f8lee Loc: New Mexico

@Sharpshooter makes the correct point that the physical length of a lens does not necessarily relate to the FOCAL length of the lens (just look at how long the lens barrel of some 35MM focal length prices are, or the 14-24MM zoom. It is also correct that the entrance element is not the critical measurement, but rather the diameter of the diaphragm is what is used in the calculation. The focal length of a lens is defined as the distance from the imaging plane (the CCD or film) to the point where the light rays converge ("criss-cross") - think of the simple graphic showing how a lens works where parallel rays of light enters the front and cross over each other to display the image upside-down on the film/chip - that crossover point is where the measurement for focal length is made. So let's put those other myths aside.

SS' second point about replacing the elements of the lens with better glass etc. has merit, but does not change the f-stop calculation as that is merely the ratio already discussed. However - and this is important to know - it can affect the amount of light transmitted through the lens (as every element of glass will absorb some small amount of light). In other words, two hypothetical 200MM focal length lenses, one designed with a single element and the other with 13 elements in 5 groups (or whatever) might both have a maximum aperture of f4, but the former will allow more photons of light to get through to the image plane. For this reason, there is a metric called the T-stop (transmission stop) - and THAT is the reading of the amount of light making it through a particular lens as measured with light-metering equipment. Cinematographers have long known about T-stops as they were using zoom lenses with many elements of glass in them long before the zooms became prevalent in still shooting. Today, if you peruse the pro video magazine you see ads for lenses that identify their T-stop rather than f-stop, as that is more pertinent to the actual use of the lens.

@romanticf16, as for how the constant-f-stop zoom lenses behave that way; it's because the diaphragm does change diameter as the lens is being zoomed.

gammaray

Greetings fellow hoggers:
When I was studying physics and specializing in optics we used certain tools to assist our understanding of optical elements, e.g. lenses, mirrors and pinholes, and image formation, distortion and quality.
One tool we used was an optical bench. This is like a lathe bed with fixtures that hold optical components in alignment concentrically to measure focal lengths and test performance, sometimes with a ground glass to observe focus.
Any element that brings an image into focus is measured with light rays that are travelling in parallel “bundles”, which is to say from infinity. With an optical bench parallel rays can be projected with a simple device called a collimator, so there is no need to focus on a distant object.
The focal ratio or “f number” of the lens is the clear diameter of the first element divided into the focal length (distance from the first element to the image). Therefor, a lens with a first element diameter of 44mm and a focal length of 80mm is an f/1.8.
With optical surfaces, transmission or reflection, smaller focal rations (faster optics) lead to more distortion...chromatic and angular. Setting aside “zoom” features, the optical elements behind the first element are to correct for some sort of distortion or aberration.
When you “stop down” a lens in any way - an iris diaphragm or a mask - you do certain things: you decrease the light gathering power, i.e. “slow it down”, you increase the depth of field, you decrease the uncorrected aberrations you might get with a wide open aperture.
You do not change the focal length. It is precisely the same as having a lens with a smaller clear diameter, but the same focal length.
Clearly, since a circular lens forms a circular image, and cameras record their image on a sensor or film with four right angles, a rectangle, you must either throw away some of the image formed by the optics, or throw away some of the sensor area - evidently convention forces us to want only images that are not circular - they fit in books better or something! But this complicates the question of whether your optical system is really performing at its full focal ratio, or you are losing sensor area.
With photography - everything is a compromise!

f8lee Loc: New Mexico

Again, f-stop is calculated as the ratio of the focal length of the lens divided by the diameter of the APERTURE diaphragm (not the front element)

Reinaldokool Loc: San Rafael, CA

The answer is simple in principle as Nikonian laid it out. However by using a variety of lens element designs and relationships, things get muddy pretty quickly. The formula works better in simple lenses and close to the answer at the shorter lenses. Really get fuzzy. If you would really like to know the definitive answers you will need to swim through the math in a lens design book.

f8lee Loc: New Mexico

This is precisely why the T-stop was invented - it is a measure of actual light transmission rather than a simple mathematical formula. So T-stop removes any "muddiness" in the conversation.

amehta Loc: Boston

I disagree with both these point.

gammaray

I am interested - Please explain

rocketride Loc: Upstate NY

The larger glass area won't directly give you a better picture.

Generally "normal" lenses (approximately 50mm f.l. for 35mm format) or moderate wide-angles and telephotos will have the smallest front elements and filter sizes. For lenses of the same f-ratio but different focal lengths, the front element size (and the necessary filter size) will differ because of two factors.

The first one is that as the focal length increases from the "normal" lens, the front glass will get bigger, as does the entrance pupil (which is the stop as imaged by all of the optics in front of it), just to maintain the geometric ratio between the aperture and the focal length that defines the f-ratio. And the lens' angle of view for a given sensor size decreases with increased focal length.

Where a 50mm lens (in 35mm film or full frame digital format) has to accept light coming in from about 24 degrees off-axis to illuminate the corners of the frame, a 200mm lens only has to accept light coming in from about about a quarter of that angle. If you look in the front of the lens from a distance, you will see big glass, with a big apreture stop "hole" behind it. (And if you look from not very close to straight ahead, you won't even see the hole.) For a long telephoto, the filter won't be that much bigger than the front element.

