Beyond "Bokeh"...

[Mark Sawyer]

We need another word.

"Bokeh" refers to the "out of focus" areas, and how the lens defines the light and shapes therein. But with some of us, especially in large format photography, there's a lot in the "personality" of the lens that goes far beyond that. It's the spherical and chromatic aberrations, the halation, the coma, the internal flare of the uncoated lens, the curved field, the "swirlies" (whatever thery are...)

If you read Merklinger's article introducing the term "bokeh" to the western photographic world, (http://www.luminous-landscape.com/essays/bokeh.shtml), it seems pretty clear that the "bokeh" phenomenon he refered to was tightly defined to how aperture shape influenced the depiction of out-of-focus edges. But we've expanded it to include so many other things. Is the nature of a Verito, a P&S, a Heliar, or a Petzval really so related to the shape of the aperture?

In our laziness, we talk about the quality of these lenses in terms of their "bokeh", when really, it goes far beyond that. And in my own snobbiness or snootiness, I silently harumph to myself when I hears dslr users argue whether Canon or Nikon lenses have "better bokeh".

Is there already a term that refers to the particulars of how one lens uniquely and distinctly depicts the world? A term that is more suitable to describing Plasticcas, Strusses, Imagons, and 99-cent lenses, as opposed to the lenses designed to all look alike?

I suppose referring to a lens' "personality" is the prevailing terminology. If it weren't for "bokeh", I'd suggest we refer to a lens' "bouquet", (not unlike a fine wine!)

Anyways... Is there already a term in use for a lens' particularly special way of seeing the world? Or suggestions for a new term?

[David A. Goldfarb]

Sometimes people talk about the visual "signature" or "character" of a lens to describe these overall effects.

Personally, I prefer "out of focus rendering," which seems more in line with descriptions of lenses from the age of soft focus lenses, to "bokeh," for that specific phenomenon.

[aphexafx]

I would agree with you Mark, except I have never perceived that anyone meant anything besides how the aperture shape influences the out of focus edges.

I can't stand the word in the first place...

[Oren Grad]

If you read Merklinger's article introducing the term "bokeh" to the western photographic world, (http://www.luminous-landscape.com/essays/bokeh.shtml), it seems pretty clear that the "bokeh" phenomenon he refered to was tightly defined to how aperture shape influenced the depiction of out-of-focus edges.

Mark, I'm afraid that's not right. Please read Harold's article again, paying attention to the section that begins with the following:

"But photographers also know that particular lens designs have individual boke character, even when diaphragm shapes are similar. The Leitz 35/2 Summicron, for example, is reputed to have "good boke" while other some other lens designs give rise to "ni-sen" (double-line streaks) and other forms of "bad boke". What makes the difference?..."

That said, I agree that OOF rendering is far from the only important aspect of optical character.

Matt, saying you can't stand the word "bokeh" is literally equivalent to a Japanese person saying he can't stand the English word "blur".

[John Cahill]

Sometimes people talk about the visual "signature" or "character" of a lens to describe these overall effects.

Personally, I prefer "out of focus rendering," which seems more in line with descriptions of lenses from the age of soft focus lenses, to "bokeh," for that specific phenomenon.

*(****
For myself, I have long thought the whole concept of "bokeh" is nothing less than a tellibly precious form of cockamamie for people who would rather talk (or email) about photography, than just doing it.
Before it had a "name" and a "cachet," it really only mattered when dealing with those little out of focus donuts which showed up with those new-fangled mirror lenses back in the late 1950s-early 1960s.
It strikes me about the same as a bunch of yuppies years back comparing the merits of the various swills coming from "micro"- breweries: take it from someone who lived and worked in Munich, most of the micro-stuff is bilge that would be flushed down the nearest town drain in "the beer capital of the world." But it's oh, so much fun to comment upon, by people who would not know a decent beer if they tasted it.
I feel the same about bokeh.

[aphexafx]

Matt, saying you can't stand the word "bokeh" is literally equivalent to a Japanese person saying he can't stand the English word "blur".

