The Watchdog notes: Sometimes (far too rarely) on usenet, there's actually some very good information exchanged between people who really know what they're talking about. One such instance occured recently on the newsgroup rec.video.production, in which a simple question inquiring about the XL1's equivalent ASA rating was responded to by a wide variety of professional videographers who debated the issue quite thoroughly. In fact, they refused to leave it alone and the resulting discussion went on for a couple of days. This is concluded in XL1 ASA Rating, Part Two.
It was just too good not to include here. You could go back and search Deja News for old usenet posts, but it's a lot of work. So, I thought I'd turn it into a Watchdog article. Trouble is, usenet posts generally happen in a non-linear fashion. Different threads will each spin off on sub-topics with a sort of fractal nature; consequently it's difficult to present all of it here in a linear manner and still make complete sense. So, what I'm going to do is to copy the posts in their logical order as an ongoing dialog between participants, and a little later, backtrack just a bit in order to divert off into the various juicy tangents.
Participants in this discussion are Christopher Lockett, Bill Farnsworth, Doug Graham, Eric DeBlackmere, S.O.C. (Society of Cinematographers), David Patterson, R. Geoff Baker and Phil Taylor.
Question: What would the ASA rating be for the XL1, assuming gain is set to zero?
Christopher Lockett: If you've got a ambient (or even a spot) meter and a gray card, you can test the camera yourself. Pretty simple, actually: Lock the camera down on a tripod, set auto iris, zoom into to a gray card that's flat-lit with a chimera or the like. See what f/stop your iris rolls to. Then, take your meter, set your shutter speed to 1/60th of a second, punch in an arbitrary ASA like say, 250, and take a reading. Manipulate your ASA until the f/stop on your meter matches the f-stop reading on your camera. Just be be safe, walk the light out and walk it in, making use of the inverse square law and re-do the test under different footcandles, low and high.
Once you've got your ASA, it makes your light setups much faster.
Bill Farnsworth: The method of figuring out the ASA is pretty well described. The ONLY thing that I have had a problem with in past discussions over the years, concerns the shutter speed setting. And because of this... is an ASA comparison even valid. Or possible. The manufacturers don't think so. I tend to agree. Here's why.
In NTSC, 30 frames are two fields at 1/60th each. Combined they make one frame every 1/30 of a second. We all know this... BUT... A frame of FILM, shot at 1/60th of a second is darker than a frame of FILM shot at 1/30th, given that the f-stop hasn't changed. One field of NTSC video isn't darker than a one frame of video. By looking at a two field image on a waveform monitor you can see that the video peaks are still the same when you look at the combined, or, one full frame on the waveform monitor.
And what about PAL? That's 25 frames a second. And that would mean a video image from an identical PAL camera would have an even faster ASA rating than an NTSC camera. And that isn't true.
That's why video cameras are rated in LUX.
Doug Graham: Yeah, but the ASA "rating" of video cameras does have some merit in comparing their low light abilities, especially for those with a film background.
Bill Farnsworth: I don't think that there is any merit to it. Or even if it is helpful at all, because it won't be as accurate as the camera itself. It might come close, but why bother if you have to use the video camera for the final exposure anyway. Use the video camera as it was intended and to it fullest capabilities then there isn't any need for an unnessessary crutch.
A video camera's viewfinder zebra pattern and/or a quality color field monitor properly set up, is all that is needed. And these two pieces of equipment are already on location or on the set. And since they are already there, then why bother adding the cost of a film light meter to the package.
Doug Graham: More importantly, knowing your camera's "ASA" allows you to use light meters that are designed for film cameras, which are MUCH cheaper than light meters designed for video cameras.
Bill Farnsworth: I understand what you are saying Doug. But a video camera is really nothing more than the worlds most expensive light meter.So using a film light meter on a video shoot is a waste of time.
Doug Graham: I don't think the field/frame thing has anything to do with it; the normal shutter speed of an NTSC video camera is 1/60 sec, and that's what you should use.
