Light meter the right tool ?

arad85 said:
I'm definitely not an expert in this, but I do have some basic understanding of physics. I can see arguments both ways, to the camera and to the light.

As far as I can see, metering to the light gives you the amount of light falling on the subject. It does not tell you anything about how much of that light is reflected to the camera as that will depend on reflectivity, incident angle of light and shape of the object being photographed.

Metering to the camera (with the lumisphere up) meters for all light in front of the object being photographed. It simulates the effects of where the light is placed in front of the object and simulates - to a certain degree - a 3D shape. It won't tell you what the worst possible case of lighting will be (e.g. a mirror that reflects the light directly into the camera) but will give you an average indication of an average 3D object with varying reflectivity for lights that are forward of the lightmeter.

Which ever one you use, you still need to interpret the results depending on the subject being photographed. I'd also be willing to bet there are cases of where Garry has metered towards the light as he knows that most of the light is going to get reflected into the camera.

As for:

I've seen this a few times and it winds me up no end. It is just not true - not even close. When you meter towards the camera, you are metering the light that hits your subject from in front of the meter. That's ANY light coming from in front of the meter. Whilst it may well be a clever one-liner put down for the "meter towards the light" enthusiasts it's just plain wrong.
Click to expand...

This is easy

Say you spent 10 mins setting up a perfect grey background, evenly illuminated, with a measured pool of light behind the sitters head. You now have to shoot 3 models, one is a black girl with matt makeup, one is a aisian girl, and the other a white girl with shiny makeup. If you meter off the first girl, the next 2 exposures will be wrong. If you adjust the exposures to suit the skin, the background will change each time. By taking the incident reading of the light falling ON the subject, each girls skin will be metered right, and correctly relative to each-other and the background wont change either
 
Last edited:
Richard King said:
This is easy

Say you spent 10 mins setting up a perfect grey background, evenly illuminated, with a measured pool of light behind the sitters head. You now have to shoot 3 models, one is a black girl with matt makeup, one is a aisian girl, and the other a white girl with shiny makeup. If you meter off the first girl, the next 2 exposures will be wrong. If you adjust the exposures to suit the skin, the background will change each time. By taking the incident reading of the light falling ON the subject, each girls skin will be metered right, and correctly relative to each-other and the background wont change either
Click to expand...

Yay. Someone who actually knows how to use their meter. Meter the light falling on the subject by pointing the meter at the light and the rest stays the same as long as the subjects sit in the same place and you don't adjust anything.
 
Garry Edwards said:
I agree, it's reasonable to expect any hire studio to be fully equipped, and there should certainly be a meter.

Let's not start this again;) But you point the meter at the subject, not at the light source.
Click to expand...

You most certainly do NOT point an incident light meter at the subject.

You post about not starting this again, but until you start telling people the right things to do instead of the wrong ones, you'll certainly get lots of people pointing out when you're wrong.

In the last thread there was an argument about light meters (that I read), you told people to point their meters at the camera, this time you're telling them to point their meters at the subject? Which is the one you really advocate?

I'll point out the site of someone who has 6 Fellowships (one of them honorary) who tells people to meter their light source. I reckon 6 international Fellowships tells me he's right. A true master of lighting who is absolutely inspirational.
 
Just to throw some more into the mix I was involved in a little experiment.
One light meter, one light set up four photographers four cameras d7000, 2 d90,and a d300.
Each with a 50mm f1.8 prime lens.
We took a meter reading and then set the histograms to as near identical as possible.
Result 5 different setting over almost 2 stops range.
My conclusion unless the meter is calibrated to the camera you are using or you have known offsets for the camera a meter is not much good, others may differ.
 
Mattys said:
You most certainly do NOT point an incident light meter at the subject.

You post about not starting this again, but until you start telling people the right things to do instead of the wrong ones, you'll certainly get lots of people pointing out when you're wrong.

In the last thread there was an argument about light meters (that I read), you told people to point their meters at the camera, this time you're telling them to point their meters at the subject? Which is the one you really advocate?

I'll point out the site of someone who has 6 Fellowships (one of them honorary) who tells people to meter their light source. I reckon 6 international Fellowships tells me he's right. A true master of lighting who is absolutely inspirational.
Click to expand...
You didn't bother to read post #24 I see...

