Measuring light

[Guest Nowitzki]

Hi all,

 

I have a request which may sound a bit stupid- but I'd appreciate

any help.

I'm going to shoot a 16mm/colour music video in future (there's

enough time left).

I already did one one year ago but that one was B/W and after all a

catastrophe since telecining was due to budget problems done at

a "cheap shop" which destroyed a lot of material(not physically-

but for that- I was psychologically destroyed).

Since there wasn't money and time left we also couldn't get the

stuff again telecined.

Well, as you guys can imagine I was pretty paralyzed. There was a lot

doubt whether !I! made the mistake or it was "done" at the telecining

lab.

By now I'm over it since I'm actually a photographer for years and

therefore I think I judged most light situations correctly.

 

Anyway, I'm going to get a Sekonic 558 Cine and for that I already read

a lot of articles about how to use incident meters correctly (at the first

shot I used an old Spectra).

One thing I recognized was that a lot of people are saying to point

the dome at the camera(what I did at the first video shoot) while

there are also enough people telling to point at the lightsource.

 

Well, I know that noone can definately say what to do since it always

depends on certain situations and desired effects.

 

So, my question is whether some of you more experienced cinematographers

may give me some hints or examples how to measure a situation correctly-

maye even a normal portrait shot, with key, fill and kicker for the beginning.

Of course as soon as I get my Sekonic I'll shoot a few rolls of slidefilm

for tests but I though hope to get some help before from more experienced

cameramen(or.. camerawoman)

 

Thanks in advance,

your truly,

Dan

[David Mullen ASC]

Use whatever method gives you the exposure you want, which is something you can test.

 

I generally point the dome at the source I want to measure, not towards the camera which will average the light and shadow more. Once I measure the source, I decide whether it should be slightly (or a lot) over or under the reading for effect.

[Laurent Andrieux]

The idea is the dome integrates all the illuminations coming on it on a pi steradian solid angle. It then gives you an "average" iris stop based on these illuminations.

 

You are supposed to point an incident light meter to the camera as to integer the illuminations that are supposed to be reflected by the subject in the directions of the camera, these being the ones that will create your image (imagine a subject totally backlit, if you want it underexposed, point at the source... if you want it correctly exposed, point to the camera)

 

The thing is, if you imagine taking a still picture and have the light in front of the subject, and want it "normally exposed" it will give you a proper setting.

 

Now, imagine that working with a motion picture camera, you would like to "turn around " the subject, either by cutting shots or tracking, pointing "at the camera" doesn't make much sense anymore, as to determine your f-stop, unless you consider changing it during the shot or along the shots, but, then, you will have a problem with matching !

 

Another point is that you don't necessarly want the subject to be "normally " exposed. Imagine the case of a backlit subject, again, it would be obvious that a contrast would occure and may be have the front of the subject underexposed. Controlling the contrast ratio being another problem, that you can solve with knowledge of how your stock will react to it, and the aesthetical plastic look you want to give.

 

In a night scene lit backlight, it's obvious no one wants a "normally" exposed subject, for instance. (even if color timing afterwards takes part in this process...)

 

So, many cinematographers also point at the different sources, as to evaluate contrast in different directions, and have an idea of "what would be the keylight if the camera was in the same direction, close to the source".

 

Then, if you imagine a subject lit with a source on its side, not in front, you can have a part of a face lit at the keylight, that you measure pointing at your side source, lets say 4 for instance, if you point the other side of the face, you might fall at 2 for instance, and if you point in front of the face (where the camera could be, why not), measure a f-stop of 2.8.

 

then it's you to decide wether you want to expose at 4, 2.8 or 2 depending on you wanting the face a bit dark on one side, and "corectly exposed" on the lit side (set at 4 - night scene why not), correctly exposed on the dark side and overexposed on the other side (set at 2 - day scene why not) or prefere to set the "average" 2.8... and again, color timing will take place afterwards...

