[lytrytyr]

We can't have cheap, fast, small and sharp at the same time. Even if we forget about cheap, we still have one term too many.

On that subject . . .

You Tube

You Tube

[stevebrot]

The explanation is simple...

And difficult as it might be to believe, that is the simple answer!! Good work


Steve

[desertscape]

The explanation is simple.

You are basically saying the same thing I just said above. BTW, most LF lenses are Plasmat designs.

---------- Post added 12-10-2017 at 12:20 PM ----------

Even in a single element design, the ability to resolve directly corresponds to degree of magnification as well as the type of sensor receiving the projected light.

The reason why telescopes are so limited by magnification has to do with the fact that they use two optical systems (the telescope's optical design combined with an eyepiece) When an eyepiece is used that magnifies too much and causes unsharpness, it is due to the aberrations from the telescope being magnified. In the photographic world, it would be like using a 10X teleconverter on the back of a long lens.

[GUB]

The explanation is simple.
The speed of a lens is not given by the diameter of the front element nor by the diameter of the "hole" inside the objective (with the iris wide open).
It is given by the diameter of FL / "entrance pupil", both expressed in the same unit (usually millimeters).
The enlarging power of the elements in front of the diaphragm comes into play. That power of course can be negative (which is what happens with retrofocus designs, IIRC).
You can easily check with any lens fitted to a standard leaf shutter. All large format lenses set in a leaf (or central, if you prefer) shutter, and all the Schneider repro and enlarger lenses which use standard barrels are built the same way: there are two cells, one in front and one behind the iris/shutter, that can be easily unscrewed.
If you remove the front cell and measure the max aperture with a ruler, you have a number.
If you screw the optical cell back, keep the diaphragm wide open, and place a ruler in front of the objective, you can look (one eye, in line with the optical axis) that the diameter of the "apparent aperture" (read, the entrance pupil) is often different from the measure you've taken before. Most of the times the difference is minimal, because most LF and enlarger lenses are almost symmetrical double gauss.
Other designs, like the wide angles for SLR cameras, would show substantially different measures.
So yes, the design of the lens makes a difference

Apart from the basic laws of optics, there is one thing that was true 160 years ago, when my oldest lens was manufactured, and is still true today: there is no perfect lens. The blanket is always too short, if you cover your head, your feet will get cold.
We can't have cheap, fast, small and sharp at the same time. Even if we forget about cheap, we still have one term too many.
Over the years the compromises have gotten better and better. They are still compromises though, the name says it all...

Cheers

Paolo

Can you clarify one thing? From common sense I have always presumed the front glass had to measure at least the Focal length divided by fastest aperture. Does that still hold true with what you say?
I agree that often it is larger presumably from design. I was experimenting with a Pentax 110 70mm and found I could put at least one stop of disc onto the surface of the front element and it did not alter the speed of the lens so I suspect it's front element was partly cosmetic.

[leekil]

I agree that often it is larger presumably from design. I was experimenting with a Pentax 110 70mm and found I could put at least one stop of disc onto the surface of the front element and it did not alter the speed of the lens so I suspect it's front element was partly cosmetic.

Did it alter the image quality or vignetting? I wouldn't suspect it would be cosmetic; that's a pretty big chunk of glass.

[GUB]

Did it alter the image quality or vignetting? I wouldn't suspect it would be cosmetic; that's a pretty big chunk of glass.

No noticeable effect. It was a while ago and I think i put over a stop of disc on it. I had wondered, with the 110 being a bit of a niche market, whether they had cannibalised another lens design for it.
EDIT
found my experiment with it (Post 993)
https://www.pentaxforums.com/forums/136-pentax-q/209474-adapted-lenses-teste...ml#post3248962

[cyberjunkie]

Can you clarify one thing? From common sense I have always presumed the front glass had to measure at least the Focal length divided by fastest aperture. Does that still hold true with what you say?

Focal length divided by the aperture gives you the speed of a lens, expressed in f/ number. The diameter of the front element is another thing, and of course it makes no sense to have lens barrels shaped like... barrels That is, with the stop wider than the front glass.
Usually the front element is somewhat larger, as can be easily seen in many lenses. Some projector optics have no stop, what in layman's term we can call the "center" of the optic is as large as the front. Others have a fixed step, a disc if you prefer, that cuts out peripheral rays.
You experiment with a 110 lens doesn't show that the lens is recycled from a larger format, and of course there were no "cosmetic" reasons to to make the lens this way.
The size, curvature, refractive index of the front element are due to the characteristics the designer wanted to get, like speed, design typology and field of view.
The internal iris, wide open, is smaller because of one main reasons. It's not coverage though. It's aberrations.
Light rays coming from the periphery of the lens have to travel a longer path (go through more glass) cause they come from a wider angle, so they don't focus on the sensor plane the same way as light rays at 90 degrees (or close to it) do.
The two main problems are chromatic and spherical aberrations.
The former was (for the most part) solved long time ago with the introduction of the so called achromatic lenses.
Spherical aberration is compensated, more or less, by the optical "tricks".
The simplest is to reduce the influence of peripheral rays with a stop, fixed or adjustable (diaphragm).
I'm sure that placing a washer in front of the lens, as you have done, made the image a little sharper.
I don't remember much of optics, so I might be wrong, as far as I remember doing something like that, in presence of an internal step, should affect aberrations (sharpness) and only marginally speed.
Please correct me if I'm wrong

Cheers

Paolo

[GUB]

The diameter of the front element is another thing, and of course it makes no sense to have lens barrels shaped like... barrels That is, with the stop wider than the front glass.

