Go to a  "printer friendly" view of this message which allow an easy print Printer-friendly copy Go to the page which allows you to send this topic link and a message to a friend Email this topic to a friend
Forums Lobby MASTER YOUR TOOLS - Hardware & Software Nikon Speedlights & Lighting topic #40627
View in linear mode

Subject: "A gedanken experiment on light meters...." Previous topic | Next topic
nl Basic MemberFri 27-Mar-09 01:45 AM
833 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
"A gedanken experiment on light meters...."


West Hartford, US
          

Hopefully the topic got your attending. I've been trying to thing something through, and I wonder if anyone can help.

For the purposes of discussion, assume a subject lit by a single light source. Let's also assume the camera has a mathematically ideal zoom lens - eg a lens that can zoom from 0 to infinite focal length, with no impact on image rendering (sharpness, resolution, aberration, etc) at any focal length. In other words, lens characteristics and imperfections have no effect in this thought experiment.

We will make an incident light meter reading by placing the incident meter at subject position, aimed at camera. Obviously, the meter reading is not dependent on the camera position.

Suppose we place the camera at distance "X" from the subject, zoom so the subject entirely fills the frame, and make an exposure. Now we move the camera to position "2X" from the subject, again zoom so the subject entirely fills the frame, and expose.

Since at each camera position we zoomed so the subject filled the entire frame, the light captured at each position can only be light reflected by the subject - in other words, there is no additional light entered the lens from around the subject.

Since we are at 2X the distance on the second exposure, the inverse square law says that the light entering the lens at position 2X is 1/4 that at position X. Therefore, incident metering should be critically dependent on the camera to subject distance, and yet we never seem to take this into account with incident metering.

There seems to be a flaw in my reasoning (and I have a geometrical argument to explain this and correct the reasoning) but I am curious to see what others think first and see if anyone can help me out with my thinking.

Thanks.
nl

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

Replies to this topic
Subject Author Message Date ID
Reply message RE: A gedanken experiment on light meters....
Noel Holland Platinum Member
27th Mar 2009
1
Reply message RE: A gedanken experiment on light meters....
ScottChapin Moderator
27th Mar 2009
2
     Reply message RE: A gedanken experiment on light meters....
Noel Holland Platinum Member
27th Mar 2009
4
          Reply message RE: A gedanken experiment on light meters....
ScottChapin Moderator
27th Mar 2009
5
               Reply message RE: A gedanken experiment on light meters....
Noel Holland Platinum Member
27th Mar 2009
10
Reply message RE: A gedanken experiment on light meters....
bclaff Silver Member
27th Mar 2009
3
Reply message RE: A gedanken experiment on light meters....
nl
27th Mar 2009
6
Reply message RE: A gedanken experiment on light meters....
bclaff Silver Member
27th Mar 2009
8
     Reply message RE: A gedanken experiment on light meters....
ScottChapin Moderator
27th Mar 2009
11
          Reply message RE: A gedanken experiment on light meters....
Noel Holland Platinum Member
28th Mar 2009
12
               Reply message RE: A gedanken experiment on light meters....
ScottChapin Moderator
28th Mar 2009
13
                    Reply message RE: A gedanken experiment on light meters....
Noel Holland Platinum Member
28th Mar 2009
14
                         Reply message RE: A gedanken experiment on light meters....
ScottChapin Moderator
28th Mar 2009
17
                              Reply message RE: A gedanken experiment on light meters....
Noel Holland Platinum Member
29th Mar 2009
18
                                   Reply message RE: A gedanken experiment on light meters....
bclaff Silver Member
29th Mar 2009
19
                                        Reply message RE: A gedanken experiment on light meters....
Noel Holland Platinum Member
29th Mar 2009
20
                                             Reply message RE: A gedanken experiment on light meters....
ScottChapin Moderator
29th Mar 2009
24
Reply message RE: A gedanken experiment on light meters....
HBB Moderator
27th Mar 2009
7
Reply message RE: A gedanken experiment on light meters....
nl
27th Mar 2009
9
Reply message RE: A gedanken experiment on light meters....
HBB Moderator
28th Mar 2009
15
Reply message RE: A gedanken experiment on light meters....
ScottChapin Moderator
28th Mar 2009
16
     Reply message RE: A gedanken experiment on light meters....
Noel Holland Platinum Member
29th Mar 2009
21
          Reply message RE: A gedanken experiment on light meters....
bclaff Silver Member
29th Mar 2009
22
          Reply message RE: A gedanken experiment on light meters....
Noel Holland Platinum Member
29th Mar 2009
25
          Reply message RE: A gedanken experiment on light meters....
ScottChapin Moderator
29th Mar 2009
23
               Reply message RE: A gedanken experiment on light meters....
Noel Holland Platinum Member
29th Mar 2009
26
                    Reply message RE: A gedanken experiment on light meters....
ScottChapin Moderator
29th Mar 2009
27
                    Reply message RE: A gedanken experiment on light meters....
Noel Holland Platinum Member
30th Mar 2009
28
                    Reply message RE: A gedanken experiment on light meters....
HBB Moderator
30th Mar 2009
29
                         Reply message RE: A gedanken experiment on light meters....
ScottChapin Moderator
30th Mar 2009
30
                              Reply message RE: A gedanken experiment on light meters....
Noel Holland Platinum Member
30th Mar 2009
31
                              Reply message RE: A gedanken experiment on light meters....
ScottChapin Moderator
30th Mar 2009
33
                              Reply message RE: A gedanken experiment on light meters....
nl
30th Mar 2009
32
Reply message RE: A gedanken experiment on light meters....
HBB Moderator
31st Mar 2009
34
Reply message RE: A gedanken experiment on light meters....
nl
31st Mar 2009
35
     Reply message RE: A gedanken experiment on light meters....
ScottChapin Moderator
31st Mar 2009
36
     Reply message RE: A gedanken experiment on light meters....
HBB Moderator
31st Mar 2009
38
     Reply message RE: A gedanken experiment on light meters....
HBB Moderator
31st Mar 2009
37
          Reply message RE: A gedanken experiment on light meters....
nl
01st Apr 2009
39
               Reply message RE: A gedanken experiment on light meters....
HBB Moderator
01st Apr 2009
40
               Reply message RE: A gedanken experiment on light meters....
ScottChapin Moderator
01st Apr 2009
41
               Reply message RE: A gedanken experiment on light meters....
ScottChapin Moderator
01st Apr 2009
42
                    Reply message RE: A gedanken experiment on light meters....
nl
01st Apr 2009
43
                         Reply message RE: A gedanken experiment on light meters....
ScottChapin Moderator
01st Apr 2009
44
                              Reply message RE: A gedanken experiment on light meters....
HBB Moderator
01st Apr 2009
45
                                   Reply message RE: A gedanken experiment on light meters....
ScottChapin Moderator
01st Apr 2009
46
                                        Reply message RE: A gedanken experiment on light meters....
Noel Holland Platinum Member
03rd Apr 2009
49
                                             Reply message RE: A gedanken experiment on light meters....
ScottChapin Moderator
03rd Apr 2009
50
                                             Reply message RE: A gedanken experiment on light meters....
gkaiseril Gold Member
03rd Apr 2009
51
                                                  Reply message RE: A gedanken experiment on light meters....
ScottChapin Moderator
03rd Apr 2009
52
Reply message RE: A gedanken experiment on light meters....
gvk Silver Member
02nd Apr 2009
47
Reply message RE: A gedanken experiment on light meters....
nl
02nd Apr 2009
48
     Reply message RE: A gedanken experiment on light meters....
gvk Silver Member
05th Apr 2009
53
Reply message RE: I feel there are four aspects
Len Shepherd Gold Member
05th Apr 2009
54
Reply message RE: I feel there are four aspects
ScottChapin Moderator
05th Apr 2009
55

Noel Holland Platinum Member Winner in the Nikonians 10th Anniversary Photo Contest Charter MemberFri 27-Mar-09 04:51 AM
2171 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#1. "RE: A gedanken experiment on light meters...."
In response to Reply # 0
Fri 27-Mar-09 04:52 AM by Noel Holland

TH
          

Camera distance to subject has nothing to do with the light levels. The light levels are all ready set no matter what distance you stand from the subject or what zoom you use.

Now what you are really asking is why is it that a lens set to 1X is F/Y for a given light level and then when I set the lens to 2X the aperture is still F/Y for that light level .... well that's how our equipment is designed to work.

When you are using non-fast zooms that's why the open end aperture changes with focal length and why fast level zooms cost so much to have constant apertures at the open end. At all other aperture settings, zoom lens are designed to change the physically size of their aperture for the same aperture setting but different zoom settings.

As you rightly state the inverse square law happens at every step of the photon packet's travel. It applies from light source to subject, it applies from subject to lens entry and with zooms we are forcing another form of inverse square law by deliberately spreading the image circle wider as we lengthen the focal length. However our equipment and our photographic decisions automatically take care of the last two issues and so we only even need to worry about the first issue. Set the light right and the rest of it is taken care of for us.

It's for this very reason that we use aperture numbers and not aperture sizes. Aperture sizes in mm or fractions of an inch would exhibit all the problems you are concerned with. It would be chaos trying to figure out the right exposure settings for a zoom. But aperture number do all the maths for use before we ever buy the lens. The aperture size is set to vary automatically to ensure that the aperture setting is preserved as the lens zooms.