That is almost true, except as you have illustrated the word is "boke". I have never understood how it turned into "bokeh" except that that is how it is pronounced, so I assume someone just bastardized it. It annoys me, I can’t help it.

So my main issue is not so much the word itself, but the fact that we have imported a word, albeit a very descriptive one, and forked it into this thing we write. So it comes across as a buzz word, and that annoys me. I'm sorry, I can't help it.

I honestly can't stand the word "bokeh" and I wish the Japanese people no ill.

....

But anyway, in the text you reference, by Harold, all of the characteristics being referred to are dealing with out of focus edge rendering, whether it be a condition of the shape of the iris, internal reflection, group placement, front element shape, or what have you, and in that sense is still boke by nature as Mark described it, no? I guess someone should explain these double line streaks to me, but if they are not an OOF characteristic, I don't think they should be considered as "bokeh" no matter what the author originally stated. I do believe that the word is used in Japan, etc. to describe just that.

[Oren Grad]

That is almost true, except as you have illustrated the word is "boke". I have never understood how it turned into "bokeh" except that that is how it is pronounced, so I assume someone just bastardized it.

Please read this post (http://theonlinephotographer.typepad.com/the_online_photographer/2009/01/what-is-bokeh.html), in which Mike Johnston explains where "bokeh" came from as well as some other context that you're missing.


But anyway, in the text you reference, by Harold, all of the characteristics being referred to are dealing with out of focus edge rendering, whether it be a condition of the shape of the iris, internal reflection, group placement, front element shape, or what have you, and in that sense is still boke by nature as Mark described it, no?

Again, that's not correct. The point of Harold's article is to explain OOF rendering in terms of the mathematical concept of convolution or, to put it in terms of the non-mathematical analogy he uses, to understand how OOF character is affected by the nature of the "brushes" used to "paint" the picture. If you think it's about the rendering of edges, you're either not reading the whole article or not understanding it.

[aphexafx]

Please read this post (http://theonlinephotographer.typepad.com/the_online_photographer/2009/01/what-is-bokeh.html), in which Mike Johnston explains where "bokeh" came from, as well as some other context that you're missing.

Will do, but the word still annoys me, sorry if that bothers you. No issue. :)


Again, that's not correct. The point of Harold's article is to explain OOF rendering in terms of the mathematical concept of convolution or, to put it in terms of the non-mathematical analogy he uses, to understand how OOF character is affected by the nature of the "brushes" used to "paint" the picture. If you think it's about the rendering of edges, you're either not reading the whole article or not understanding it.

I understand this and I understand convolution. However, all of this is referencing how the lens handles OOF rendering, thus it is the lens' OOF characteristic, whatever the source of this characteristic, be it the shape of the iris or the type of cement used to hold the groups together. etc., that is referred to as "bokeh". I agree with this. I don't agree with anyone using the term to refer to anything other than a component characteristic of OOF rendering, is all.

Added: I see the issue now, and I did say "OOF edge definition" in the last post (yipes, even the one before that, but I was led on my Mark and I am not responsible). :D I agree that it is not tied only to OOF edge rendering, and that this is just a very visible part of the "bokeh" spectrum.

Now I have to explain. I DO agree that "bokeh" refers to OOF rendering beyond edge definition. However, in my haste, and laziness, I got caught up in the fact that this is how people generally see "bokeh". My point has been that it shouldn't be used to refer to things such as contrast, coloration, tint, vignette, or anything else that is not an OOF characteristic of a given lens. I thought Mark was refering to people referring to these non-OOF characteristics as "bokeh" and I was noting that I had never encountered such a misuse, even though I was cutting myself short by my own references to edge definition which is only part of the picture.

Whew.

[John T]

Many Japanese people spell their names with an "h" when the vowel is drawn out longer. For instance, some people spell their name Ito while others spell it Itoh. The Japanese spelling is the same.

Because English speaking people would normally rhyme the terminology "boke" with smoke, the "h" helps with pronunciation. I can't see how this should cause you such consternation.

