itvc on flim

[gavo]

Is it worth it to itvc or does it itvc everthing includeing menus and interlaced special features etc.

[Trahald]

Originally posted by wmansir
DVD-RB recreates the progressive/interlaced state of the original EXACTLY. It basically copies the original TFF/RFF flags, encodes the RAW video, then applies the flags from the original to the output. That means you don't have to worry about progressive/Interlaced, pulldown, IVTC or even Field Order.

[jdobbs]

Couldn't have said it better myself... :)

[gavo]

o ok thanx but wouldnt itvc give better quatily? j/ wondering

[manono]

Hi-

...but wouldnt itvc give better quatily?

Yes, for 2 important reasons:

1. By encoding at 23.976 after IVTC, and running pulldown during or after the encoding, you'll be encoding 20% fewer frames when compared to encoding already telecined content at 29.97fps. Also, encoding interlaced telecined content requires a higher bitrate for the same quality. Thus, with the bits available being distributed among 20% fewer, and all progressive, frames, the result will be a considerable gain in quality. This becomes even more significant when you're trying to compress, for example, 6GB of movie video down to 3GB or so. This assumes a typical case of another couple of GB of extras compressed down to 1 GB or so. Even if it's a movie only backup, compressing 6GB to under 4GB (audio excluded) can be helped by encoding all progressive frames at 23.976fps.

2. Although standard interlaced television sets won't have any problem displaying interlaced content, many of us are now feeding progressive displays (computer monitors, DLP, LCD, Plasma, projectors) from Progressive (Scan) players. All of the software players, and many, if not most, Progressive Scan DVD players are flag readers. As such, they'll deinterlace interlace encoded DVDs, with the resulting loss of quality.

Check out this DVD Player Comparison (http://www.hometheaterhifi.com/cgi-bin/shootout.cgi?function=search&articles=all). The 3rd column from the left shows which ones can't handle 3-2 Cadence, Video Flags, which is what you get with interlaced encoding of movies. Then scroll down to see the amount of red pick up as you go down, and how none of the software players can handle the problem correctly.

In my opinion, progressive encoding, after IVTC if necessary, is the only way to go, and interlaced encoding of movies is to be avoided at all costs. Just because the DVD was incompetently encoded from an already telecined source, that's no reason to repeat the error.

OK, you DVD Mods, hit me (us) with your best shots. :)

[Sir Didymus]

Hi manono, :)

Your post deserves respect, since it is clearly focused on preserving (or improving) the quality of the original content, so no reason to worry about talking on the subject.

By the way I think I remind the question has been already posed some other times in this forum... Maybe worthwile trying to search a little bit on the matter...

IMHO, what is missing from your side is the underlying philosophy of the Rebuilder approach. I mean, I think the Rebuilder purpose (well, one of the majour purposes...) is to produce a backup copy of the source title, keeping all of its features... And replicating its bad authoring issues, if present...

I am from PAL land, so I have no idea in general if the process of IVTC can really be performed in all situations in a clean and lossless manner; I don't think it's so easy to detect properly which fields or frames have been duplicated for increasing the frame rate from 24 to 30 fps. As you know (much better than myself...) by performing IVTC you have to cut away 20% of the source content. I can't even imagine the disasters that can be produced by wrongly applying an hypothetical IVTC option in DVD-RB by unexperienced users...

I think if a movie has been telecined, the damage is done...
Suggest (if feasible...) to try to apply some off-line or manual pre-processing for recovering...

Cheers,
SD

OT & personal: love your location... Being there years ago... - Are you from the Big Island ?

[jdobbs]

I think there is a fundamental misunderstanding of telecining going on here.

The only reason IVTC was being used in old methods was to pull the original frames from the telecined output. In DVD-RB the original is fed correctly, so there is nothing to remove. What would you invert?

The point to remember is that if the source was FILM -- it will be FILM in the output -- you don't have to IVTC because the frames are being fed properly.

If anyone actually believes you can run an IVTC on an interlaced source -- bad idea, the quality will suffer terribly and the output becomes a jumpy mess.

As for films that were telecined before they were even written to the DVD... they are so rare that they aren't worth considering.

[manono]

Hi Sir Didymus-

Yes, I've read all the other threads and posts on the subject before. Gavo's new thread on the subject gave me the opportunity to reopen old wounds, so-to-speak.

And replicating its bad authoring issues, if present...

