Interlaced AVI 2 SVCD

[manolito]

Two questions about converting interlaced AVI sources (analog VCR captures) to SVCD:

Usually when I do this I deinterlace the movie for better compressability. For better quality I also did a few interlaced encodes. In those cases I just chose the "Interlaced" template for CCE, and the result looked pretty good. Now I read that the resizer will scramble the fields unless you separate the fields before resizing and then weave the fields together again. The script should be something like this:

SeparateFields()
Resize (with half of the original height)
Weave()
AddBorders

Now my questions:

1. Is this really necessary? I did a few encodes (source 768 x 576, target 480 x 576) with and without separating the fields, and the results looked pretty much the same to me.

2. If I do use SeparateFields, should I insert the command just before the resizer, or does it make sense to separate the fields before applying preresize filters like undot, deen, fluxsmooth and unfilter?

Cheers
manolito

[Kika]

1. No! If the vertical resolution isn't changed, you don't need to separate the Fields. But... look at 2.

2. If you want to use some Filters/Resizers, always first separate the Fields. All temporal filters do need the field separation.


So: First separate the fields, than filter, than resize...

BTW: If you are capturing at 768x576, you should letterbox it first to 786x576, then do a resize to 480x576.
Or capture at 464x576 and letterbox it to 480x576.

[manolito]

Thanks for the clarification.

Cheers
manolito