 

You see it's all a matter of contrast, and control of lighting and exposure...

 

I'll give you another trick : considering backlight as an effect on a scene, when you measure - wherever you're pointing to the keylight or to the camera - protect the dome from the backlight, otherwise it will be integrated also and then make your f-stop increase, and maybe underexpose the face...

 

Hope I helped you with this without making too many mistakes in my attempt of explanation. Forgive my bad english also,

 

Regards

[Guest Nowitzki]

Dear David & Laurent,

 

thank you much for taking time to help me.

 

I'll go on and practise when I get the lightmeter- with your hints

and especially your detailed illustrations, Laurent, I'm

sure a lot of things will be easier to understand for me.

 

Thanks a lot again :)

 

Dan

[Laurent Andrieux]

Thanks, Dan, you're welcome :)

 

Just wait until David and others correct the mistakes I make so often (think I'll soon be famous just because of them !) before you try to test what I'm saying ! :lol:

[Guest Nowitzki]

Thanks, Dan, you're welcome  :)

 

Just wait until David and others correct the mistakes I make so often (think I'll soon be famous just because of them !) before you try to test what I'm saying !  :lol:

<{POST_SNAPBACK}>

 

OK, then I'll be better waiting :D

 

Good night,

Dan

[Sam Wells]

You can also think of the hemisphere as a sort of stand-in for a human face (usually one of or The principal 3D subjects in photography).

 

-Sam

[Paul Bruening]

Hello,

 

You can also use a peice of black construction paper to block the light selectively off of the globe for the key or fill to measure ratios. While the rule is to compare key plus fill to fill, seperating them often matters when the lights are split for dark-center subjects. Also, keep in mind that meters fresh out of the box can still have sensativity variation. I've seen five guys holding farely new meters, compare them and find three stops of difference. It can be a little maddening. I just got my old Calcu-lite calibrated at the factory and it is 1.5 stops off from my brand new 558 Cine.

 

I have reverted to shooting a grey card and getting a densomiter reading from the lab I use. Though, there may be error in the lab's processing, at least the meter is checked against something.

 

I hope that helps a little.

[Guest Nowitzki]

Thanks for your advices Sam & Paul.

 

About the sensitivity differences between new meters.. this is an issue

I haven't thought until yet. Before I shoot I'll better compare it to a calibrated

spectra as a precaution.

I'm sorry I didn't understand completely(sorry for my english) what you meant

with:

 

I have reverted to shooting a grey card and getting a densomiter reading from the lab I use. Though, there may be error in the lab's processing, at least the meter is checked against something.

 

Will you now have to compensate 1.5 stops all the time with your sekonic ?

 

Greets,

Dan

[Sam Wells]

I don't know about the 558 model but the booklet that came with my Sekonic spot meter actually had a pretty good guide to using it. I mean I knew that stuff already but I looked at it and I think it would help get someone going in the right direction.

 

I have 3 meters currently, a Sekonic spot, Sekonic incident, and Spectra incident. They play nice together agree on most days :D although the Spectra is a weird one, a Combi-II and is only right on one scale, after 2 trips to the factory I don't care, it's a magic talisman as much as a scientific instrument......

 

-Sam

[Sol Train Saihati]

A respected pro once told me a story about an old school cinematographer who never used a lightmeter. He would lay his hand out palm-up, bend his middle finger over and use the resulting shadow to judge the stop! Madness? Genius!

[Ed Moore]

A respected pro once told me a story about an old school cinematographer who never used a lightmeter. He would lay his hand out palm-up, bend his middle finger over and use the resulting shadow to judge the stop! Madness? Genius!

<{POST_SNAPBACK}>

 

That's possibly the best thing I've ever heard, serious respect from your crew if you pull that off :) I guess another advantage is, for exteriors, if you held your palm out flat facing upwards you could use your finger's shadow to work out whether it's time for lunch yet :D

Edited May 23, 2005 by EdMoore