PAL DVD to NTSC avi, video only - no audio

[jcat]

From my reading on these forums, the big problem with PAL->NTSC conversion seems to be audio sync. But what I need is to convert the *video only* from a PAL DVD to NTSC avi. I want to use clips from the footage in a video editing project (using Adobe Premiere Pro), and I need to edit using NTSC project settings, since most of the source material for my project is from NTSC DVDs (and I'll eventually want to author a NTSC DVD with the final edited product), but I'll be stripping out and ignoring the original audio - I'll be adding my own audio during the editing process. So audio sync in the original source is not an issue.

This is an old UK TV series, so it wasn't (I assume) originally film. It is not interlaced - well, actually, DGIndex shows interlacing, but the footage doesn't look interlaced, and I think this means it was encoded one field at a time, but both fields are part of the same frame. I understand this isn't unusual for PAL and I can treat it as progressive and not worry about the field-by-field encoding (though I'm kind of confused about that).

What's the best way to do this? DGPulldown avoids the audio sync problem - but if I don't care about audio, is DGPulldown the best thing to use, or is there a more direct way? Will a simple size change and AssumeFPS from 25 to 29.97, into VirtualDubMod (or directly into my editing software), get me where I want to go, or will the footage be "sped up" or something?

I've read every post I could find in the forums about converting PAL DVD to NTSC, but I haven't seen this particular question, and I'm not sure if the fact that I don't want audio changes the answer.

[Guest]

What's the best way to do this? DGPulldown avoids the audio sync problem - but if I don't care about audio, is DGPulldown the best thing to use, or is there a more direct way? Will a simple size change and AssumeFPS from 25 to 29.97, into VirtualDubMod (or directly into my editing software), get me where I want to go, or will the footage be "sped up" or something? Welcome to the forum. But please read and follow the forum rules, specifically rule 12: "Do not ask what's best." Thank you.

If you are going to finish up on an NTSC DVD, and your content is progressive, DGPulldown still brings benefits even if the original audio is going to be discarded. Your DVD will be encoded at 25fps, leaving extra bit budget for real data, as well as keeping uncorrupted frames should you or someone else later want to recover the original frames from the DVD.

AssumeFPS will not do the job, you'd have to use ChangeFPS or ConvertFPS. ChangeFPS will duplicate frames, reducing the usable bit budget. ConvertFPS will modify the source frames.

[jcat]

I did read the rules, but obviously not carefully enough! What's best was a bad way of asking anyway - I was even thinking that when I typed it. Sorry about that.

So I think I understand how I'd use DGPulldown on my final exported product - I'd use PAL avi as my source in Premiere, export the video as PAL avi (and the audio separately), use TMPGenc to encode to PAL mpg2, resize (actually, can I combine the mpg2 encoding and the resizing?), run the video through DGPulldown 25fps->29.97fps, and then mux the pulldown file with my audio.

But could I use DGPulldown before editing, on the video only? It works only on mpg2 files, and I need to edit avi. Generally for editing I use DGIndex on my VOBs and then frameserve either into VDubMod, where I make clips and export them as uncompressed avi for use in Premiere, or directly into Premiere. So in order to use DGPulldown before editing - on just the video - what would I have to do? Extract the mpg2s from the VOBs, then apply DGPulldown, then DGIndex, then frameserve the resulting mpg video into VDubMod or Premiere as I've been doing?

Would that work? Though maybe a better question is, even if it does, is there any advantage to doing that over using DGPulldown on my final product? I just assumed that it would be easier and/or work better to do any necessary conversions before editing, since then I don't have to worry about audio sync, and I suppose that's true if I'm going to use ChangeFPS, but since audio sync isn't an issue with DGPulldown, maybe I just shouldn't worry about doing any pre-project conversions.

Finally, just to be clear - it sounds to me like the big advantage of DGPulldown over ChangeFPS (I'd rather not modify the source frames, so ConvertFPS doesn't sound like what I want) is space - are there others? Are there any disadvantages to DGPulldown over ChangeFPS?

Thanks for your help - I have the DGPulldown theory thread bookmarked and go back to it constantly, but my understanding is still pretty shaky.

[Guest]

There are existing threads answering your questions, so please use search (forum rule 1!).

Do your project in PAL and apply DGPulldown at the very end. You do not have to worry about audio sync. If it is in sync before applying DGPulldown, it will still be in sync after applying it.

I would crop and resize with Avisynth in the same script that invokes MPEG2Source().

DGPulldown repeats fields, so the result will be smoother than ChangeFPS, which repeats frames.

[Paulcat]

If you are using TMPGEnc and not AVISynth, you can select the NTSC DVD option, as long as you remember to set the framerate to 25fps (TMPGEnc will repeat every 5th frame otherwise and give you a jittery mess), and your encoding to progressive (non-interlaced). You should output the audio and video elementary streams sperately, then you can apply DGPulldown to your final m2v (mpeg2 video) file right after TMPGEnc is done.

If your source is PAL AVI, then it will be progressive from the start. You can use the same technique, just keep the framerate at 25 fps in TMPGEnc

[jcat]

I've been using Avisynth with TMPGenc - using it to resize and frameserving into TMPGenc. But if I'm starting with an avi (PAL, progressive), is there any advantage to using Avisynth? I can do the resizing just as easily in TMPGenc; I can see why I'd use DGIndex and Avisynth if I were starting with VOBs, but I'm not sure what Avisynth adds if I start with an avi.

[Guest]

In that case, the advantage to Avisynth is the superior resizing alternatives and the possibility to invoke other sometimes very useful Avisynth script functions and filters.