About Ev

I thought it might be time again for the little lecture on what Ev is, and isn't.

APEX (the Additive System for Photographic Exposure) is a system that defines base 2 logarithmic values for a number of quantities of interest in photographic exposure. Its original intent was to make it easy for photographers to work the standard exposure equation without having to take their shoes off.

The emergence of exposure meters (with their little circular slide rules that worked the exposure equation) pretty much removed that motivation, and overall, APEX never came into wide use. Today, outside the actual practice of camera design, we rarely encounter any of the APEX values other than Ev: exposure value. We find it used a lot, sometimes correctly.

Ev (exposure value) describes, in base 2 logarithmic form, the quantity photographic exposure, which recognizes the joint effect on exposure of aperture (as an f-number) and exposure time (shutter speed).

An exposure of 1 sec at f/1 has an Ev of Ev 0. At 1/2 sec and f/1.0, we have Ev 1. At 1 sec. and f/8, we have Ev 6. Each "one stop" change in either factor increases or decreases the Ev by one unit (the larger numbers representing less exposure).

The APEX value describing object luminance (brightness) is Bv (brightness value). Bv 0 denotes a luminance of 3.4 candelas per square meter; Bv 1 denotes a luminance of twice that (6.8 cd/m^2). And so forth.

Sadly, there has come into widespread use a bastard convention in which object luminance is expressed as an "Ev" value. There is of course no need nor justification for this; there is a perfectly good base 2 logarithmic representation of object luminance, Bv.

The bastard convention works this way. When a luminance is said to be "Ev x", it means the luminance for which, with an exposure index of ISO 100, the recommended photographic exposure (per the "standard exposure equation") would be Ev x.

That means, if we are operating at ISO 400, then for a luminance of "Ev 10", we would use an exposure of Ev 12. Is that great or what!

I will use the symbol Ev for this bastard value. Then, it is always so that:

Ev=Bv+5, or

Bv=Ev-5

The numbers here are unitless; that "5" is not "5 Ev" or "Ev 5" or "5 Ev units" as we often hear; it is just 5.

I urge all to avoid any use of this bastard convention.

Another curiosity in this area is that the shutter speed priority and aperture priority metering modes on Canon cameras are labeled Tv and Av, respectively.

These are the APEX symbols for the quantities time value and aperture value, which are base 2 logarithmic expressions of shutter speed and aperture, respectively.

Indeed, in those modes in a Canon camera, shutter speed or aperture are set. But of course not as Tv or Av - rather, in the conventional form (in seconds and f-number). (If in Tv mode we actually set Tv, then for a speed of 1/16 sec we would set "4".)

Go figure!

Best regards,

Doug

An especially sad playing out of the inappropriate convention of using the APEX value Ev to indicate scene luminance (and in fact, ambient illuminance as well) recently came to my attention.

I had occasion to review the manual for the Minolta Flash Meter IV, a versatile exposure meter.

In its specifications are indicated the measuring ranges for both reflected light technique (where illuminance is measured) and incident light technique (where illuminance is measured) in terms of a range of Ev.

Ev is of course a base-2 logarithmic measure of photographic exposure: the joint effect of f-number and exposure time (shutter speed). An exposure of f/1.0 at 1 s (or any equivalent combination, such as f/2.0 at 4 s) is Ev 0, and it works by "stops" from there, smaller exposures having larger Ev numbers).

But under the convention I find so appalling, those specifications mean "the luminance [illuminance] for which this meter would, if set to ISO 100, recommend a photographic exposure that would be expressed as such an Ev". In other words, "Ev 5" would be that luminance for which this meter, in the reflected light mode, if set to ISO 100, would recommend a photographic exposure of f/2.0 at 1/16 s or any other equivalent.

But it gets worse. This meter will in fact report the recommended photographic exposure not just in f-number and shutter speed (where the photographer can set one to a value of his liking and the other will come out after the measurement) but also their recommended joint effect. This is the quantity properly spoken of (in a logarithmic scale) as Ev (exposure value).

But it is not labeled or spoken of as Ev - that designation has already been hijacked for a wholly different quantity. So the Ev indication is labeled "ExIN".

What does "ExIN" stand for? "Exposure Index".

Sadly, "exposure index" has a well recognized, and quite different, meaning in the field of exposure metering. It means, as I like to explain it, "what we tell the meter is the ISO sensitivity of the film or digital sensor". (The use of this term accommodates the fact that what we set the "ISO" dial on the meter to may in fact not be the ISO speed of the film or digital sensor, for any of several reasons.)

Now in the related product, the Minolta Auto Meter IV F, we have the usual abuse of Ev in the specifications to denote the limits of the range of luminance and illuminance over which the meter will make a proper measurement. But this meter will not deliver an exposure recommendation in terms if actual Ev (so it will not have to use some made-up name for that).