In this case it's not all done with mirrors but little slivers of metal irising in and out as you vary the focal length to ensure that the aperture number stays the same and your head doesn't explode with the math.

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

    
ScottChapin Moderator Awarded for his high level skills in various areas, including Aviation and Birds Photography Charter MemberFri 27-Mar-09 09:49 AM
8047 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#2. "RE: A gedanken experiment on light meters...."
In response to Reply # 1
Fri 27-Mar-09 09:55 AM by ScottChapin

Powder Springs, US
          

Aber ja doch! Aber ich bin verwirrt.

It's OK to talk to yourself, but you mustn't answer . Zoom has nothing to do with it. This is a Bill Claff question. I don't know why, but the exposure is always dependent upon the light at the subject.

The proper exposure for the moon is sunny bright f/16 regardless of the lens you use. Going wide and including more of the universe doesn't increase the exposure for the moon.

This seeming contradiction to the inverse square law has always befuddled me. I used to think that the ratio of 1AU plus the distance to the moom to 1 AU is virtually 1, so there is no significant light loss. This doesn't explain why flood lit buildings have a relatively stable exposure regardless of your distance from them though.

Deshalb bin ich ganz verwirrt.

Scott Chapin
Powder Springs, GA, USA
Nikonians Team Member

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

        
Noel Holland Platinum Member Winner in the Nikonians 10th Anniversary Photo Contest Charter MemberFri 27-Mar-09 12:53 PM
2171 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#4. "RE: A gedanken experiment on light meters...."
In response to Reply # 2


TH
          

>This seeming contradiction to the inverse square law has always befuddled me.

Not really.

Inverse square law means that as the distance doubles the light quarters. But it can equally be stated that if a subject is projected onto a sensor at one half it's previous height the intensity of the light on any one sensor element will quadruple.

The two situations cancel each other out since for a given focal length, as you move back twice the distance the subject image becomes half the previous height.

Dimmer by factor of 4 for distance, brighter by a factor of 4 for image size reduction - ergo no change overall.

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

            
ScottChapin Moderator Awarded for his high level skills in various areas, including Aviation and Birds Photography Charter MemberFri 27-Mar-09 01:58 PM
8047 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#5. "RE: A gedanken experiment on light meters...."
In response to Reply # 4


Powder Springs, US
          

I think where I ran aground here was that I did not interpret you correctly. I knew the trade off between increasing focal length and proportionally increasing the aperture's diameter to maintain the same exposure.

When you mentioned that doubling the subject to camera's distance would reduce the light by the inverse square, I felt this was not true and therin lies the rub. A factor not considered is the light source to subject distance which has to be added in and of course subject reflectivity accounted for. I thought this was the conundrum.

When shooting in a theater z.B.(For the German proclivity), I would get the same basic exposure on stage as I would sitting in the second row. That is because of the source light to subject distance being greater than the variance in subject to camera distance. I was reading to deeply into your original statements and being, well, anal.

Scott Chapin
Powder Springs, GA, USA
Nikonians Team Member

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

                
Noel Holland Platinum Member Winner in the Nikonians 10th Anniversary Photo Contest Charter MemberFri 27-Mar-09 05:29 PM
2171 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#10. "RE: A gedanken experiment on light meters...."
In response to Reply # 5


TH
          

>I was reading to deeply into your original statements

ROFLMAO

Been there, done that and we obviously both shop at the same chain of T-shirt shops.

No worries.

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

bclaff Silver Member Awarded for multiple contributions for the Resources Nikonian since 26th Oct 2004Fri 27-Mar-09 12:32 PM
9681 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#3. "RE: A gedanken experiment on light meters...."
In response to Reply # 0


Vancouver (WA USA not Canada), US
          

Neal,

I think you have already got your answer(s)... but here is another.

The amount of light gathered is determined by the area of the entrance pupil.
The area is proportional to the square of the diameter (D) of the entrance pupil since area = pi * (D / 2 )2.
Where the diameter of the entrance pupil is equal to f / f# where f is focal length and f# is the f-number.

Except at close distances, m = f / S where m is magnification, and S is the Subject distance.

Let's say you double magnification by doubling the focal length.
But, at the same f#, you also double the diameter of the entrance pupil.
So the "larger" image is stretched out over a larger entrance pupil and these effects cancel each other out.


Bill

Visit me at My site

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

    
nl Basic MemberFri 27-Mar-09 02:17 PM
833 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#6. "RE: A gedanken experiment on light meters...."
In response to Reply # 3


West Hartford, US
          

Thanks all for chiming in. The answer I felt was correct is that, provided the distances are "relatively" small, using a constant aperture means that with change in focal length with changing distance required to keep the image size constant results in a change in the aperture size that exactly compensates for the light fall-off due to the inverse square law. I believe the first response was getting to this, although there was a lot of the text that I did not fully understand.

The reason that I comment that the distances must be "relatively" small is that in the extreme, you might argue that if the camera position were to be infinitely far from the subject, then there be, asymptotically, zero light entering the lens, which would imply that the exposure cannot be constant with varying distance from the subject.

There are two key points here. First, you are zooming at a distance to ensure constant subject size, which has the discussed effect on aperture _size_ as a function of f/stop and focal length relationsihp.

Secondly, the flaw in my argument re infinite distance above can be easily understood if you assume the subject functions as a point light source (reasonable at large distances). The geometric / non quantum mechanical explanation for the inverse square law of light falloff is that the light from a point source must expand to cover the surface of a sphere and distance r from the point source. As the surface area of a sphere is proportional to r-squared, the light per unit area falls off as the square of radius. This argument is not, btw, correct under the quantum mechanical view of light, but we can work with it here.

Anyway, at relatively short subect to lens distances, the fact that the aperture is geometrically planer does not matter much, but at very large distances, the different geometry between an aperture of given size (infinite in the limiting case of an infinite focal length) that is planar vs a sphere of radius equal to subject to camera distance becomes important. Essentially, in the limiting infinite distance case, the lens should be in a sphere around the subject, keeping exposure contant.

nl

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

    
bclaff Silver Member Awarded for multiple contributions for the Resources Nikonian since 26th Oct 2004Fri 27-Mar-09 04:30 PM
9681 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#8. "RE: A gedanken experiment on light meters...."
In response to Reply # 3


Vancouver (WA USA not Canada), US
          

My initial answer is for a fixed distance and varying magnification by varying focal length.
The case of changing distance to the subject is slightly different and has already been covered.

It's important to remember that most of our subjects are reflecting illumination that originates elsewhere.
It's the total length of the light path that matters.
I haven't had to deal with this issue indoors, but clearly when your light source is something like the Sun, it doesn't matter you far you are from the subject!

So, let's modify the question: what about photographing a non reflected image such as a light blub or computer screen; does distance matter?


Bill

Visit me at My site

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

        
ScottChapin Moderator Awarded for his high level skills in various areas, including Aviation and Birds Photography Charter MemberFri 27-Mar-09 11:17 PM
8047 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#11. " RE: A gedanken experiment on light meters...."
In response to Reply # 8


Powder Springs, US
          

>So, let's modify the question: what about photographing a non
>reflected image such as a light bulb or computer screen; does
>distance matter?

Of course it does, but since the light isn't reflected, do you use an incident meter at the camera, pointed towards the bulb/screen? Actually, in micro photography, that might be a real issue. I remember my OMs had special focusing screens for more accurate micro photography metering, I think.

Scott Chapin
Powder Springs, GA, USA
Nikonians Team Member

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

            
Noel Holland Platinum Member Winner in the Nikonians 10th Anniversary Photo Contest Charter MemberSat 28-Mar-09 11:13 AM
2171 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#12. " RE: A gedanken experiment on light meters...."
In response to Reply # 11


TH
          

>Of course it does

Sorry but no it doesn't.

ALL subjects are lit subjects. Where the light started from is irrelevant. What matters solely is how much light starts the journey from subject to camera. Whether the subject emits the light itself or it reflected the light from another source, it makes no difference. The principals are the same.

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

                
ScottChapin Moderator Awarded for his high level skills in various areas, including Aviation and Birds Photography Charter MemberSat 28-Mar-09 12:11 PM
8047 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#13. " RE: A gedanken experiment on light meters...."
In response to Reply # 12


Powder Springs, US
          

Then why does my metering change while increasing the distance from a lamp while keeping the crop the same? That area is subtended by a smaller angle and contains less photons at greater distances.

Why do dimmer stars appear brighter when they are closer to the earth and visa versa? I think the key is the source to subject to camera distance that matters.

Scott Chapin
Powder Springs, GA, USA
Nikonians Team Member

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

                    
Noel Holland Platinum Member Winner in the Nikonians 10th Anniversary Photo Contest Charter MemberSat 28-Mar-09 01:24 PM
2171 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#14. " RE: A gedanken experiment on light meters...."
In response to Reply # 13


TH
          

>Then why does my metering change while increasing the distance from a lamp while keeping the crop the same?

Don't use a lamp as an example. As you move away the subject becomes a point source and the balance of the bright subject versus the background starts to fool your metering system.

