View Single Post
Old 10th September 2009, 19:28   #5106  |  Link
jdobbs
Moderator
 
Join Date: Oct 2001
Posts: 20,975
Quote:
Originally Posted by Furiousflea View Post
You are correct. Although it's not the fault of x264 but the switches that are being used for encoding.

The problem seems to be that BDRB is not deinterlacing to 60fps and then performing interlaced encoding as it should.

It's missing out the deinterlacing part so we just get some crappy zig zag video.

This is basic stuff!
If its so basic, maybe you can tell me how to fix it?

I remember when I was asked to add the option to keep interlacing -- and I hesitated for exactly this reason... and it's also the reason it isn't the default and it is a hidden option.

I'd especially like to hear how deinterlacing somehow identifies whether the source is TFF or BFF so it can be reconstructed properly (which is the source of the problem we're discussing) -- especially since 90% of the sources are stored as frames rather than fields in the source even when interlaced.

I know I can scan the source, look at flags and determine field order (assuming it isn't a hybrid source and there isn't any pulldown)... but it's anything but basic. Oh... I forgot to mention... there isn't any guarantee that the field order won't change in the next scene -- so I'd have to scan THE ENTIRE SOURCE. At least that was my experience on the MPEG-2 side (and a major cause of my DVD Rebuilder headaches) where I had to scan and record the flags for every single frame so I could restructure it after encoding.

I just love the concept of adding another complete scan of 40 Gigabytes and recording of flags of the source followed by another complete scan/flag setting of the output to what is already too-long a process.

Everybody's a critic.
__________________
Help with development of new apps: Donations.
Website: www.jdobbs.net

Last edited by jdobbs; 10th September 2009 at 19:59.
jdobbs is offline   Reply With Quote