Yes, but that's my point. It's often possible to improve on the DVD mastering, encoding, authoring. Well, authoring is something different, and I don't pretend to be an expert in that area. However, with over 1000 DVDs under my belt, encoded mostly for AVI, with a good number for SVCD and DVD, I do know my way around VDubMod and CCE. And with perhaps over half of them having required IVTC, I'm also pretty experienced in that area. I also played a small part in the development of both Decomb and SmartDecimate. Also, having seen perhaps over 100 DVDs where a PAL master was used for the R1 NTSC DVD, quite a few film sources that were field blended to get to 29.97fps, rather than being telecined, either prior to, or during encoding, and having examined a number of other screwy mastering jobs, I've also had some experience in that area.

I think if a movie has been telecined, the damage is done...

A lot of that damage can be undone. They're all telecined for output, if encoded as progressive originally. It's when film sources are encoded as interlaced that they can be improved upon.

Suggest (if feasible...) to try to apply some off-line or manual pre-processing for recovering...

If it is possible, I'd like to try. However, every post on the subject that I've read says that with any kind of preprocessing (like encoding the main movie vobs myself), that DVD Rebuilder is no longer guaranteed to work properly. If there's a guide out there that tells me how to reinsert an encoded and pulled down .m2v and the AC3 back into the process, then please point me to it.

Are you from the Big Island ?

Yes. :) Right up the hill from the start and finish of the Ironman World Triathlon championship. Kailua-kona, the big game fishing capital of the world, although the Aussies might dispute that. The Big Island, home of Hawaii Volcanoes National Park and the world's most active volcano. In nearly continuous eruption since 1983.

jdobbs-

Thank you for your reasoned reply. First, I'm not talking about truly interlaced sources; such things as DVDs of sporting events, most concert DVDs, some TV series DVDs, DVDs made from DV sources, low budget movies shot with video cameras, hybrids with a mix of 24fps film and 30fps interlaced or progressive. Of course those can't be IVTC'd without producing jerky output to some degree. I'm talking about what were originally 24fps movies as shown in a movie theater.

Now, say you run the movie files from the DVD through DVD2AVI (DGIndex). It says FILM 100%. Fine; it's been encoded as progressive, the proper flags have been set, the Movie.d2v says 0 1 2 3 0 1 2 3, and DVDRebuilder does a fine job on it. But say you run a movie (originally 24fps) through DVD2AVI, and it says 0.00% FILM (=100% NTSC), and the .d2v is filled with all 0s and/or 2s. It's been encoded as interlaced. That's the kind I'm talking about. Those, even though the original source was a 24fps movie, haven't been encoded progressively, don't have the proper film flags, can't be properly output by most Progressive Scan DVD players, but can be improved greatly by the use of IVTC prior to encoding.

I may not have understood your post fully, but I don't think you addressed my main point. You said:

The point to remember is that if the source was FILM -- it will be FILM in the output -- you don't have to IVTC because the frames are being fed properly.

Sure, but I'm not talking about that kind.

If anyone actually believes you can run an IVTC on an interlaced source -- bad idea, the quality will suffer terribly and the output becomes a jumpy mess.

Sure, I thought I was clear about what I was talking about in my first post. If not, then the earlier part of this post should clear it up. I'm not talking about interlaced 30fps sources either.

Now, you did say:

The only reason IVTC was being used in old methods was to pull the original frames from the telecined output. In DVD-RB the original is fed correctly, so there is nothing to remove. What would you invert?

If I understand you correctly, you're saying that the DVD RB analysis can tell when progressive frames are stored on the DVD, even if incorrectly encoded as interlaced/video/NTSC. Do I have that right? If so, why are you then reencoding them as interlaced, when that can wreak havoc for people with progressive setups? But I may not be understanding you correctly, and you are using that statement as a preface for your next statement about FILM content not needing IVTC.

As for films that were telecined before they were even written to the DVD... they are so rare that they aren't worth considering.

Huh? If you back up DVDs only from the major Hollywood studios, then all you might see is progressively encoded DVDs (except for studio and production company logos, interlaced encoding around chapter breaks, bad edits and the like). But I assure you, these things are only too common, and can benefit from IVTC. They are exactly the kind I'm talking about. Perhaps the majority of Asian movie DVDs from Asian production companies are of this type. I still fail to understand why IVTC isn't allowed by DVD RB, if the user wishes to do so.

