Why Never Stop Down?

[MrFriendly]

Advertisement (gone after registration)

Excuse my ignorance, but I'm very new to Leica.

I've heard Peter Karbe talking about light flux and how we should never stop down if we don't need the depth of field.  He also says how if we stop down to, say, 5.6 only a very small percentage of the lens performance is used.  So why is that in the MTF charts the performance of the lens actually goes up when you stop down?  Or in other words, besides narrow depth of field, what else do we lose when we stop down?  It certainly isn't contract, so what is he talking about?

[Per P.]

Don't worry about it. Stop down if you need the depth of field for whatever you're photographing. But don't worry about needing to stop down if you don't need, or want, depth of field.

This "don't stop down" thing is a storm in a teacup from a statement by Peter Karbe in a specific context, for specific lenses, that might have lost a bit in translation to begin with, and since then has been misunderstood, misinterpreted, and taken out of that context. If Leica wanted to avoid you stopping down they'd produce a one f-stop lens.

[wda]

Welcome to the forum MrFriendly. I share the sentiments of post #2. I regard Peter Karbe's remarks as predominantly theoretical. The main aim is to use  Leica lenses, or any lenses, to get the results you envisage.

For much of my work, limited these days, I need good depth of sharpness at closer distances at up to 3 metres. Rarely do I need full wide open apertures at close distances. Leave theory in the classroom. Understand it, but work out you own solutions. 

Edited October 8, 2020 by wda

[evikne]

With my eyes I can only focus on one thing at a time. I like that this is reflected in my photos as well. I rarely see any reason to stop down to get more into focus. And it's great to have lenses that I can keep wide open without having to worry about the image quality.

[mmradman]

Typically lens are at its best performance when stopped down as it eliminates or minimizes any lens  anomalies.  Peter Karbe is saying  that current Leica lenses are already good at maximum aperture so closing down to eliminate anomalies is not required.  Close down only for increased DOF.

It is also interesting  to hear from P Karbe that difference in size between M and L lenses is due to need to create  lens which is easier and cheaper to manufacture while providing maximum image quality, hence L lenses are larger.  Apparently it is easier to assemble and achieve required tolerances in a larger lens barrel.  Larger lens barrel is also required to accommodate AF motors etc.

[jaeger]

• Good fast lens' reviews always read  "sharp on subject, creamy background when wide open"
• "Wanna Be" fast lens' reviews always read "wide open is too soft and bad CA but stopping down problems are gone" 
• Either lens takes sharp images when stop down to f/5.6. 
• Pinhole lenses have f/250+ cost only $50 also take sharp images. 

Do these 4 lines speak to you?  If not then I don't have a clue.

Edited October 8, 2020 by jaeger

[adan]

Advertisement (gone after registration)

Ther is a grain of core scientific truth in the idea that:

- the wider the aperture or opening of any system used for imaging with electromagnetic radiation (anything from gamma rays to radio waves, with human-visible light being a tiny band in the middle), the higher the resolution.

- The narrower the aperture, the more diffraction around the edges of the aperture degrades the resolution of details. What should show up as a sharp point is instead projected as a spread-out blur.

https://en.wikipedia.org/wiki/Diffraction

At extremes, this is why radio telescopes have contantly been pushed to be bigger and bigger dishes, or even multiple dishes spread out across the landscape (the Very Large Array in New Mexico being one). Or even a Very Long Baseline Array (in this era of computerized communication) where a radio telescope in Hawaii and one in Puerto Rico can be synchronized to produce one image from an "aperture" of thousands of km across.

And even optical telescopes pride themselves on, and are measured by, their diameter (rather than focal length) to indicate precision as well as light-gathering ability:

https://en.wikipedia.org/wiki/List_of_largest_optical_reflecting_telescopes

Aircraft or spacecraft doing radar imaging and mapping go even further, creating synthetic aperture images by taking one image of the surface, and then moving on to take another image of the same area from a point 10 to 5000 km away (imagine a Synthilux lens a mile in diameter ). Or even multiple combined images from intermediate points.