On the other end, as we go shorter, the size of the aperture for a given f# decreases proportionally. The front elements, however, also eventually start getting bigger as the lenses get shorter and "wider" because the lenses have to accept light from progressively farther off-axis. A 20mm lens for the same format has to accept light coming in from around 50 degrees off-axis. If you look in the front of the lens from a distance, you will see big glass, with a little hole behind it. And as you look in from progressively farther off axis, you will keep seeing the little aperture stop "hole". It may even look like it is turning towards you behind the front glass. For an extreme wide-angle, the filter will be a lot bigger than the front element because of those very far off-axis rays. And even bigger if the front element is steeply curved.

So, for a given lens, the size of the each element is dictated by the focal length and f-ratio of the lens and the angular coverage it is expected to provide.

For a zoom lens, this determination has to be made over the entire focal length range. Big parts of some lenses aren't doing anything at some focal lengths.

f8lee Loc: New Mexico

Disagree all you like, but:

The focal length of a lens is defined as the distance in mm from the optical center of the lens to the focal point, which is located on the sensor or film if the subject (at infinity) is "in focus".

http://www.dpreview.com/glossary/optical/focal-length




and:
The ratio is between the diameter of the aperture in the lens and the focal length of the lens.

http://www.uscoles.com/fstop.htm

amehta Loc: Boston

Those are two different statements. The second is what I almost agree with, though it is an approximation for the case where the subject distance is at infinity. The "criss-cross" is not a factor.

SharpShooter Loc: NorCal

f8, I did not read your link, I'll do that after work.
But let's not forget that we are not talking about Galalleo's lenses here, which were simple long lenses, with simple formulas to go with them.
One look at the Sigma Art 50mm, because of the design, is much longer than a Canon 50mm.
And a telephoto lens is so named because it has a telephoto element group that allows it to be much shorter than its actual length designation.
Modern lens designated lengths have little to do with actual light path distance as did simple lenses of yesteryear.
Lens designation is assigned by a standard that's made by comparing a lenses field of view to that of a lens used by the ISO as a standard, or very close to it.
So if a lens is two inches long or ten feet long, but both have the field of view of a 500mm standard, then they are both 500mm lenses, irregardless of their actual physical length, as designated by a formula.
I'm sure there are many formulas, and all are correct and accurate for a particular type of lens, whether the lens is a simple long lens, a telephoto, a D.O. or a mirror type lens. The patch that the light waves take, is not the focul length.
So for focul length there are for sure several right answers.

Now on f-stop, my suggestion from earlier, was to possibly take into account similar advancements in modern design.
But on f-stop I simply at this point and time just don't know if there are multiple variables. Yes, we all know the basic formula. But does it apply to all modern lens degins?
So far I need to rely on the explanation from those that are way smarter than I am.
But hey, I'm a whirlwind with my camera set to the little green square!! :lol:
SS

amehta Loc: Boston

I think he is wrong also. The f-stop is about the light entering the lens, and the objective lens is the main factor in the maximum f-stop. The diaphragm is placed in the optical path in a way that it can reduce the light in a "clean" way when stopped down. That is not the same as the size of the opening used to calculate the f-stop.

f8lee Loc: New Mexico

Golly, this is getting silly. The "optical center" of the lens is the fancy term for the "criss-cross point" I mentioned in my earlier post. This point does not need to be within the confines of the physical lens; it can be in front of it (as it can be for some telephotos) or behind the rear element (as it is for any wide angle lens with a focal length of less than the distance the mirror sticks out in front of the sensor - for instance, an 8 MM lens where the mirror swings out 10MM - thus the need for so-called retrofocus designs). It is the distance from that optical center back to the sensor itself (film or electronic) that defines focal length.

And T-stop is an actual mechanical measurement of the light transmitted through the actual lens, as opposed to a mathematical formula, because, as you alluded, different glasses and different lens designs will indeed cause more or less light to actually make it through the lens for a given focal length and aperture size. It's been used in the movie and video worlds for decades, because that's where huge zoom range lenses with way more elements and groups were being commonly made long before zooms became prevalent in still imagery.

I really don't know how else to describe this stuff.

SharpShooter Loc: NorCal

F8, I see what you're saying. And yes, retrofocul was what I eluded to with the Sigma example.

It actually is so much simpler to put my camera on the little green square, and let it do it's thing!

Thanks for that explanation. As you can see, I know just enough to get myself into serious trouble, both with and without a camera!! ;-)
SS
SS

Nikonian72 Loc: Chico CA

So far, you have disagreed with every reference cited by three different posters, yet have not posted a reference supporting your contention that the front element diameter is used to determine aperture number (f/stop). Please cite legitimate references.

Query: If the front element diameter, which does not change, is used to calculate maximum f/stop of a lens, what measurements are used to calculate ALL OTHER f/stop (aperture) settings? ONLY the diaphragm iris diameter changes between different f/stop settings.