Damn, this reply got long. Sorry for the long post.

[manono]

To make this even longer, although dividing it into 2 parts:

Here are some examples of what legal flag sequences look like. Imagine that we have a sequence of 4 film frames that we want to
convert to video and store as an MPEG-2 stream.. We need to turn those 4 frames into 10 fields (3 + 2 + 3 + 2).

First, the most common way, using 4 MPEG pictures and all the flags:

Example 1:

MPEG Picture ...Film Frame... Picture_Structure... Progressive_Frame... Repeat_First_Field... Top_Field_First
.....1 ...........1 .............Frame ...............True .................True .................True
.....2 ...........2..............Frame................True .................False ................False
.....3 ...........3..............Frame................True .................True .................False
.....4 ...........4 .............Frame ...............True .................False ................True

However, it would be perfectly acceptable to encode that same sequence of film like this, using 10 MPEG pictures:

Example 2:

MPEG Picture ...Film Frame... Picture_Structure... Progressive_Frame... Repeat_First_Field... Top_Field_First
.....1 ............1 ...........Top_Field...............False..................False...............False
.....2 ............1 ...........Bottom_Field............False..................False...............False
.....3 ............1 ...........Top_Field...............False..................False...............False
.....4 ............2 ...........Bottom_Field............False..................False...............False
.....5 ............2 ...........Top_Field...............False..................False...............False
.....6 ............3 ...........Bottom_Field............False..................False...............False
.....7 ............3 ...........Top_Field...............False..................False...............False
.....8 ............3 ...........Bottom_Field............False..................False...............False
.....9 ............4 ...........Top_Field...............False..................False...............False
....10 ............4 ...........Bottom_Field............False..................False...............False

Or like this, using 5 MPEG pictures (note that MPEG pictures 2 and 3 contain fields from two different film frames):

Example 3:

MPEG Picture ...Film Frame... Picture_Structure... Progressive_Frame... Repeat_First_Field... Top_Field_First
.....1 ............1 ...........Frame....................False..................False...............True
.....2 ............1&2..........Frame....................False..................False...............True
.....3 ............2&3..........Frame....................False..................False...............True
.....4 ............3 ...........Frame....................False..................False...............True
.....5 ............4 ...........Frame....................False..................False...............True

These are all real examples, actually used on real DVDs. Example number 3 is extremely common, especially on smaller releases, and
even on trailers and supplements on major releases. And there are dozens of other legal variations. In each case, exactly the same
sequence of fields will be produced at the decoder output, even though the flags and the number of pictures actually stored on the
disc will be different. The compression factor will be best with the first variation, which is the only reason it’s the most popular.
(It’s certainly not for the purpose of making progressive DVD players work better.)

Damn, that took a long time. Examples 2 and 3 are what I'm talking about. Already telecined 30fps is stored on the DVD. When reencoding for DVD-5, it's important that IVTC take place, in order for 24fps all progressive frames be stored (as in example 1). Note that near the bottom it says, "The compression factor will be best with the first variation", and " It’s certainly not for the purpose of making progressive DVD players work better." I see examples of 2 and 3 all the time. They are anything but,
"so rare that they aren't worth considering." This chart is from:
http://www.hometheaterhifi.com/volume_7_4/dvd-benchmark-part-5-progressive-10-2000.html
Look about 40% of the way down to find it. The rest of the page is also a good read for anyone wanting to learn something.

[Sir Didymus]

Originally posted by manono
If it is possible, I'd like to try. However, every post on the subject that I've read says that with any kind of preprocessing (like encoding the main movie vobs myself), that DVD Rebuilder is no longer guaranteed to work properly. If there's a guide out there that tells me how to reinsert an encoded and pulled down .m2v and the AC3 back into the process, then please point me to it.


Hey manono, your posts and links are very instructive [even for a poor brain running at 25 fps, like the one of the writer...]. At least at the end I finally understood what is the pourpose of these damn TFF and RFF flags!!!

I think one way for doing what you are looking for may be this one:
http://forum.doom9.org/showthread.php?s=&threadid=88266

I posted it in another forum, since it is definitely OT in this nice place. Almost sure it's full of mistakes... But it would be interesting the opinion of some master on the subject, like you...