I was hoping that someone else would draw a pretty picture because I'm hopeless at drawing pretty pictures, but as nobody has, here is my effort. Please don't be critical of my drawing skills:)

I've kept it as simple as possible so that EVERYONE can understand it. Obviously, extra lights would make it more complex, and especially if there was an on-axis fill - but simple is good...
metercamera.jpg


The example at the top is pure theory and assumes a light source that doesn't spread out, and where all of the light will reflect at exactly the same angle as it hits - let's call it a laser.
And the subject is completely flat. For this to work in absolute terms, the subject would have to be a mirror.

Now, if you meter from the subject to the light in this example you'll get a true reading of the amount of light reaching the subject but, as you can see, none of that light is reaching the camera, so a meter reading to the camera (which tells you what to set on your camera) accurately tells you that NO light is reaching the camera - so a meter reading to the camera gives you the correct info, and a meter reading to the light source gives you a correct reading but wildly wrong info.

Now, let's move to the bottom diagram. Here, the light is more practical, we'll call it a softbox but it could be just about any kind of light that spreads over distance.
And our subject is 3 dimensional, maybe someone's head.

Again, the angle of reflectance equals the angle of incidence, but there are two major differences this time.
1. The light has spread out and so is striking several different parts of the subject
2. The subject isn't flat, so is reflecting light away at multiple different angles, some of which are going towards the camera.

If you meter towards the light you will again get a true reading of the amount of light reaching the subject but you won't get a true reading of the light reaching the camera, because at least some of that light isn't going towards the camera.

Again, pointing the meter towards the camera will produce a reading that approximates the amount of light that reaches the camera, it is therefore a correct reading.

I'm not really interested in your mentor's views, but he is a respected and successful photographer and I can't help wondering whether you have misunderstood what he has told you...

Edit: Hang on, image isn't showing, I'll try to sort that out.
 
Last edited:
I was taught in college to measure with the meter pointing towards the camera.
Also, I will point the meter towards each light in turn, purely in order to check the relative values of each light source, in order to balance them the way I want.
When the lights are all balanced the way i want them, then I do my reading with the meter's dome usually(although not always) held just under the subjects chin, but pointing towards the camera.
It's always worked.
 
Mattys said:
Yay. Someone who actually knows how to use their meter. Meter the light falling on the subject by pointing the meter at the light and the rest stays the same as long as the subjects sit in the same place and you don't adjust anything.
Click to expand...

Did I say point it at the light?

I might point it at the light(s) to understand the ratio between one light and other(s). However If I want the cumulative contribution for all of the lights, from the camera's POV, then I point the meter at the camera. In essence that gives me the averaged reading, and irons out issues such as what angle the light(s) are at


Extreme example If I had a very small directional light, say a 5 degree honeycombed snoot, and directed that light in right from right under the camera, at an angle where If I popped a mirror in place of the subject ALL of the light was directed at the camera, I would expect to meter about 100% of the light

If I took that light and lowered it to a 45degree angle to the mirror, in hitting the studio ceiling above the camera somewhere

If I take a larger more diffuse light (softbox, Umbrella) a light that is spreading and from a large light source. Then light from the whole surface of the softbox is spreading in all directions. The conceptual trick here is to think about the surface of the softbox as a grid of smaller light sources.

If we take a inch by inch patch in the top left corner of the softbox, light from that patch is radiating outwards in all forwards directions (not evenly). Light from that patch will fall on the same small spot that was originally lit by the honeycombed snoot. Lets imagine we have a softbox positioned in order that the light from that inch by inch patch is 45degrees from the same spot that my snoot was pointing at previously. Light from that patch will hit that spot and bounce off and miss the camera, just like the light from the snoot did

However this is a big soft box. If we consider a inch by inch patch in the centre of the soft-box, light radiates from that patch in all forwards directions. Light from that patch will hit the same spot we measured the snoot at, and bounce straight to the camera. As we look at all of the inch by inch patches of the soft-box, some will and some will not contribute to the light falling on the point we metered at. Others will partially contribute

In a real world lighting situation, with a huge soft box to the top left, and a smaller beauty dish say above the camera, and a reflector to the right, the contributions from all of the lights, at the point we are metering from is dependant not only on the power of the lights, but the size and location of the light source (as demonstrated above) Simply pointing a light meter at a light, only tells you, at that point, how that light is illuminating that spot at the angle of your meter

