Avoid using VirtualDub to prevent format conversions?

Lyris · 30th May 2012, 17:36

Hey all,

Sometimes when I'm working on DVDs, I'll need to do things in stages. For example, taking a 4:4:4 TIFF image sequence, and using AVISynth scripts to produce a PAL or NTSC res downconvert of this.

Then I'll save this file in VirtualDub (as a Lagarith AVI file, usually in YV12 because that's all many of the AVIsynth filters operate in anyway). Downstream, I'll load this AVI file in another AVISynth script to do more processing.

The reason for the multiple steps is that combining everything into one big script is just too slow.

But, VirtualDub only processes in RGB, correct? So each time I do this, I'm adding a pixel format conversion. Just today I saw one case where the chroma resolution had been mangled by this process. Red text on a nearly black background had colour bleed everywhere.

So, my question is, how should I be doing this to preserve quality wherever possible?

30th May 2012, 17:36	#1 \| Link
Lyris Registered User Join Date: Sep 2007 Location: Europe Posts: 602	Avoid using VirtualDub to prevent format conversions? Hey all, Sometimes when I'm working on DVDs, I'll need to do things in stages. For example, taking a 4:4:4 TIFF image sequence, and using AVISynth scripts to produce a PAL or NTSC res downconvert of this. Then I'll save this file in VirtualDub (as a Lagarith AVI file, usually in YV12 because that's all many of the AVIsynth filters operate in anyway). Downstream, I'll load this AVI file in another AVISynth script to do more processing. The reason for the multiple steps is that combining everything into one big script is just too slow. But, VirtualDub only processes in RGB, correct? So each time I do this, I'm adding a pixel format conversion. Just today I saw one case where the chroma resolution had been mangled by this process. Red text on a nearly black background had colour bleed everywhere. So, my question is, how should I be doing this to preserve quality wherever possible?