It is thus theoretically best to use the widest aperture possible. to minimize even the tiniest diffraction blur, if one wants one specific thing to be as sharp as possible.

However...

in photographic lenses with multiple elements with some spherical surfaces (even the Leica ASPHs are ~80% spherical surfaces), one gets aberrations at the widest aperture that are eliminated by stopping down. They are not "diffraction-limited" (they have other imperfections that may be "fuzzier" than diffraction blur, until one stops down a lot).

Karbe really means that the 21st-century Leica APO/ASPH lenses are significantly better at the widest aperture, such that it is not as necessary to "stop down two stops" (and risk diffraction blur) for most photographic uses.

My experience with the 75m APO-Summicron-M ASPH is that it is not as sharp at f/2.0 as it is stopped down a bit. At f/2.0 it shows a faint touch of fuzziness and blue/purple fringing of fine lines, that goes away by f/2.8.

It is not "perfect" at f/2.0, but it is extremly usable, and better than most other lenses of similar focal length and aperture. but if I want to copy someone's painting and capture every brush stroke everywhere across the image with maximum clarity - I stop down to 5.6 or so.

That is called "the real world."

[Ko.Fe.]

18 hours ago, MrFriendly said:

Excuse my ignorance, but I'm very new to Leica.

I've heard Peter Karbe talking about light flux and how we should never stop down if we don't need the depth of field.  He also says how if we stop down to, say, 5.6 only a very small percentage of the lens performance is used.  So why is that in the MTF charts the performance of the lens actually goes up when you stop down?  Or in other words, besides narrow depth of field, what else do we lose when we stop down?  It certainly isn't contract, so what is he talking about?

 

I did read original Puts' compendium and it was very obvious from the read (to me) what most of the Leica lenses he reviewed are better on f5.6. If not at optimum for performance/bokeh.

Also several Leitz lenses I used to own were focus shifting right after w/o and up to f4, while spot on w/o.

It could be also good sales pitch. If you paid for thousands dollars lens, you got to use it w/o.   

Personally, I like to use all lenses between f5.6 and f8. Optimum performance, smooth bokeh and enough DOF.

 

 

 

[pippy]

13 hours ago, Per P. said:

...This "don't stop down" thing is a storm in a teacup from a statement by Peter Karbe in a specific context, for specific lenses, that might have lost a bit in translation to begin with, and since then has been misunderstood, misinterpreted, and taken out of that context. If Leica wanted to avoid you stopping down they'd produce a one f-stop lens...

^This^...

1 hour ago, adan said:

...Karbe really means that the 21st-century Leica APO/ASPH lenses are significantly better at the widest aperture, such that it is not as necessary to "stop down two stops" (and risk diffraction blur) for most photographic uses...

...^and this^.

Adjusting aperture is used, primarily, to control D-o-F. If you need minimal D-o-F shoot wide open (if your lens / focussing ability can cope). If you require more D-o-F then stop-down. This part of the story really isn't rocket science but shooting at minimum apertures WILL lead to poorer IQ due to diffraction as explained by Andy in post #7. You will just have to experiment and see what is - and what is not - acceptable for your individual needs.

Philip.

Edited October 8, 2020 by pippy

[250swb]

13 hours ago, mmradman said:

 

It is also interesting  to hear from P Karbe that difference in size between M and L lenses is due to need to create  lens which is easier and cheaper to manufacture while providing maximum image quality, hence L lenses are larger.  Apparently it is easier to assemble and achieve required tolerances in a larger lens barrel.  Larger lens barrel is also required to accommodate AF motors etc.

Not really. SL lenses are telecentric so a larger diameter mount means a straighter light path to the sensor, and the photo-sites on the sensor are designed to accept this for improved performance. But that isn't the only thing. Increased performance in one area can be accentuated with software performance in another, so the lenses are now no longer purely optical in design, they rely on the firmware to improve distortions etc. So yes they are cheaper to produce, but you aren't getting unadulterated optical performance anymore.