Instead, think of the surface of a light table, flat wide and emitting an even illumination. Now think of a sheet of white paper illuminated evenly from either side. Picture yourself looking through the viewfinder and ignore everything outside of the frame.

I put it to you that if you didn't know that the subject was a light table or a sheet of white paper, you would never be able to tell just by looking through the viewfinder.

Now do the same exercise in your mind by moving away from the subject.

So if the sheet of white paper doesn't change exposure as you move away, why should the light table surface?

Got me now?

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

                        
ScottChapin Moderator Awarded for his high level skills in various areas, including Aviation and Birds Photography Charter MemberSat 28-Mar-09 11:45 PM
8047 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#17. " RE: A gedanken experiment on light meters...."
In response to Reply # 14


Powder Springs, US
          

why not use a lamp? That was Bill's question and illuminated objects do not behave with the same magnitude as luminous objects. I think that was Bill's point.

There is reflected light loss on illuminated objects, which effectively (mathematically) makes the source light to be much further away than it is according to the inverse square rule. The object would have to reflect 100% of the incident light in order for light fall off along the total path follow the inverse square law. This makes relatively small differences in subject to camera distances unnoticeable, relatively, since the subject to light source distance is much greater.

The total light path has to be 1.41 times as great as the light source to subject distance in order to see a stop's difference.

Scott Chapin
Powder Springs, GA, USA
Nikonians Team Member

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

                            
Noel Holland Platinum Member Winner in the Nikonians 10th Anniversary Photo Contest Charter MemberSun 29-Mar-09 10:27 AM
2171 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#18. " RE: A gedanken experiment on light meters...."
In response to Reply # 17


TH
          

The major difference between illuminated and luminous subjects is merely the amount of light travelling towards the camera. That is why I suggested the comparison of a light table and a sheet of paper. With those two objects you can adjust the total amount of light travelling towards the camera so that the two can be matched for this thought experiment.

>There is reflected light loss on illuminated objects, which effectively (mathematically) makes the source light to be much further away than it is according to the inverse square rule.

No that is irrelevant. Once the light reflects from the surface the source of light stops being the original source and becomes the subject. The light we photograph comes from the subject. The only situation where that comment would be true is if you are photographing a mirror. In all other situations you can ignore the original light source for the purposes of examining the movement of light between the subject and camera.

For all subjects we impose a tonal judgement on the exposure. That tonal judgement has nothing to do with mathematics but with aesthetics. When photographing a light source we often deliberately underexpose to prevent the subject blowing out and bring out it's detail. In reality there is NO correct exposure for any one situation. This is the crux of the zonal system. As photographers our exposure judgements are aesthetic judgements determined by where we want to place each part of an image on a tonal range.

When we discuss the technical aspects of photography we must always exclude those aesthetic and artistic judgments from our evaluation of technical issues. Hence the suggestion of a light table versus a sheet of paper. There is no baseline exposure possible for an effective point source like a light bulb except by use of an artistic decision as to what it should look like. But for the wide diffuse emitted light of a light table you can use mid grey as your baseline, just as you would for a sheet of paper using reflected light.

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

                                
bclaff Silver Member Awarded for multiple contributions for the Resources Nikonian since 26th Oct 2004Sun 29-Mar-09 04:42 PM
9681 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#19. " RE: A gedanken experiment on light meters...."
In response to Reply # 18


Vancouver (WA USA not Canada), US
          

Noel,

The only situation where that comment would be true is if you are photographing a mirror.

How do you justify this distinction?
How is 100% reflection different from 50%?

Bill

Visit me at My site

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

                                    
Noel Holland Platinum Member Winner in the Nikonians 10th Anniversary Photo Contest Charter MemberSun 29-Mar-09 06:18 PM
2171 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#20. " RE: A gedanken experiment on light meters...."
In response to Reply # 19
Sun 29-Mar-09 06:30 PM by Noel Holland

TH
          

Because in Scott's comment he treated the light fall off as originating at the source and continuing as it bounced off the subject and toward the camera. To my mind that only works with a 100% reflective surface. I contend that a surface that diffuses the light, as are almost all surfaces except mirrors, acts in the same manner as a translucent surface with the light behind it for the purposes of the inverse square law. It's the very fact that a diffuse surface scatters the light about that means that it has to be treated as a separate step. You measure the light reaching the surface, take a judgement call on tonal relation based on the colour and apparent surface qualities of the surface and from there on out to the camera the exposure is set. No matter how far the subject is away then the exposure is still the same. That concept is the same no matter whether the subject you are photographing is lit externally or emitting light internally.

For example, if you measure the light hitting a recently appeared chesterfield sofa on the cricket pitch of Lords. If the incidence meter reads f/5.6 at 1/500 sec that exposure works if you are photographing the sofa with a 10.5 mm fisheye at 2 inches just as much as it does for a 2,000 mm from over half a mile away.

The distance between the subject and the camera makes no difference. Also whether the subject is a diffuse surface lit by reflected light or a translucent surface lit internally, again there is no loss of exposure due to distance.

Scott has already mentioned star light and questioned why distant starlight is so hard to photograph. Well I agree there is a problem with huge distance but that is not due to deficiencies in the exposure calculation but due to the wave particle duality relationship for light. Exposure is based on the concept that light is a wave and therefore constant. Exposure at the sensor / film is dependant on light as a particle in order to cause the photoelectric effect or in the case of film the latent image. However it all falls down with starlight. the light leaving a distant star is spread in a sphere around it. It is even nearby and acts like a wave but as the photons spread out, gaps in the wave appear. At great distances there are more gaps than wave and we are dependant for our ability to see distant stars no longer on Christiaan Huygens' wave theory but on the relationship of Newton's corpuscular theory of light..... and I'm going to stop there before this thread deviates completely from photography and we venture into particle physics.

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

                                        
ScottChapin Moderator Awarded for his high level skills in various areas, including Aviation and Birds Photography Charter MemberSun 29-Mar-09 09:38 PM
8047 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#24. " RE: A gedanken experiment on light meters...."
In response to Reply # 20


Powder Springs, US
          


>Scott has already mentioned star light and questioned why
>distant starlight is so hard to photograph.

I didn't mention photographing them. What I said was that dimmer stars can have greater apparent magnitudes due to their proximity to earth. Brighter stars can have lower magnitudes due to their greater distance from earth. When photographing light sources, their brightness diminishes with increased distance.

Not accounting for atmospheric conditions or loss from reflecting off something, the light falls off as an inverse square. Let's say at a distance of X from the source you have 1 unit of light. If at a distance of 2X from the source, a mirror reflects a distance of of 2X to the camera, you would have (1/4x)**2 light, or 1/16th units of light,

Now replace the mirror with a reflector that reflects half of the light hitting it and absorbs the rest. Now you have only 1/32 units of light reaching the camera, since the inverse square law has been interrupted so to speak.

For the inverse square law to hold true with the 50% reflector (and have 1/32 units of light), the light path would have to be 5X long, so the source light is "effectively" 3X from the reflector (subject), not 2X.

So in Hal's experiments, we have to take this source to subject distance adjusted for reflectivity into consideration. Light does fall off from reflecting objects, we just don't see it normally, because it is not usually significant.

Zone 5 is 5 stops different from zone 10. I would guess that somewhere in the vicinity of 5 stops are lost when light is reflected off an average zone 5 object. That makes the light source 32 times as far away as it appears to be. If Hal's light is 10' away, it might effectively be 320 feet away. Now our distance ratios in Hal's experiments might be more like 322':346', which is way less than one stop difference.

As Bill mentioned in his earlier post, it is the total light path distance that matters. I add that it has to be adjusted for reflected light absorption that "accelerates" the inverse square law.

Scott Chapin
Powder Springs, GA, USA
Nikonians Team Member

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

HBB Moderator Hal is an expert in several areas, including CLS Awarded for his excellent article contributions to the Resources. Charter MemberFri 27-Mar-09 03:20 PM
8415 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#7. "RE: A gedanken experiment on light meters...."
In response to Reply # 0


Phoenix, US
          

Neal:

Yet another perspective, expanding a bit on Noel's comments in response #4 to introduce angle of view of subject changes with focal length.

The size of the subject and the size of the camera sensor remain constant throughout this exercise.

The lens is zoomed to fill the frame with the same subject dimensions at all focal lengths within its range.

As the distance from the subject to the lens increases, the lens focal length is increased and the angle of view decreases in order to fill the frame with the subject.

The same photon density, assuming constant illumination, coming from the subject at 1X distance is now contained in a proportionally smaller bundle as the focal length increases. Ergo, the density per unit area increases in an inverse square of the distance manner as Noel mentioned. When spread over the sensor by the lens, the density at 1X distance and focal length is maintained.

Nice question Neal. Any more??

Regards,

HBB in Phoenix, Arizona
Nikonian Team Member

Go here for a list of membership upgrade benefits.

Photography is a journey with no conceivable destination.

