[MarkJerling]

not fudging aperture numbers, but fidging focal length. f2.5 at 52mm gives a focal length of 130mm. We've had models reject the photos because the lens was not a true 135mm lens

The plot thickens!

[pschlute]

The premise of the comment in the OP was simply that a 135/f2.5 lens with filter thread of 52mm couldn’t physically fit it’s theoretical maximum aperture (54mm)

The focal length of a lens is not simply about using a tape measure on it. It is an optical measurement which may or may not correspond to a physical equivalent. f-stop is a ratio which refers to the FL. If you require exactness beyond what the engineers of prior decades or current ones provide then I say good luck.

[stevebrot]

The premise of the comment in the OP was simply that a 135/f2.5 lens with filter thread of 52mm couldn’t physically fit it’s theoretical maximum aperture (54mm).

The assumption being that the front element is forward to the lens principal planes and entry pupil, though with telephoto design, this is purposefully not the case.

* Strangely, entry pupil diameter is independent of the front element diameter.

By back calculation, the entrance pupil size of a 135mm f/2.8 lens should be a little over 48mm. I have a 135mm f/2.8 on the counter in front of me with a 45mm front element and another with a 46mm. The magic comes courtesy of telephoto design.

Addendum: I checked my 200/4.0 and 200/3.5 and found similar.


Steve

[nickthetasmaniac]

By back calculation, the entrance pupil size of a 135mm f/2.8 lens should be a little over 48mm. I have a 135mm f/2.8 on the counter in front of me with a 45mm front element and another with a 46mm. The magic comes courtesy of telephoto design.


Steve

Out of interest, which lenses are you talking about?

[stevebrot]

Out of interest, which lenses are you talking about?

Komura 135/2.8 pre-set in Exakta mount and Vivitar (Komine) 135/2.8 in M42. The Komura is very petite in general.


Steve

[Lowell Goudge]

That’s why you call the lens a 200 and not a 200mm. Today the tolerances are tighter. 1% Offset I think. Aperture has different offset. Filter diameter is always wider than lens diameter, so differences are even worse.

No I was careful in my statements that I measured the actual diameter of the lens itself, as opposed to using the nominal filter diameter.

---------- Post added 02-10-20 at 06:30 PM ----------

Snip

* Strangely, entry pupil diameter is independent of the front element diameter.

I think this is true only i one sense, the front element diameter can be larger than the entry pupil diameter, and this is often the case on wide angles lenses, but it can’t be physically smaller as far as I know

Edit note:

I should probably reword my last comment slightly to something along the lines of the ratio of front element to entry pupil diameter with respect to the magnification of the front element / group. But at the end of the day, You cannot take in more light than the diameter of the front element.

[Ontarian50]

That’s why you call the lens a 200 and not a 200mm. Today the tolerances are tighter. 1% Offset I think. Aperture has different offset. Filter diameter is always wider than lens diameter, so differences are even worse.

I think some would argue that tolerances were tighter back in the day. Under the Japan Camera Inspection Institute, there were tight tolerances applied to products for export. This was done to make sure Japan's reputation for making cheap tin toys and cheap radios in the 40's and 50's was transformed into a reputation for precision and accuracy. A lens had to be within a few percent of its marked focal length to meet approval. Same for apertures, and shutter speeds, etc. etc.

The JCII stickers disappeared around 1990 (anyone know the exact year?). This was era of the early wonderzooms, and lenses with internal focusing, which changed their actual focal length depending on focusing distance. While a traditional unit focusing lens extends its focal length as you focus closer, internal focus and front focus zooms can do weird things - but they usually lose focal length. So your current 18-270mm is likely a pretty accurate 270mm when focused at infinity, up close it may be lucky to break 200mm. To the owner and user of such a lens, that doesn't likely matter much, but it would have driven the JCII mad.

But technically, yes, things could be "fudged" back in the day, or today. For marketing reasons, a lens measuring closer to a 1.8 aperture could be sold as a 1.7, and still be within JCII tolerance.

[bertwert]

The JCII stickers disappeared around 1990 (anyone know the exact year?).

Looks like '89.

when did Pentax stop putting a " passed " sticker on lens and camera bodies - PentaxForums.com

[Lowell Goudge]

Did a search and found this on lens tests

Popular Photography - Google Books

Read the beginning of the article, “one popular 50/1.4 tested at 53 mm”

I also found depending on the year modern photograph used 5% on focal length and aperture , and at other times had tolerances on focal length that were tightest around 50 mm and the farther away you got on either side from 50 mm the tolerances got wider any where up to 4%.

So in reality it is take it or leave it.

[GUB]

And I did a quick measure up over here--
The poor old much maligned Takumar Bayonet group. - PentaxForums.com
(Post 13)
At 6 metres the Takumar Bayonet 135 2.5 was 94% of the focal length of the M135 3.5 ie 128mm

[stevebrot]

And I did a quick measure up over here--
The poor old much maligned Takumar Bayonet group. - PentaxForums.com
(Post 13)
At 6 metres the Takumar Bayonet 135 2.5 was 94% of the focal length of the M135 3.5 ie 128mm

This one? The poor old much maligned Takumar Bayonet group: #13 - PentaxForums.com

In all due respect, I don't see frames for the M 135 where the plane of focus is the same as for the TB 135. (The M 135 appears to be back-focused from the face of the center lens cap.) It is probably best to do this sort of comparison at a greater distance, say 100 meters, since focal length is defined when focused to infinity and the effective focal length would have been a fair bit longer anyway at 6m.

