• Please use real names.

    Greetings to all who have registered to OPF and those guests taking a look around. Please use real names. Registrations with fictitious names will not be processed. REAL NAMES ONLY will be processed

    Firstname Lastname

    Register

    We are a courteous and supportive community. No need to hide behind an alia. If you have a genuine need for privacy/secrecy then let me know!
  • Welcome to the new site. Here's a thread about the update where you can post your feedback, ask questions or spot those nasty bugs!

Some things about "expsure metering"

Doug Kerr

Well-known member
A question asked in another thread led me to make some observations about "exposure metering" (or to be more apt, "automatic exposure control").

So, that part of my brain having been awakened from a comfortable sleep, I thought I would review some matters that might be of interest in that area.

I will start with the ISO standard for automatic exposure control systems. This contemplates a "simplistic" system, one that operates upon the measured average luminance of the scene (just as basic free-standing exposure meters do).

In such a system, suppose we set the exposure index (an input to the exposure calculation process, which I describe as "what we tell the exposure control system is the ISO Speed of the film or digital sensor) to truly reflect the ISO Speed of the film or digital sensor.

If we do that, we find that with the camera regarding any scene, for the shutter speed and aperture that the automatic exposure control system sets for us, for any given exposure index setting, the average photometric exposure on the film or digital sensor will be a fixed value .

Photometric exposure is essentially the product of the illuminance on some spot on the film or digital sensor and the time for which it persists the shutter time). It is basically the physical phenomenon to which the film or digital sensor responds.​

From here on I will only consider digital cameras, as some of the details to follow only pertain as described to digital sensors, not film.

Now, the rest of the story assumes that, in the scene being regarded by the camera, the ratio of the maximum luminance to the average luminance is about 5.6. Said another way, the assumption is that the average luminance is 18% of the maximum luminance. (Why? Because somebody once decided that this is "typical", and the operation of the automatic exposure control system is predicated on that.)

Then, if that happens to be true, the maximum photometric exposure on the sensor will be 1/2 stop below the saturation level (the level above which the sensor cannot actually report changes in photometric exposure).

One way to look at all this is that 1/2 stop "gap" is a margin against possible overexposure (where the photometric exposure on some part of the scene is above the saturation level). This margin is sometimes, by parallel with the similar concept in sound recoding, the "headroom" of the metering system.

But this headroom only obtains if in fact the ratio of the maximum luminance within the scene to the average luminance of the scene is the assumed 5.6. If that ratio is greater than 5.6, then the "headroom" is eroded, and if the ratio gets as high as 7.9, we are on the verge of overexposure of the highlights.

Of course, our real concern here is with modern automatic exposure control systems. They rarely follow the concept of "simplistic" systems which work on the measured average luminance of the scene. Rather, they will use some more specialized, and in many cases "intelligent", evaluation of the observed luminance across the scene.

But it is very common to bring the "calibration" of such a more sophisticated automatic exposure control system into conformity with the ISO standard for the "calibration" of simplistic automatic exposure control systems. By that I mean that typically, these systems are "calibrated" so that if the camera is regarding a test scene of uniform luminance (so the "cleverness" of the system gets a rest), the exposure that is set would be the same as prescribed by the ISO standard for (simplistic) automatic exposure control systems.

But the "cleverness" of these modern exposure control systems means that, for the most part, we need not allow any "headroom" to avoid overexposure. And if could do away with the headroom, we will generally be able to enjoy 1/2 stop more exposure, typically with an improvement in noise performance in the darker portions of the scene.

So camera manufacturers decided to give the exposure ordained by their automatic exposure control systems a "boost" of about 1/2 stop compared to that prescribed by the ISO standard. Now there are several ways this could be done:

A. Change the "calibration" of the system by 1/2 stop in the "more exposure" direction.

B. Set the exposure index (an input to the exposure calculation in the system) to 1/2 stop less than the properly-rated ISO speed of the sensor.

C. Rate the ISO speed of the sensor 1/2 stop less than the proper rating and set the exposure index to that.

Now the problem with scheme A is that if the user, on some occasion, uses a free-standing exposure meter, its exposure recommendations will be different from the exposure enacted by the integrated automatic exposure control system, so it will seem as if the integrated system is "wrong".

Scheme B has the very same problem (it is really just scheme A, looked at in a different way).

Scheme C has the problem that the user may reconstruct what the ISO speed of the sensor seems to be from the working of the exposure control system and will conclude that the value published by the camera manufacturer is "wrong".

So what was decided upon was to introduce a new metric of digital sensor sensitivity, the ISO Standard Output Sensitivity (ISO SOS). This is 1/2 stop less than the ISO Speed.

This is what is published for the camera (hopefully, with a note telling what metric it is). And it is this value that is automatically set as the exposure index into the automatic exposure control system.

And this is how we "eat" the headroom.

Best regards,

Doug
 

Jerome Marot

Well-known member
But the "cleverness" of these modern exposure control systems means that, for the most part, we need not allow any "headroom" to avoid overexposure. And if could do away with the headroom, we will generally be able to enjoy 1/2 stop more exposure, typically with an improvement in noise performance in the darker portions of the scene.

So camera manufacturers decided to give the exposure ordained by their automatic exposure control systems a "boost" of about 1/2 stop compared to that prescribed by the ISO standard.


Which I find to be a triumph of daft marketing over the real need of the photographer. In practice, blocked shadows are not a problem, because we expect them to be black. It is even a part of the standard tool kit to give the famed "film look". On the other hand, over exposed highlights, even slightly so, quickly look posterised.

Now I understand why, for some newer cameras, I need to adjust the meter down. Thank you.
 

Doug Kerr

Well-known member
Hi, Jerome,

Which I find to be a triumph of daft marketing over the real need of the photographer. In practice, blocked shadows are not a problem, because we expect them to be black. It is even a part of the standard tool kit to give the famed "film look". On the other hand, over exposed highlights, even slightly so, quickly look posterised.

Now I understand why, for some newer cameras, I need to adjust the meter down. Thank you.

Thank you.

Best regards,

Doug
 
Top