[ Post New Message | Post Reply to this One | Send Private Email to D Gary Grady | Help ]

Response to Is there any current DV system that handles both ntsc,pal and HDtv format editing?

from D Gary Grady (dgary@mindspring.com)
I presume we'll eventually see a multiformat system (in fact I suspect Apple's Final Cut is most of the way there already), but a quick comment on HDTV vs current DV data requirements:

As I understand it, standard resolution NTSC digital video is 720 x 480 pixels per frame. Amazingly enough, this is true of everything from Digital8 all the way up to D1. The different formats differ only in how much they compress. D1 doesn't compress at all. All forms of DV compress by throwing away some of the color resolution (4:1:1 for everything but Panasonic's costly DVCPRO-50) and then by applying a digital compression algorithm to squeeze each resulting frame by a factor of 5 to 1.

To the best of my knowledge, HDTV simply doubles the horizontal and vertical resolution to 1440 by 960. It's basically four standard resolution images stuck together into a 2x2 grid. (The 16:9 aspect ratio is achieved by using short, fat pixels rather than the tall, thin pixels of conventional 4:3 television. If they used square pixels they'd get 3:2.) This is the main reason a broadcaster can send four standard programs or one HDTV program in the same channel.

By using bigger CCDs and four DV codecs working in parallel, it should be possible to put an HDTV image on a standard DV tape running four times as fast as normal. Hence I suspect it won't be long before we see HDTV camcorders. (I also expect they will ludicrously overpriced at first...)

I also expect us to see standard-to-HDTV digital "filter" software that interpolates pixels to give the illusion of higher resolution, so that what you shoot today will still look cool on HDTV.

About MPEG-2 vs DV compression: MPEG-2 is best thought of as a delivery format. It achieves much higher compression ratios than DV because it doesn't just compress each frame separately as DV does, it compresses groups of frames. (That's how you can get a whole movie on one side of a DVD.) But the editable version of MPEG-2 doesn't get any better compression than DV. If you shoot DV there is no point in recompressing to another format for editing unless you absolutely have to.

(posted 9035 days ago)

[ Previous | Next ]