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

    
nl Basic MemberFri 27-Mar-09 05:05 PM
833 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#9. "RE: A gedanken experiment on light meters...."
In response to Reply # 7


West Hartford, US
          

Hal:

Actually, I am not sure I follow you. I do not think that the photo density / unit area increases as distance increased; rather it should decrease (hence the inverse square law). I think that the key issue is that by zooming to keep the subject at a constant image size (in my hypothetical scenario, equal to the sensor size for convenience) you are effectively changing the aperture size (at constant f/stop) to exactly compensate for the 1/r-squared falloff of photon density. For instance, at 2X distance the photon density is reduced to D/4 (D being the photon density at distance X). However, ASSUMING that the required change in focal length results in a doubling of the aperture radius at constant f/stop leading to a 4X change in aperture area, exactly compensating the 1/4 photon density. This is, I think, what was meant by the idea of the optics of lens solving this problem for us!

Another point is that I am unclear on why the light to subject distance is a concern here. In other words, the illumination of the subject is measured via incident meter, and yes, that measurement depends on light to subject distance. However, once light source and subject are positioned and exposure determined, the light to subject distance does not affect exposure as the camera is moved (assuming the same incident angle to the subject). In my hypothetical situation, the light source could be placed at any point and intensity adjusted to achieve the desired incident reading without changing anything further in my reasoning (assuming point vs large light source issues, etc, were to be ignored).

Certainly camera perspective matters, which is why there remains controversy over whether the key light should be metered by pointing the incident meter at the light or at camera. I tend to meter by aiming at camera for the incident but for background and hair lights of course I aim at the light and/or from background to camera.

And yes, if these types of questions are interesting and helpful to other Nikonians as discussions points (you know I am always trying to develop an understanding for the underlying principles which helps me to figure out how to solve problems in lighting) I have more questions!

nl

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

HBB Moderator Hal is an expert in several areas, including CLS Awarded for his excellent article contributions to the Resources. Charter MemberSat 28-Mar-09 09:53 PM
8415 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#15. "RE: A gedanken experiment on light meters...."
In response to Reply # 0


Phoenix, US
          

Neal et al:

In an effort to throw some light on this thread (Sorry ... I couldn't resist!), I set up a simple test. Below are four images of the interior portion of a 24 by 36 inch test chart I printed up for use with lenses. The chart was taped to a door in our kitchen.

My Sekonic meter at an ISO of 100 in incident mode indicated 1/13 sec at F/2.8 in the middle of the chart. My Minolta color Meter IIIF indicated a color temperature of 2760 K in the same place. All four images were taken at 1/13 sec at F/2.8, an ISO of 100 and Auto White balance. Camera was a D2X and all images were shot in RAW.

The camera/lens was mounted on a tripod and positioned such that the test chart completely filled the frame at each focal length. Three lenses were used: 14-24 mm F/2.8; 24-70 mm F/2.8; 70-200 mm F/2.8. The only illumination on the chart was a set of overhead tungsten track lights, which were not changed prior to or during the test.

There was no RAW or CS4 processing of any kind. All images are as they came out of the camera. Images were cropped slightly to correct slight camera alignment issues, then sized and compressed for posting.

In descending order below, the lens focal lengths and chart to film plane dimensions are:

14 mm: 27.5 Inches
24 mm: 45.5 Inches
70 mm: 114.5 Inches
200 mm: 318.0 Inches

I'm going to violate our three image per post rule, but I think it is worth it for this discussion.

Comments?

Regards,

HBB in Phoenix, Arizona
Nikonian Team Member

Go here for a list of membership upgrade benefits.

Photography is a journey with no conceivable destination.











Attachment #1, (jpg file)
Attachment #2, (jpg file)
Attachment #3, (jpg file)
Attachment #4, (jpg file)

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

    
ScottChapin Moderator Awarded for his high level skills in various areas, including Aviation and Birds Photography Charter MemberSat 28-Mar-09 11:30 PM
8047 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#16. "RE: A gedanken experiment on light meters...."
In response to Reply # 15


Powder Springs, US
          

Hi Hal,

The images keep getting darker the further away that you get, as expected, but more than expected, since the subject is illuminated. But then, the source light might have been fairly close. It shows that you do need to open up as you get further away.

In Bill's question regarding a light bulb, it would be more exaggerated, as the light bulb is luminous and there is no loss due to reflection.

It only stands to reason that you need to open up as you move away from a light bulb. Otherwise flood light exposures would be the same whether they are 5' or 10' away from the subject. We wouldn't need guide numbers for flash units either.

Scott Chapin
Powder Springs, GA, USA
Nikonians Team Member

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

        
Noel Holland Platinum Member Winner in the Nikonians 10th Anniversary Photo Contest Charter MemberSun 29-Mar-09 07:06 PM
2171 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#21. "RE: A gedanken experiment on light meters...."
In response to Reply # 16
Sun 29-Mar-09 07:14 PM by Noel Holland

TH
          

I took all four images into PS and carried out the following actions.

1) Mode = grayscale
2) Select info pane
3) Select move tool

Hovering over the centre section of each image the following K levels were noted or each of the four circular patches starting top left and going clockwise

Image 1) 44%, 69%, 59%, 63%
Image 2) 48%, 72%, 65%, 67%
Image 3) 48%, 72%, 65%, 64%
Image 4) 52%, 72%, 68%, 64%

Now lets look at the distances:

1) 27.5 inches
2) 45.5 inches
3) 114.5 inches
4) 318.0 inches

So that means using the inverse square law a spread in stops of:
1) Base distance
2) 1.77 the base distance = 1.25 stops of light loss
3) 4.46 the base distance = 3.15 stops of light loss
4) 12.37 the base distance = 8.75 stops of light loss

Formula = (Image distance / base distance) / sqrt(2) = loss in stops

Now I'll agree that the images appear to show a very tiny amount of light loss but that does not prove anything but that lens differ. These images do not show that the inverse square law applies for the distance between subject and camera - it shows the reverse. If the inverse square law applied then there would be more than 8 stops of exposure difference between the first and last image and the last image would be little more than a black frame.

Do the math folks.

In fact just use your heads - if you moved your eyes closer to that target would it have appeared brighter? Of course it wouldn't have, so why think that it would appear any brighter or darker as you move your camera?

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

            
bclaff Silver Member Awarded for multiple contributions for the Resources Nikonian since 26th Oct 2004Sun 29-Mar-09 07:26 PM
9681 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#22. "RE: A gedanken experiment on light meters...."
In response to Reply # 21
Sun 29-Mar-09 07:27 PM by bclaff

Vancouver (WA USA not Canada), US
          

Noel,

The apertures (entrance pupil diameters) are different.

Bill

Visit me at My site

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

                
Noel Holland Platinum Member Winner in the Nikonians 10th Anniversary Photo Contest Charter MemberSun 29-Mar-09 11:06 PM
2171 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#25. "RE: A gedanken experiment on light meters...."
In response to Reply # 22


TH
          

>The apertures (entrance pupil diameters) are different.

The aperture sizes are indeed different, but the aperture settings are not. I've been saying all along that as the focal length increases the aperture size must increase, of that I have no quibble. But I've also been saying that the aperture setting remains the same.

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

            
ScottChapin Moderator Awarded for his high level skills in various areas, including Aviation and Birds Photography Charter MemberSun 29-Mar-09 08:49 PM
8047 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#23. "RE: A gedanken experiment on light meters...."
In response to Reply # 21
Sun 29-Mar-09 08:57 PM by ScottChapin

Powder Springs, US
          

>Now lets look at the distances:
>
>1) 27.5 inches
>2) 45.5 inches
>3) 114.5 inches
>4) 318.0 inches
>
>So that means using the inverse square law a spread in stops
>of:
>1) Base distance
>2) 1.77 the base distance = 1.25 stops of light loss
>3) 4.46 the base distance = 3.15 stops of light loss
>4) 12.37 the base distance = 8.75 stops of light loss
>
>Formula = (Image distance / base distance) / sqrt(2) = loss in
>stops
>


Huge red flag here, since your distance does not include the total light path distance from the source light to the subject and back to the camera, the math is all wrong. It needs to be redone. You have to add the source to subject distance into the distances used AND adjust for the effective source to subject distance due to light loss from reflection which accelerates the inverse square law.

So the minor but noticeable drop in Hal's photos are due to a slight light fall off, since I am sure that the ratio of total light paths is less than the square root of 2.

As to the light bulb. Your logic says that Studio floods that meter f/8 when 10' from the subject should read f/8 at 5' from the subject, since the subject does not care how far away the light source is.

I don't think Hal's, loss of light was due to the lens. As I understand it, the f numbers were all the same. He just used different lenses as he moved back, to maintain the crop, so that light reflected from other objects did not affect the exposure.

Scott Chapin
Powder Springs, GA, USA
Nikonians Team Member

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

                
Noel Holland Platinum Member Winner in the Nikonians 10th Anniversary Photo Contest Charter MemberSun 29-Mar-09 11:24 PM
2171 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#26. "RE: A gedanken experiment on light meters...."
In response to Reply # 23
Mon 30-Mar-09 12:00 AM by Noel Holland

TH
          

Big green flag, I say that the distance from the light source is irrelevant.

The distance from the light source only affects the light arriving at the subject. Once that exposure is set the distance from the subject to the camera makes no difference. Otherwise incidence light meters would have a distance scale marked on them. They don't therefore I contend that it's because they don't need them and that is because the distance from the subject to the camera makes not one jot of difference to the exposure measurement.

