PDA

View Full Version : 3/2 pulldown question


Mounir
19th April 2009, 22:55
Is it wise to use a 3/2 pulldown scheme with a 30fps/progressive video (think youtube video which have been deinterlaced)

Thanks in advance for your inputs

Leak
20th April 2009, 00:39
Is it wise to use a 3/2 pulldown scheme with a 30fps/progressive video (think youtube video which have been deinterlaced)
Ummm... 3:2 pulldown is used to take 24FPS video to 30FPS (well, roughly) - if you apply it to 30FPS video you'll get 37.5FPS video, and that's quite... irregular, to say the least...

manono
20th April 2009, 03:36
He may be talking about trying to IVTC a 30fps progressive stream, and the answer is still "no".

shoopdabloop
20th April 2009, 04:16
30p-to-24p'd video is extremely annoying to watch, even with ConvertFPS(). Don't do it.

Mounir
20th April 2009, 15:45
Well actually, i got canopus procoder 2 and there is a "pulldown filter" (not 3/2) available.

Take a look at the picture & description: http://img168.imageshack.us/img168/9633/pulldown.jpg

It seems to be different than a 3/2 pulldown the which change the frame rate (24fps->30fps) & frame structure (progressive -> interlaced)

Well i just needed confirmation i tried this filter & the result is very good to me

neuron2
20th April 2009, 16:37
Looks like standard 3:2 pulldown for film rate to NTSC video rate.

You applied this to an existing 30P source? Why would you do that or think you need to?

Mounir
20th April 2009, 16:49
Why would i need to.. ? To reinterlace properly a video that was deinterlaced for the web

neuron2
20th April 2009, 20:37
Reinterlacing (converting double rate progressive to single rate interlaced) is not pulldown. And the PNG you posted talks about changing film rate to broadcast rate, which has nothing to do with reinterlacing double rate progressive back to single rate interlaced. So you continue to make no sense.

If you reinterlace 30P you will have 15i. What's the point?

Mounir
20th April 2009, 22:34
Yeah you're right, i just got confirmation from canopus.