How to Determine Original DVD Frame Rate & Frame Type

[lomaidala]

I have a DVD I'm turning into an AVI using AviSynth. I have two questions:

1) How do I determine what the original or base framerate of the DVD is?
2) How do I know whether to deinterlace the video or inverse telecine?

Attached is a sample of the DVD source (made using DGIndex and Mpg2Cut2).

[hech54]

Why don't you just experiment with that little sample you provided and see what looks best?

[LMotlow]

1) How do I determine what the original or base framerate of the DVD is?

Quickest way is to open the MPG in VirtualDub and click "File..." -> "file information" for the frame size, frame rate, and other info.

Get more info using MediaInfoXP to get a text report on the frame structure and other data. Get MediaInfoXP here: https://www.videohelp.com/software/MediaInfoXP. A MediaInfoXP report is posted below.

If you can't open MPG diectly in Virtualdub, you need FCCHandler's MPEG2 VirtualDub plugin. Get it here: https://sourceforge.net/projects/fcchandler/files/Virtualdub%20Mpeg2%20plugin/mpeg2.zip/download.

The zip contains 4 files. Look over the "ReadMe.txt" file for instructions. If you're using 32-bit VirtualDub, use the 32-bit version of the plugin. If you're using the 64-bit version of Virtualdub (I wouldn't do it) use the 64-bit version.

2) How do I know whether to deinterlace the video or inverse telecine?

Open the MPG in Virtualdub and scroll thru the video using frame-by-frame. You'll see interlace effects (fuzzy edges or combing) in 2 of every 5 frames during motion. This indicates pulldown (telecine, also called 2:3 pulldown).

If you see interlace effects on every frame during motion, it's interlaced.

MediaInfoXP reports the frame rate here as 23.976, but pulldown is used to get 29.97 fps playback. Sometime you'll just see 29.976 if pulldown can't be detected accurately. Text report from MediaInfoXP:

General
Format                                   : MPEG-PS
File size                                : 10.9 MiB
Duration                                 : 9s 600ms
Overall bit rate mode                    : Variable
Overall bit rate                         : 9 530 Kbps

Video
ID                                       : 224 (0xE0)
Format                                   : MPEG Video
Format version                           : Version 2
Format profile                           : Main@Main
Format settings, BVOP                    : Yes
Format settings, Matrix                  : Custom
Format settings, GOP                     : M=3, N=12
Duration                                 : 9s 359ms
Bit rate mode                            : Variable
Bit rate                                 : 8 444 Kbps
Maximum bit rate                         : 9 800 Kbps
Width                                    : 720 pixels
Height                                   : 480 pixels
Display aspect ratio                     : 16:9
Frame rate                               : 23.976 (24000/1001) fps
Color space                              : YUV
Chroma subsampling                       : 4:2:0
Bit depth                                : 8 bits
Scan type                                : Progressive
Scan order                               : 2:3 Pulldown
Compression mode                         : Lossy
Bits/(Pixel*Frame)                       : 1.019
Stream size                              : 9.40 MiB (86%)

Audio #1
ID                                       : 189 (0xBD)-128 (0x80)
Format                                   : AC-3
Format/Info                              : Audio Coding 3
Mode extension                           : CM (complete main)
Format settings, Endianness              : Big
Muxing mode                              : DVD-Video
Duration                                 : 9s 600ms
Bit rate mode                            : Constant
Bit rate                                 : 448 Kbps
Channel(s)                               : 6 channels
Channel positions                        : Front: L C R, Side: L R, LFE
Sampling rate                            : 48.0 KHz
Frame rate                               : 31.250 fps (1536 spf)
Compression mode                         : Lossy
Delay relative to video                  : -82ms
Stream size                              : 525 KiB (5%)

Audio #2
ID                                       : 189 (0xBD)-129 (0x81)
Format                                   : AC-3
Format/Info                              : Audio Coding 3
Mode extension                           : CM (complete main)
Format settings, Endianness              : Big
Muxing mode                              : DVD-Video
Duration                                 : 9s 568ms
Bit rate mode                            : Constant
Bit rate                                 : 448 Kbps
Channel(s)                               : 2 channels
Channel positions                        : Front: L R
Sampling rate                            : 48.0 KHz
Frame rate                               : 31.250 fps (1536 spf)
Compression mode                         : Lossy
Delay relative to video                  : -82ms
Stream size                              : 523 KiB (5%)

[jagabo]

1) How do I determine what the original or base framerate of the DVD is?
2) How do I know whether to deinterlace the video or inverse telecine?