>You have to add the source to subject distance into the distances used AND adjust for the effective source to subject distance due to light loss from reflection which accelerates the inverse square law.

No, inverse square law does not apply to subject to camera distances only light to subject distances. That was what I was trying to prove with the above post.

If you have a spot meter then try this experiment. Place a large light source one metre away from the wall. Place a black flag (not a red or green one )in front of the light source to ensure it illuminates the wall but no light strays forward except reflected from the wall. Now spot meter the wall from a few centimetres away (or as close as you can get without shadowing the area being read). Now spot meter the same spot of wall from 3 metres away.

I suggest that the reading would be the same in both positions. You suggest that the reading would be affected by the light having to travel 4 metres instead of just 1 metre.

Would I be correct in stating our current respective stances?

Edit - if so I've just tested it.

I taped a sheet of A3 graph paper on a door. I put an SB800 set to manual, 1/32 power and a 105 mm spread. I set it on a light stand 1 metre away from the paper and put a Honl flag on the side. I then placed a mark on the graph paper and directed the strobe right at the mark. Tot eh strobe I attached a PW multimax and I used a Sekonic Dualmaster L-558R meter. I took measurements of the reflected light at the marked spot from 1 metre away from the paper and 3 metres away. The Sekonic was set to spot meter, ISO 200 and 1/250 sec.

The meter reading at 1 metre distant from the paper was F/20 and zero tenths... anyone want to guess what the meter reading was from 3 metres distance (a total light path distance of 4 metres)?

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

                    
ScottChapin Moderator Awarded for his high level skills in various areas, including Aviation and Birds Photography Charter MemberSun 29-Mar-09 11:57 PM
8047 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#27. "RE: A gedanken experiment on light meters...."
In response to Reply # 26
Sun 29-Mar-09 11:57 PM by ScottChapin

Powder Springs, US
          

Probably the differences are stated accurately. My concern is that once the light has reflected and lost its oomph, there will be no huge noticeable differences within reasonable working distances.

Incident light meters themselves have a white dome that cuts the light down by as much as a zone 5 reflective subject would. After that, perhaps 5 stop loss, the additional loss within reasonable working distances is not huge.

I do know that as I metered the same area of a lamp, which was Bill's original question, the meter changed as I walked away from it. As I walked back, I continued to zoom in to maintain the same crop. It wasn't huge, but it was there.

I just don't think that once light has been absorbed and reflected that the difference will be noticeable within normal working distances. The original question was in reference to photographing luminous sources as opposed to illuminated subjects though.

Oh my head hurts Noel. We may need to agree to disagree on this one. I don't want hard feelings to come of it.

Scott Chapin
Powder Springs, GA, USA
Nikonians Team Member

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

                        
Noel Holland Platinum Member Winner in the Nikonians 10th Anniversary Photo Contest Charter MemberMon 30-Mar-09 12:05 AM
2171 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#28. "RE: A gedanken experiment on light meters...."
In response to Reply # 27


TH
          

> I don't want hard feelings to come of it.

Certainly not mate. I wouldn't have pursued this thread if I didn't think it was worth getting to the bottom of it.

>Oh my head hurts

LOL, know what you mean.

>We may need to agree to disagree on this one.

Well before you do, go up to my post above and repeat the experiment I described in my edit and see what results you get.

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

                    
HBB Moderator Hal is an expert in several areas, including CLS Awarded for his excellent article contributions to the Resources. Charter MemberMon 30-Mar-09 12:06 AM
8415 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#29. "RE: A gedanken experiment on light meters...."
In response to Reply # 26


Phoenix, US
          

Noel, Scott, et al:

I just metered a small section of the white door where the above test chart is mounted using my Sekonic L758DR in one degree spot mode and under the same overhead tungsten track lights used yesterday. The spot meter was set to F/2.8, as in yesterday's test.

At all distances from the door, from 27.5 inches up to and including 318 inches, the spot meter gave a constant 1/125 second at F/2.8, three-plus stops faster than the 1/13 second used yesterday. We would expect this because the door is painted white and the meter, now measuring reflected light instead of incident, is trying to make it gray. Hence, the underexposure.

I could not go beyond this distance for two reasons:

1) A jog in the line of sight due to a hallway offset.

2) Even without the hallway offset, the one degree spot extended beyond the area of the door I was metering into darker shadow areas.

Great discussion! Thanks to all!

Regards,

HBB in Phoenix, Arizona
Nikonian Team Member

Go here for a list of membership upgrade benefits.

Photography is a journey with no conceivable destination.

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

                        
ScottChapin Moderator Awarded for his high level skills in various areas, including Aviation and Birds Photography Charter MemberMon 30-Mar-09 01:05 AM
8047 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#30. "RE: A gedanken experiment on light meters...."
In response to Reply # 29


Powder Springs, US
          

The thing here though, is that as you step backwards from the wall, the area reflecting light back to the meter gets larger and larger. So the decreased light due to distance is compensated for by more light being reflected back from the source. This might be what Noel is really getting at?

Now, if your spot meter measured the exact same area on the wall as you moved back, you would be measuring the original source as its light fell off by the inverse square. I don't think you are.

When you double your distance you get 1/4th the intensity of the light from 4 times the original source. That makes it a wash......I think.

I will have to try Noel's test, but I have to figure out how to keep my spot meter on exactly the same area of the subject as I back off.

The other thought is that keeping the lens focal length the same will yield constant exposure. This is because when you are twice as far away, the subject exposes 1/4 as much of the sensor, so it's a wash.

Scott Chapin
Powder Springs, GA, USA
Nikonians Team Member

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

                            
Noel Holland Platinum Member Winner in the Nikonians 10th Anniversary Photo Contest Charter MemberMon 30-Mar-09 01:31 AM
2171 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#31. "RE: A gedanken experiment on light meters...."
In response to Reply # 30
Mon 30-Mar-09 01:32 AM by Noel Holland

TH
          

The thing here though, is that as you step backwards from the wall, the area reflecting light back to the meter gets larger and larger. So the decreased light due to distance is compensated for by more light being reflected back from the source. This might be what Noel is really getting at?

Yes, absolutely.

If you keep the lens at the same focal length as you do with a spotmeter then you need to meter from a large surface area so that when you back up the meter still covers an area of the illuminated source. That means the meter is indeed reading from a larger area of the subject. A direct reading works in that situation.

For a subject that doesn't entirely cover the metering area then direct metering no longer works and then you have to either use indirect metering such as an incidence meter or else zoom the lens. And it the zooming of the lens that causes light loss due to magnification, not the distance.

Distance and reduction in relative size on the sensor cancel each other out and hence there is no exposure change over the distance between subject and camera.

Now, there is just one small jump to make..... you need to make the swap in your head of thinking of a luminous subject from being a light source to being the subject. If the reflected light subject does not change in respect to distance to the camera then neither can the luminous subject, because the basic principal is the same - it's no longer a light source but the subject.

Yeah I know it makes your head but it's such a little step, a light source that is the subject - is treated as a subject not a light source.


>but I have to figure out how to keep my spot meter on exactly the same area of the subject as I back off.

Put a big red x on the paper using a marker pen.

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

                                
ScottChapin Moderator Awarded for his high level skills in various areas, including Aviation and Birds Photography Charter MemberMon 30-Mar-09 10:58 AM
8047 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#33. "RE: A gedanken experiment on light meters...."
In response to Reply # 31


Powder Springs, US
          

Actually, for me, it's a big jump. It's the jump to believing the same holds true when you keep the subject identically cropped/framed at all distances. Then all reflected light is from the same source and falling off at an inverse square. There is no offset from an area growing at the same rate that the light falls off.

I can put an "X" on the wall, but my handheld meter will not stay cropped on it as I back away from it. If I were to change the zoom on a lens, then Neal and you would say any differences are attributable to the lens. There is no laboratory certified measuring equipment in my house.

I could take a reflected reading of the source at the wall and compare to a reflected reading from the wall to see how much my assumption of inverse square acceleration holds true. I think the problem here is that the meter would have to crop enough to see only the source AND you would have to meter the wall such that the meter would only see the same area metered from the light source at the wall. But that would be futile, since we already know that a gray card reflects substantially less light.

There's nothing in my arsenal that measure these levels that accurately. It's hard for me to accept that it is just a lens variance of Hal's, when it is behaving as I would expect.

It would be interesting to hear Bill's comments regarding his earlier statement that the total light path is what matters. To my way of thinking, it does, providing the image cropping remains the same.

Scott Chapin
Powder Springs, GA, USA
Nikonians Team Member

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

                            
nl Basic MemberMon 30-Mar-09 01:36 AM
833 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#32. "RE: A gedanken experiment on light meters...."
In response to Reply # 30


West Hartford, US
          

Well, I didn't expect to generate so much discussion, but I'd glad to see such an active debate on this topic!

I have drawn my own conclusions, having given this a great deal of thought, and I am fairly confident in my conclusions. That being said, the discussion has been extremely helpful to me in thinking this through.

For those interested in the underlying physics and intrigued by the classical vs quantum mechanical treatment of light falloff, I'd be happy to join a new thread on that topic, but that my be way too OT for this forum. I'll just leave it by noting that even at long distances, one should not neglect wave-particle duality, but remember that the photons aren't anywhere until they are measured. (And no, that's not intended to be funny.)