Best regards,

Doug

A fair question in the face of my recurrent grousing about the bastardization of Ev to be an indication of scene luminance (and, it turns out, of incident illuminance as well) is: "Well, Kerr, what would you have us do?"

The sand only runs down through the hourglass of time, and at this point it would be wholly unreasonable and impractical to suggest that the "proper" notation be adopted across the field.

But for the record, I'll state what that would be, not in the sense "what should we now do instead" but rather, "what should 'we' have done all along"?


Wow! Won't that be really complicated?

Not at all. The luminance which is (improperly) stated as Ev 10 is properly stated as Bv 5. It's just that simple.

Well, won't we no longer be able to recognize intuitively what a certain luminance (expressed in Bv) is?

For example, what might be the luminance (expressed as Bv) of a typical sunlit scene (perhaps one that matches the "sunny 16" model). A good question. You of course recognize intuitively what that luminance, to which we often refer indirectly, is expressed in "bogus Ev".

Well, here's a quiz.


The well-known "sunny 16" scene luminance is, in terms of bogus Ev, "Ev 10.6".

Yeah, you knew that.

And to realize that in terms of Bv it would be Bv 5.6 would require folk to relearn what they all knew already. Just like Coke in liters. And all numeric telephone numbers.

"I hate those all-numeric telephone numbers. I used to know where SPring 7 numbers were, but I have no idea where 777 numbers are".​

But in reality, I hardly expect Canon to start expressing the limit of operation of its AF systems in terms of Bv.

So all I ask at this point in time is that we recognize that "Ev" as used (bogusly) to describe scene luminance, and Ev used (legitimately) to describe photographic exposure (not so often seen), are two different things, improvidently joined by a shared designation.

And, when I mention photographic exposure in Ev in some discussion of metering equations or such, please don't get derailed because you thought Ev was a measure of scene luminance and so what could I possibly be speaking of.

As is my wont, I'll close with some more telephone anecdotes - metaphors, actually.

A good friend and neighbor: "I hate that we have to dial ten digits for all the telephone numbers in Dallas now. I used to be able to remember the seven-digit numbers, but not these."

Kerr: "So, what was my seven-digit number?"

"Well, I have no idea. I have it on a speed dial button".​

A few decades earlier:

Sam: "I hate that we have to dial seven digits for all the telephone numbers in Dallas now. I used to be able to remember the six digit numbers, but not these."

Pete: "So, what was my six-digit number?"

"Well, I have no idea. Maw knows".​

Best regards,

Doug

Hi Doug,

Actually I probably wouldn't bother to reply, if it weren't you: known for precise terminology
Now, what I wonder is, how come you're using "..exposure time (shutter speed).." as being the same thing?

Greetings,
Bogdan

Hi, Bogdan,

Bogdan Hrastnik said:

Actually I probably wouldn't bother to reply, if it weren't you: known for precise terminology
Now, what I wonder is, how come you're using "..exposure time (shutter speed).." as being the same thing?

Click to expand...

Actually, exposure time is the quantity of interest. That quantity is usually called, in actual photographic practice, "shutter speed", so I put that in parentheses so people don't think I'm speaking of some new quantity they hadn't heard of before!

Now, there are those who think that for an exposure time of 1/500 seconds, the shutter speed is "500". (After all, isn't speed the inverse of elapsed time?)

But in any normal exposure equations that ask for "shutter speed", one must put in "1/500 second" for the case I mentioned. (There are of course "calculators" in which one would enter "500" into the "shutter speed" box, but that's not what I am speaking of.)

This is of course why it is better not to ever say "shutter speed" in technical work.

Best regards,

Doug

What I was thinking of, was that "shutter speed" is actually "speed of the shutter" (or curtain, if you prefer) which has constant speed for given (slr) camera -independant of exposure time.
Am I complicating things too much?

Wish you the best,
Bogdan

Hi, Bgdan,

Bogdan Hrastnik said:

What I was thinking of, was that "shutter speed" is actually "speed of the shutter" (or curtain, if you prefer) which has constant speed for given (slr) camera -independant of exposure time.
Am I complicating things too much?

Click to expand...

Oh, you are speaking of the travel velocity of the shutter curtain, which of course can certainly be described as the "shutter speed". It is not quite truly constant; but typically its velocity profile is consistent regardless of the exposure time that has been set.

Or perhaps you are speaking of the time for the curtain to complete its travel across the frame, which could perhaps be spoken of as the "shutter speed" (although "speed" does not properly come in units of time).

However, there is no practice I am aware of to use "shutter speed" to mean either of these rather than to mean the exposure time provided by the shutter.

When I speak of the velocity of travel of the shutter curtain, or the time for complete travel of the curtain, I always use such a complete description to make clear what I am speaking of. "When all else fails, call it what it is."

Thanks for your thoughts on this.

Best regards,

Doug