Interlaced, progressive - give it to me straight

doin wrote on 8/28/2004, 7:16 AM
I have read and read about this but I still don't totally get it.

I am using Vegas 5 to capture DV from my camcorder. I am then frameserving it to cinemacraft SP to create a DVD compatible files.

The DVD will be played on a regular 32" television right now. However, there will be a time when I will play it on an HDTV ready television

My question is: How do I render/encode the file for compatibility on both? What settings do I make in CCE now? Does pulldown come into effect at all?

As I've understood bought movie DVD's, when you play them on a HDTV ready television using component cables, you can press progressive on the DVD player and it will play in progressive mode (480pi) for a better picture.

This is basically what I want to do now - create home movie DVD's that show perfect on a regular television now and compatible with an HDTV television later.

Much appreciated!

Comments

John_Cline wrote on 8/28/2004, 7:39 AM
Just encode them at the standard interlaced 720x480 (or 720x576 if you're in PAL-land). Anything else will just compromise the quality of the original footage. Progressive does NOT necessarily produce a better picture unless the footage was progressive to begin with! Read my post at the following link. I'm talking about 24p but the same arguments about temporal vs. spatial resolution still hold for 30p.

Progressive vs. interlaced

John
farss wrote on 8/28/2004, 7:41 AM
Given that you've most likely shot in interlace you're best off staying that way. The only options you have are standard NSTC DVD or 24p, which is what Hollywood mostly author in. However they're coming from film to start with.
You could convert your interlaced video to 24p and end up with a 'film look' of sorts but in terms of image quality you're going backwards.
The only other option is buy a DVX100 or XL2 and shoot progressive. But even though these are both excellent cameras and shooting progressive does give a boost in vertical resolution you're still going to face another issue, you're still stuck in SD resolution. You could consider a HDV camera, maybe Sony will make a decent job of their new one.
Or you could simply be happy with what you've got, the DVDs you've made will probably fail before you don't have a way to play them anyway.

Bob.
farss wrote on 8/28/2004, 7:49 AM
John,
whilst everything you say in that post is true it does ignore one little know fact. A camera shooting interlaced doesn't entirely throw away every second line in each field, doing so wastes data that could be used for noise reduction. So line averaging is bought into play, this gives you an extra 6dB of gain however it does drop the vertical res back around 30%.
In progressive scan they cannot pull this trick, so you pickup 30% more resolution, albiet at the expense of gain.
Of course this only applies in the camera, once the footage is shot as you said nothing is gained converting to progressive and much is lost.

Bob.
johnmeyer wrote on 8/28/2004, 12:15 PM
Pulldown is only an issue if you start with 24p (or film). In that case, encode at 23.976 with the pulldown flag set. Your DVD player will do the right thing (i.e., insert extra frames if you display on NTSC 29.97 interlaced display). However, if you start with 29.97 video, you do NOT want to eliminate frames in order to get to 24p. You are just throwing away information (frames), and that won't help.

John, I just read your post that you linked to. Awesome! I really appreciate the quality of your essay. Much, much better than most articles in the various video magazines I get.

Question: You mentioned the difference in depth of field between film and video. You said it has to do with the small sensor and the lenses used in video cameras. I have read many articles written by Hollywood types that talk about this same problem, but I'm still not sure I understand the physics of what causes this problem. If I shoot at an f2 aperature, then I should get great depth of field. However, to do that, I have to increase the shutter speed. Is the problem with video that increasing the shutter speed isn't done because it changes the "feel" of the video? The sensor size just doesn't seem right to me because I think you can get depth of field in the film world whether you are shooting 70mm, 35mm, 16mm or even Super8.
John_Cline wrote on 8/28/2004, 3:56 PM
John,

It all has to do with the size of the sensor, whether it's a camera chip or a film frame. Assuming you use the same focal length and focus distance, the larger the image sensor area, the shallower the depth of field for a given F-stop. Someone wrote a great article on the issue, I can't seem to find it, but I'll keep looking. In the meantime, check this article written by D Gary Grady:

Depth of Field

John
johnmeyer wrote on 8/29/2004, 8:06 AM
check this article written by D Gary Grady

I read it cover to cover. It has exactly the information I was looking for. I now understand. Thank you!!

BTW, I just ordered the Nikon D70 still camera (which I haven't yet received) and was curious how much of an issue this will be. Went back and checked the specs, and it has a very large sensor that begins to approach 35mm size (although it is still smaller). Thus, I should still be able to take pictures with things in the background out of focus.