Pulldown Removal Question

Stumbly wrote on 4/13/2003, 8:47 PM

Hi,
Is the pulldown removal accurate?
I ran a little test to compare 24pA DV Avi to 24p Quick Time. I captured exactly 10 seconds of 24pA DV footage. Then I converted this footage to a 24p Quicktime. I set Vegas to NTSC DV 24p, and checked the "remove pulldown" option. Then I imported both clips and put them on tracks, one over the other. At this point I moved the cursor to different positions and proceeded to "mute" and "un-mute" the top track to compare frames from the different files. I noticed two things:
1) The frames were almost never the same. Shouldn't the AVI and the MOV be exactly the same if the pulldown removal is being porperly performed?
2) I noticed that when I changed from "Draft" to "Best," the avi looked terrible, like 2 frames had been merged into one. Why should viewing as "best" look any different than "draft" exept for the resolution?
3) Is this just a display error? Will it affect my final output?
Thanks!
Mike

Comments

SonyDennis wrote on 4/13/2003, 10:02 PM

If your AVI frames didn't look right, something is wrong. Try it on another 24pA clip. This does work, we were showing it all week at NAB.
///d@

Stumbly wrote on 4/14/2003, 1:20 AM

Thanks Dennis,
I tried another clip from another piece of 24pA footage, with similar results. I'm converting my captured avi to QT via DV Film Maker ( I can't get VV to render avi to QT correctly). Curiously, when I zoom waaay in to the end of the clips the avi ends perfectly at 10 seconds while the QT overshoots by a nose hair; not sure if that means anything... But the image quality of the AVI still looks iffy to me, while the QT looks great.

There's still a strange discrepancy between Draft and Best with the avi. In Draft it seems when I advance frame by frame, every third or fourth frame is held. In Best, the held frame seems to be blurred. Sometimes, like if I ask a friend to look at the problem, it seems to miraculously work, without having changed any settings or even restarted the program or the project.

Stumbly wrote on 4/14/2003, 2:42 AM

Ahhh,
The discrepancy between "draft" and "best" disappears *sometimes* when I change "deinterlace method" to "none" under project properties...
Mike

SonyDennis wrote on 4/16/2003, 11:19 AM

What are the media properties on this clip, specifically, the Format, Frame rate, and Field order.

What are your project settings, specifically, Field order and Frame rate.

///d@

mikkie wrote on 4/16/2003, 2:05 PM

FWIW, with ivt an awful lot depends on your source, how/if you grabbed it, the codecs you use, the original method used to convert the film from 24 fps and so on. In my experience I've seen one prog. work better then another on one video , & the opposite on the next - I do confess to not having run VV4 through it's paces with ivt yet so I can't comment other then to say it likely does very well.

trivia: I have had an occassion or two where the end result was better if I let the encoder do the ivt, probably because the source fed to the encoder was cleaner. Microsoft reccomends ivt whenever possible/needed before encoding to winmedia 9 - the varying motion can drive the encoder nuts resulting in jerky motion in the encoded video - makes sense and might work the same way with mpeg2. Finally, I've seen films where the original conversion was not uniform throughout, & understand it's not that uncommon.