squeeze or no squeeze?

Jim McQuaid · March 28th, 2005, 07:54 AM

I honestly have no idea what John was referring to with "the clear advantage of lighter compression." So let me explain what I think and you can educate me.

First of all, I am only looking at the miniDV result from the DVX. I'm not trying to make any general statements about other tools and processes, how it should work in HD or going to film.

Second of all, I'm only thinking of the standard camera setup, not the anamorphic adapter.

Within these limits, it seems to me that with most miniDV cameras and with the DVX, you can choose to lose resolution in horizontal or vertical by cropping or stretching. There are still only so many pixels to go around. And if I'm going to end up discarding (or interpolating) some pixels in the process to achieve the 16:9 ratio, I'd rather do it in a way that provides the greatest freedom for making choices (and undoing them if necessary).

That's why matting in post appears to be the best choice.

John McManimie · March 28th, 2005, 02:52 PM

I think what Jon Fordham is referring to is the 5:1 compression of DV. When you crop 4x3 video you are throwing away pixels AFTER the 5:1 compression stage so that 1/4 of the data space on the tape is unused. While the same number of pixels from the CCD end up in the frame, shooting 16x9 in camera allows those pixels to be interpolated across a much greater data space and should result in a better image. Of course, it might depend on your target - 16x9 or 4x3. If 16x9 TV is the target, I would think that in-camera cropping would be the better choice (from a quality standpoint --- you do, of course, lose the abilty to do things like re-framing).

Jon Fordham · March 29th, 2005, 01:45 PM

Jim,

John is correct. By eliminating the top and bottom areas of the image that are going to be cropped out in camera, the compression can focus on only the area of the picture that matters. Thus you get lighter, better compression which results in a noticeably superior picture.

I never have understood the idea of "undoing" or re-framing compositions. And I've never understood the fear of being "stuck with" a composition. I've met plenty of so called directors who can't make a decision to save their life. But I've never met a shooter that was too indecisive to figure out how he wanted to frame a shot.

If it's freedom you're worried about, consider that by achieving an image with lighter compression you'll have better source material to stretch and manipulate. Like you said, it's being discarded anyway. So throw them out to start with so that you have better material to manipulate.

Daniel,

If you're interested in doing 2:35, then the DVX with Anamorphic adapter can do it. By using the anamorphic lens AND the in camera letterbox on the DVX100, you'll get a 2:35 image when viewed on a 16/9 monitor or leterboxed on 4/3. If you have the DVX100A, you could also use the in camera squeeze with the anamorphic lens to achieve 2:35. With the squeeze mode you'll have to "double" the letterboxing in post.

March 28th, 2005, 07:54 AM	#16
Jim McQuaid Regular Crew Join Date: Nov 2003 Location: Durham, North Carolina Posts: 61	16:9 in post I honestly have no idea what John was referring to with "the clear advantage of lighter compression." So let me explain what I think and you can educate me. First of all, I am only looking at the miniDV result from the DVX. I'm not trying to make any general statements about other tools and processes, how it should work in HD or going to film. Second of all, I'm only thinking of the standard camera setup, not the anamorphic adapter. Within these limits, it seems to me that with most miniDV cameras and with the DVX, you can choose to lose resolution in horizontal or vertical by cropping or stretching. There are still only so many pixels to go around. And if I'm going to end up discarding (or interpolating) some pixels in the process to achieve the 16:9 ratio, I'd rather do it in a way that provides the greatest freedom for making choices (and undoing them if necessary). That's why matting in post appears to be the best choice.

March 29th, 2005, 01:45 PM	#18
Jon Fordham Major Player Join Date: Sep 2003 Location: New York, NY Posts: 242	Jim, John is correct. By eliminating the top and bottom areas of the image that are going to be cropped out in camera, the compression can focus on only the area of the picture that matters. Thus you get lighter, better compression which results in a noticeably superior picture. I never have understood the idea of "undoing" or re-framing compositions. And I've never understood the fear of being "stuck with" a composition. I've met plenty of so called directors who can't make a decision to save their life. But I've never met a shooter that was too indecisive to figure out how he wanted to frame a shot. If it's freedom you're worried about, consider that by achieving an image with lighter compression you'll have better source material to stretch and manipulate. Like you said, it's being discarded anyway. So throw them out to start with so that you have better material to manipulate. Daniel, If you're interested in doing 2:35, then the DVX with Anamorphic adapter can do it. By using the anamorphic lens AND the in camera letterbox on the DVX100, you'll get a 2:35 image when viewed on a 16/9 monitor or leterboxed on 4/3. If you have the DVX100A, you could also use the in camera squeeze with the anamorphic lens to achieve 2:35. With the squeeze mode you'll have to "double" the letterboxing in post. __________________ Jon Fordham Director of Photography www.jonfordham.com

March 28th, 2005, 02:52 PM	#17
John McManimie Major Player Join Date: Mar 2004 Location: Portland, OR Posts: 314	I think what Jon Fordham is referring to is the 5:1 compression of DV. When you crop 4x3 video you are throwing away pixels AFTER the 5:1 compression stage so that 1/4 of the data space on the tape is unused. While the same number of pixels from the CCD end up in the frame, shooting 16x9 in camera allows those pixels to be interpolated across a much greater data space and should result in a better image. Of course, it might depend on your target - 16x9 or 4x3. If 16x9 TV is the target, I would think that in-camera cropping would be the better choice (from a quality standpoint --- you do, of course, lose the abilty to do things like re-framing).