These are a complex issues and there are many many threads here that deal with them. Search.

[manono]

Attached is a sample of the DVD source (made using DGIndex and Mpg2Cut2).

You could have done it in a single step by using the [ and ] buttons in DGIndex to isolate the section you wanted and then File->Save Project and Demux Video.

Yes, it can be difficult. but not with your sample. You can set the DGIndex Field Operation to Forced Film, make the D2V project file and get an already progressive 23.976fps video from it to be used in your AviSynth script with MPEG2Source. This doesn't say the rest of it is like that. As mentioned by LMotlow, your best way to determine what you have is your eyes by examining different parts of the video. The DGIndex Preview (or examination of the resulting D2V file) can be helpful as well (although it shows true interlace the same as hard telecine), to let you know if it's entirely soft telecined or a mix of hard and soft telecine as is common with anime.

[lomaidala]

Thank you, LMotlow and manono!

Opening up each VOB file individually in VirtualDub (using the MPEG2 plugin), I am seeing 2:3 pulldown everywhere I look. Sometimes it looks like "1:4," but I think those are just similar frames that make the telecine difficult to detect. So it looks like it's safe to say my source is not interlaced.

But I'm still not clear on the frame rate. Opening up the VOB in MediaInfoXP, it says the frame rate is 29.970, as opposed to the 23.976 it says when I open the sample. I assume this is because of what LMotlow says about pulldown not always being detected accurately.

MediaInfoXP reports the frame rate here as 23.976, but pulldown is used to get 29.97 fps playback. Sometime you'll just see 29.976 if pulldown can't be detected accurately.

So does that mean the original framerate is 23.976 or 29.97?

You can set the DGIndex Field Operation to Forced Film, make the D2V project file and get an already progressive 23.976fps video from it to be used in your AviSynth script with MPEG2Source.

That's my current process exactly. But if the source is 29.97fps, wouldn't I be dropping frames by doing this?

The DGIndex Preview (or examination of the resulting D2V file) can be helpful as well (although it shows true interlace the same as hard telecine), to let you know if it's entirely soft telecined or a mix of hard and soft telecine as is common with anime.

Is the difference between hard and soft telecine something I should understand for what I'm doing?

Why don't you just experiment with that little sample you provided and see what looks best?

I experimented beforehand. The problem is, all of my results looked good. So in that sense, I'm more interested in doing what's correct as opposed to doing what looks good--but I also want the video to look as good as possible while still being done properly.

These are a complex issues and there are many many threads here that deal with them. Search.

I searched before posting, but the most relevant thread I found was this, which wasn't very helpful:
https://forum.videohelp.com/threads/166869-How-do-you-correctly-determine-the-original-FPS-of-video

[manono]

Opening up each VOB file individually in VirtualDub (using the MPEG2 plugin), I am seeing 2:3 pulldown everywhere I look.

That tells you virtually nothing since VDub will always show the interlaced 29.97 framerate. What you want to know is whether it's soft telecine, hard telecine, or a mix. One way is to open all the VOBs in DGindex and run the Preview. Another is to make a D2V file and then look at the bottom to check the film (or video) percentage.

So does that mean the original framerate is 23.976 or 29.97?

I could be wrong but I don't believe MediaInfo parses the video so it's telling you only how the first frame was encoded.

That's my current process exactly. But if the source is 29.97fps, wouldn't I be dropping frames by doing this?

If it's soft telecine then all you'll be dropping is repeated fields.

Is the difference between hard and soft telecine something I should understand for what I'm doing?

Yes, especially for anime. Soft telecine just means the telecine is performed by flags (TFF/RFF) where the underlying video was encoded as progressive 23.976. Hard telecine means the telecine is actually encoded into the video and has to be IVTC'd to get it back to progressive 23.976fps.

[jagabo]

These are a complex issues and there are many many threads here that deal with them. Search.

I searched before posting, but the most relevant thread I found was this, which wasn't very helpful:
https://forum.videohelp.com/threads/166869-How-do-you-correctly-determine-the-original-FPS-of-video

Search for "DVD pulldown telecine".

All NTSC DVDs output 59.94 fields per second. They can be encoded as progressive at any frame rate from 19.98 to 29.97 fps with soft pulldown (flags in the stream that tell the player how to produce 59.94 fields per second from the frames), or any frame rate can go through hard pulldown to 29.97 interlaced frames per second. Beyond that there are frame rate conversions with field blending and/or frame blending -- sometimes more than one conversion.