Finally, Hal: great experiment. I think the slight differences in the images is indicative of the different lenses that you used. I wish I had a 14-200/f2.8 ... but even then, unless it were a hypothetically perfect lens....The spotmeter approach does get you as close as you can to an ideal response, although Scott is correct that the metered area of the wall changes even with a spotmeter as distance changes, although presumably far less so that for a non-spot reflected meter.

nl

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

HBB Moderator Hal is an expert in several areas, including CLS Awarded for his excellent article contributions to the Resources. Charter MemberTue 31-Mar-09 05:33 PM
8415 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#34. "RE: A gedanken experiment on light meters...."
In response to Reply # 0


Phoenix, US
          

One Last Quick and Dirty Example:

Below are four new images of a portion of the test chart used earlier. They were taken under the same lighting, using a Nikon D40X with a Nikkor 18-200 mm F/3.5 - 5.6 zoom. The incident meter reading was 0.4 sec at F/5.6, and all images were captured in RAW mode at those specs. As before, no post processing done with the exception of minor cropping for alignment issues.

Notice the slight barrel distortion at 18 mm and the slight pin cushion distortion at 35 and 70 mm.

I suspect we will agree that the exposure consistency here using a single lens is greater than the prior test which used three different lenses. Interesting.

From top to bottom, the focal lengths and distances are:

18 mm, 23.00 inches
35 mm, 41.00 inches
70 mm, 80.00 inches
200 mm, 199.00 inches

Regards,

HBB in Phoenix, Arizona
Nikonian Team Member

Go here for a list of membership upgrade benefits.

Photography is a journey with no conceivable destination.











Attachment #1, (jpg file)
Attachment #2, (jpg file)
Attachment #3, (jpg file)
Attachment #4, (jpg file)

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

    
nl Basic MemberTue 31-Mar-09 07:00 PM
833 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#35. "RE: A gedanken experiment on light meters...."
In response to Reply # 34


West Hartford, US
          

Hal:

Great data, and (I think) clearly supports the idea that (at least over the distances we are using) the incident light meter results provide exposure in a distance-independent manner. I am not suprised to see this result compared to your earlier findings; while the variability in exposure was greater than I would have expected between multiple pro-caliber Nikon lenses, I did think that there would be at least some inter-lens variability.

I have been thinking further on the geometry/physics involved here, and without trying to reopen the debate as to light path from source or object, here are a few more thoughts:

We should remember that the inverse square law for light both assumes and requires that the light is a point source, radiating equally in all directions.

For an illuminated subject, that really isn't true, so while the inverse square law works pretty well for our strobes which tend to approximate to a reasonable degree a point source with a light spread that is not spherically symmetric but "close enough for government work" the inverse square law does wind up working well enough. If you use a non-point source, when you are about a distance of 5 x the longest source dimension (this works for an object reflecting light as well) the error from 1/r2 is about 1%, so if a strobe light is (say) 6" in diameter, once you are about 30" from it the error from the inverse square law is minimal.

I am still stuck on the geometrical relationship between the size of the aperture, focal length, lens position, and source size exactly balancing so that the required f/stop remains contant as distance varies. I need to get a better picture in my head (and learn to think in steradians) to ensure the math works out...but I assume it will.

If you use a large softbox and meter (incident) at a given softbox to subject distance, then move the softbox in to 1/2 the distance, you expect a 2 stop exposure difference. In practice, however, when you meter it is often not exactly 2 stops. This may be due to changes in the softbox angle relative to subject (more or less feathered), non-uniform illumination of the softbox front diffuser (sometimes 1/2 - 1 stop across the front of the softbox) and so on, but could also be due to the relative large size of the softbox light source compared to source-subject distance and hence failure of the inverse square law due to a non-point source.

At very large distances we will probably not see the same uniformity of exposure - which is at least in part due to the optics of long telephoto lenses as well as atmospheric absorption, which we are ignoring at short distances but cannot at longer distances - at an longer distances, are also affected by the color of the light source as the atmospheric scattering is wavelenght dependent as well.

I find it a bit interesting that I cannot google any other discussions of this nature. I would have thought this would be something that would have come up before. I will have to go back to some of my books and see what there is...

nl

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

        
ScottChapin Moderator Awarded for his high level skills in various areas, including Aviation and Birds Photography Charter MemberTue 31-Mar-09 08:35 PM
8047 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#36. "RE: A gedanken experiment on light meters...."
In response to Reply # 35


Powder Springs, US
          

Ach was! Mein Kopf tut weh.

First:

I think it unreasonable to assume that reflected light no longer obeys the law of inverse squares. From what I can find it does. The angle of incidence equals the angle of reflection and the photons are dissipated. Light does not reflect perpendicularly to the plane of inflection.

I think the reason we do not see this from the various camera positions is due to the accelerated light loss from the perhaps 5 stop loss at reflection and the loss as it is filtered through the light's globe or enclosure, unless Hal has bare bulbs in his kitchen. I believe the mathematical implication here is such that the ratio of light path lenghts between the two camera positions is much nearer to 1:1 than you think.

Second:

This discussion stems from Bill's (who remains mysteriously mums, and is enjoying this I'm sure) question regarding photographing light sources, not illuminated objects.

Now I guess we need to meter cropped white monitor screens from various distances. I believe here it must be significant, or guide numbers are meaningless. There has to be more light entering a lens as it gets closer the the source, just as there is more light falling on a person's face as they get nearer to a light source.

Scott Chapin
Powder Springs, GA, USA
Nikonians Team Member

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

            
HBB Moderator Hal is an expert in several areas, including CLS Awarded for his excellent article contributions to the Resources. Charter MemberTue 31-Mar-09 11:27 PM
8415 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#38. "RE: A gedanken experiment on light meters...."
In response to Reply # 36


Phoenix, US
          

Scott:

The kitchen has an eight foot ceiling.

When the camera was at 18 mm focal length and 23 inches from the chart, a track light was directly over the camera and pointed at the wall to the left of the chart, not directly at it. The track light is a tungsten flood bulb, 3.5 inches or so in diameter, mounted in a can that is fastened to the track. Ergo, the light on the chart is mostly indirect, bounced off the wall to the left and around that area of the kitchen, with very little of it being direct.

The distance from the center of this bulb to the center of the chart is 42 inches.

The next track light is about 43 inches from the chart, within a couple inches of the camera's position at 35 mm focal length. This light is also pointed at the area to the left of the chart, which is a cooking surface.


More data:

I turned on some dim track lights in a hallway leading to the bedrooms. The walls are painted a very light, neutral gray. All light on the wall at the end of the hall is indirect, reflected off walls, floor and ceiling. With the Sekonic meter in the one degree spot mode, I measured a spot at the end of the hall, about three feet from the wall. I then backed up taking measurements of the same area as I went, until I reached fifty feet from the wall and had to stop because my back was against the chart we have been discussing here. At every point I measured, the meter gave me 1.0 seconds at F/5.6.

At three feet, the one degree circle in the meter encompasses a circle 0.6283 inches in diameter, and an area of 0.3019 square inches.

A fifty feet, the one degree circle in the meter encompasses a circle 10.4722 inches in diameter, and an area of 86.0961 square inches.

Comments?

HBB in Phoenix, Arizona
Nikonian Team Member

Go here for a list of membership upgrade benefits.

Photography is a journey with no conceivable destination.

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

        
HBB Moderator Hal is an expert in several areas, including CLS Awarded for his excellent article contributions to the Resources. Charter MemberTue 31-Mar-09 08:50 PM
8415 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#37. "RE: A gedanken experiment on light meters...."
In response to Reply # 35


Phoenix, US
          

Neal et al:

I was a bit surprised at the lack of consistency when using three quality Nikkor lenses in the earlier images ... and, for now, have no explanation for it. The often maligned 18-200 mm shows remarkable consistency across the focal length range. I actually captured images at eight focal lengths, but posted only four because they were all essentially identical. I was also careful to use shutter speeds well below sixty Hertz, to avoid any possible sine wave effect on illumination. This effect would be minimal however, because the tungsten filaments do not cool off or dim appreciably during the cycle as voltage decreases and passes through zero. AC gas discharge lamps are another matter entirely.

I also searched the web for additional discussion, with the same result you experienced.

I keep waking up in the middle of the night with angle of view, focal length, subject to lens distance and sensor size diagrams in my head. I have long been a fan of cultivating the unconscious mind as a problem solving tool. I guess I'm going to have to sit down, draw some diagrams and do the arithmetic so I can get some uninterrupted sleep.

It also keeps popping up that we are not dealing with collimated or polarized light. Not exactly sure what that means at the moment.

I haven't tried metering my soft boxes at full stop distances yet, but will do so the next time I have them up. My soft boxes are about 24 by 36? inches, with a circular xenon tube about three inches in diameter. The inside surfaces are coated with a very reflective, mottled silver foil. Not exactly a point source.

Agreed, at longer distances, other factors that you mention emerge.

If nothing else, we are keeping the gray matter stirred up a bit, eh?

Thanks for your comments Neal. Let us know if you have an epiphany of sorts.

