Metering Question

[Helen B]

The amount of reflected light which reaches your lens decreases the further away you are from the subject.

I think we have all agreed on that.

This can be easily demonstrated by fitting a manual zoom lens and setting your camera to A priority. Choose a subject, set your aperture and make a note of the calculated shutter speed. Now zoom in and out without changing the aperture and you'll notice that the shutter speeds will be recalculated by the camera's metering system. To keep the shutter speed the same as the original setting, you have to adjust the f-stop.

Have you read my explanation earlier in this thread about why you do not need to adjust the exposure settings at constant f-number? If you find that you have to adjust in practice, then something else is going on. The meter might be reading a different area, most likely.

 

[Helen B]

But does the distinction further the understanding of exposure and the F-number?

It should.

I'm unable to see how. It would be helpful to me (for future explanations) if you could describe the insights on exposure that are gained from knowing that the size of the physical hole in the lens may not change when manipulating the lens.

Here are my reasons for trying to use the concept of the entrance pupil, even though it is often tempting to use incorrect, but possibly initially simpler, explanations.

People often want to know the answers to more than one question. An oversimplified answer might save time with one question, but may then cause confusion when the next question comes up.

Why not try to do things properly from the outset?

The concept of the entrance pupil, once grasped, is very useful. It affects a few issues, such as stitched panoramas, the properties of variable aperture and constant aperture zooms, lens hood design, and why ND grads work (see the recent thread) that cannot be easily explained by thin lens theory.

If you tell someone during a discussion about exposure that the physical aperture changes as they zoom, and they actually look at a lens (who would think of trying it in practice instead of believing what they read on the internet?) while zooming and notice that it doesn't change over most of the zoom range, what credibility do you have?

Most of all: Do as you would be done by. How do you like to learn? Would you like someone to take the time and trouble to explain things properly, with an eye on giving the whole picture?

 

[Fred Berg]

@ Helen B: Yes, I read with interest your very nice explanation of the theories and my intention was not to contradict this, but rather to agree - albeit by coming from the other side of the problem. It can be difficult to grasp that when moving back (physically) from a subject and then zooming in so that what is seen in the viewfinder is the same image there should be no reason to change any settings (assuming no major changes in conditions) even though there is indeed now less reflected light reaching the lens. By doing a small exercise with your camera, that is choosing a subject with your lens at its widest setting (say 35mm) and then zooming to its narrowest setting (here 70mm) then if the f-stop remains the same and the camera is in A mode, you might reasonably expect the shutter to slow down by half (perhaps from 1/250 to 1/125). You can check this by zooming back to 35mm and then walking forward until what you see in the viewfinder is more or less what you had when zoomed to 70mm, and then taking a meter reading - it will normally be a slower shutter now.

This really is no different than starting the other way round and using the inverse square law to explain things; but it's more hands-on and perhaps easier to get to grips with.

It isn't 100%, though, as the example of the wall given by Graystar shows. I did this exercise this morning in my living room and twice there was the difference I expected and once there was no difference (also a wall with uniform lighting - perhaps the difference at the distances in the confined space were too small for the meter to pick up on). I always carry a hand-held meter in case I encounter difficult lighting situations.

I agree absolutely that there is no reason not to try and explain things as fully as possible, but also think that if someone can also give a practical way to put the theory into practice, then this is also often very helpful.

 

[Helen B]

By doing a small exercise with your camera, that is choosing a subject with your lens at its widest setting (say 35mm) and then zooming to its narrowest setting (here 70mm) then if the f-stop remains the same and the camera is in A mode, you might reasonably expect the shutter to slow down by half (perhaps from 1/250 to 1/125). You can check this by zooming back to 35mm and then walking forward until what you see in the viewfinder is more or less what you had when zoomed to 70mm, and then taking a meter reading - it will normally be a slower shutter now.

Fred, there's something odd happening here. In the case you describe the shutter speed should always be the same, if the aperture is the same (ie same f-number). I've given explanations with numeric examples earlier (for both moving away and zooming in and moving away and keeping the focal length the same), and it is what we find in practice. You can take a reflected light reading from close-up and use the indicated exposure at any distance and any focal length. How do you think that incident reading light meters work? If what you said was true (ie that distance and focal length affected exposure settings), you couldn't use an incident meter.

 

[Robin Usagani]

He is right only if you are in a smokey bar, foggy room, underwater etc.. where visibility drastically change over distance. .


Just to make sure I cover everything, he is also right if the shooter is wearing white and reflect some of the light to the subject LOL

 

[Helen B]

Yes, it's worth mentioning those situations. I think that we excluded the 'light source at the camera' earlier in the thread, and the question was about a subject in constant illumination, but no harm in a reminder.

 

[Robin Usagani]

Yes, it's worth mentioning those situations. I think that we excluded the 'light source at the camera' earlier in the thread, and the question was about a subject in constant illumination, but no harm in a reminder.

I know that. What I said is still valid. Constant illumination wont do very well if you go farther in a foggy day. Your exposure will be higher if you go closer even though the framing is identical. You have more fog between sensor and the subject when you are farther.