[Guest]

I just think karbe was pointing out that a lot of effort has gone into making certain lenses outstandingly good wide open and that nobody should be afraid to use them as such.

 

[earleygallery]

It's a little like a chef saying that their dish is so good you don't need to add any seasoning or sauces/mustard.

[AndreasG]

What do we know about Peter Karbe as a photographer? Are there any photos available?

[mmradman]

9 hours ago, 250swb said:

Not really. SL lenses are telecentric so a larger diameter mount means a straighter light path to the sensor, and the photo-sites on the sensor are designed to accept this for improved performance. But that isn't the only thing. Increased performance in one area can be accentuated with software performance in another, so the lenses are now no longer purely optical in design, they rely on the firmware to improve distortions etc. So yes they are cheaper to produce, but you aren't getting unadulterated optical performance anymore.

Nothing wrong with what you said, i still recall that Peter Karbe stated that achieving tight tolerances in assembly is easier in a lens with bigger barrel than a smaller one like M series. 

From memory some M lenses are also identified as telecentric to a degree, WATE  (Tri Elmar M 16-18-21mm ) may be one of them and it is a real dwarf compared to anything in L line.

[logan2z]

4 hours ago, AndreasG said:

What do we know about Peter Karbe as a photographer? Are there any photos available?

My guess is he wouldn't be a particularly great photographer. Just like I wouldn't expect a top-notch aeronautical engineer to be a great fighter pilot or a top-notch automobile engine designer to win Le Mans.  Very different skill sets.  

Edited October 9, 2020 by logan2z

[mmradman]

4 hours ago, logan2z said:

My guess is he wouldn't be a particularly great photographer. Just like I wouldn't expect a top-notch aeronautical engineer to be a great fighter pilot or a top-notch automobile engine designer to win Le Mans.  Very different skill sets.  

Absolutely agree,  not sure what was the purpose of query as majority of LUF users could hardly call themselves great photographers (me included), even less so great lens designers.

[lct]

My experience with the M 50/2 apo is that it has more CA at f/2 than above, somewhat expectedly, so i won't certainly refrain from using any of my lenses stop down when i need so. Good thing about modern lenses they can be used at full aperture more easily than earlier ones though.

[oldwino]

Technical and use requirements aside (yes, sometimes you need that f0.95. But honestly, most of the time you just end up carrying a heavy chunk of glass around with you for those rare occasions), I find the “Always Wide Open” (AWO) thing to a bit of fetishism and a way to hide compositional failings behind a wall of “Extra Creamy Bokeh” (ECB). All these photos look the same, and they’re quite boring because of that. 

[Dennis]

Welcome to the forum.

On 10/7/2020 at 7:44 PM, MrFriendly said:

besides narrow depth of field, what else do we lose when we stop down?

I'm not loosing, I'm earning.
When I shoot and put the camera (with a 35) in the bag, I set the aperture @ f/5.6, always. It's my base aperture. And when I need more shallow DOP, I shoot wider open. Easy
But 5.6 is my start. Why? Mostly to add context to the photo, layers, information. More DOP, usually better IQ and fewer issues.

If you are new to Leica, you will find many interesting threads in this forum, it's an amazing source of passion, knowledge and entertainment. Seek, read, be inspired, but always use your own judgment to make choices.

 

 

[Advertisement]

Hallo,

Look here: Leica M

[tommonego@gmail.com]

I have never been a wide open photographer, I have photographed paintings big and small flat things but the images always looked better at f5.6 or f8. I have also had 2 lenses where f2 was useless, an f2 Nikon 24mm (mid 70's) and an f2 1959 Nikon 85mm LTM, both lenses were excellent f4 - f11. With the 24mm in an SLR configuration the f2 gave a brighter view than the f2.8. Anyway most of my mid range lenses are f2 or better and perform well at max aperture, so time to get out of my comfort zone and start using lenses wide open.