The camera only records light travelling towards it (that's obvious) so to equalise say the lighting contribution of that soft-box I split up into imaginary squares, or the snoot, or the beauty dishes , or the reflectors you need to point the light meter at the camera. Pointing the meter at a gridded snoot will give you 100% reading, but in reality how much of that light will travel back to the camera? Not much, especially if the subject is a mirror. If the light source was a huge softbox, if the subject is a small mirror how much will bounce back at the camera - Some of it - draw the diagrams it will make sense

talk.jpg


Appologies for the crappy diagram

Pointing the light meter at the light makes no account for the size or directionality of the light, or the angle of he light (compare the gridded snoot to a huge soft-box)

Pointing the light-meter at the camera, remembering angle of incidence = angle of reflection takes into account the size and directionality and evenness across the surface of the light source

When you throw into the mix the fact that the qualities of all of the light sources, the sizes and positions will all be different, that's why you meter towards the camera
 
Last edited:
Garry Edwards said:
You didn't bother to read post #24 I see...


Big snip.
Click to expand...

Nope. I saw something which is blatantly against what you wrote elsewhere, in fact it's the opposite, and quoted that and asked why you were telling different things.
 
Garry Edwards said:
I'm not really interested in your mentor's views, but he is a respected and successful photographer and I can't help wondering whether you have misunderstood what he has told you...
Click to expand...

Nope. Not misunderstood him. In fact he tells people that metering the camera is wrong unless the camera itself is a light source - such as you've poured petrol on it and set it on fire.

Pretty difficult to misunderstand that.
 
Garry Edwards said:
Sometimes, the results from measuring to the light can be way out - 2 stops, 4 stops or even more. It depends on the position of the light in relation to the subject, the reflectivity (or otherwise) of the subject but, mostly, on how much of that light, when reflected off of the subject, goes towards the camera.
Click to expand...

Actually, now I see what you've posted, (and you highlighted it in bold) I wonder why you use an incident meter. You talk about the light which is reflected off the subject towards the camera, so why bother with an incident meter when metering the reflected light is what you're trying to do. Even measuring the light reflected off, I would have the meter turned towards the subject and not the camera as that isn't measuring the light reflected off the subject - the body of the meter is between the subject and it's measuring sensor

I must be misunderstanding you. There can be no other explanation for it
 
Mattys said:
No you didn't. I did. What you said was that measuring the light is better than measuring the subject (paraphrased).


Thankyou for explaining your thinking though. Enlightening.
Click to expand...
It is if you are not say shooting a landscape. Horses for courses
 
Mattys said:
Actually, now I see what you've posted, (and you highlighted it in bold) I wonder why you use an incident meter. You talk about the light which is reflected off the subject towards the camera, so why bother with an incident meter when metering the reflected light is what you're trying to do. Even measuring the light reflected off, I would have the meter turned towards the subject and not the camera as that isn't measuring the light reflected off the subject - the body of the meter is between the subject and it's measuring sensor

I must be misunderstanding you. There can be no other explanation for it
Click to expand...
if you meter what is reflected, it takes the subject into account> That's OK for a landscape, but not OK if you need to say take a sequence of 200 wildly different product shots on the same illuminated background
 
Richard King said:
if you meter what is reflected, it takes the subject into account> That's OK for a landscape, but not OK if you need to say take a sequence of 200 wildly different product shots on the same illuminated background
Click to expand...

And that is very much my point in querying what was typed (and I quoted)
 
Garry Edwards said:
I think we all know that there are a lot of talented people out there who do things a certain way because it suits their style, or because it produces the results that they like, - or, perhaps, simply because they don't understand the basic physics - and that's fine, even though it doesn't make them right.
Click to expand...

And with the greatest respect that doesn't mean you are right either. You may think that way - just like Frank does his :)

There are many in both camps though.
 
I think nowadays this is a little bit academic, as

1. people chimp and use the histogram, or shoot straight to a monitor, then just tweak out
2. most people don't have the need to shoot consistently, shot on shot on shot
3. a lot of people are very reliant on correction in PP
4. many people have a standard set-up that they get working and alter it very slightly when a problem pokes its head up

However. it becomes very important if...