Bill Farnsworth: Video doesn't have a normal shutter speed. Professional cameras have a Mode called "shutter speed" that starts at 1/100th and goes up to 1/2000th at 30 frames and 2 fields. When the shutter mode is turned off it reverts back to 30 frames per second/2 fields. That doesn't make it a 1/60th sec. "exposure".
Again, NTSC video cameras do not have a normal shutter speed of 1/60 sec. They have a two field system that scans every 1/60th of a sec to make one frame 30 times a second. It's not the same. Shutter speed affects exposure. This doesn't. Otherwise, (like I said earlier) a PAL video camera 25 frames would require less light to be exposed at the same f-stop as a NTSC 30 frames camera. Professional cameras (and most others) have an "elecrtonic shutter" mode that goes from 1/100th to 1/2000th.
(BW) In my experience, unless one is fiddling with the shutter for effect, 1/60th of a second IS the normal exposure time on any NTSC video camera.
Bill Farnsworth: As I said earlier, if you are using shutter speed as a viable factor, then why is a PAL camera's light sensitivity the same as a NTSC camera's, when both cameras are identical models and manufacturer? Just some more thoughts.
Eric DeBlackmere, S.O.C.: I've heard the argument before that a video camera is the world's most expensive light meter so there is no need to ever use a light meter with video. I have to disagree. Some reasons to use a light meter:
First, Auto Iris can be fooled. A light meter gets you closer to your mark and lets you see what is going on with regards to lighting.
Bill Farnsworth: Auto iris is for the weak of heart and the inexperienced. I understand what you are saying Eric but the zebra pattern in the viewfinder and how to interpret that information is what really counts. Unless there is a waveform monitor on the set. This is the best way for video.
Eric DeBlackmere, S.O.C.: Second, a light meter helps you maintain consistency in your lighting set-ups from scene to scene.
Bill Farnsworth: It certainly can't hurt. And definitely can speed up a shoot if the gaffer can jump ahead to the next set up. And all you want to do is match footcandles But there is no way that a film light meter should be used for the final critical exposure.
Eric DeBlackmere, S.O.C.: Third, and most importantly for me, when I am scouting locations I take my meter with me. Since it is calibrated for the EI rating of my camera, I can use it to check the lighting levels at my locations and to quickly figure out how much lighting I will need and where I will need to place it before I ever bring one piece of gear onto the location. Throw in my director's viewfinder and I have all I need to plan my set-ups without having to haul my camera around with me.
Bill Farnsworth: Understandable, but after 20 years, I still use my camera for scouting. They have certainly gotten lighter over the years and I see no reason to use anything other than the bottom line tools.
Eric DeBlackmere, S.O.C.: A light meter is not an essential tool to have but I strongly recommend them as a great tool to have to help in planning, set-up, and to help maintain consistency in your lighting.
The Watchdog notes: It will be seen that what Bill Farnsworth says is very good at prompting well-considered, intelligent responses from a variety of knowledgable folks. Here's the first of several.
Bill Farnsworth: But a video camera is really nothing more than the worlds most expensive light meter.So using a film light meter on a video shoot is a waste of time.
David Patterson: Well, video cameras are also used for shoots other than news. Let's say you are doing a quality industrial on a well-lit set. The set walls are visible behind the foreground action which involves various props and actors over a period of days. As each dark or light prop is placed on the set, you can't just set exposure based on the zebra or the auto iris function of the camera, because the set walls will get brighter and darker accordingly. You have to keep the iris constant, and alter your lighting on the foreground action and/or props. Thus, one need for a light meter.
Another case: you are shooting a series of interviews in many different towns. You want all of them to have the same key to fill ratio for a constant look and style. The light meter makes this must easier to accomplish.
It is certainly possible to shoot video without a light meter. I do it all the time. But on a shoot where you are going for exact matches and a consistant look, the light meter and the waveform monitor are both very useful.
Also, it is generally accepted that NTSC video cameras have an effective "shutter speed" of 1/60th of a second with the electronic shutter off. And yes, I know that there is no shutter and yes, I know that ASA refers only to film stocks and not to video cameras. But using the ASA idea IS helpful to videographers in some situations.
