Frameserving DV files kept interlaced to SVCD

[Ivarsson]

This might be the wrong forum but I'll risk it since it is primarily DV related.

I have an Avisynth script that I think alters the field order and keeps the source interlaced. Please comment if there is anything wrong with the script.

Script:
____________________________________
Video = DirectShowSource("D:\file.avi")
Audio = WavSource("D:\file.wav")
#This audio was exported from the editing program.
#I have a Type1 file
Audiodub(Video,Audio)

FixBrokenChromaUpsampling

DoubleWeave.SelectOdd
#This should alter the field order as I understand it

SeparateFields()
#To work with the fields separately

Crop(8, 0, 704, 285)
BicubicResize(448, 272)
AddBorders(16, 8, 16, 8)
#Add overscan to your SVCD

Weave()

ConvertToYUY2(interlaced=true)
_______________________________________

Have I done this script correctly?

[bb]

Instead of DoubleWeave.SelectOdd() I'd use SeparateFields().Trim(1,0), and drop the single SeparateFields() command. You have the Weave() command in the end of your script, so there's no reason to change the field order by creating a frame based output and then split it up into fields again (and weave back).

bb

[Ivarsson]

Thanks for the input.

[FredThompson]

Why are you doing this? I'm guessing it's some kind of space-saving method. Also, why are you removing the overscan source?

[Ivarsson]

I want to keep my files my source interlaced thereby being able to encode it interlaced (which my original script above didn't do).
I remove the overscan (720 -> 704) to get the right aspect ratio. In the future I plan to resize to a little bit lower resolution and add a little more border at the bottom to get the aspect ratio right without losing any of the picture.

[FredThompson]

I see.

You might find this page very helpful: http://www.uwasa.fi/~f76998/video/conversion/

Also, if you're doing passthru or copy-to-tape conversion, you'll probably find most source isn't perfectly centered and just lopping off the edges, while it works, might not be the best approach if you want to preserve as much of the source as possible.

If your source is actually full resolution, why downsize the resolution.

[Ivarsson]

I downsize to see the full picture when playing on a standalone DVD (with overscan).

[Ivarsson]

Thanks for the link by the way.

[FredThompson]

You're welcome.

I meant, if you go from 702 to 480, you're getting rid of 1/3 of the picture information. Why not just keep the original resolution?

This is something I'm struggling with right now, also. That's why I'm so interested in your idea.

Bought a bunch of large drives to hold all the DV I'm working with. Even DVDR is a somewhat limited resource for storing full DV because only about 20 minutes will fit on a disc.

I was toying with the idea of very high bitrate MPEG-2 to store an hour per DVDR. That seemed to make sense because there would be a 1:1 relationship between the DVDR and the reference tape. Should be high enough resolution that the DV tape could just be archived and I'd work from the MPEG-2.

[Ivarsson]

I'm planning to keep my clips in full resolution also for future use with a DVD burner (discs are too expensive at the moment IMO). I will just delete the lines with cropping/resizing/addingborder (and change the script according to the advice I got from bb).

[FredThompson]

Sounds like a good plan. Sure looks like burner prices are going to drop again and maybe the discs will be like CDRs are now by Christmas. If I made burners, even the consumer ones, I'd make it happen to get business.