1. you have to shoot consistently week by week and have your shots match up, in terms of the background, overall exposure, F-stop and subject contrast, illumination etc.
2. you are problem solving an issue - like too much wrap around light
3. you shoot film
4. you are trying to reduce or increase the dynamic range in your lighting to match a film or sensor
5. vary your lighting mix / tools to suit the subject, shot on shot
 
Last edited:
Garry Edwards said:
Sometimes, the results from measuring to the light can be way out - 2 stops, 4 stops or even more. It depends on the position of the light in relation to the subject, the reflectivity (or otherwise) of the subject but, mostly, on how much of that light, when reflected off of the subject, goes towards the camera.
Click to expand...

If you are wanting to measure how much light is reflected off the subject - why do you point the meter at the camera? Surely the back of the meter is blocking the light being reflected off the subject. I'm confused :thinking:
 
SimonTALM said:
If you are wanting to measure how much light is reflected off the subject - why do you point the meter at the camera? Surely the back of the meter is blocking the light being reflected off the subject. I'm confused :thinking:
Click to expand...

My thoughts exactly Simon...

though I was struggling to put it in such simple to understand terms as you have (y)

Do pray tell, how can you measure reflected light at the source where its relfected with the back of the sensor facing the source your measuring?
 
Last edited:
No, it doesn't block the light at all. You simply place the meter so that it's immediately in front of the subject, with the receptor pointing towards the camera. The only potential problem is if you block the light with your own body, obviously you position yourself so that you don't do that.
 
SimonTALM said:
If you are wanting to measure how much light is reflected off the subject - why do you point the meter at the camera? Surely the back of the meter is blocking the light being reflected off the subject. I'm confused :thinking:
Click to expand...

Yes..... Another point that confuses me is that if you wanted say to expose for the highlight in the scene, metering towards the camera is not likely to provide that (unless the subject is lit from the front)? Maybe I'm wrong.....
 
BarryDawsib said:
My thoughts exactly Simon...

though I was struggling to put it in such simple to understand terms as you have (y)

Do pray tell, how can you measure reflected light with the back of the sensor facing the source your measuring?
Click to expand...
You don't. If the light source is more or less where the camera is, the sensor is seeing light from it.
If the light source is at an acute angle, the light that skims across the subject also skims across the sensor.
If the light source is behind the subject (say a hair light), you either don't measure it at all or you measure it in a different way, e.g. by reflected light.
 
Garry Edwards said:
No, it doesn't block the light at all. You simply place the meter so that it's immediately in front of the subject, with the receptor pointing towards the camera. The only potential problem is if you block the light with your own body, obviously you position yourself so that you don't do that.
Click to expand...

That doesn't measure any reflected light though. It still takes an incident reading of the light falling on the subject. If you have the dome out then it will measure the light in a wider arc possibly giving a slightly different measurement than if you pointed it at the light source.
 
Garry Edwards said:
You don't. If the light source is more or less where the camera is, the sensor is seeing light from it.
If the light source is at an acute angle, the light that skims across the subject also skims across the sensor.
If the light source is behind the subject (say a hair light), you either don't measure it at all or you measure it in a different way, e.g. by reflected light.
Click to expand...

Why not point the meter at each light source and take each measurement? I've never heard anyone say don't measue the hair light. if you don't measure it how do you know it's right?
 
Last edited:
Garry Edwards said:
No, it doesn't block the light at all. You simply place the meter so that it's immediately in front of the subject, with the receptor pointing towards the camera. The only potential problem is if you block the light with your own body, obviously you position yourself so that you don't do that.
Click to expand...

So your not measuring the light reflected from the subject; you are measuring the light coming from the direction of the camera at the point where the subject is.

Why? I don't get why you don't need to measure the lights - still confused :thinking:
 
Simon, it's simple. You don't measure the light directly because that is not the amount of light the sensor sees. It really is as simple as that.
 
EOS_JD said:
That doesn't measure any reflected light though. It still takes an incident reading of the light falling on the subject. If you have the dome out then it will measure the light in a wider arc possibly giving a slightly different measurement than if you pointed it at the light source.
Click to expand...

That's right, it takes an incident reading of the light reaching the subject but which is reflecting towards the camera, and this is what makes it the most accurate method - as proved by both Richard King's pretty pictures and my not-so-pretty ones:)
Why not point the meter at each light source and take each measurement?
Click to expand...
Do that if you want to. It will measure how much contribution each light is making, and if you like to know about lighting ratios and so on that info will help you - but it won't give you an accurate reading that you can set on the camera, for all the reasons already given.
I've never heard anyone say don't measue the hair light. if you don't measure it how do you know it's right?
Click to expand...
Surely that's just personal choice? You might want a very subtle hairlight, or one that jumps out and hits you. Or somewhere in between. Usually, with digital it's much better to chimp and see whether you like the effect or not than to measure it and to have to rely on a purely theoretical figure.