In my opinion the best way to determine the frame rate (and detect field/frame blending -- which requires very different handling) is to apply a simple bob filter (after soft pulldown, if necessary) and count the number of unique (ignoring bob artifacts) frames per second during a panning shot.

Some material (cartoons and anime, the intro of a TV series, music videos, etc.) is a mix of frame rates (8 fps, 12 fps, 24 fps, 30 fps, 60 fields per second, etc.). For example, cartoons often have characters animated at 12 fps panning shots at 24 fps, and title overlays at 60 fields per second.

Your sample was 23.976 fps with soft pulldown.

[LMotlow]

Soft telecine just means the telecine is performed by flags (TFF/RFF) where the underlying video was encoded as progressive 23.976. Hard telecine means the telecine is actually encoded into the video and has to be IVTC'd to get it back to progressive 23.976fps.

For what it's worth, DGIndex information panel for this segment displays "View Type" as "Film", so it's 100% soft telecine -- for this sample, anyway. That might not be true for the entire video.

[jagabo]

Soft telecine just means the telecine is performed by flags (TFF/RFF) where the underlying video was encoded as progressive 23.976. Hard telecine means the telecine is actually encoded into the video and has to be IVTC'd to get it back to progressive 23.976fps.

Manono knows this, but soft telecine can be used on any progressive frame rate from 19.98 fps to 29.97 fps. Hard telecine can be used on any frame rate. A progressive frame rate of 23.976 is most common, but there are other possibilities.

[lomaidala]

Thanks, everyone. This is making a lot more sense now.

In my opinion the best way to determine the frame rate (and detect field/frame blending -- which requires very different handling) is to apply a simple bob filter (after soft pulldown, if necessary) and count the number of unique (ignoring bob artifacts) frames per second during a panning shot.

Which bob filter would you recommend for that?

Search for "DVD pulldown telecine".

Ah, now I see. That kept me busy reading for quite a while. I can see you guys have gone over this many times. Different noobs, same answers.

[jagabo]

In my opinion the best way to determine the frame rate (and detect field/frame blending -- which requires very different handling) is to apply a simple bob filter (after soft pulldown, if necessary) and count the number of unique (ignoring bob artifacts) frames per second during a panning shot.

Which bob filter would you recommend for that?

I usually just use Bob() in AviSynth, or Bob Doubler in VirtualDub. It gives lots of aliasing artifacts but I just ignore those, it's otherwise pretty foolproof. You could also use Yadif(mode=1). Even QTGMC() -- but that's very slow, and sometimes its filtering masks or adds motion.

[lomaidala]

The DGIndex help file says 95% is the threshold for selecting Forced Film rather than Honor Pulldown Flags, but I've also read on this forum that the threshold is 99%. Is the contradiction here on the forums a precaution to be safe?

What you want to know is whether it's soft telecine, hard telecine, or a mix. One way is to open all the VOBs in DGindex and run the Preview.

How would I know if my source is a mix of soft telecine and hard telecine?

In my opinion the best way to determine the frame rate (and detect field/frame blending -- which requires very different handling) is to apply a simple bob filter (after soft pulldown, if necessary) and count the number of unique (ignoring bob artifacts) frames per second during a panning shot.

Which bob filter would you recommend for that?

I usually just use Bob() in AviSynth, or Bob Doubler in VirtualDub. It gives lots of aliasing artifacts but I just ignore those, it's otherwise pretty foolproof. You could also use Yadif(mode=1). Even QTGMC() -- but that's very slow, and sometimes its filtering masks or adds motion.

I tried the VDub bob doubler. I can't make any sense of the pattern in the result. It doesn't appear to resemble anything I've seen discussed.
See attached sample.

[jagabo]

The DGIndex help file says 95% is the threshold for selecting Forced Film rather than Honor Pulldown Flags, but I've also read on this forum that the threshold is 99%. Is the contradiction here on the forums a precaution to be safe?

Yes.

What you want to know is whether it's soft telecine, hard telecine, or a mix. One way is to open all the VOBs in DGindex and run the Preview.
How would I know if my source is a mix of soft telecine and hard telecine?

That's what DgIndex is telling you above. "Film" is progressive frame with pulldown flags (soft telecine). "Interlaced" is hard telecine (or true interlaced video from a video camera).

I tried the VDub bob doubler. I can't make any sense of the pattern in the result. It doesn't appear to resemble anything I've seen discussed.
See attached sample.