Cheers,
SD

[jdobbs]

@manono

I fully understand the telecining process (apparently so do you) and I admittedly aimed the software for doing commercial discs. But I have to say "Where are you digging up these discs?"

Of the examples you've used pretty much 99.99+% of the movies people encounter fall into example 1 (except of course the ones that are interlaced originals).

In the last several years I think I can only remember having seen one DVD that had been authored using example two (field based picture structure) and it wasn't telecined, it was a truly interlaced source. It's completely legal, just uncommon. DVD-RB will work correctly with that format -- but it will be frame based when CCE gets done with it.

As for example three, like I said from my experience they exist but are exceptionally rare -- you say they are common, maybe I've been in a cave the whole time, or we just encode different types of movies. Pretty much the only time I see those are from homemade DVDs authored by hacks (it happens a lot when people transfer from VHS to DVD and don't IVTC while transferring as they should -- I've done that myself on titles that haven't been released on DVD, but I did it right) or with movies released quickly when DVD was first introduced. I guess I need to get my hands on some of the Asian titles...

I guess I just never structured DVD-RB to be used to clean up somebody else's mess. I aimed at making a one-to-one backup that matched the original exactly and followed the one-click transcoder paradigm.

Let me think about it -- actually I know there is a workaround that will allow you to IVTC those kinds of sources -- I just haven't done it in a long time.

[Added] Big Island... hmmm... I haven't been there in years -- I went to Chaminade back in the 80s -- even taught a couple of semesters at Hawaii Pacific.

[manono]

Hi guys-

Sorry to be late in the response. I was kind of burnt out from all the typing several days ago, and wanted to wait a bit before starting again.

Sir Didymus-

Thanks so much for taking the time to write the guide for me (and any others that might be wondering how to do similar things). It looks like it can do the job very well, and I intend to use it soon. That was very kind of you. I can easily do the encoding myself for movie-only backups, and then create a menu on my own and author the DVD with DVDAuthorGUI, but your guide will allow me to also keep the original (motion) menus along with any extras I might want.

jdobbs-

...I have to say "Where are you digging up these discs?"

Well, they're not bootlegs, or porn, or anything else particularly unusual. Every single one (with the exception of a very few I've imported from elsewhere) is available from Amazon.com.

Of the examples you've used pretty much 99.99+% of the movies people encounter fall into example 1

Maybe yes; maybe no. In absolute terms, I daresay that there are more DVD titles with the peculiarities I'm about to list, than the perfectly encoded and mastered ones out of the major studios. In addition, most of South America, Japan, South Korea and Taiwan are also NTSC countries. Hong Kong also produces many NTSC DVDs. I haven't had much experience with DVDs from other countries, but I would guess that most aren't of the quality of those of the US Hollywood blockbusters. Some, of course, are even better. PAL DVDs have the higher resolution (along with the usual PAL speedup). And quite a few Japanese and Korean DVDs are also better. Although DVDs of the blockbuster films are certainly the best sellers, not all of us are satisfied with viewing only the pablum dished out by Hollywood. In addition, Doom9 is a site visited by people from all over the world, and many, I'm sure, would like to be able to use DVD Rebuilder along with AviSynth to improve upon the DVDs produced by their home countries.

I've saved some vob samples of some of the stranger examples I've come across. I don't usually keep the ones (there are a couple, I believe) that stored the 30fps telecined output (more commonly called "hard telecine") because, frankly, I see too many of them to think of them as being out of the ordinary. Except where noted, don't pay attention to the field order, as it can get reversed when a vob sample is cut.

