Lens design "optimizations" -- real or myth?

[bensyverson]

You always hear about different lenses being designed or optimized for certain conditions. For example, Leica glass is supposedly designed to perform best at wide apertures and close focus (at the expense of infinity and small aperture performance?).

I understand that lens design, like anything, must involve a series of compromises... but I'm also slightly skeptical.

Any opinions?

[Huck Finn]

Check out the lens ratings at www.photodo.com. Link to the old site. Ignore the overall ratings & just compare the performance wide open. Leica performance speaks for itself. Nothing really comes close to the 50/2 Summicron-M except maybe the M-Hexanon 50/2, which is almost a clone of the Summicron. Clearly Konica wanted to match the Summicron. No one else is even close.

The photodo ratings aren't the be-all & end-all. There are otfactors that lens designers have to factor in. Some lenses are more flare prone than others. Some have better tonality. Etc. As you said in your post, there are compromises to be made.

[Gabriel M.A.]

Real. To the noncritical eye (aka the "who cares!" persona), a myth.

[bensyverson]

Originally Posted by Huck Finn

Check out the lens ratings at www.photodo.com.

MTF charts are definitely helpful, but as far as I know, they're all taken at infinity focus. In my ideal nerdcore world, I'd love to see three sets of MTF curves for each lens: closest focus, 3 meters, and infinity. It would be interesting to see which lenses are weighted to one side or the other...

[Bryce]

I just compared the Pentax SMC-F 50mm F/1.4 to the Leica equivalent there, and I don't see what the fuss is all about. Both are excellent, but the small difference made the Pentax a better lens anywhere but wide open....

[Nick R.]

I'd be more surprised if there wasn't a difference. For example check out this LINK or a good optics book and you'll see that there are many difficult problems involved when designing a lens. Correcting all would be incredibly expensive if even possible. Each manufacturer chooses which problems to concentrate on and how much to invest and pass on as higher costs. After a certain point the differences in lenses between quality manufacturers may be small, but it seems more unlikely that each manufacturer would hit upon the same lens solution and purchase the same raw materials.

[NickTrop]

Read a book on low-light photography and the author talked about how some manufacturers of less expensive off-brand glass would take an existing lens designed for f2.8 and "make it" go to 1.7, 1.8 to get that spec even though they were virtually unusable at that aperture. He didn't mention names, though.

Other lenses were designed from the get-go to be fast lenses with usable perfomace at wider apertures but have more elements and are more difficult to manufacture and cost more. If this is the same as being "optimized", seems this is more more a marketing term.

Film companies do the same thing all the time with their speed ratings. Sure you can shoot Ilford 3200 at 3200 but it's real rating is 800-1000 by all accounts. Not to knock that film, I like it a lot and always have a few rolls in 120 but it's another example of it being all about that eye-popping spec.

f2.8? ho hum
f1.4? Hmmmm, fast lens - and cheap!
ASA 640 - just another medium speed b/w film
ASA 3200? Whoa! That's a fast MF speed film, and it's available in 120!

All manufactures do this, especially on the lower end, when they're trying to hit a spec (performance be damned) so they can put that number on the box at a lower price point than the legit manufacturer.

Post deleted by posters request

[bensyverson]

So does anyone have any pictures from, or anecdotes about, lenses which perform better focused close? It sounds possible, but I haven't encountered any references online...

[Xmas]

The Mamiya 140mm macro lens has a moving internal group that you set dependent on the working distance.

The asp lux 50mm has an internal element that moves as you focus... lots of extreme performance lenses have moving internal elements to optimise performance.

Noel

[rxmd]

Hi Ben,

try, for example, any 50mm macro lens vis-a-vis a generic high-speed normal lens for the same focal length at close range, such in as reprophotography where even performance and good sharpness across a flat field is required.

Philipp

[bensyverson]

Interesting... I had never considered that macro lenses would be formulated to ensure
performance on flat planes, but that makes sense...

[rxmd]

The best lenses in that respect (flat field with uniform sharpness at close range) are usually enlarger lenses, for obvious reasons. A good enlarger lens on a bellows will beat any standard lens and many macro lenses in most aspects except convenience. That's why they are usually used for digital reproduction of slides with DSLRs, for example.

[bensyverson]

Wow, crazy. Why don't they build that performance into taking lenses? Because they'd be bigger and heavier?

[wyk_penguin]

Originally Posted by NickTrop

Read a book on low-light photography and the author talked about how some manufacturers of less expensive off-brand glass would take an existing lens designed for f2.8 and "make it" go to 1.7, 1.8 to get that spec even though they were virtually unusable at that aperture. He didn't mention names, though.

That is usually associated with manual focus SLR lenses where the etra stop would allow better focsing in dim environment.

Leica prefers to use lens elements with strange curvatures and refractive indexes to achieve large apertures without too much lens elements. (p.s. they usually went all the way to f/1.4 instead of some lowly f/1.8) With the advancement of computers and coatings, lenses are made with more and more elements. Not sure if they are better. The oldies suit me fine.

[dll927]

Herb Keppler's article in the latest Pop Photo compares the old Zeiss Biotar and a current Nikkor. Even shows resolution charts, which are darned near identical. And the Biotar was 50 years old.

Lens tests probably can't give every possible statistic. Pop Photo's usually seem to be done at closest focus, and in the case of zoom lenses, usually at two or three settings. The main comment about the aperture is usually in terms of light fall-off (edges and/or corners) from wide-open down 2-3 stops.