Regards,

HBB in Phoenix, Arizona
Nikonian Team Member

Go here for a list of membership upgrade benefits.

Photography is a journey with no conceivable destination.

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

            
nl Basic MemberWed 01-Apr-09 01:27 AM
833 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#39. "RE: A gedanken experiment on light meters...."
In response to Reply # 37


West Hartford, US
          

Hal:

No epiphanies yet, but I'm working on it! In any case, I have to turn in early tonight as I have an early morning ahead of me, but perhaps tomorrow night I can sit down with some paper and pencil and work on some diagrams and trigonometry. I also have a copy of Wilde's book on lenses around here somewhere, so I have to try to find that and see if he provides any help for us.

As soon as I have time I will run a softbox experiment too.

Scott:

I am not trying to repeal the laws of physics. However, the inverse square law is valid for a point source. Now, while the sun is obviously far from a point source, at astronomical distances (like on the Earth) we are far enough away that it behaves like a point source to such a high approximation that any deviation can certainly be ignored. This is NOT the case for a subject of say, 6 feet in height when you are 10 or 12 feet away shooting a portrait, and in such a setting, the inverse square law does not exactly describe the light distribution around that subject. It is pretty close, still, so we can use it as a very good approximately. The inverse square law applies to situations in which the source dimension is small compared to the observer to source distance.

I understand if you are not convinced. However, google for the energy of a dipole radio transmitter (valid since light and radio are both EM waves, just different frequencies). You will find that along the axis of the dipole, the energy density falls off as r cubed, not squared. This is a function of the large source to measurement distance.

nl

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

                
HBB Moderator Hal is an expert in several areas, including CLS Awarded for his excellent article contributions to the Resources. Charter MemberWed 01-Apr-09 03:40 AM
8415 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#40. "RE: A gedanken experiment on light meters...."
In response to Reply # 39
Wed 01-Apr-09 03:58 PM by HBB

Phoenix, US
          

Neal, Scott, et al:

It occurred to me that if I am using the sun as my light source and want to decrease its intensity by one stop, without changing aperture or shutter speed, I will have to back up 131.5 million miles or so. Sort of takes the fun out of available light, eh?

I will be dropping off line Thursday through next Monday, while attending a two-day nude photography workshop at the Edward Weston Studio in Carmel, California.

Regards,

HBB in Phoenix, Arizona
Nikonian Team Member

Go here for a list of membership upgrade benefits.

Photography is a journey with no conceivable destination.

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

                    
ScottChapin Moderator Awarded for his high level skills in various areas, including Aviation and Birds Photography Charter MemberWed 01-Apr-09 10:49 AM
8047 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#41. "RE: A gedanken experiment on light meters...."
In response to Reply # 40


Powder Springs, US
          

I'm jealous.

The sun distance was what I was referring to when I was saying that the ratio of 1 astronomical unit to 1 astronomical unit plus the distance to the moon is still 1:1. That's why the moon is a sunny 16 shot.

Scott Chapin
Powder Springs, GA, USA
Nikonians Team Member

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

                
ScottChapin Moderator Awarded for his high level skills in various areas, including Aviation and Birds Photography Charter MemberWed 01-Apr-09 11:12 AM
8047 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#42. "RE: A gedanken experiment on light meters...."
In response to Reply # 39


Powder Springs, US
          

Hi Neal,

I don't disagree with anything you have said, with exception to the concept that light does not fall off according to the inverse square when it is reflected. It does fall off, and if the light source were a point, then it would by the inverse square (accounting for reflective light loss).

Last night other issues started to haunt me in my sleep...LOL. Some of the light reflecting off of Hal's targets come from reflecting off the surrounding walls, floor and ceiling. Therefore the reflected light is highly softened and probably the lion's share of that reflected light might very well be perpendicular to the targets. It certainly would make an "apparent" light source seem a long distance away, much further away than the fixtures in the ceiling.

I think the fact that the inverse square is based on a point would hold up to a single light source. In the case of a light bulb, it would be a point inside the bulb at the center of the emanating photons. Where the law would fall apart is at the surface of the bulb, since that's not the center of the source.

Another interesting concept, that I should have deduced is the Cosine Law of Illumination that says reflected light falls off at the cosine of the angle of incidence. IOW, at 60 degrees, a cone of light projects an ellipse having twice the area of a circle that would be cast when perpendicular to the surface. In that case there is a loss of 50%. Add that to the 5 stop drop from the reflection from a gray card, and you've got some really soft light.

Scott Chapin
Powder Springs, GA, USA
Nikonians Team Member

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

                    
nl Basic MemberWed 01-Apr-09 12:26 PM
833 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#43. "RE: A gedanken experiment on light meters...."
In response to Reply # 42


West Hartford, US
          

Scott:

I agree. Light certainly does fall off the farther you get from the source. We are agreed that the fall off is inverse square from a point source. Where we disagree is that the fall off is NOT inverse square if the source is not a point source. HOWEVER, once you are a distance of about 5x the longest dimension of the source, the deviation from inverse square is so small as to be immaterial. That's just not the case for a large subject, whether it is lit by self-luminance or reflection.

In any case, the fact that we are agreed that the light DOES fall of (whether inverse square or not) as you move away from the source was the genesis of my question in the first place! When you make an incident reading, you do not take into account the subject to camera distance, and I was trying to understand why. My hypothesis was that IF you are adjusting focal length to keep the same apparent subject size, then by keeping a constant f/stop, the larger effective aperture size resulting from the required increase in focal length compensates for the light falloff, as my assumptive explanation. The fact that both the inverse square law and the f/stop relationship are based on powers of 2 accounts for this exact compensation.

We have ignored and/or assumed a lot in all of this. The effective aperture of a lens is not the same as the physical opening in the diaphragm when the aperture is behind one or more lens elements (which is, of course, a requisite factor in designed constant aperture zoom lenses!). This becomes a significant factor when focusing at <10 times focal length and/or when the image plane is not a 1 focal length distance from the lens. The lens opening does not exactly subtend a portion of the "sphere of light" from a point source, but is small enough to work as an approximation, especially at longer distances. And so on.

Still, it is not an uncommon practice to take an incident meter reading and then chimp test images and adjust exposure based on histogram, blinkies, polarioids, etc. Whether this practice is required because of small but important effects of light falloff, or because the metered "optimal" exposure is not necessarily the optimal "artistic" exposure is another point for discussion.

Finally, you are correct about a light bulb. A bare filament, being quite small, behaves like a point source at relatively small distances away. However, a light bulb requires a greater distance for the same behavoir because once the light emitted from the filament strikes the bulb's glass suface (which itself is often treated to modify the light), that surface becomes the effective source and is much larger than the filament. It is still small enough (say about 6") that once you are about 30" away, the deviation from a point source is negligible, so the inverse square law works just fine for light bulbs, flash bulbs, strobes, etc.

nl

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

                        
ScottChapin Moderator Awarded for his high level skills in various areas, including Aviation and Birds Photography Charter MemberWed 01-Apr-09 01:53 PM
8047 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#44. "RE: A gedanken experiment on light meters...."
In response to Reply # 43


Powder Springs, US
          

Neal,

I cannot take exception with anything you are saying, except for the focal lenght vs. aperture relationship.

If you have a huge uniformly lit constant reflective wall and you stand a fixed distance from it, then I think what you say is true. As you zoom in and out with the wall filling the frame, the change in aperture compensates for the change in focal length.

I "think" that when you are indoors, the light is highly diffused. It comes from multiple sources including bounced light of the surrounding walls, ceilings and floors. This effectively delocalizes (did I just make up a word?)the light source.

What fools us a lot, I think, is that our eyes are fixed focal length lenses. As we back off from a subject, the image exposes less and less retina. The photons per square unit remains the same even though the photon density is droping, like a fixed focal lenght lens and the film/sensor.

If light falls off from a reflected surface, exposure has to be increased, ceterius paribus. All things are not the same though, going back to "effective" camera to source distance and reflected and bounced light in the room.

Next step Hal, can you get your wife to let you paint everything in your kitchen flat black, except for your target of course?

Scott Chapin
Powder Springs, GA, USA
Nikonians Team Member

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

                            
HBB Moderator Hal is an expert in several areas, including CLS Awarded for his excellent article contributions to the Resources. Charter MemberWed 01-Apr-09 04:26 PM
8415 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#45. "RE: A gedanken experiment on light meters...."
In response to Reply # 44


Phoenix, US
          

Sure ... Do all of you remember that my wife is a vegetarian? She loves carrots ... preferably brilliant cut and at least two per serving. Donations to her diet cheerfully accepted.

I can turn off all lights in the hall but one, and all other lights in the house, close all doors, blinds, shutters, etc. elsewhere and take the measurements at night. Close enough?

This will have to wait until I return from the photo workshop mentioned earlier.

I have a nuclear physicist friend (Bill) who is a department head at George Washington University in Washington, DC. Bill spends most of his time traveling the world between various particle physics experiments/laboratories he is involved with, including the Large Hadron Collider (LHC) in Europe.

Some years ago when it was discovered that neutrinos have mass, I called Bill and asked him: "Bill, if neutrinos have mass, does that mean they are Catholic?" There was a very long pause before he said: "You called me just to ask me that?"