I think the bigger problem with the terminology is the bastardization of the word. Some people use it to describe shallow depth of field, selective focus or characteristics of the focus plane of soft focus lenses.

[aphexafx]

I can't see how this should cause you such consternation.

Well, I still sleep at night. ;)


I think the bigger problem with the terminology is the bastardization of the word. Some people use it to describe shallow depth of field, selective focus or characteristics of the focus plane of soft focus lenses.

I agree with this completely, does Oren Grad???

[Oren Grad]

I agree with this completely, does Oren Grad???

I'll ask him next time I see him.

Do read Mike's post. Then you can finish tidying up here. ;)

[aphexafx]

lol (I think...:confused:)

But I am seriously asking if you agree? I want to know for my sake. (I was loose with my wording and now I'm paying for it, I think.)

I will read everything referenced here, absolutely.

[Bjorn Nilsson]

While this word (I will not mention it because of some people hating it. :) ) have been misused a lot, it have established itself. The strange thing about it is that the concept of *** is as fuzzy as what it is trying to describe.
So, in order to making myself/yourself understood I recon I will have to accept the different understandings of what *** means. Inventing a new word "###", would simply confuse matters further.

A similar problem comes up in non-english spoken countries. E.g. France where they "must" come up with their own french word for more or less every new thingie on the market. I don't know if this craze is still in full force, but it had many computer programmers busy with translating interfaces for programmes, while inventing new french names for every computer concept there was.
But an even worse example must be when someone in Sweden came up with the name "Freestyle" for a new gadget which was introduced in the 80'ies. Need a clue? (You should also know that Freestyle isn't/wasn't part of the swedish vocabulary until then. It translates into "fristil" in Swedish. Same word really but different pronunciation.)
Anyhow, in Sweden we took another english word for the "Walkman". It was very funny to travel these days, when swedes walked into tax-free shops and asked for a Freestyle and the salesman didn't have a clue despite the shop being full of "Freestyles".

//Björn

[aphexafx]

Bjorn, that is very funny. :)

To all involved, I have reinvented myself and now think that the work "bokeh" is just fantastic. I have written it on my forehead with a Sharpie and I am going to replace the word "nice" in my vocabulary with "bokeh". I will say, "you are a very bokeh person." "What a bokeh job you've done." And, "Hey, I think that your new girlfriend is real bokeh."

It is now ok to continue using the word "bokeh" because I am of the opinion that it's...well, it's just swell.

sigh...shoot me, shoot me now.

:D

What I am more concerned with is whether or not my impression that bokeh refers to the non-sharp, or OOF, rendering characteristics of a lens, and nothing else, etc. I will read the linked articles, for sure, but I've examined this before and this was my conclusion. If this is still wrong, in your opinion, anyone, please tell me. I'm not trying to start a bokeh war here. I can change. I have the will and the power. But so far this appears correct to me.

Bokeh.

[David A. Goldfarb]

Before it had a "name" and a "cachet," it really only mattered when dealing with those little out of focus donuts which showed up with those new-fangled mirror lenses back in the late 1950s-early 1960s.

Not really. Here's a quote from the Wollensak catalogue for the Verito that I've posted before--

"a specially designed double lens... which, while it gives the desired diffused or soft optical effect, shows no distortion, double lines, or other optical imperfections, and being rectilinear gives an even diffusion over the whole plate... Will not make sharp negatives with wiry definition unless stopped down to f:8."

This sort of description is typical for the period. They didn't call it "bokeh," but they were well aware of "double lines, or other optical imperfections" in the out-of-focus area as an aesthetic problem.

I suspect that the general predominance of the 35mm SLR in the 1970s shifted the emphasis in the design and marketing of lenses almost exclusively toward questions of sharpness over the rendering of the out of focus area, since the subject of a photograph is more important than the background in general, and with a small format like 35mm there isn't much room for compromise. Increase the format size and you can have the luxury of a sharp subject and a smooth background. The contemporary interest in "bokeh" is a return to an earlier photographic concern.