I tend to take the view that nobody can sell lenses that don't cut the mustard, so a lot of the hype about special glass, shapes of elements, etc. is exactly that - hype. Nowadays they are all designed by computer, so it's probably true that many current lenses couldn't have existed decades ago. But if they could design great lenses 50 years ago, present-day differences can only be a matter of degrees.

Probably the biggest difference is the existence of zoom lenses. Any photographer my age knows they didn't exist 50 years ago. The first zoom I ever had was a Vivitar from the early 1970's. It was a start, but nothing compared to what we have now. So if those special glasses and elements are an improvement, it may affect zooms more than 'prime' lenses.

[Doctor Zero]

I think there are two different issues. I am quite convinced that lenses are optimised to different purposes. Macro and non-macro is one example. There are several forms of aberration and correcting one can worsen another. The manufacturer decides which way to go. Hence also the different optical designs of lenses, the different bokeh etc.

Can you see this difference? For almost everything you can describe an S-curve of quality versus price. In the middle region, a small increase in price will get you a large increase in quality (however you define that, of course!). Most people would want to be here. At the high end, a small increase in price gets you a small increase in quality. I've always imagined Leica is at this top end. Does it make a difference compared to the top end Nikons and Canons? For at least some lenses I'm sure it does - but I doubt this difference is very obvious unless you know what to look for and print larger than 4x6 inch. Thus is it worth it? This is very subjective.

I'm as guilty as anyone. I enlarge very few of my photos (too embarrassed to put them up, I guess). Heck, I don't even take that many pictures: in the last 10 or so years, 400 or so rolls (I number my negative strips, hence I know). And yet - I want a Leica, and Leica lenses. In part because I notice a difference between my CV 21 and my dad's Biogon 21. But with my photography, my skills, my frequency, there is absolutely no need to spend that much money. But I will, because (in the near future!) I can. In my case, it's like the leather seats in a car, as someone mentioned in another thread. Necessary? No. Wanted? Personal decision.

And now, I'm off to the pub ! Drink anyone?

Doctor Zero

[ferider]

Originally Posted by bensyverson

So does anyone have any pictures from, or anecdotes about, lenses which perform better focused close? It sounds possible, but I haven't encountered any references online...

The classic Nikkor 50/1.4 is a great example. 1.4 is only usable up close, otherwise the lens vignettes pretty heavily.

Roland.

Post deleted by posters request

[schaubild]

Typically "digital" optimization with cheaper brands means reduction of inner reflections in lenses. Coating inner surfaces with anti refelective material costs money, and as these areas aren't directly visible it is a common way of saving money. As sensors reflect light more than film material, these shortcuts have to be eliminated. What you gain with it is more contrast. So it's not an optimization but just delivering something that works.
On the other end of the market, Alpa as an example, lenses are adjusted specifically to focus perfectly at infinity, the normal setting of these large format lenses is for studio distances. Digital optimization with these brands means specifical adjustment of the focus plane to include the additional sensor glass. Every lens is calibrated individually.

So there is some flexibility in using the term "optimization" :-)

[Finder]

Well, that is an interesting statement about Alpa. I asume you are talking about the medium-format company, not the older 35mm one. Since Alpa does not make its own optics, I doubt they do any more than buy the model that is sold by the manufacturer. To make changes to a production lens would be too costly.

As far as optimized object distances, why would you optimize something for infinity (accept for astrophotography)? Most photography is done at something less than infinity. I am not sure what "studio distance" means. I have used anything from about a meter to twenty meters in a studio. These distance including infinity are easy to manage in a lens design.

[yarinkel]

"Optimized for X" is a marketing word meaning "among all the compromises we had to make, we have cut less a bit less costs on the X aspect".

Everyone working in engineering will know from experience what I mean.

[Finder]

Lenses can be "optimized" for certain conditions/tasks and "looks." mostly the lens designer is looking to eliminate as many problems as he can within the limits set on the lens. Those limits could be economics, specific use, format, the lens type, or the "look" (usually a control of abberations and where the lens falls between resolving power and accutance). And usually it is a combination of those factors.

How much it matters to a photographer is a complex question. In optics, you usually get what you pay for. Can a difference be percieved between two lenses with similar specs and a big difference in price? Maybe. Maybe not. Are German Optics "better" than Japanese? Maybe. Maybe not. I find that when superlatives used to describe a particular lens or optical manufacturer, there is some emotional attachment involved. This leads to very subjective opinions even when cloaked in technical jargon. All lenses are a compromise. Get the best you can afford and enjoy them.

[visiondr]

Finder, you are absolutely correct.

Also, consider this when looking at a photograph: do you really notice the technical perfection of the image, or are you caught by the emotional? Certainly, a better lens is better to have, all things being equal. But, I am no HCB or Winogrand. Neither of these fellows used anything near what we call state of the art in optics. Then again, given the opportunity, would they have used the best of today? Of course they would.

[Pherdinand]

macro lenses are made to be very sharp at close focus.
Also, enlarger lenses are optimized for very close focus, magnifications over 1; while camera lenses are optimized for magnifications less than 1:10 or even 1:20 in many cases. E.g. a 50mm lens could easily be made to focus closer than the usual 80-90 cm (almost 1:16-1:18 magnification), but the image quality (sharpness, light falloff) would be pretty bad.

Lens design is a very complex thing for photographi images. Resolution, rectilinearity and color are all important, making life hard for the designers.