The Watchdog notes: recall earlier this exchange between Doug Graham and Bill Farnsworth, which will now bring new responses by R. Geoff Baker and Phil Taylor.
Doug Graham: More importantly, knowing your camera's "ASA" allows you to use light meters that are designed for film cameras, which are MUCH cheaper than light meters designed for video cameras.
Bill Farnsworth: A video camera IS a light meter.
R. Geoff Baker: You do raise some questions... but I'm left believing the pursuit of a video ASA equivalent is valid.
First, I don't follow your 1/60 vs. 1/30 argument: In fact I find it incorrect. If I lock the aperture on my VX-1000 and then switch to 1/30 of a second (from the 1/60 default) the image IS brighter... by exactly one stop.
As to why shouldn't there be a difference between PAL & NTSC: I won't know 'til I compare two otherwise identical cameras. But I'd expect that a variation would be entirely reasonable, derived from the different 'shutter speed' and the different pixel density per frame.
Next, you could rate film in lux; the terminology predates video and is not video specific. It is the reverse approach - ASA describes sensitivity but not light level, lux describes light level required and only implies sensitivity.
Finally, once you establish the ASA equivalent for your video device, it holds for any light level measurement you care to make, given the shutter speed / gain factor... so why not?
Phil Taylor: Yes, you can establish a relationship between a film light meter reading and the exposure setting for your video camera. But there are different variables that will affect the relationship as you go from one camera to the next (I guess you all agree to that). But if I were attempting to find such a relationship I would go to the back yard when the sun was bright but not glaring and stay out of the shade, hook up my camera to a good monitor, set the aperture to manual at f/16, white balance the camera using a white card, then video an 18% grey card while on manual aperture. Then take a reflected light meter reading from the grey card with the light meter setting at 1/60th for shutter speed and determine what the aperature setting would be to zero the light meter needle. Then remove the grey card from view and observe your monitor to assure yourself that the video image is good. If it is then you have established a relationship that can be relied on in the bright light.
Overcast skies will change the relationship somewhat and the brighter or dimmer the light the more variation in the relationship. Therefore, it is much safer to manual white balance using a white source to reflect the available light than it is to rely on a film light meter. Even if you take light meter readings out on some location you will still have to shoot it with a video camera that has been white balanced. So, keep the video camera on manual... and as the light changes do another white balance. You'll be amazed at the good color saturation you will lay down in all instances. The manual white balance is truly the video camera light meter. Use it often.
The Watchdog notes: once again, an excellent additional thread was spun from another of Bill Farnsworth's posts, this time involving Phil Taylor, Eric DeBlackmere, S.O.C and David Patterson:
Bill Farnsworth: That's why video cameras are rated in LUX.
Phil Taylor: I agree with Mr. Farnsworth on this but go one step farther. ASA (American Standards Association) applies to the capturing entity, i.e. film. It is in effect the speed of the film and is governed by the emulsion placed on the acetate. Bigger clumps of chemicals... higher the ASA, smaller clumps, lower ASA. The shutter speed nor the lens iris setting have nothing to do with ASA. As all video tape is essentially the same (not manipulated to increase or decrease light sensitivity) its ASA doesn't exist. And cameras certainly do not have ASA ratings.
Eric DeBlackmere, S.O.C.: The proper terminology in this discussion is EI (Exposure Index) as opposed to ASA. And video cameras do have an EI. Remember, the equivalent to film in the video world is not tape, but the chips in our cameras. That is what we are determining an EI for.
David Patterson: And we can take this a step further, and make an educated guess as to the equivalent ASA or EI of a particular video camera by manipulating an old film formula. Film cameramen know that the rule of thumb is that at 24fps with a 100 speed film and 100 foot candles, the exposure is f/2.8. At 24fps with a 175 or 180 degree shutter your effective shutter speed is about 1/48th of a second, or rounded off to 1/50 (pretty close to a 1/60th).