[russyoung]

David, you're dead on - your Verito quote (which can also be found in my dissertation on soft focus) very clearly indicates that the turn-of-the-century photographers were fully cognizant of the character of out of focus image areas. Two key Pictorialists, Alvin Langdon Coburn (in America) and Heinrich Kuhn (in Austria) used the term "woolly" for a particularly undesirable bokeh.

Although the evidence is incidental, I firmly believe that their choice of one soft focus lens over another was heavily influenced by the bokeh. Having tested about 30 different period soft focus lenses at studio distances, perhaps 95% possess beautiful bokeh; they have numerous aperture leaves AND residual spherical aberration, both variables which contribute to an enjoyable form of bokeh.

In addition to the ascent of miniature formats, I believe there is another reason for the loss of the concept after the 1930s and that is the rise of the f/64 School in California. Since virtually everything in their images was in focus, the issue of bokeh became moot. Stopping the lens to its smallest aperture eliminates many aesthetic decisions and simplifies the process... and negates many of the tools available to the informed photographer.

OK, I'm ready to be burned alive now.

Russ Young

[Ken Lee]

I suppose referring to a lens' "personality" is the prevailing terminology. If it weren't for "bokeh", I'd suggest we refer to a lens' "bouquet", (not unlike a fine wine!)

Anyways... Is there already a term in use for a lens' particularly special way of seeing the world? Or suggestions for a new term?


Boquet: brilliant idea !

There's a Sanskrit word: rasa (the "a" is pronounced as in "father") - which means flavor, taste, as well as sap, juice, and... delight. It's a bit like flavor + savor at the same time. Perhaps other languages have an equivalent term. English may not be the most poetic or musical of languages.

In any event, you're right: There's certainly more to it than blur rendition. Even when the image is fully sharp (as sharp as the lens will get anyhow), there are still subtle differences. The best way to see them is with side-by-side comparison.

How do musicians characterize the differences between instruments of the same basic design - for example, Violins ?

The last time I heard, no one has been able to fully duplicate the sound of Stradivarius instruments. Each one sounds a little different from the next, and the important ones have individual names. I imagine that they also change over time.

[Mark Sawyer]

Mark, I'm afraid that's not right. Please read Harold's article again, paying attention to the section that begins with the following:

"But photographers also know that particular lens designs have individual boke character, even when diaphragm shapes are similar. The Leitz 35/2 Summicron, for example, is reputed to have "good boke" while other some other lens designs give rise to "ni-sen" (double-line streaks) and other forms of "bad boke". What makes the difference?..."

That said, I agree that OOF rendering is far from the only important aspect of optical character.

Agreed, but even when Merklinger writes of the Imagon, the only sf lens he tested, he talks primarily of its aperture shape and edge definition. And that's what most photographers are talking about when the use the term "bokeh".

Maybe it's just that I'm tired of hearing photographers getting an f/1.4 lens to go along with their f/4 zoom, so they can get "more bokeh".

It seems, though, that when we're talking about the differences between a Verito image and a Vitax image, we're in some other ballpark, and "bokeh" as most photographers know it isn't what we're talking about...

On the other hand, his final paragraph reads: "To summarize then, your camera paints its image with a repertoire of brushes whose characteristics are determined by the shape of the diaphragm opening and the details of the lens design's aberrations. Some brushes are softer-edged than others, and that's what makes the difference in boke."

So maybe I don't know what I'm talking about...

[Oren Grad]

Agreed, but even when Merklinger writes of the Imagon, the only sf lens he tested, he talks primarily of its aperture shape and edge definition.

I think I understand where part of the semantic problem may be here. Harold discusses two factors - the aperture shape, and the distribution of light intensity across the blur disk. As far as the latter is concerned, a "bright ring" pattern, where there's greater intensity at the periphery of the blur disk - you could say, more prominent defnition of the edge of the blur disk - tends to produce unpleasant effects. But the fundamental concept isn't edge definition in the sense of definition of the edges of the pictured subject or the edges of the diaphragm blades, it's the shape of the spread function that describes the light distribution across the blur disk and how that affects the OOF appearance. That's also what he's talking about when he refers to how soft-edged the "brush" is.

