# Estimating Light Requirements

7 replies to this topic

### #1 Michael Collier

Michael Collier
• Basic Members
• 1262 posts
• Gaffer
• Los Angeles, CA

Posted 18 September 2006 - 04:15 AM

I want to estimate how much power and lights will be needed for a scene where I have to blow out a window that is 8 feet tall and runs the length of the cafe we are shooting in. I would like to have 12-20 feet blown out. tracing paper will be placed over the windows, and the lights will be about 5 feet back. We will have several smaller 500-1000 watt units.

First (and most importantly) I need a way to calculate how much light needs to hit the paper from the outside to get the paper to read about 4-5 stops over key from the inside. We are shooting 200T film, but will rate it at 160.

If I can get a solid number to light the paper to from the outside by incident reading, I can take the lights and test to get a ballpark on how many to rent (no bigger units are availible up here, so 500-1000 is my limit. I can arrange for up to 30 lights though)

My question I guess is transmissive light similar to incident light? My figuring is if light were to be striking the paper from camera side, it would be very easy to calculate. Also the light would spread evenly, since at that point its reflective. Now if it is going through the paper, I can calculate the 2/3 stop light loss (or whatever it may be for that particular paper) and throw more light to compensate and the exposure would be the same. Since at that point it would be very diffused (large paper area, with a large light source origin) the light quality should be similar to reflective light and therefore would measure the same when checked with a spot. Is this right? or am I missing something key about the difference between transmissive light and reflected light. I have never had to light such a large area to look like day, and if it will take more light than the generator or availible lights can provide, I will just drop the whole idea and shoot around the windows, or reduce the amount of onscreen windows there are. It would be a shame though. For the story big windows with no detail through them would be ideal.

(if there are typeos or things that don't make sense in this post, don't be to harsh, I have had a killer weekend of work. I had to shoot a CD release party for 15 hours and I woke up hung over. It was fine until we got to the club and the base started to hit. I am exausted. I need to get out of ENG for good. no more on the fence stuff.) I need a long week shooting news just to relax and prep for the short.
• 0

### #2 Tim Terner

Tim Terner
• Sustaining Members
• 340 posts
• Producer
• Prague, CZ

Posted 18 September 2006 - 05:06 AM

What stop do you intend shooting at Michael
• 0

### #3 Michael Collier

Michael Collier
• Basic Members
• 1262 posts
• Gaffer
• Los Angeles, CA

Posted 18 September 2006 - 09:14 PM

I would like to get to an f4, but I can go as high as a 2.8 (the lens is a 2.2), if its the difference between seeing windows and not seeing windows. By my calculations that means I need around 100 FC spot on the papers reverse side to get to key, so by that estimation I need 1600 FC to get 4 over?.

Edited by Michael Collier, 18 September 2006 - 09:16 PM.

• 0

### #4 Michael Collier

Michael Collier
• Basic Members
• 1262 posts
• Gaffer
• Los Angeles, CA

Posted 18 September 2006 - 09:53 PM

OK, so I did some tests with my spectracine IV-a and some tracing paper that was layer between filters of a gell kit I bought (looks like the paper loss is around 1 stop, I think I can find tracing paper with less stop loss and more consistent transmission of light (it seems to be very textured from the backlight side.)

Check my work and see if I did anything wrong in my assumptions.

I read my spectra owners manual and it says if you remove the photosphere, it becomes a 45 degree reflected light meter. Is this the same as a spot meter (with the exception that its s45 degrees and not one degree?)

I lit the paper until the back side read an F16, with the meter very close to the paper. Then I replaced the photosphere and measured the incidence from the light side to see how many FC was hitting. It read around 760 or so. I figure if I can get a 760, that means the paper will be 4 over key on the camera side, is this right? If it is then I can start calculating how much I need.

Also, is there a good place to get tracing paper at least 4ft wide? 8ft wide would be better, but between windows I don't want to have seems, since that will take more time in rigging to get the seems not visible to the camera.

I ask because I figured the paper would look much brighter to my eye.

wait, did the test again with 216 and read under 200 FC falling on the 216 (with the photosphere on) and measured an F22 (5 stops over key) from the other side with the photosphere off. Is that a more accurate reading or do I have more testing to do to find less disparity between tests?
• 0

### #5 Hal Smith

Hal Smith
• Sustaining Members
• 2280 posts
• Cinematographer
• OKC area

Posted 18 September 2006 - 10:17 PM

First (and most importantly) I need a way to calculate how much light needs to hit the paper from the outside to get the paper to read about 4-5 stops over key from the inside. We are shooting 200T film, but will rate it at 160.

Harry Box's book mentions 1000H vellum tracing paper but doesn't give any transmission figures. I looked at Clearprint's web catalog ( http://www.clearprintpaperco.com ) for figures but they don't have any listed. You might try emailing Clearprint technical support at info@chartpak.com to see if they have transmission figures for their products.
• 0

### #6 Michael Collier

Michael Collier
• Basic Members
• 1262 posts
• Gaffer
• Los Angeles, CA

Posted 18 September 2006 - 10:19 PM

I am posting pics so you can see my test, and maybe you can see something there I missed. The first pic is with the photosphere off from the camera side. Note the reading of 22.3

The second is with the photosphere on, switched to FC reading. I got a reading of 125 FC (much smaller than the 760 I thought it would take, I may have reversed the test on accident)

The meter was set to 120 and 24fps. 22 would be 5 stops over a 4.0 key if I rate at 120. Is this correct, or is my test completly backwards. Please let me know before I spend hundreds to rent equipment that won't quite cut it.

I think I may see my problem. Is that reading 125 FC or 1250 fc. I read it again and noticed it reads much higher from farther away (560 but without that small 0 afterwards. Is that what I am overlooking?)

Edited by Michael Collier, 18 September 2006 - 10:23 PM.

• 0

### #7 Hal Smith

Hal Smith
• Sustaining Members
• 2280 posts
• Cinematographer
• OKC area

Posted 18 September 2006 - 10:37 PM

I read my spectra owners manual and it says if you remove the photosphere, it becomes a 45 degree reflected light meter. Is this the same as a spot meter (with the exception that its s45 degrees and not one degree?)

Yes, I've used that principle to check the calibration on my spot meter. I lit a white wall as evenly as possible, then took a reading with my Spectra IV-a in reflected mode, then looked at the wall with my Spectra Combi-II with my Photospot II on it.

PS: Check your Spectra manual, I think you'll find that the IV-a is only calibrated for footcandles with the Photodisc on it, not the Photosphere.
• 0

### #8 Michael Collier

Michael Collier
• Basic Members
• 1262 posts
• Gaffer
• Los Angeles, CA

Posted 18 September 2006 - 11:19 PM

Yes you are right, it does need the photodisk, no sphere. I found that shortly after I posted last. I found that the difference was 1250fc or 1100 fc, so I didn't feel a need to repost.

If nobody sees anything wrong with this test (other than I can't read) then I think i have my answer. 350FC gives a stop of 16 when rated at 160, which if I did shoot at a 2.8, I would be 5 over. Given I need 12 feet by 8 feet covered, I estimate we need between 6 and 15 omnis (depending on exactly how many windows will be in shot.) but at least I have a rough estimate to rent lights and rent genny on.
• 0