 

[Fred Berg]

Hmm, what I'm saying is the same as you, Helen....I think. Suppose I stand reasonably close to a vase in a room and take a reading with my camera of, say, 1/125 at f3.5 and a focal length of 35mm. Then I go further into the room and focus from the new distance on the same vase with the same f stop and the same focal length, now the shutter speed will be different (almost certainly faster, unless I'm in the underwater bar mentioned above). But the situation has changed because of the physical distance (and so I am picking up light from a larger source (the image in my viewfinder is larger) and might now have a shutter speed of 1/250), but if I zoom from my new position until I have reduced the image in my viewfinder until it is the same as I had close up to the vase at 35mm, and if for instance that means I now have a focal length of 70mm, then the shutter speed will/should return to the original 1/125 (or very near to this) because I have also reduced the light source to be metered. The settings remain the same because the ratios (image size to light available) are the same or similar whether I'm up close at 35mm or further away at 70mm, even though the real distance and actual amount of light have increased/decreased. This assumes TTL metering, of course. And, yes, unless this were so then incident metering wouldn't work. BTW, I tried the wall in my living room again this afternoon since the light was more subdued and I got a reading at 35mm of 1/125 f3.5 at distance which became 1/60 f3.5 zoomed to 70mm. I didn't check with my hand-held meter but I think this would have been very near the incident reading .

Or am I completely on the wrong road here?

 

[Helen B]

Fred, you are on the wrong path somewhere, and I wonder if it has to do with your lens' actual aperture vs marked aperture.

The following applies to evenly lit walls, bright objects in a dark room and most diffuse light sources.


The only thing it doesn't apply to is true point sources of light such as distant stars, or a few other very small light sources (there's a sensel size / AA filter effect dependency) and focused light sources, and specular reflections of a distant light source.


Assumptions: The ISO and lighting remain the same. Because the lighting remains the same, the brightness of the objects remains the same. I want the brightness of the image of the object to remain the same as I move.




It's a rework of my earlier example.


Initial condition: Light meter reading 1/125 at f/3.5, 35 mm lens. Let's say 1 m object distance.
Second condition; same focal length and aperture, 4 m object distance.


As you move from 1 m to 4 m, with the same focal length of lens, the area of the image of any small element of the object becomes 1/16 of what it was.


The light power (lumens) coming from that area element that passes through the lens' fixed aperture is also 1/16 of what it was, by the inverse square law. Therefore the illuminance (arriving light power divided by area) of the film or sensor by the object element is the same (lumens per square metre). Therefore for that image element to get the same amount of energy per area (lumen seconds per square metre) the shutter speed has to stay the same.

This is borne out in experience and practice. I take a meter reading and I do not have to adjust it for distance.

I've already given a fairly detailed version of the situation where the focal length changes, so no need to repeat it. There is no reason why you should need to change settings when going from 35 mm to 70 mm, so something else is happening when you try it as I said before. Is your lens a true constant aperture lens? When I try it with my good old f/4 35-70 there is no need to change the shutter speed at constant aperture as I zoom in. If I take a spot meter reading (hand-held spot meter) I do not need to adjust it for lens focal length. Never have done. How would you explain that if you think that the exposure changes as you zoom in at constant f-number?


Schwettylens, I wasn't disagreeing with you, but the exposure for the whole smoky / foggy / hazy thing is more complicated, the obscuring medium is usually illuminated itself, so it doesn't simply darken the distant object, it adds overall light in the path between the object and the camera. My remark about constant illumination was with respect to your mentioning a photographer wearing a white shirt that is reflecting light onto the subject. In that case the illumination is not constant as the photographer moves.

 

[Tiberius47]

Okay, I've actually gone and done this.

I set up a softbox with a 430EX firing through it.

I used a 7D with a variety of lenses, both the Tamron 17-50 f2.8 and the Tamron 70-200 f2.8.

The entire set up was on manual. The 430 was firing through the softbox at 1/128 power, and the camera was set at ISO 100, 1/250 and f5.6. This left the softbox under exposed by about two stops. This was done intentionally to leave the softbox as a grey instead of a white, to make it easier to see any changes in brightness.

I took four photographs, with the exposure staying constant for each one. Each photograph was taken at a different distance, dictated by the field of view of each lens at its longest and its widest. I moved between shots so that the softbox's height just about filled the frame for each shot.

The focal length and distance are marked on each shot below:

As you can see, despite hte fact that the 200mm shot was taken about 14 times further away than the 17mm shot, there was no significant change in exposure.

Apart from resizing and putting all of them into one image, no photoshop work was done.

 

[chuckdee]

I never use my camera's metering.....

 

[Spiffyinferno]

After reading the full thread, I'm curious-- can something like atmospheric/aerial perspective impact metering even on smaller distances like this? Obviously we wouldn't be seeing something noticeable to the human eye, but assuming the subject's illumination is truly unchanging, I would think there would be some impact with the atmospheric perspective factor. Is this what you are talking about regarding the intensity of the light changing? Perhaps I need to read up on the inverse square law and some of the other concepts mentioned here before I ask questions like this.. Thanks for the interesting thread though!

 

[chuckdee]

Do some test shots outside of your house and see what you come up with