It may be that the TB 135 was shorter and that it is actually an f/2.8 at a shorter focal length, but that would be hard to determine without measuring the entry pupil.


Steve

[GUB]

This one? The poor old much maligned Takumar Bayonet group: #13 - PentaxForums.com

In all due respect, I don't see frames for the M 135 where the plane of focus is the same as for the TB 135. (The M 135 appears to be back-focused from the face of the center lens cap.) It is probably best to do this sort of comparison at a greater distance, say 100 meters, since focal length is defined when focused to infinity and the effective focal length would have been a fair bit longer anyway at 6m.

It may be that the TB 135 was shorter and that it is actually an f/2.8 at a shorter focal length, but that would be hard to determine without measuring the entry pupil.


Steve

Yes the M135 wide open is slightly back focused but going by the cracks in the timber only by an inch or two. (The other two caps are about 500mm forward and back of the centre one). May be enough oof to upset the result but certainly not enough to change the focal length calculation. Note that the Tak m42s share the fov of the M135.
In other comparisons with my M135 it has fallen short of the Taks in sharpness. Mine might be a poor example.

Testing old glass out at infinity is not a good option as many have slipped out of adjustment (or never were in) and won't reach infinity without adjustment.

[Astro-Baby]

Focal ratio as already stated is the ratio between a lens units barrel length and its widest possible aperture. I am more at home with telescope optics than camera lenses but the principles are very similar.

The difference as I would see it with a camera lens is that as you focus the lens the lens is in effect changing length so the focal ration will change slightly. I have a lens sat here whose max length will shift by 15mm between closest focuser and infinity. That would account for about .25 of a stop. So at one end of focus it would be bashing 3.5 and at the other end it would be at 3.26. Do I care when I am looking through the lens and watching my meter ?

Further there may be factors in the lens design which limit the effective diameter of the primary lens (the big one at the front). This could be for instance a baffle plate to limit internal reflections or on a wide angle lens a very much smaller correcting lens at the rear that might introduce a clipping effect These additional lenses could create a clipping effect or the clipping may be a natural by product of the lens design.

For instance you could have a 3" lens front but the light path could be clipped off by the way the internal lenses affect the overall path of the light in the lens.

As I say I am more familiar with telescope optics which tend to have fewer internal elements but are not dissimilar in many ways to a telephoto lens. The surest way I would think to assess the lens actual effective diameter would be a similar test for a Maksutov scope which can suffer light path clipping when the primary lens is the same size as the corrector plate.
The test is this. Place the lens flat on a surface and hold a piece of white card to the rear of the lens a few inches away would be about right. Now illuminate the front of the lens using a powerful light. A strong red torch would be good. Measure the diameter of the light cone coming from the back of the lens onto the white card. That will give you rough idea as to how close the lens maximum aperture actually is. It will probably be slightly smaller than the measured diameter of the primary lens if clipping or aperture fudge is apparent.

That would give you a good idea as to whether the front face of the lens suffers clipping from a too small filter ring and give you the true state of the lens aperture.

My bet is the difference would be tiny and given the latitude a camera lens would have to work with (i.e. its length is changing as you focus) the small difference would be pointless to worry about and really more a theoretical issue than any practical issue thats going to reduce the resolving power of the lens.

From a practical standpoint these discussions are fun but they will not substantively affect performance. All lenses have to make a compromise because no one wants to shoot with a lens that operates at off f stops. The lens could be engineered to be precisely f2.73 but for practical purposes f2.8 is close enough.

Ditto with shutter speeds - when you have a continuously variable shutter the camera may be telling you its firing at 500th but it may well be shooting at 523rd but none of us are bothered by that kind off detail in real life and even if the camera was nice enough to say 'hey bud - this ones at 534th of a second' you aren't likely too be able to repeat it so its pointless.

At the end of the day the cameras meter will be the ultimate arbiter as that will be assessing the light fall onto the film or sensor and thats the important bit.

[stevebrot]

The test is this. Place the lens flat on a surface and hold a piece of white card to the rear of the lens a few inches away would be about right. Now illuminate the front of the lens using a powerful light. A strong red torch would be good. Measure the diameter of the light cone coming from the back of the lens onto the white card. That will give you rough idea as to how close the lens maximum aperture actually is. It will probably be slightly smaller than the measured diameter of the primary lens if clipping or aperture fudge is apparent.

What you will get is the (probably OOF) image circle of the lens at that flange distance.


Steve

[Astro-Baby]

You might well be right Steve, I havent bothered to test it, given that a camera lens will have corrector lenses to manage the image to ge it the right way up and to manage focus to a fixed film plane etc etc there will be considerations that would make testing a lot harder. You should get a result though if the card were at the approximate distance of the film plane. Its all kind of theoretical though because theres nothing you can do about it....it just is. If you had a lens stated as 300mm f4.5 and you went to all the hassle of testing and measuring and you found it was actually 298mm f4.6 its not as if you can do anything about it. Probably every camera lens ever made has these kinds of compromises.