Is DGPulldown a good way to convert PAL to NTSC DVD?

[digicube]

I'm using avstodvd and notice DGPulldown is way faster than MCJMFPS. is it a good idea to stick to DGPulldown for all PAL to NTSC conversions?

[manono]

DGPulldown doesn't convert; it adds pulldown flags. Using it to apply 25->29.97 pulldown is certainly easy. You don't have to convert the audio. You only have to resize the video to 720x480 and encode at 25fps.

If from a film, I prefer slowing the video to 23.976fps and slowing the audio to match. Then you encode for 23.976fps and apply the standard 23.976->29.97 pulldown, aka 3:2 pulldown, during the encode or afterwards with DGPulldown.

This assumes a progressive and film source. If the source is interlaced, you'd never use DGPulldown to begin with. I have no idea what MCJMFPS is.

[LMotlow]

I have no idea what MCJMFPS is.

Apparently a doom9 script by johnmeier and manolito, posted somewhere in doom9 over a year ago, and appears somewhere in a list of AVStoDVD plugin scripts by manolito. Maybe the O.P. could save us a few hours of searching by posting a helpful link.

[dannyboy48888]

I use dgpulldown all the Time. I usually resize PAL to NTSC. On my progressive scan player and PC it ignores the flags and plays at 25fps. On my CRT it honors them and still looks ok on motion scenes. A lot easier than trying to speed up slow down etc or add frames. I also use it on my actual 24fps downscales (snowden, ip man) instead of relying on hcenc pulldown as sometimes it gets the motion jerky.

[jagabo]

MCJMFPS uses motion interpolation via mvtools. It will create lots of distortion with some material. DgPulldown is far better.

[digicube]

My video source is interlaced DV AVI file. I'm using a avisynth script like this Directshowsource(DV.avi).AssumeBFF().ConvertToYV12 (interlaced=true)
Does this converts the interlaced video to progressive? I haven't noticed any bad parts after the conversion PAL to NTSC using DGPulldown.
I also noticed avstodvd imports it as a 5:4 video and adds 40 pixels borders on the left and right side. I just changed it to 4:3 and it get rids of the borders.

[jagabo]

My video source is interlaced DV AVI file. I'm using a avisynth script like this Directshowsource(DV.avi).AssumeBFF().ConvertToYV12 (interlaced=true)
Does this converts the interlaced video to progressive?

No.

I haven't noticed any bad parts after the conversion PAL to NTSC using DGPulldown.

Your source probably doen't contain interlaced frames, even if it's encoded interalaced.

I also noticed avstodvd imports it as a 5:4 video

Because the aspect ratio information is lost in an AviSynth script. All that's left is the 720x576 frame size, a 5:4 ratio.

[digicube]

My source is a VHS capture. I thought all DV videos are interlaced regardless of source.

[manono]

My source is a VHS capture. I thought all DV videos are interlaced regardless of source.

It was encoded as interlaced. The content can be (and usually is, if a PAL film) progressive. There's a difference.

[digicube]

What kind of media does a PAL film uses? video tape?

According to wikipedia, All DV variants except for DVCPRO Progressive are recorded to tape within interlaced video stream.

[manono]

Have you examined the individual frames? Are they interlaced? Do you know what interlacing looks like? You could always cut out a 5-10 second sample for upload here so someone can have a look.

"What kind of media does a PAL film uses? video tape?"

Maybe I should have been more precise. Film, almost by definition, is 24fps. Even in PAL countries a movie, as seen in a movie theater, is projected at 24fps (or integer multiples of 24fps). However for PAL home media and broadcast, that 24fps must be converted to 25fps (or 50 fields per second). And that 25fps is almost always encoded as interlaced, whether the content is interlaced or not.

[jagabo]

Interlaced PAL video is broadcast as 50 fields per second. Each field comes from a different point in time, at 1/50 second intervals. Usually when film is broadcast as an analog PAL signal the film is sped up from 24 fps to 25 fps and each frame of the film is broadcast for the duration of two fields. When you digitize that signal pairs of consecutive fields are woven together into frames for storage. When two fields come from the same film frame the image is progressive -- even if the encoder treats it as interlaced (which is about how the encoder handles the frame internally).