XL1 ASA Rating, Part Two
compiled by the Watchdog with Don Palomaki
The Watchdog notes: Here's the long overdue conclusion to an earlier article, XL1 ASA Rating, Part One. Well, hopefully it's a conclusion, but you never know; the discussion about light meters for DV cameras and applying the photographic term "ASA" to video may indeed continue, in which case there may be a Part Three... we'll see. In the first part, some discussion centered around whether a light meter is appropriate for use on video shoots (Bill Farnsworth likes to say that in essence your camcorder already is a light meter and you really don't need another one). Here, we'll talk about how a light meter can indeed be useful in certain ways. Don Palomaki opens, the observations of other pro shooters follow.
Don Palomaki: ISO/ASA/Film speed ratings do not have a precise meaning for video cameras. CCDs have a very different response than film (higher contrast). Also, because video gives immediate feedback you can judge exposure using a field monitor and zebra pattern in in the EVF rather than having to trust a light meter and wait for film to be developed to see the results.
That said, the incident light level required to just hint at zebra on a piece of white poster board at 0 dB of gain corresponds to a film speed of in the range of ISO 160 to 200 when measured with an incident light meter. Using +6 dB gain corresponds to ISO 320 to 400 and +12 dB to ISO 640 to 800. You can push it even 'faster' in post, but image noise may become objectionable for the task at hand.
If you feel a need to use a light meter with the XL1, do your own
calibration tests to find the speed setting that most pleases you
for the subjects you are shooting.
Steve Randall: Keep in mind though that videocameras can add gain to the signal effectively rendering the ASA inaccurate. You'll need to calibrate for each gain level. A companion item to using the a lightmeter for lighting is the waveform monitor and the zebra bars.
The American Cinematographer Video manual, 1st edition, page 137 gives the exact procedure in detail for establishing an expose index. If you don't have the American Cinematographer Video Manual, I recommend that you obtain a copy. It's an excellent reference. American Cinematographer magazine is sold on the newsstands. Although they cater to film shooters, DV and HD video are slowly creeping in.
There are some who believe that a light meter is not needed as the videocamera itself automatically adjusts to give the best picture. I beleve it can be a useful tool, especially when the camcorder is not available.
Doc Rock: Steve's advice is good. I'd just like to add that you use the meter to determine the ratio between key light, fill light, back light et al. I use an incident meter most of the time, which means that, from the subjects position, you point the meter at the light source, and establish a ratio of 2 for key, 1 for fill and 2 or more for the back light (to give great highlights and rim). Measure the key without the other lights, the fill by itself, etc. I'd recommend a meter that lets you measure either incident or reflected light according to the situation.
George Walker: In one sense, it doesn't matter what is the ASA rating of your CCD chips. Lets say you set your light meter to an ASA rating of 100. Typically you would then go to the position where the talent will stand and you take a reading, using the white incident light cover over the sensor, with all the lights on. It'll show something like f/8. Now move around the set to see what the other readings are. Bigger numbers mean there is less light reaching the sensor (f/11, f/16, f/22) and smaller numbers mean there is less light reaching the sensor (f/5.6, f/4, f/2). For still photographers, the reading tells them that with ASA 100 film, to perfectly expose the film at that position with that lighting, set the camera to what's shown on the meter, e.g. f/8... usually with 1/10ths of a stop... f/8 +7/10ths.
With light from lamps, the further away you are, the less light there is. Take a reading six feet from a lamp and then twelve feet away, and you will see that the light avaliable at the second reading is much lower.
A still photographer would take readings at the talent's position with the left lights only to see what that reading should be and then compare it with the light from the right lights. The balance between left and right gives just enough variation to give a look that has some depth. For men, the difference is usually set greater to emphasis a ruggard appearance; but with women, they are closer to soften the face, etc. Once you have played with a light meter, you have a reference point and can reproduce a set or improve on it.
In lots of posts, I have read "for broadcast, pay attention to lighting." Generally I assume that they want a brighter image without hot spots. The XL1 has viewfinder zebra stripes that can be turned on to show where the over-exposed areas are. Once you know they are there, you can adjust your exposure to correct these hot spots until the zebra stripes go away.
Finn Nesgaard: There is one big difference between the DV camera's built-in light metering and using a hand-held light meter as indicated in Georges answer. Traditional light metering mesures the average light and gives an f-stop according to a standard subject with 18% reflection. In the case of exposing an image for DV, it is extremely important to care about the highlights because theres is nothing after the last bit is set. That is why you have the zebra pattern option in the viewfinder.
This means that the light meter's evaluation and the camera's evaluation of the light level is made on quite a different basis. This also means that the two ways of light metering is not identical, but rather supplementary.
I've had the problem of shooting video in India. Because of people's dark skin, I had to expose as light as possible. But still I had to be very carefull about the high lights. The two ways of mesuring gave extra security.
Bill Pryor: You can use a standard gray bounce card to get a reading. Light it, zoom in and adjust aperture till you see the zebra pattern in your viewfinder. Let's say that it comes out to a f/5.6. Take a reading with your light meter and adjust it so you make it read at f/5.6 under the same conditions.
D. Gary Grady: The problem with that is the zebras indicate overexposure (brightness of 95 IRE or more), or roughly caucasian skin tone (70 IRE or over), not the middle of the exposure range which is what the light meter is going for. A better bet would be to set the camera shutter speed for 1/60 (or 1/50 for PAL) and use Shutter Priority automatic exposure (Tv mode in the XL1). Note which f-stop the camera uses, then find an ASA setting on the light meter that gives you the same exposure.
There are people who question the utility of using a light meter for videography anyway. Using it for setting exposure is certainly questionable. You're better off using viewfinder zebras to avoid any areas of overexposure (a few small highlights are okay), and then fill in to bring up the shadow areas.
Rather than use a light meter to determine image exposure, in video a better application is to use it when lighting a set. Periodically, someone will complain that it's a pain to have the camera shut down in standby mode while in the middle of setting up lights, because they want to check exposure with the camera. People coming from a film background chuckle at this, especially when the same video people insist that there is no possible application for a light meter in shooting video.
If you use a light meter (preferably an incident meter or a spot meter) for that, then you don't really care about the ASA equivalent of the camera... only its dynamic range in stops.
Finally, it's useful to read some basic information about the zone system -- not the darkroom part, but the basic ideas of mapping different brightness levels in front of the lens to brightness levels in the captured image. Once those ideas are understood, a lot of questions suddenly answer themselves.
Back to the XL1 Articles Menu
See also XL1 ASA Rating, Part One
Thrown together by Chris Hurd