Low light in HDV vs DV?

[Kevin Shaw]

Has anyone tested the FX1/Z1U to see if they get a better low-light image in DV mode versus HDV? I've been assuming it would be about the same, but someone with a different HDV camera says he sees a noticeable difference between the two recording modes.

[Matt Vanecek]

I don't see a difference, really, on my FX1. I just shot about 25 hours SD on my FX1--I'll have to go back and compare it to the HD footage, but after bringing the footage into PPro and editing, I can't say it struck me as any different visually than HD I captured and worked with. But perhaps that's just my equipment (my computer monitor and a, um, a, um...."client monitor", yeah, that's it...).

I generally shoot between 1.8 to 2.4, 1/60, +3db for what I *usually* shoot, although I've bumped up to +6db in strenous situations.

HTH,
Matt

[Lou Bruno]

No difference that I can observe. I also don't see any difference when I downconvert from HDV to SD. But, maybe it's my old eyes.

[Boyd Ostroff]

If you're shooting DV then you can drop the shutter speed to 1/30 sec without a significant loss of quality. If you do this in HDV mode then you'll notice lost resolution as the result of field doubling. You also might not notice a gain boost as much in DV mode.

But other than that, I'd have to agree with the others here.

[Kevin Shaw]

Originally Posted by Boyd Ostroff

If you're shooting DV then you can drop the shutter speed to 1/30 sec without a significant loss of quality. If you do this in HDV mode then you'll notice lost resolution as the result of field doubling.

Would motion look the same at 1/30 in DV as it does in HDV? Can you elaborate on the technical difference at 1/30 between the two formats?

[Boyd Ostroff]

If you set the shutter for 1/30 second then it remains open for both 60i fields. As a result of this, since the Sony cameras don't have progressive CCD's, the same data gets written to both interlaced fields. This is called field doubling, and since every other line in the image is the same you have cut your vertical resolution in half - from 1080 to 540. But if you shoot in NTSC DV mode, only 480 lines are needed for a full resolution image so the camera can sample those from the 540 unique lines.

At 1/30 sec shutter speed you will get motion that looks more "filmic" because of the motion blur - I used this on my DV camera from time to time, but the image looked soft due to the field doubling. On the FX1 and Z1 I'd assume you get something which looks more like 30p with this trick in DV mode, but I haven't played with it myself.

See the Cineframe article on Adam Wilt's website: http://www.adamwilt.com/HDV/cineframe.html

Because the field-tossing happens in HD, the downconverted SDTV output looks considerably better than field-doubled standard definition. Instead of being half as good as normal interlaced mode, Cineframe looks only slightly worse than interlaced. I definitely prefer the FX1 to any of my SD DV cameras if I have a need to shoot slow shutters or "the cineframe look" for standard-definition release.

[Ervin Farkas]

Boyd,

have you done any testing on this? Can you conform from experience (not just theory) that it goes as described? I will do some testing myself but I'm trying to get some additional info.

This trick would come in handy as I'm shooting sometimes in very low light situations with the Z1, and my final product is in most cases standard definition anyway.

What would be the better workflow in this case? Shoot and edit HDV and resize for DVD, or shoot DV (in which case if I understand this right, the field doubling would be done in the camera, before the downconversion)?

Thanks,