How (exactly) does my camcorder generate interlaced output?

[wayback]

I have a Canon Elura 100 miniDV NTSC camcorder, and the DV output is interlaced, the manual says. I'll need to convert a number of tapes from this camcorder to DivX format, and the resulting files will be viewed pretty much exclusively on computers. So, I've reluctantly accepted the need to produce progressive DivX files.

But as has been pointed out here many times, deinterlacing has its price in picture quality. I'm going to be using AviSynth and VirtualDub (in fast recompress mode), so I hope I'll have a fair amount of control over exactly how things are done. My assumption is that I'll get a better deinterlaced view that way than relying on on-the-fly deinterlacing during playback on a variety of relatives' computers.

But I wonder whether a better understanding of how the interlaced DV is generated by the camera might help me pick the right filter in AviSynth. In particular, if it isn't really true interlacing, maybe I can pick a deinterlacing filter that's less destructive.

When I think of interlacing, I go back to the electron gun. In old CRT cameras, it would scan alternate fields in succession, one line at a time. So not only are the fields separate, but time actually flows evenly throughout the scanning process at 1/30 sec. per frame.

Do modern camcorder sensors allow things to be done the same way? Or do they have to use tricks to approximate the same thing - such as actually sampling the *same* pixels on each "field" instead of different ones? Do they scan along from one pixel to the next, or do they latch each line all at once? Or even each field all at once?

Well, I guess what I'm hoping is that I can use tricks to get back to a progressive scan instead of a full, official deinterlace, with better quality.

I would appreciate any thoughts on this. Also, if there's a link to a detailed description of exactly how these cameras generate the interlaced output, that would also be helpful.

One other question. When I simply view in Media Player the DV avi file captured by WMM, I don't see any interlacing artifacts. Assuming the file on the tape is really interlaced, where is the deinterlacing being done - by WMM during the capture, by the MS DV codec, or by MP? (When I open the file in TMPGEnc 3, it scans it, and then confirms it's interlaced, bottom field first, at least in structure.)

Thanks for any help.

[lordsmurf]

It does not generation interlace, it shoots natively in interlaced formats. Your playback may have an smart de-interlace filter being dragged over it live during playback. A few exist, for playback only.