1. Jungle Siren (Field Blended to 29.97fps, instead of being telecined)-really is bottom field first, quite rare in my experience. These can usually be IVTC'd back to 23.976fps.:
http://www.imdb.com/title/tt0034929/
http://www.amazon.com/exec/obidos/ASIN/B0006L0LI4
Pic. structure: Frame
Field topfirst: No
DCT type: Field
Quantscale: Linear
Scan type: ZigZag
Frame type: Interlaced
2. White Huntress (Hard Telecine-must be IVTC'd)-really is bottom field first, a real rarity for a DVD:
http://www.imdb.com/title/tt0051190/
http://www.amazon.com/exec/obidos/ASIN/B0006L0LI4
Pic. structure: Frame
Field topfirst: No
DCT type: Field
Quantscale: Linear
Scan type: Alternate
Frame type: Interlaced
3. Entrails Of A Beautiful Woman (Normal progressive, but pic structure=field):
http://www.imdb.com/title/tt0224102/
http://www.amazon.com/exec/obidos/ASIN/B0002CX15W/
Pic. structure: Field
Field topfirst: Yes
DCT type: Field
Quantscale: Linear
Scan type: ZigZag
Frame type: Progressive
4. Kamikaze Taxi (field blended from 23.976 to 29.97fps, and I was able to restore 23.976 progressive fps):
http://www.imdb.com/title/tt0113523/
http://www.amazon.com/exec/obidos/tg/detail/-/B0002UB2XM/
Pic. structure: Frame
Field topfirst: Yes
DCT type: Field
Quantscale: Nonlinear
Scan type: Alternate
Frame type: Interlaced
5. No Quarter-Page And Plant Unledded (PAL to NTSC-field blended from 25fps to 30fps. These can be encoded for 25fps for AVI, or "IVTC'd" using RePAL or Restore24 and then slowed to 23.976 for NTSC DVD). If I've seen one of these, I've seen a hundred. They can come from European, Japanese, or Chinese (and HK) films. I've even seen it done to American films which were transferred for PAL DVD, and that same PAL master was then used for the R1 NTSC DVD. Inexcusable:
http://www.imdb.com/title/tt0324384/
http://www.amazon.com/exec/obidos/ASIN/B0002MGY5G/
Pic. structure: Frame
Field topfirst: Yes
DCT type: Field
Quantscale: Nonlinear
Scan type: Alternate
Frame type: Interlaced
6. Parasite Eve (24fps movie, with a considerable amount of interlaced encoding, along with the progressive. Needs IVTC. Quite a lot of anime episodes are also like this. Typically, after running them through DVD2AVI, they'll show 65% or so FILM, and the rest is NTSC. While newer anime often includes some 30fps CG segments, it's usually so little that I much prefer to get them back to 23.976fps):
http://www.imdb.com/title/tt0119860/
http://www.amazon.com/exec/obidos/ASIN/B00005JXXX/
Pic. structure: Frame
Field topfirst: No
DCT type: Field
Quantscale: Nonlinear
Scan type: Alternate
Frame type: Interlaced
7. Red Cherry (24fps field blended to 30fps. If you know what you're doing, this kind can be returned to 23.976fps):
http://www.imdb.com/title/tt0114243/
http://www.amazon.com/exec/obidos/tg/detail/-/1572525754/
Pic. structure: Field
Field topfirst: No
DCT type: Field
Quantscale: Nonlinear
Scan type: Alternate
Frame type: Interlaced8.
8. Spies (PAL to NTSC, and zigzag scan type together with interlaced encoding):
http://www.imdb.com/title/tt0019415/
http://www.amazon.com/exec/obidos/tg/detail/-/B00064AEWY/
Pic. structure: Frame
Field topfirst: Yes
DCT type: Field
Quantscale: Nonlinear
Scan type: ZigZag
Frame type: Interlaced
9. Woman In The Moon (Hard Telecine, and zigzag with interlaced encoding):
http://www.imdb.com/title/tt0019901/
http://www.amazon.com/exec/obidos/tg/detail/-/B00064AEXI/
Pic. structure: Frame
Field topfirst: Yes
DCT type: Field
Quantscale: Nonlinear
Scan type: ZigZag
Frame type: Interlaced
10. Cloak And Dagger (it is FILM, but without the progressive flag being set for every frame. The DVD2AVI Preview flips back and forth quickly between Interlaced and Progressive. The page linked in my last post says this:
Some flag readers blindly follow the flags on the disc exactly as they are encoded, which fails miserably on a wide variety of films. The most common problem with this is the hundreds of discs that have a common encoding error where the progressive_frame flag alternates off and on every other frame. This causes the simple flag readers to constantly slip in and out of film mode, and potentially in and out of progressive chroma upsampling mode, both of which are very bad. The result is a constantly strobing image, sometimes subtly and sometimes grossly. This problem in the flags was caused by a bug in an MPEG encoder, but that MPEG encoder was extremely popular, and is still in use, for example for a variety of recent Disney releases. Players ignore this problem at their peril.
I don't know how DVD Rebuilder handles this kind, as I haven't run any of them through it. I would hope that the progressive_frame flag would be set, but I doubt it, if it duplicates the original DVD structure.
http://www.imdb.com/title/tt0038417/
http://www.amazon.com/exec/obidos/tg/detail/-/B00008RV0H/
Pic. structure: Frame
Field topfirst: Yes
DCT type: Field
Quantscale: Nonlinear
Scan type: Alternate
Frame type: Interlaced