If I can catch Bill home long enough to talk to him I will run this discussion by him and report back.

Regards,

HBB in Phoenix, Arizona
Nikonian Team Member

Go here for a list of membership upgrade benefits.

Photography is a journey with no conceivable destination.

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

                                
ScottChapin Moderator Awarded for his high level skills in various areas, including Aviation and Birds Photography Charter MemberWed 01-Apr-09 05:01 PM
8047 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#46. "RE: A gedanken experiment on light meters...."
In response to Reply # 45


Powder Springs, US
          

LOL....no Hal. You HAVE to paint the hallway flat, non reflective black. There can be NO chance that non direct light from the source can make it to the target via bouncing of the walls, ceiling, or floor. You probably need a very tiny, yet quite bright bare bulb with no globe to filter the light source.

I would be interersted to hear what Bill says about it, when you get a chance to talk to him.

Scott Chapin
Powder Springs, GA, USA
Nikonians Team Member

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

                                    
Noel Holland Platinum Member Winner in the Nikonians 10th Anniversary Photo Contest Charter MemberFri 03-Apr-09 09:28 AM
2171 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#49. "RE: A gedanken experiment on light meters...."
In response to Reply # 46


TH
          

Or erect a 40 foot high platform on a mountain, at night, with no moon, under heavy cloud, line the platform and everything on it in dense black counter brushed velvet,.......

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

                                        
ScottChapin Moderator Awarded for his high level skills in various areas, including Aviation and Birds Photography Charter MemberFri 03-Apr-09 11:09 AM
8047 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#50. "RE: A gedanken experiment on light meters...."
In response to Reply # 49


Powder Springs, US
          

LOL.....yeah, that'll work.

Scott Chapin
Powder Springs, GA, USA
Nikonians Team Member

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

                                        
gkaiseril Gold Member Nikonian since 28th Oct 2005Fri 03-Apr-09 01:38 PM
6739 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#51. "RE: A gedanken experiment on light meters...."
In response to Reply # 49


Chicago, US
          

Too much light polution in the Arizona nights, Kitt Peak National Observatory, like may astromical sites, has to deal with this increasing problem.

George
My Nikonian Galleries

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

                                            
ScottChapin Moderator Awarded for his high level skills in various areas, including Aviation and Birds Photography Charter MemberFri 03-Apr-09 03:41 PM
8047 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#52. "RE: A gedanken experiment on light meters...."
In response to Reply # 51


Powder Springs, US
          

George, that IS sad. I remember being in a Murfreesboro valley 20 years ago and it took me 5 minutes to find URSA Major. The skies were so dark that there were a gazillion stars.

In metro Atlanta, it's hardly worth getting a telescope out.

Scott Chapin
Powder Springs, GA, USA
Nikonians Team Member

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

gvk Silver Member Nikonian since 16th Feb 2006Thu 02-Apr-09 01:47 AM
257 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#47. "RE: A gedanken experiment on light meters...."
In response to Reply # 0


Mystic, US
          

Chapter 5 in _The Manual of Photography_ (Focal Press, 2000) entitled "The photometry of image formation" contains a discussion of how image illuminance is determined. Of course, the "bible," _Principles of Optics_ by Born and Wolf (Pergamon Press, 1975), also has a brief discussion of photometry in section 4.8.

Gerry

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

    
nl Basic MemberThu 02-Apr-09 01:52 AM
833 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#48. "RE: A gedanken experiment on light meters...."
In response to Reply # 47


West Hartford, US
          

Gerry:

I actually read that chapter last night. I did not find an answer to our question there. Did you find something that I missed?

nl

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

        
gvk Silver Member Nikonian since 16th Feb 2006Sun 05-Apr-09 05:28 AM
257 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#53. "RE: A gedanken experiment on light meters...."
In response to Reply # 48


Mystic, US
          

Sorry for the late response, but I have been traveling and have only had limited internet access.

There seem to be at least three questions discussed in this thread. The first question relating to the, apparently intentional, incorrect hypotheses in the original post that exposure should depend on camera to subject distance is, I believe, answered by the reference that I gave. In fact, it was also answered in the first couple of posts by Noel Holland without the mathematical details. The second question, whether the answer is different for a luminous rather than reflecting subject, is also implicitly answered, at least for diffuse surfaces.

The final question that seems to be under discussion is whether light fall off from extended sources follows the inverse square law. The answer is obviously no, in general, since counter examples have already been posed, at least for subjects near the source. However, such situations may be analyzed following Huygens, Fresnel, and their successors by summing individual contributions due to point sources representing small areas of the extended source. Each of these contributions thus obeys the inverse square law. Integrating contributions from these infinitesimal sources, also recognizing that the light ray path from source to subject may include nearby reflecting surfaces, can provide detailed answers for illumination from specific sources. Relying on an incident light meter is a much simpler way to get an answer relevant to the photographer. You are correct that extended sources of illumination are not covered directly in the references I supplied, although techniques for carrying out the integrals that I mentioned are discussed at length elsewhere in Born and Wolf.

Gerry

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

Len Shepherd Gold Member Nikonian since 09th Mar 2003Sun 05-Apr-09 12:16 PM
12722 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#54. "RE: I feel there are four aspects"
In response to Reply # 0


Yorkshire, GB
          

In this reply I am embracing comments for beginners as well as advanced photographers.
1/ When light leaves a photo flood or flash unit the angle of the light spreads out.
The further the subject is from the light source the lower the percentage of light received by the subject - which is where inverse square law normally comes in between flash and subject.
What inverse square law means is if you double flash distance from 6 feet to 12 feet the exposure is the equivalent of 4 apertures (not two) slower.
It is possible with some spot light units to prevent the light spreading normally, though rare to light a subject with a single narrow angle light source. With very narrow angle spot lights inverse square law does not necessarily apply.
Inverse square law does not apply to sunlight as the suns average distance from the earth is 150,000,000 and the chances of being 300,000,000 miles from the sun (when inverse square law would apply) are low
2/ Once light is reflected from a subject it is fixed in brightness and, provided the camera can frame the illuminated area, the light output from the subject (but see below) is constant and inverse square law does not apply.
This is why an incident light reading (normally calibrated to 12 or 14% reflectance Arther factory) is generally accurate if the subject has the same reflectance.
For beginners an incident meter reads the light falling on a subject. Camera meters are different reading light reflected from the subject.
3/ In a studio set up there is usually little ambient light (because normal room lights are likely to be turned off) to influence an in camera recording at different zoom settings.
If there is considerable ambient room light not recorded by the incident meter the incident reading may not be accurate.
Also the amount of ambient light seen by the camera lens can vary with zoom setting and focus distance.
4/ What can get overlooked is whilst lenses have a correct focal length and aperture at infinity focus many change effective focal length and effective aperture the closer you focus.
We are used to super fast AF, 1:1 macro, maybe a 12mm wide angle availability, or a compact 300mm lens. These features are generally only possible if effective focal length and effective aperture vary with focus distance.
What this means on none macro lenses is f8 at infinity can transmit a different amount of light to f8 at minimum focus - with differences of a full stop being common with variable aperture zooms.
This means an incident light meter reading might need a correction factor depending on the focus distance of the taking lens, and is probably why some of the test shots are showing moderately different exposure readings at different focus distances.
Switching lenses can also make a difference, sometimes as much as 7.5%, as generally lenses transmit between 87.5 and 94% of light reaching the front element - depending on the optical design.
***
Digressing to the Nikon macros f8 on the top screen at infinity can change to f16 just by changing focus to 1:1, partly because these lenses have big changes in effective focal length and effective aperture between infinity and 1:1 focus.

Photography is a bit like archery. A technically better camera, lens or arrow may not hit the target as often as it could if the photographer or archer does not practice enough.

Len Shepherd

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

    
ScottChapin Moderator Awarded for his high level skills in various areas, including Aviation and Birds Photography Charter MemberSun 05-Apr-09 06:22 PM
8047 posts Share on Facebook Share on Twitter Share on Linkedin    Click to send email to this author Click to send private message to this authorClick to view this author's profile
#55. "RE: I feel there are four aspects"
In response to Reply # 54


Powder Springs, US
          

Hi Len,

I guess the leap that I find hard to make is understanding why a 36 sq meter wall and a 9 sq meter section of the same wall would reflect the exact same quantity of light. Inverse square or not.

It seems that doubling the distance and focal length would cause the lense to see 1/4th as much light, even if the light were not decreasing with distance at all.

Scott Chapin
Powder Springs, GA, USA
Nikonians Team Member

  

Alert Printer-friendly copy | Reply | Reply with quote | Top

Forums Lobby MASTER YOUR TOOLS - Hardware & Software Nikon Speedlights & Lighting topic #40627 Previous topic | Next topic


Take the Nikonians Tour and learn more about being a Nikonian Wiki /FAQ /Help Listen to our MP3 photography radio channels Find anything on Nikon and imaging technology - fast!

Copyright © Nikonians 2000, 2014
All Rights Reserved

Nikonians®, NikoScope® and NikoniansAcademy™ are trademarks owned by Nikonians.org.
Nikon®, Nikonos® and Nikkor® are registered trademarks of Nikon Corporation.