNTH.

Manono check your pm

edit: I was using premeire to check to see runtime but when opening up within dgindex and tmpeg itself via clip source i was able to see they were in fact both the same size and the origional was 25 fps and the other one that was pulldowned lol was 29. So again i am sure the problem is witihin tmpg

[FulciLives]

Try reading these two threads:

1.) THE HOLY GRAIL? ... A new method of PAL to NTSC conversion!

2.) PAL to NTSC problem

These both cover how to do the DGPulldown method of PAL to NTSC using AviSynth. I should point out that both of these links are provided in my TMGPEnc Plus guide.

I'm starting to think you didn't read my TMPGEnc Plus guide very well.

- John "FulciLives" Coleman

[jagabo]

Hmm i just looked at both the pulldown file and the original file (25 fps) in TMPg source range and they are both the same size but when i loaded them up in elements they considerably werent.

Maybe Elements doesn't understand 25->29.97 fps pulldown.

[manono]

Hi-

Maybe Elements doesn't understand 25->29.97 fps pulldown.

Which is why I said earlier:

I sure as heck wouldn't trust Adobe to tell me though, if that's what you're using.

and asked him twice how he was determining the relative lengths. It's OK, though. I got him straightened out with some private tutoring.

[jagabo]

Hi-

Maybe Elements doesn't understand 25->29.97 fps pulldown.

Which is why I said earlier:

I sure as heck wouldn't trust Adobe to tell me though, if that's what you're using.

and asked him twice how he was determining the relative lengths. It's OK, though. I got him straightened out with some private tutoring.

So was it an Adobe issue? I'm always amazed at how limited Adobe software is.

[manono]

Hi jagabo-

I don't know for sure it was an Elements issue. And the fellow was a little vague about what he did, so I'm still not sure what the original problem was. One thing I do know though, was that PGCDemux didn't report an over 2 second delay. When muxing, since he didn't know of it, it went into the Vobs with this huge delay. Maybe he should have known that this was a fixable constant delay. I'm not sure I would have spotted it myself, actually, since that's kind of a big delay.

I walked him through the process, using an AviSynth script for frameserving to CCE, and gave him all the settings. Curiously, when trying to figure what max bitrate to set, I asked him for the audio bitrate, and he gave me the full name of the AC3 file as generated by DGIndex, which had this 2156 ms delay in the name. He (we) fixed the delay using DelayCut. When all done and muxed, it was in synch. Part of the problem (and maybe the entire problem) was this delay not taken into account. Whether or not he messed up in his original encoding, I don't know. Whether or not Adobe Elements gave him the wrong length for his DGPulldowned M2V, I don't know. I suspect his original encoding was OK, and I also suspect Adobe Elements did report the length wrong, and you were right to suspect it didn't handle the unusual pulldown correctly.

[jagabo]

Thank you for the details manono.