You used the wrong field order in the Bob Doubler filter.

[lomaidala]

What you want to know is whether it's soft telecine, hard telecine, or a mix. One way is to open all the VOBs in DGindex and run the Preview.
How would I know if my source is a mix of soft telecine and hard telecine?

That's what DgIndex is telling you above. "Film" is progressive frame with pulldown flags (soft telecine). "Interlaced" is hard telecine (or true interlaced video from a video camera).

So what would be a mix between soft and hard telecine? Interlaced Film?

I tried the VDub bob doubler. I can't make any sense of the pattern in the result. It doesn't appear to resemble anything I've seen discussed.
See attached sample.

You used the wrong field order in the Bob Doubler filter.

Thanks! I switched it. The pattern of unique (excluding anti-aliasing discrepancies) frames I count after applying the bob doubler is now 4-6-4-6. I was expecting 2-3-2-3 based on what I've read, but since the bob doubler doubles the framerate, I guess I should have seen this coming. 4-6-4-6 after using a bob doubler means that the video's base framerate is 23.976 fps, and that it uses 2-3 pulldown (soft telecine), right?

[manono]

So what would be a mix between soft and hard telecine?

Assuming no true interlaced content, any film percentage less than 100.

Interlaced Film?

Film, by definition, is progressive. DGIndex calls 'interlaced film' as Video.

and that it uses 2-3 pulldown (soft telecine), right?

What you see can come from either hard or soft telecine. Gotta sample of this 4-6-4-6?

[jagabo]

The pattern of unique (excluding anti-aliasing discrepancies) frames I count after applying the bob doubler is now 4-6-4-6.

It should be 3:2. Did you bob the video in your AviSynth script and in VirtualDub? That would show a 4:6 pattern. Your AviSynth script should only contain a source filter and maybe a colorspace conversion:

Mpeg2Source("VTS_01_6.d2v") 
ConvertToYUY2(interlaced=true)

Check the frame rate in VirtualDub after opening the script (File -> File Information). It should be 29.97 fps. Then apply the Bob Doubler filter.

Or better yet, perform the bob in AviSynth and use VirtualDub only to view the result:

Mpeg2Source("VTS_01_6.d2v") 
Bob()

[lomaidala]

Did you bob the video in your AviSynth script and in VirtualDub?

No, just in VirtualDub.

Your AviSynth script should only contain a source filter and maybe a colorspace conversion:

How would I know if it should have a colorspace conversion?

Check the frame rate in VirtualDub after opening the script (File -> File Information). It should be 29.97 fps.

Yes, 29.97 fps.

Then apply the Bob Doubler filter.

That's what I did.

Or better yet, perform the bob in AviSynth and use VirtualDub only to view the result:

Same result. See attached file: bob (AviSynth).avi

What you see can come from either hard or soft telecine. Gotta sample of this 4-6-4-6?

See attached file: bob doubler (VirtualDub).avi

[jagabo]

How would I know if it should have a colorspace conversion?

VirtualDub doesn't handle interlaced YV12 properly. So you usually want to convert it to YUY2 or RGB before using VirtualDub. It doesn't matter here if you ignore the chroma artifacts (barely visible in your test clip because the motions are so small).

See attached file: bob (AviSynth).avi
See attached file: bob doubler (VirtualDub).avi

Both show a 3:2 repeat pattern.

By the way, use a panning show with more motion -- it will be much easier to see.

[LMotlow]

By the way, use a panning show with more motion -- it will be much easier to see.

Not only is it difficult to see in the posted samples, it's very slow motion and very small motion. Animation can fool you, too, as it's often created with duplicate frame patterns long before any form of pulldown is applied.

Attached is a DVD action scene from an old Errol Flynn movie. Fast motion, lots of figures, cross pans and vertical pans. Even without bobbing, the 3:2 pattern is easy to see (along with Turner Classic Movies' sloppy encoding). The DVD sample is hard telecined, meaning that the video shows the 3:2 telecine pattern but there are no telecine flags: the telecined frames are physically interlaced. DGIndex will report this video's "Frame Type" as Interlaced. However, DGIndex will report your soft-telecined VOB sample's "Frame Type" as "Film".

Another tool you can use is the old GSpot. No longer developed, but it's still good for fast analyzing many standard formats.