Video cameras are spec'ed out in terms of sensitivity by giving the aperture f/stop for a given light level in lux (usually 2000) at a 90% reflectance.
Example: a camera might be rated as f/8 at 2000 lux.
Rough equivalent: 2000 lux = 200 footcandles
So if:| f/2.8 | at 100fc | = ASA 100 |
| f/4 | at 200fc | = ASA 100 |
| f/5.6 | at 200fc | = ASA 200 |
| f/8 | at 200fc | = ASA 400 |
My main camera is rated at f/8 at 2000 lux, and in fact it works out in the real world to have a working "ASA" of about 320... about a third of a stop from the calculations. And that never took into account the difference between 1/48th and 1/60th, or that 10 lux is not exactly 1 footcandle. So it's pretty close.
Look at your camera's sensitivity rating. These "ASA's" will be approximate, but a good start.
| f/4 | at 2000 lux | = ASA 100 |
| f/5.6 | at 2000 lux | = ASA 200 |
| f/8 | at 2000 lux | = ASA 400 |
| f/11 | at 2000 lux | = ASA 800 |
This is all assuming zero gain and tungsten balance. Each 6db of gain will double your sensitivity, therefore double your ASA. A plain daylight filter position on the filter wheel is the equivalent of an 85 filter and costs you 2/3rds of a stop or would lower a 400asa to a 250. Various cameras have other values of neutral density or ND filtration built into the other daylight filter wheels. e.g. A 1/4 exposure cut would be equivalent of two stops; therefore cutting 400 to 100, plus the 85 filter brings it down to asa 64.
Also, keep in mind that professional cameras for broadcast use are rated pretty realistically. Manufacturers tend to play fast and loose with specs for consumer cameras. When they say that a camera can shoot a picture in a certain light level, they may mean that it can only do so with the gain at plus 18db AND they almost certainly don't mean that the picture will be 100 units on a waveform monitor.
This is a good thread... I'm glad somebody brought it up!
Christopher Lockett: Since I helped start this mess, I'll just end my part of it by explaining why I use a light meter, for film or video, and why I "ASA rate" my video cameras. If these reasons may mean nothing to you, no sweat... I'm sure you do good work, too. But this is how and why I do what I do:
On location scouts, it's easier to carry my meter, check to see what kind of ambient light levels we have in the room. As I am not a news shooter, I sometimes have the luxury of scouting locations. My meter is much lighter than any camera I've ever used.
Ambient vs. 18% gray: my ambient meter tells me things I need to know in a way that an 18% gray reflective meter won't. Yes, a video camera is a light meter, but so are my still photography cameras. And guess what, I still use my handheld ambient meter when I want an exact lock on what the light is doing to what I'm shooting. I started in stills and film and have to shoot video to pay bills. I know using a light meter works, for me, maybe not the next guy, but the next guy ain't shooting my projects.
EI vs. ASA: Someone stated that "Exposure Index" was what we should be talking about instead of ASA. I disagree. When I pick a film stock, of course I'm choosing one based on its light sensitivy, RMS granularity, how black the blacks are rendered, how many stops it will hold, how say, Kodak Vision 500T sees the color blue versus how Fuji's 500T sees it, etc... but when I'm on set and I've got a meter in my hand, the only thing ASA means to me and to my meter is "How much light do I need here?" So, when the guy asked about "ASA rating" his XL1, I assumed he was asking "How senstive is this camera to light - how much light do I need to shoot?" Also, EI, by common usage in film, is a combination of shutter angle, ASA rating, frames per second and footcandles, which gives you an f/stop rating, the "index" part of Exposure Index that tells you how to expose the stuff properly. So, minus the granularity and other stock characteristics, indeed ASA is what we should be talking about.
I shoot a heck of a lot of corporate and industrial stuff these days. I can whip out the meter, take a reading, figure out how many lights I'll need, etc. before the grip-trician can finish loading the magliner and haul everything to the corporate boardroom or wherever it is I'm shooting.. It's how I came up working and for me, it's fast.
Bottom line - I know how to use a meter and for me, my cameras and my clients, it works.