I could go on and on. Don't even get me started on silent film DVDs. Anyway, they're out there, and "IVTC" of various kinds can restore and improve many of these strangely encoded and mastered DVDs. I'm not even really asking that IVTC be made possible with DVD RB. That's for you to decide. I have no idea what would be involved, or if you would be opening the door for people to really screwup their DVD backups, with the resulting headaches when they come whining about it here in this forum. My main point in reopening this can of worms was to help demonstrate that the people that have asked for the IVTC option in the past weren't necessarily all idiots. When the combined force of jdobbs, wmansir and Trahold all came down on them for what they thought were reasonable requests, well, it's pretty intimidating.

Thank you for reading.

[jdobbs]

You've kinda' proven my point. In the examples you've used I can tell you that most (other than the hard telecined) would be better left for DVD-RB to take care of rather than IVTCing.

I don't want to argue -- and you obviously like typing more than I do, so I'd rather drop this discussion.

[SkVid]

Very interesting all of that above ... but maybe too heavy for me.

But maybe you can help me with my problem.
I have 2 sources where the material is interlaced TTF. This says Bitrate-Viewer and Procoder2 when you add the original VOB as source.
After the prepare phase of DVD-Rebuilder the TTF flag in the ECL-file is set to 0 (top_first=0). Starting the encoding phase Procoder2 says the source (AVI-Synth file) is BFF and ecodes it as BFF.
Isn't that a problem when DVD-Rebuilder sets back the original flag when rebuilding?
If it is a problem … what can I do?

Thanks for your answers.

SK

[jdobbs]

Is there a problem in the playback? If the flag is set incorrectly you will see it very obviously (wildly jittering movement -- not minor either, you don't have to study the picture to see it). Note: You have to review it on a standalone unit, not a PC.

[SkVid]

This is what is surprising me. There is no jittering on a standalone player.
But in other threads you get told to watch whether your source is TFF or BFF. And now you get told that it doesn’t matter.
When I use "mpeg2source("…").info" in a AVI-Synth script it tells me that the source is frame based and TTF. Opening another script and loading the first AVI-Script in it with .info it tells me that the source (the AVI-Synth script file with the mpeg source in it) is frame based BFF. This is why I think the encoder (Procoder2) handles the source as BFF and encodes it that way. The new video stream is BFF.
I demuxed the source into the elementary streams. The video is TFF.
After the rebuild phase the new created vob is frame based TFF. That must be the flag that is set afterwards.

[jdobbs]

I think what throws all the confusion in is the fact that AVISYNTH feeds the picture to the encoder as a frame. The encoder has no idea whether it was originally stored TFF or BFF.

[G_M_C]

Sorry to kick up this thread, but before i backup one of my DVD's i need to be shure of one thing;

Can DVD-RB handle 29,97i NTSC --> 23,976p deinterlacing/IVTC'ing ?

In other words; Can I set DVD-RB to do this ? Or would everything be OK if i'd do it myself by editing/using scipt editor ?

My intention is to use the latter option, to include the EEDI2()/tDeint()/tfm()-trio to go from 29,97i (59,94 fields/sec) to 23,976 progressive. I'm afraid that the rebuild-process will fail, because i change framerates and/or the NAVPACK's will be "screwed".

[jdobbs]

No. You cannot use any filter that would change the frame count. But as I said earlier -- you should never need to.

By saying you want to iVTC -- you imply that the source is telecined. In DVD-RB the telecined source is encoded at 23.976 anyhow... because DVD-RB already recognizes a telecined source.

I'm sorry if I misunderstand, but I'm getting the impression you may not fully understand telecining and iVTC. A telecined source is stored on the disc at 23.976fps, not 29.97fps. The additional fields to bring it up to 29.97 are created upon playback using the pulldown sequence via TFF/RFF flags. If the program doing the reencoding is aware of how that works (and DVD-RB is) it encodes at 23.976 and reinserts the TFF/RFF flags to reestablish pulldown.