With your soft telecined-flagged VOB_6 example, GSPot reports that "Pics" = 23.976, "Frames"=29.970, and "Fields" = 59.94 (follow the arrow in the image below):


With the attached "Hard Telecine Demo B.mpg" and its hard-interlaced telecine, GSPot reports that "Pics" = 29.970, "Frames"=29.970, and "Fields" = 59.94 :


I don't have soft-telecined DVD samples around at the moment. Whether hard or soft, telecine looks the same way in VirtualDub and other apps that don't pay attention to pulldown flags.

[manono]

The DVD sample is hard telecined, meaning that the video shows the 3:2 telecine pattern but there are no telecine flags: the telecined frames are physically interlaced. DGIndex will report this video's "Frame Type" as Interlaced.

I was wrong to say DGIndex reports hard telecine as Video. Instead, it reports the Frame Type as Interlaced and Video Type as NTSC. For soft 3:2 telecine it reports Frame Type as Progressive and Video Type as Film.

[jagabo]

Whether hard or soft, telecine looks the same way in VirtualDub and other apps that don't pay attention to pulldown flags.

While it's true that soft and hard telecine look the same in VirtualDub, it's because VirtualDub does pay attention to pulldown flags. Ie, it performs the pulldown to create interlaced frames from the progressive frames stored in the MPEG video. The old VirtualDubMod ignores the pulldown flags and displays progressive frames.

[LMotlow]

So noted. Thanks.

[lomaidala]

See attached file: bob (AviSynth).avi
See attached file: bob doubler (VirtualDub).avi

Both show a 3:2 repeat pattern.

How were you able to arrive at this conclusion? And would you say it's hard or soft telecine? I ask because manono said it could be either.

[jagabo]

See attached file: bob (AviSynth).avi
See attached file: bob doubler (VirtualDub).avi

Both show a 3:2 repeat pattern.

How were you able to arrive at this conclusion?

I stepped through your videos frame by frame in VirtualDub.

And would you say it's hard or soft telecine?

You cannot tell with this method. This is only for determining the base frame rate (and for looking for field/frame blending). You can use DgIndex or GSpot to determine if you have soft or hard pulldown. This has been explained already.

[lomaidala]

See attached file: bob (AviSynth).avi
See attached file: bob doubler (VirtualDub).avi

Both show a 3:2 repeat pattern.

How were you able to arrive at this conclusion?

I stepped through your videos frame by frame in VirtualDub.

So I take it then that if a video goes 4-6-4-6, then that means it is actually 3-2-3-2, right?

[manono]

It only means it's not 4-6-4-6. That's why I asked for a sample as I, like jagabo, supposed you either had bobbed in your script before bobbing again in VDub, or something screwy was going on. You can get that pattern with animations where there are often lots of duplicate frames.

[jagabo]

So I take it then that if a video goes 4-6-4-6, then that means it is actually 3-2-3-2, right?

No. That means it's 4:6. Just open the video in VirtualDub, apply no filters, and step through it. You should a 3:2 repeat pattern. If you're still seeing 4:6, go to Video -> Filters... and verify there are no filters (VirtualDub can be set up to automatically apply desired filters at startup).

[lomaidala]

It only means it's not 4-6-4-6.

That means it's 4:6.

Now I count 3-2-3-2. I have no idea how I was seeing 4-6-4-6 before.

Thanks everyone for all of your help.

I've made an attempt organize the information I've gathered from this thread and others into a guide I can use as reference. Maybe it can also help other people who stumble across this thread. If any part of it is wrong, please let me know.

For determining the base framerate of the DVD and whether or not it is soft telecined or hard telecined, here are several methods as they become available throughout the process of making an AVI from DVD.

After ripping the DVD to your hard drive, use DGIndex to open the VOB (or VOBs if there are more than one).

File > Preview

"Video Type" will either say "Film" (soft telecine), or "NTSC" (hard telecine).
If the percentage is lower than 99%, it's hard telecined.
If the percentage is 99% or higher, it's soft telecined.
"Frame Type" will either say "Progressive" (soft telecine) or "Interlaced" (either hard telecine or [if from a video camera] true interlaced video).

Video > Field Operation > Honor Pulldown Flags
File > Save Project

Make an AviSynth script script and open it in VirtualDub.
Bob the video. One way is to use the VirtualDub filter "bob doubler."