It's unfortunate that Harold's article is posted in isolation. It was only one part of a suite of articles on bokeh that Mike ran in Photo Techniques magazine; if it had been possible to post the whole set, the context, the meaning, and our intent might have been clearer. At any rate, in his post over at TOP Mike tries to clear up some of this confusion.

Matt, when we wrote those articles about bokeh for PT, we were talking specifically about the character of the out-of-focus parts of a picture. Nothing more, nothing less.

David and Russ, thanks for the helpful historical perspective. That Verito quote is a gem!

[Ole Tjugen]

The bokeh of the Heliar was an important marketing point for Voigtländer in the 1930's, although that word was of course not used. They used words like "dunftig" instead, when describing the transition from sharp to unsharp.

[Chauncey Walden]

So, the dunftig of a German lens will blow away the bokeh of a Japanese one?

[David A. Goldfarb]

The bokeh of the Heliar was an important marketing point for Voigtländer in the 1930's, although that word was of course not used. They used words like "dunftig" instead, when describing the transition from sharp to unsharp.

"Duftig," isn't it?

[Jim Galli]

Somehow I missed this post. Let me add a few thoughts.......

While my Ross Univesal is dignified, stylish, rich in fruit, with vibrant acidity and wonderful finesse, when compared to my Pinkham Smith I must say the Pinkham is dark, rich, exotic, smoky, with complex herb, beefy Syrah-laced fruit and, for me, quite elegant and stylish.

The Cooke Portellic on the other hand is pure rich english walnut with dashes of vanilla laced with pomegranite.

I'll have to brush up on my french to even begin to describe the qualities of my Eidoscop.

[Steve Hamley]

I'm getting hungry...:D

Cheers, Steve

[Nathan Potter]

I also missed this post - it's very informative so far because in all my industrial photography I had never heard of the term. I've been trying to figure out what the hell you guys have been talking about since I came to this forum. I now see that you've been confused but to a lesser degree than me. But I have come to understand and recognize some interesting examples of fine Bo-Ke thru the efforts of Jim Galli and others and enjoy the effect.

But there seems to be inadequate explanations of the source. Oh its all clearly related to OOF effects but all lenses can be defocused and not show what one would consider good Bo-Ke. But I don't think it's so difficult to nail it if one thinks a bit about it.

The swirls and blur patterns are only enlarged Circles Of Confusion that lie beyond the point of best focus. Since each resolvable point on the subject is seen by every part of the lens and ideally wants to be imaged exactly at the film plane the lens designer does the best job possible in achieving just that. But it is impossible with even modern lenses and even more so with more primitive types. Using more primitive designs at maximum aperture exacerbates the problem of correction at the film plane due to peripheral lens curvature etc. Thus the shape of the OOF Circle Of Confusion indicates the degree of correction that the lens designer has achieved. I would guess that the shape and character of the OOF blur pattern results principally from the failure of the lens design in simpler lens types. Essentially the skewness and coma in the airy disc, not normally visible at the point of best focus becomes blatantly visible in its grossly enlarged COC at the OOF condition. Additional variation is found depending on the aperture blade design but I bet that is more minor compared to lens aberrations. Very fine highly corrected lenses will show perfectly circular and uniformly illuminated OOF COC from a roughly infinity point source subject. There may be some other minor lens effects that contribute the bokeh that we either love or hate but IMHO lens aberrations dominate. :) :)

Nate Potter, Austin TX.

[Mark Sawyer]

The Cooke Portellic on the other hand is pure rich english walnut with dashes of vanilla laced with pomegranite.


And a lingering finish!

(Personally, I find the Ross Universal a bit heavy on the tannins. But, no doubts, these lenses can be... intoxicating!)

[Oren Grad]

In Japanese one refers to the character of bokeh as the "bokeh-aji" - literally, the taste or flavor of the bokeh... :)

[Struan Gray]

I too think of bokeh as being about *all* the characteristics of the out-of-focus parts of the image, and not just those dominated by aperture effects.

