F/stops are only one of many possible scales that can be used to measure light, but the confusion results because they do not measure the specific quantity. They are just proportional representations that only have meaning because of about 100 years of industry coordination.BetterSense wrote:You are quite wrong on several points.
No, they are not. Fstops ARE a geometrical measurement that is based on the physics of the optical system. It's very simple...it's the focal length divided by the iris diameter. You make it sound like F/ratio was invented in the context of photography. ... Any two optical systems with the same f/ratio will have the same DOF characteristics; the amount of light coming through the lens may be different between them. If you really want to be precise in comparing the light transmission of different optical systems, you can use T/stops, which Hollywood invented because in their application, exposure was more critical than DOF.f/stop calculations are just arbitrary designations of the proportion of light that passes through the lens. It's merely an easy way of mathematically dividing up light in some meaningful way.
It's true that in theory the same f/stop should pass the same amount of light in every lens. But it never works that way, due to light loss and the fact that many lenses in the past were wierd designs that didn't pass light the same way. The problem for most users is that the f/stop scale works so well now because of standardization that it seems as if a reading of F/8 means that the meter has detected an F/8 amount of light, so you just set the lens to F/8 and it's all ready. But that's not what's happening at all, and this has led to much confusion, because when it doesn't add up, the question is, why not?
Most of the hard work rests upon the exposure meters themselves. Whether they derive their f/stop readings from circuitry or calculator dials, they only read "correctly" because of years of time-honored agreement about what works best with what films and standardized lens designs. I'm talking about the whole system, not just the lenses. F/stops mean nothing on their own, because they don't measure light at all. They just divide it up proportionally in ways that can be used consistently by the tools that do.
The stranger thing here is that even though f/stops are based upon a mathematical ratio, the lens barrel markings aren't always accurate. Either they have been approximated to whole numbers for ease (f/8 is actually f/7.7) or they have been tweaked by the lens maker to give results that are consistent with standard film stocks and exposure meters. Sometimes, in the case of the old Westons, the meters would be the ones to compensate for lens and film combinations, if you followed Weston's film-speed demands.
In many lenses, if you measure the actual ratio at each aperture it won't match the markings, and the question is, why is that?
So, my point is that it's all very hodge podge. In the past, some cameras with odd lens designs wouldn't even have barrel markings at all-- the Bolex 150 series is a great example. People complained when it came out that you didn't know the f/stop, but it would have been meaningless anyway, because the Bolex f/8 would not have corresponded to any other f/8 in terms of the quantity of light passed by its lens. And that's the part that counts.