Video > Filters... > Add... > bob doubler
Field order > Top field first

Or you can do it in an AviSynth script:

Bob()

Or:

(Yadif(Mode=1,Order=1))

From VirtualDub, save as AVI.
Open the AVI in VirtualDub and count the number of unique (ignoring aliasing artifacts) frames per second during a panning shot.
This quote from manono tells how to interpret the count:

If it's goes 2-3-2-3 with clean frames, it just needs an IVTC. If PAL but looked interlaced when just examing frames but goes 2-2-2-2 with clean frames when bobbed, it needs the fields aligned properly (TFM() by itself). If the bobbed fields are blended/ghosted then you know some sort of a funky conversion was done on the video. For PAL DVDs it's almost always a conversion from 24fps Film. For NTSC DVDs it's usually a conversion from 24.975fps or 25fps PAL. But sometimes they had an NTSC source and it's field-blended from 23.976->29.97fps. To figure which, start counting. If 2-3-2-3, the "base" framerate is 23.976fps. If 2-3-2-2-3-2-2-3, the "base" framerate is 24.975fps or 25fps. To figure which, generally the one that produces a video with fewer resulting blended frames is the better one, but that can be real tough to figure out. In such cases, although 24.975fps is by far the more common (and what the RePAL filter produces, which is what I usually use), 25fps may be the safer one as it's better to get a duplicate frame every now and again than to remove a unique frame. Plus, 25fps allows for some unblend screwups.

That covers the vast majority of movie DVDs. If not a movie but shot on video, then you might have other problems. If anime, and especially if from an anime TV series, you might have a hybrid (a mix of film and video) which can present its own special set of problems. If a silent film, then the "base" framerate can be something other than 23.976fps (NTSC) or 25fps (PAL), and they're sometimes blended as well. Sometimes (rarely, thank goodness) you'll get something that's gone through a couple of conversions and is doubly blended. There are no perfect solutions for that. And there are other strange things out there as well.

Open up the new D2V file created by DGIndex to check the film percentage at the bottom of the file (the same 99% threshold applies).
If necessary, redo the DGIndex process, this time selecting the correct Field Operation (Forced Film vs. Honor Pulldown Flags).

If your source is soft telecined, select this option:
Video > Field Operation > Forced Film
This results in progressive 23.976 fps.

If your source is hard telecined, select this option:
Video > Field Operation > Honor Pulldown Flags
This results in interlaced 29.97 fps.

If your source is hard telecined, your D2V file will now be ready to IVTC (inverse telecine) in your AviSynth script.
After the IVTC, it'll be progressive 23.976 fps.

[manono]

For determining the base framerate of the DVD and whether or not it is soft telecined or hard telecined, here are several methods as they become available throughout the process of making an AVI from DVD.

After ripping the DVD to your hard drive, use DGIndex to open the VOB (or VOBs if there are more than one).

File > Preview

"Video Type" will either say "Film" (soft telecine), or "NTSC" (hard telecine).
If the percentage is lower than 99%, it's hard telecined.
If the percentage is 99% or higher, it's soft telecined.
"Frame Type" will either say "Progressive" (soft telecine) or "Interlaced" (either hard telecine or [if from a video camera] true interlaced video).

If I'm sure there's no true video (It's a movie) and it's not field blended, if the film percentage is anything less than 100, then I always make the D2V project file using 'Honor Pulldown Flags' and set up my IVTC like this:

TFM(d2v="Movie.d2v").Tdecimate()

That way the progressive 23.976fps with soft pulldown will be treated as if you had used Forced Film and only the hard telecined parts will undergo the full IVTC. It's faster than a full IVTC and much less prone to making mistakes. The TFM - Readme.txt file explains all about the D2V setting.

If necessary, redo the DGIndex process...

Although that's one way, it's not really necessary to make a new D2V as the original one can be edited. For example, if you made it using 'Honor Pulldown Flags', at the top it says this:

Stream_Type=1
MPEG_Type=2
iDCT_Algorithm=6
YUVRGB_Scale=1
Luminance_Filter=0,0
Clipping=0,0,0,0
Aspect_Ratio=4:3
Picture_Size=720x480
Field_Operation=0
Frame_Rate=29970 (30000/1001)


If at the bottom it says '100.00% FILM' and you want to use 'Forced Film' instead, just edit the D2V project File so it says at the top:

Stream_Type=1
MPEG_Type=2
iDCT_Algorithm=6
YUVRGB_Scale=1
Luminance_Filter=0,0
Clipping=0,0,0,0
Aspect_Ratio=4:3
Picture_Size=720x480
Field_Operation=1
Frame_Rate=23976 (24000/1001)

And vice-versa.