It's natural to want to explain and quantify something that is being touted as desirable, and the effect of the aperture shape on fully-out-of-focus backgrounds is the simplest thing to explain to jobbing photographers without a strong background in science or mathematics. Anyone who can follow the standard arguments about depth of field can see how the circle of confusion becomes a polygon of confusion, with simple resultant effects on the projected image.

To go further you need to get to grips with the mental concept of convolution, and you need to develop a feel for how lens aberrations affect the 'kernel', i.e. the shape that the perfect image is convoluted with. The concept of a circle or polygon of confusion is still useful, but now the light intensity is not constant across its surface, but varies from place to place. Lens designers use 'spot diagrams' to plot these intensity distributions, and they are not so very hard to understand visually, but today's lack of a technically-competent photographic press means that most articles stop before they reach even this simple point.

Eric Beltrando's excellent dioptrique (http://dioptrique.info/) site has spot diagrams for many classic designs. A detailed reading takes years of experience, and I don't claim to be even an armchair expert, but even a simplistic look of the diagrams for, say, a classic Petzval (http://dioptrique.info/objectifs/00003/00003m.htm) and a 'sharp' lens like the Ziess Ortho-Protar (http://dioptrique.info/objectifs8/00395/00395M.HTM) is instructive. The Petzval has a much larger spread of light for quite small angles (i.e. it loses sharpness fast off-axis) but the blur forms a nice rounded blob rather than the tight, yet angular shapes made by the ortho-protar.

So, my personal feeling is that 'bokeh' is an ugly but useful word which does indeed mean *all* the non-sharp aspects of a lens' performance, but that non-technical gossip tends to emphasise only one subset of the whole. It is similar to how 'colour temperature' is used as a shorthand for the colour characteristics of a light source whether or not is has a spectrum for which the 'temperature' simplification is valid.

[Struan Gray]

...

[Archphoto]

Just an other stupid question from a technical guy: one expects the shape and other properties of bokeh to be dependant of:

- the amount of aperture blades
- and the shape of the aperture baldes

So a 4 straight blade aperture will give a diferent bokeh than a (near) perfect circular one........

Peter

[Ken Lee]

The Petzval has a much larger spread of light for quite small angles (i.e. it loses sharpness fast off-axis) but the blur forms a nice rounded blob rather than the tight, yet angular shapes made by the ortho-protar.

Struan, are these diagrams always made with the lenses wide-open ?

As we stop-down the lenses moderately (IE, not past the point of the diffraction limit), do most of these aberrations tend to go away ?

I recently performed a test (http://www.kenleegallery.com/images/forum/4lenses2.png) of some of my lenses, and noticed that even by f/8, the differences between lenses of similar design become hard to judge (http://www.kenleegallery.com/images/forum/4lenses1.png), at least when the apertures are round. A Heliar, 2 different Tessars (one in shutter, another in barell), and a Fujinon A, are basically indistinguishable in this test.

What makes the Heliar and Tessar advantageous for blur, it seems to me, is less any special optical design, but more their wider apertures with many blades.

The Fujnon 240A was shot wide open at f/9. The rest of the lenses were f/4.5 vintage designs, with very round multi-bladed apertures stopped down to f/8.

Are there any lenses for large format, which give very uniform bright disks (shot wide-open, or with a round diaphragm) ?

[Ken Lee]

Just an other stupid question from a technical guy: one expects the shape and other properties of bokeh to be dependant of:

- the amount of aperture blades
- and the shape of the aperture baldes

So a 4 straight blade aperture will give a diferent bokeh than a (near) perfect circular one........

Peter

Yes, definitely.

See this article (http://www.luminous-landscape.com/essays/bokeh.shtml) by Harold Merklinger. He shows triangular apertures and the resulting triangle-shaped blur. He even show how depending on whether they are in front of, or behind the plane of focus, the blurs will appear right-side up, or upside-down.