Yeah that is another way of what I mean the front element has to be at least FL/f. Regarding that I have always thought the Niknon 70 - 200 2.8 was cutting it fine - a 77mm filter thread and requiring 71.4mm to do 200mm/f2.8. Leaves 5.5mm (2.75mm) for edging.

[UncleVanya]

Yeah that is another way of what I mean the front element has to be at least FL/f. Regarding that I have always thought the Niknon 70 - 200 2.8 was cutting it fine - a 77mm filter thread and requiring 71.4mm to do 200mm/f2.8. Leaves 5.5mm (2.75mm) for edging.

There are a lot of 200mm f/2.8 lenses with the 77mm filter size. DA* 200, D FA 70-200, etc.

What am I misunderstanding about your comment?

[photoptimist]

There are a lot of 200mm f/2.8 lenses with the 77mm filter size. DA* 200, D FA 70-200, etc.

What am I misunderstanding about your comment?

That, in theory, a 200 mm f/2.8 absolutely must have at least a 71.4 diameter front element which leaves little room for mounting structure and lens barrel. In reality, such a lens needs a larger front element to avoid vignetting.

There's also the issue that the lens might say "200 f/2.8" but actually be 195 mm f/2.9 (and needing only a 67 mm front element).

[UncleVanya]

That, in theory, a 200 mm f/2.8 absolutely must have at least a 71.4 diameter front element which leaves little room for mounting structure and lens barrel. In reality, such a lens needs a larger front element to avoid vignetting.

There's also the issue that the lens might say "200 f/2.8" but actually be 195 mm f/2.9 (and needing only a 67 mm front element).

Yes - I got that but you singled out Nikon which made me curious as it implied that was an outlier to me. However virtually every 200 f/2.8 seems to fit this model. Even the venerable Canon FD SSC f2.8 had a 77mm filter. And then there's the Canon USM II 200mm f/2.8 which has a 72mm filter thread. That seems like a likely candidate for slightly overstated specs.

What's more interesting to me are people who create hoods out of filter and step down rings and then refuse to see that they may have created a light restricting baffle in some cases and claim improvements that actually may be strictly due to more restrictive f/stops.

[GUB]

Yes - I got that but you singled out Nikon which made me curious as it implied that was an outlier to me. However virtually every 200 f/2.8 seems to fit this model. Even the venerable Canon FD SSC f2.8 had a 77mm filter. And then there's the Canon USM II 200mm f/2.8 which has a 72mm filter thread. That seems like a likely candidate for slightly overstated specs.

What's more interesting to me are people who create hoods out of filter and step down rings and then refuse to see that they may have created a light restricting baffle in some cases and claim improvements that actually may be strictly due to more restrictive f/stops.

Sorry but I had only noticed it on the Nikon - I think it was when I had noticed some discussion about an earlier variety being a bit shy of 200mm or something . Your other examples are equally or more valid. You may understand now why I asked that original question to cyberjunkies post querying whether the front element had to be at least FL/f . It is pretty clear now everyone accepts that.

[GUB]

I'm sure that placing a washer in front of the lens, as you have done, made the image a little sharper.

The point was with the 110mm 70mm was that reducing the front element a stop utilizing a washer made little or no difference to the sharpness nor the speed of the lens. In short the lens did not seem to be utilising the outer area of the front element. Of course this was on the Q - I think 110 film is considerably larger so maybe there could be some vignetting. But the washer was hard down on the glass so should be close to optically invisible shouldn't it?

[photoptimist]

The point was with the 110mm 70mm was that reducing the front element a stop utilizing a washer made little or no difference to the sharpness nor the speed of the lens. In short the lens did not seem to be utilising the outer area of the front element. Of course this was on the Q - I think 110 film is considerably larger so maybe there could be some vignetting. But the washer was hard down on the glass so should be close to optically invisible shouldn't it?

There's one sure way to tell if a part of a lens is being used: create a opaque disk the size of the front element and punch a small hole in the part you think is not being used. If the result is a totally black image, then that part of the lens is not being used by the image circle of the camera. If there's any image at all, then that's the contribution of that part of the lens to the image.

[GUB]

There's one sure way to tell if a part of a lens is being used: create a opaque disk the size of the front element and punch a small hole in the part you think is not being used. If the result is a totally black image, then that part of the lens is not being used by the image circle of the camera. If there's any image at all, then that's the contribution of that part of the lens to the image.

Yes I might have to fire up my Q batteries and try that.