Knowing what lights to bring

[Joe B Jowers]

I shoot video and have attempted to determine the working speed rating for my DVX100a [i dare not say ISO or ASA] without the use of a waveform. I've found a few different methods but have not been able to come to a definitive number or guide. I read that using a light meter for video work is impractical because video sensors respond to light differently at various apertures. I took this to mean that at f4 my camera sensitivity would be 320 and say at f 5.6 it could be 800 or greater. If my testing methods have been correct I've experienced similar results. I've tried to eliminate ND filters and gain switches as variables.

 

More to the point, I want to know how to go about determining how much light I will need for any given shot. Lets say I go on a scout with my video camera [not always practical] and find that wide open the exposure is adequate but I want to shoot at f5.6... What tools do I bring?

 

Often I can bring or rent only enough to get by because of budget constraints.

 

Do I need to think in terms of foot candles and fixture output? I'm sure there must be a systematic way to go about knowing that I need a 575 and a 200w joker or a 1200w HMI or that I can get by with a couple of 650w and softboxes.

 

Determining what to bring to accentuate lighting in bright rooms with flat existing light is sometimes challenging for me also.

 

I apologize for the awkwardness of the question. Any help would be appreciated.

 

Joe Jowers

[Adrian Sierkowski]

Thinking in footcandles is pretty useful, I suppose, but won't help you with video exposure IMHO.

Basically the camera will respond differently to different scene contrasts, or so I understand it. What I would recommend would be to use your camera/meter in as many different contrast situations and then average an ASA, so you'll have something to work off of.

For example, I know such and such HVX I use in low light responds at let's say 160, and in bright light at 250. Well, a good ballpark for that might be 200, depending on my taste.

These things come with experience, of course, and how big a location you're working on, but once you get a nice little rough guestimation, a DSLR or even a point and shoot where you can set the asa, can be useful on location scouts.

[Brian Dzyak]

I read that using a light meter for video work is impractical because video sensors respond to light differently at various apertures.

 

 

Hmm, interesting. I'd love to know where you read that. It doesn't seem right somehow. :unsure:

 

You can determine the effective ASA of the camera so given some testing, you could figure out the exact lights you might want based on a thorough scout. Just bring along the light meter and a tape measure. :)

 

I think that in general, the "basic" kit that will suffice in most situations is something in the range of a 650w to a 1K for a key then work around that. Of course that is going to vary depending on a host of other factors, so really, if you know precisely what you intend to shoot and figure out the base ASA of the camera, with a little math you should be able to figure out almost every light you would need.

[Andrew Koch]

First thing, Go to "My Controls" toward the top right of the web page and change your screenname to your First name, a space and then your last name. This is a requirement of this forum and you need to change it so you don't get banned.

 

I have a proposed test for you that might work for your situation. If you don't have a waveform monitor, get an 18% greycard and fill the frame with it under flat light that is the color of light you will be shooting with that you want to appear as white light (whether it be daylight, tungsten, florescent, etc.) White balance to that greycard with auto iris. Once you are white balanced, turn back to manual iris. The next step worked "okay" with the HVX200, I'm not sure if it is the same for the DVX100. Set your Zebras to 50%. Close down your iris until there are no zebras then gradually open the iris until it just begins to zebra. Take a reading of the card with your meter. Change the ISO of your meter until it matches the camera's ISO. This is a ballpark figure and by now means a subsitute for a waveform monitor.

 

I just thought of another way that might be more accurate, but will take quite a bit of time and energy. Shoot an 18% Greycard and white balance the same way I just described. Set the Iris until card appears properly exposed to your eye on the viewfinder. This is not necessarily the actual "correct" exposure for the card since there are so many variables. This is just the starting point for the test. Write down the f-stop that appears on the camera at this point. Take a meter reading and change the iso until the stop on your meter matches the stop on the camera. Write this ISO down. For demontration purposes, lets pretend the ISO is 320. Record this to tape for about 10 seconds. Next, open the iris by 1/3 of a stop. Notate this (put some kind of note like "plus 1/3" in the corner of the frame for easy identification, but keep any text away from the majority of the frame or the center.) Record this. Open up the iris another 1/3 and have something in the frame say "plus 2/3". Record this. Keep increasing in thirds until you reach 2 stops over your original exposure. (This might be overkill, but what the heck, try it.) Go back to your original exposure and shoot it again. No do the same thing in reverse. Close down your iris in 3rds untll you reach 2 stops under.