So trying to iVTC the source doesn't make any sense. iVTC would only be applied to something that had already gone through the pulldown and had already been converted to 29.97fps. That pretty much only happens on video that have been captured from a TV signal following field insertion (at the station) -- but not on a DVD source. There are very rare exceptions (called hard-telecined sources), but they are very, very few and far between.

[G_M_C]

No. You cannot use any filter that would change the frame count. But as I said earlier -- you should never need to.

By saying you want to iVTC -- you imply that the source is telecined. In DVD-RB the telecined source is encoded at 23.976 anyhow... because DVD-RB already recognizes a telecined source.

I'm sorry if I misunderstand, but I'm getting the impression you may not fully understand telecining and iVTC. A telecined source is stored on the disc at 23.976fps, not 29.97fps. The additional fields to bring it up to 29.97 are created upon playback using the pulldown sequence via TFF/RFF flags. If the program doing the reencoding is aware of how that works (and DVD-RB is) it encodes at 23.976 and reinserts the TFF/RFF flags to reestablish pulldown.

So trying to iVTC the source doesn't make any sense. iVTC would only be applied to something that had already gone through the pulldown and had already been converted to 29.97fps. That pretty much only happens on video that have been captured from a TV signal following field insertion (at the station) -- but not on a DVD source. There are very rare exceptions (called hard-telecined sources), but they are very, very few and far between.

Well, you are right (offcourse); I asked this question for a totally different reason, to be exact... The situation is as follows:

A friend of mine has a scolarship for e few years doing research in the America's. Every once in a while he returns here in NL for a holiday. Approx a year ago he brought me the (then brand new) NTSC version of King Kong; Not the extended edition but the first edition wich is equal to the theatrical version. It has 29,97 fps telecined material on it ...

Last Christmas he brought me something else; Seems he's bought himself some fancy equipment, and is capable in capturing Satellite broadcasts, and to show this oof he brought me (on 2 DVDR9's) the 1080i version of .... King Kong, and not only that but the Open Matte verion of it, as broadcast on some HD satellite in the US i guess.

Now we came to the novel idea that we could replace the 2,35:1 video on my DVD with the (resized to 720x480) Open Matte version; You could say that we wanted to give my old DVD new life with new very sharp and progressive ultra-high quality picture in real 16:9 !!

We compared the two versions and found out that they were equal in length etc. too, because the audio-track of the DVD fits exactly onto the HD-satellite-capped version. So, we thought, that it could, at least in teory, be done ... Not beeing experts and all, we had notived that DVD-RB splits its DVD's up in seperate chapters throught avisynth scrips / Trim() commands. And our initial thought was simple; Just replace the Mpeg2Source (xxxxx) to point to the capped-version, do the resizing and let DVD-RB do its job as usual. But resizing from 1080i to 720x480 involved de-interlacing, because a seperatefields.resize.weave will not give very good results (besides other problems that could arise).

And from this point on, i got my question, as i posed earlier; Can i change framerate in the scrips. Might better have asked for a idea on how to exchange the video too, both probably not as easily possible with DVD-RB (wich wasnt made for that anyway).

So you were right that having to deinterlece and decimate comes mostly on broadcasts, as you said in your previous posts. We werent to bright in assumin that is would be this easy offcourse, we'll look into IFOedit in combination with some authoring programms now. Basically the "only thing" we want to do is change the picture on the DVD ... i get the feeling that "the only thing" turns out to be the understatement of the week ... ;)

[seekrtz]

There is a HiDef King Kong rip on newsgroups, H264 version with ger/eng audio tracks - it's by far the best source I've ever seen for this movie and I'm sure it's superior to the MPEG2 version you are probably trying to convert right now. However, it's like 24.5GB and it's also pal(which is nice because that means it's 23.976fps progressive sped up to 25fps), currently working on re-encoding it to fit 1 DL DVD as vc-1 using the orig dvd's 6 ch audio:P

Anyway, if you want to IVTC before converting it to a dvd, just use doubleweave().Pulldown(0,2) - keep trying different combinations if that doesnt work, one of them will work; also - sometimes movies change patterns half way thru, so just use trim and append both pieces.

[G_M_C]

Hmmm, the problem isn't iVTC'ing at all, its replacing the video on my DVD. And the source i've got has been captured by ourselfs (actually that friend of mine); So i dont need to go entirely "illegal" (because the DIY capturing, and the owning of the DVD).

PS: Especially here on Doom9 i wouldnt pose questions about non-legal sources if i were you ...