With film, it's different. You need to rely on experience to judge the intensity that will work at any given level of exposure, and you need to measure by reflected light to take account of the reflective qualities of the hair - for example, a natural blond may well reflect 5 stops more light than afro hair.
 
SimonTALM said:
So your not measuring the light reflected from the subject; you are measuring the light coming from the direction of the camera at the point where the subject is.
Click to expand...
NO! With the lumisphere up, you are measuring any light coming at the subject from anywhere in front of the meter position. The lumisphere measures across the whole 180 degrees, not just where the meter is pointing....
 
arad85 said:
NO! With the lumisphere up, you are measuring any light coming at the subject from anywhere in front of the meter position. The lumisphere measures across the whole 180 degrees, not just where the meter is pointing....
Click to expand...

Are we talking about studio flash here?
AKA a singular directional modified light source?

And taking meter readings with s Flash Meter?

If so, what has 180* got to do with it? when you are metering each light individually?
 
Are we talking about studio flash here?
Click to expand...
We're talking about any form of light. Light is light.
AKA a singular directional modified light source?
Click to expand...
Any number of light sources, it makes no difference as long as the reading is taken from the subject to the camera
If so, what has 180* got to do with it? when you are metering each light individually?
Click to expand...
We're not metering lights individually, we're metering the total amount of light that will reach the camera - which of course is why we're pointing the meter sensor towards the camera.

This really isn't complicated...
 
BarryDawsib said:
If so, what has 180* got to do with it? when you are metering each light individually?
Click to expand...
The question was about measuring with the meter pointing at the camera. When you do that, with the lumisphere up, you are metering for any light in front of the lightmeter. You are NOT metering light coming from the camera or metering the camera itself.

Nothing to do with pointing at the light and yes, we're talking about studio flash which is semi-directional (i.e. it isn't focused like a laser beam but the modifier used makes it more or less directional).
 
arad85 said:
NO! With the lumisphere up, you are measuring any light coming at the subject from anywhere in front of the meter position. The lumisphere measures across the whole 180 degrees, not just where the meter is pointing....
Click to expand...

so 180 degrees centred around the direction of the camera - I think we agree that the meter will not measure any light reflected from the subject which was where I was getting confused with Gary Edwards statement that you measured how much of that light, when reflected off of the subject, goes towards the camera.
 
BarryDawsib said:
Garry, light is not light, its either controlled or it is not, flash or ambient.

So to ask the same question but with more figure to try and get a STRAIGHT answer:

Are we talking about metering for studio flash?
Click to expand...

We're talking about studio flash.
Or hotshoe flash
Or tungsten light
Or HMI
Or fluorescent
Or candles
Or daylight, light is light and as far as metering is concerned, it makes absolutely no difference.

The only practical difference is that, with any form of continuous lighting, you have the extra option of metering with your camera.

That IS a straight answer. It's the only kind I give.
 
Not wishing to get into the heated debate but this thread has answered something that really puzzled me a couple of weeks ago. I went outside with the express intention of experimenting with the lightmeter taking a few portraits in the afternoon sun. I used various locations/angles to the light and metered directly at the light source (the sun), transferred the readings to the camera and took the pictures. All the faces were underexposed by up to two stops!!!

jeff
 
Jeff.B said:
Not wishing to get into the heated debate but this thread has answered something that really puzzled me a couple of weeks ago. I went outside with the express intention of experimenting with the lightmeter taking a few portraits in the afternoon sun. I used various locations/angles to the light and metered directly at the light source (the sun), transferred the readings to the camera and took the pictures. All the faces were underexposed by up to two stops!!!

jeff
Click to expand...
Yes, that's what happens, and that's why I and other people have been trying to explain why you need to measure from the subject to the camera. And why I told BarryDawsib that light is light.

Debate is healthy, but I really don't understand why it's become heated - the principle is obvious enough.
 
SimonTALM said:
I think we agree that the meter will not measure any light reflected from the subject
Click to expand...
Yes, we agree. But we're talking about a lightmeter used as an incident light meter, not as a reflective light meter.
 
You must log in or register to reply here.
Back
Top