 

Now that you have all of this on tape. Transfer the footage to a non-linear editing program that has waveform monitors built in. Even the Xpress versions of Avid have it as well as final cut pro when doing color correction. I believe Adobe Premiere has it as well. If you can't find a program with a waveform monitor, then this test won't help you. If you do, observe the waveforms of each exposure. Find the one that is closest to 50 IRE (I have heard some place middle grey at 40 or 45, so ask around and determine which one is best and go with that, but for the sake of this post, lets just say 50 IRE). If the Waveform monitor reads 50IRE on your original first exposure, then use your original ISO off the meter (our example of 320). If the waveform reads at 50IRE when you recorded 1/3 of a stop over original, decrease your ISO by a third (ex: 320 becomes 250. 320 becomes 200 if 2/3 over, and so on.) If your waveform reads 50 IRE when you are 1/3 under original exposure, increase your ISO by a third (320 becomes 400. If 2/3 under then 320 becomes 500, and so on). These increments will help to overcome the lack of accuracy from exposing by eye off of an LCD.

 

This method would be a pain in the a** to do, but at least it would help you find a ballpark rating for your camera, at least under the lighting setting you will use. To be honest, I'm not even sure it is the greatest idea, but it's just a thought. A simpler method would be to purchase DV Rack, which has a software based waveform monitor that you can use live on set if you have a laptop and a firewire cable.

 

At the very least, having a rough ISO for your camera will allow you to bring your meter to a scout. The most you would be off by would probably be a stop. Don't get me wrong, a stop is significant, so I would recommend erring on the side of a bit more light than you think you will need, and if it is too much, drop some scrims in the light. Much easier than having lights that are not strong enough and having to build everything up.

[Serge Teulon]

.....Shoot an 18% Greycard...... Set the Iris until card appears properly exposed to your eye on the viewfinder and monitor.....Write down the f-stop that appears on the camera at this point. Take a meter reading and change the iso until the stop on your meter matches the stop on the camera.

 

S

Edited September 5, 2008 by Serge Teulon

[Andrew Koch]

What does "S" mean? Also, you quoted me, but that's not exactly what I wrote. I made it very clear that setting exposure by eye to determine ISO was not necessarily the most accurate way to go, but the way I was quoted implies otherwise. However, I will agree with the change that you made about using a monitor, provided that it is properly calibrated. I would always use the monitor over the crappy viewfinder, as long as it is available (I wasn't sure if he had a monitor because he posted under students and first time filmmakers)

[Walter Graff]

Hmm, interesting. I'd love to know where you read that. It doesn't seem right somehow. :unsure:

 

Probably from me. I've been saying it an writing it in articles for years. The MTF of a video camera is not as uniform in slope as film. That means as light increases exposure index does indeed change. The more light the greater the Exposure index. A waveform, grey scale, two lights and a light meter will show this. And with little light you have a second problem which involves what are effectively little lenses placed on the CCD (called micro lenes) to give the camera more sensitivity. They are like magnifying lenses. Problem is they help least when you need them most. You can get an effective exposure index for a range of light, but it will change with a few stops more or less.

 

As for lights, with the camera you mention any range of 200w at 1k will always be able to do what you want. Indoors most folks get an exposure index of 300-400 wide open and that range of light will do everything you need.

[Brian Dzyak]

Probably from me. I've been saying it an writing it in articles for years. The MTF of a video camera is not as uniform in slope as film. That means as light increases exposure index does indeed change. The more light the greater the Exposure index. A waveform, grey scale, two lights and a light meter will show this. And with little light you have a second problem which involves what are effectively little lenses placed on the CCD (called micro lenes) to give the camera more sensitivity. They are like magnifying lenses. Problem is they help least when you need them most. You can get an effective exposure index for a range of light, but it will change with a few stops more or less.

 

Interesting! Thanks. I hadn't heard that before. :)

 

How much does that exposure index change? Under what conditions would we start feeling it and have to compensate in some way (exposure, lighting)?

[Joe B Jowers]

Problem is they help least when you need them most. You can get an effective exposure index for a range of light, but it will change with a few stops more or less.

 

So if I am willing to do the testing in a variety of lighting conditions I can at least have an appox. ISO. Cool.

 

Now as for the second part of my question. Ordering the lights. I lit a large area in a retail store. I used a 1200 par, a 400w HMI and a 200w joker, accenting hear and there with some dedo's. I was really flying by the seat of my pants and the limits of the budget. It worked out ok and the biggest problem was when I tried to light a small room with that 1200. The talent had to walk of screen towards the light. Ugh. Edit.

 

I'm not new to the game, I'm starting to light bigger sets and projects I'm looking for tips on how to anticipate the lighting needs. I know that's not very specific.

 

So during the scout I'd, determine the area to be lit, the mood/ quantity and quality of light, look at the photometrics and determine fixtures...how do you guys do it?

 

And by the way thanks for all the answers!