One of the most confusing sections of Read ErMaC & AbsoluteDestiny's Friendly AMV Guides is the part about aspect ratios and resizing. I feel that much of this complexity comes from the fact that it gives you a choice of editing at the original resolution of 720x480, or resizing to square pixels. My removing the parts about editing at the original size, this could greatly simplify the guide.
My opinion is that situations where you can edit at the original size are few and far between, and even when you CAN do that, there isn't much benefit from doing so.
First off, editing at 720x480 is rather situational, for instance:
- you can't really mix different series, because they all need to at least be cropped differently, and if they use different aspect ratios then it simply isn't an option at all
There are two primary arguments for editing this way that I have heard. I would like to attempt to debunk them.
1) If your source is 4:3, you lose a lot of resolution by downsizing to 640x480. Then when you send it to a convention, you have to resize back to 720x480. It would look much nicer if you had the original 720x480 quality.
It probably wont look any better. Seriously. Take your 720x480 footage, save a still image from it. Then do a Spline16Resize(640,480).Spline16Resize(720,480)
I can't see any difference at all unless I zoom in pretty close, and even then its still a VERY minor difference. The truth is, the resizer compensates for that lost resolution pretty darn well.
2) If you are editing anamorphic footage, your lossless intermediate files will be a lot bigger.
From my tests, lossless lagarith files in yv12 colorspace at 848x480 tend to be about 10% bigger than 720x480 files. In the grand scheme of things, that's not very much at all, really.









