ISO for Mini DV

[Christopher Dean Harring]

I couldn't find reference to this, so I figured I'd put it out there: I use a light meter to set up lighting ratios and find hotspots in the frame, and I was curious if there is an ISO setting (simulating, I guess, the ISO of the tape or the CCDs) that might accurately reflect the F-stop readings I am taking. I know this is a stretch, but I come from a film background where the light meter is God.

[Charles Papert]

Christopher, you might try searching the archives here, it has indeed been touched on. Each camera has a slightly different rating, of course; most current DV cameras seem to be in the 400 to 640 ASA range at 0 db. Easiest way to check yours is to shoot a grey card, see what aperture the camera selects and then true your meter to that by changing the ASA setting.

That said, I too come from a film background and I rarely pull out my meter when shooting digital, except for pre-lights and special occasions. Video has a different exposure curve than film and what the meter reads sometimes doesn't apply. The best choice is to have a good broadcast monitor (and if possible a waveform monitor) and judge the image itself for hotspots, ratios etc.

[Barry Green]

Yeah, search the archives, because it's not as easy as just calibrating the meter to the camera, because CCD's apparently respond to different light levels at different sensitivities (the response is not directly linear).

In testing the DVX100, at light levels low enough for the camera to require f/2.8 in order to deliver 55 IRE to a gray card, the corresponding ISO was about 400. At light levels bright enough such that 55 IRE required f/16, the corresponding ISO was 800 or 1000, depending on the gamma curve selected! For most of the exposure range it was ISO 640, but at the limits the response was not linear, which makes using a lightmeter of somewhat limited value.

[Walter Graff]

<<<-- Originally posted by Barry Green : Yeah, search the archives, because it's not as easy as just calibrating the meter to the camera, because CCD's apparently respond to different light levels at different sensitivities (the response is not directly linear).

In testing the DVX100, at light levels low enough for the camera to require f/2.8 in order to deliver 55 IRE to a gray card, the corresponding ISO was about 400. At light levels bright enough such that 55 IRE required f/16, the corresponding ISO was 800 or 1000, depending on the gamma curve selected! For most of the exposure range it was ISO 640, but at the limits the response was not linear, which makes using a lightmeter of somewhat limited value. -->>>

Wow Barry! one of the smartest posts I've seen on ten boards in five years!
You may be the only one to speak the truth. Even on very pro boards I belong to folks walk around talking light meters and EI ratings (only still cameras can use an ASA rating) like there is some stability to video. In technical terms every camera has a skiable MTF(modulation transfer function). This slope tells use how well a camera sees white to black. Problem is light falling on CCDs isn't linear along that slope. And the microlenses(little pieces of plastic on the front of CCDs that act like a lighting magnifier) they add to the front of CCDS (the call them HyperHAD chips), that are supposed to make the chips more sensitive (but they help least when yo need them most, when you are wide open). So yes, you can use a light meter with video but no you will not get a completely accurate reading. That doesn't stop some folks though. Either they came from film and feel insecure or they like to use one because they always wanted to shoot film but never could and it makes them feel good. Your monitor is your best light meter and a waveform monitor is the only accurate light meter you will find in video.

[Christopher Harring]

Thanks for the information. It sounds like I can continue using my meter to establish ratios and the like, but I'll definitely start dragging my monitor out to the set with me.