As they say, "your mileage may vary", depending on the subject you are shooting: If the out of focus region contains distinct highlights, you will see the shape of your aperture blades. It it's just a uniform blue sky, you'll see little effect.

[Struan Gray]

Struan, are these diagrams always made with the lenses wide-open ?

If you click on the individual plots the f-number is given at the bottom. The first column seems to be always wide open, the others stopped down to 'typical' apertures (for example, the ortho-protar has spot diagrams for f22, but the Petzval only has f11 as the smallest aperture).


As we stop-down the lenses moderately (IE, not past the point of the diffraction limit), do most of these aberrations tend to go away ?

There are seven classical first-order aberrations. Spherical aberration, coma, astigmatism, field curvature, distortion, axial colour and lateral colour. All of them except distortion and lateral colour are reduced by stopping down, but by different amounts. This page has a useful summary (as well as a more nuanced discussion of bokeh than is usual: http://toothwalker.org/optics.html


...are basically indistinguishable in this test.

That makes perfect sense. After all, the goal of nearly all photographic lens designs is to reduce aberrations as much as possible. Those that are allowed to creep in are nearly always there as necessary consequences of the desire for a larger aperture for particular applications.

I agree with those who say most bokeh discussions are like arguing about how many angels can dance on the head of the pin; but sometimes I take photographs where it matters tremendously, and I am glad that I *know* how my lenses will behave in those situations. My undergrowth obsession in particular has made me very aware that I need to look out for foreground bokeh as well as background bokeh.


Are there any lenses for large format, which give very uniform bright disks (shot wide-open, or with a round diaphragm) ?

I don't know of any. You can trade aberrations off against each other when adjusting the lens design, and for photographic objectives I can see that allowing some spherical aberration to creep in so as to quench the uglier asymmetric aberrations like coma or astigmatism makes sense. The most expensive photographic lenses like aerial survey objectives or top-end repro-lenses tend to be made for sharp imaging at a fixed reproduction ratio, so I'm not sure bokeh concerns enter much in their designers' thoughts.

I have seen spot diagrams for on-axis performance of superachromat objectives (including a 2.8 Petzval :-) and they were remarkably even across the defocus disc, but these were objectives with an imaging angle of a few degrees, so not much use for LF. I would imagine the superachromat lenses for Hasselblad have similarly even distributions, at least at the centre of the field. If anyone feels like loaning or buying me one I am happy to do a test :-)

[Ken Lee]

Struan - Thanks so much for your articulate, informative, and patient reply. This subject gets a lot of attention, but it's very instructive.

I have seen movies where candle lights in the OOF regions, appear as though they were luminous gold coins, perfectly circular and even.

Given that high-end lenses for cinematography are considerably more expensive than the ones we use, is that sort of rendering, one of the reasons they are so costly ?

When I compared a Tessar, Heliar, and Solinar-S wide open, I found that the Solinar had the most distortion (lateral coma), and the worst blur rendering. But considering that it wasn't really intended to be shot wide open, I brushed that aside. At f/22, the differences I saw, were only a matter of coverage.

[Dan Fromm]

"Given that high-end lenses for cinematography are considerably more expensive than the ones we use, is that sort of rendering, one of the reasons they are so costly ?"

No, they cost so much because of low production volumes, complexity, and precision of manufacture. Its what they are, not what they do, that makes them expensive.

[rdenney]

Outside this forum of real experts, most people confuse selective focus as a technique with bokeh as a characteristic. The relevance of bokeh is simply this: We use selective focus as a means of isolating the subject from the background. If the rendering of the background is distracting, it defeats that purpose.

Some lenses have a reputation for stunning bokeh that I find to be extremely distracting. They think it's artistic, sort of like applying a pointilist or impressionist filter in Photoshop. And that reveals another confusion between measurement and evaluation. For the lenses I use in conjunction with selective focus, I want the smoothest rendering of the background possible. That usually means a faded edge to the airy disk, which is a sign of undercorrected spherical aberration, according to my reading. That aberration is reduced by stopping down, and I think a good lens design couples the reduction in spherical aberration to the increase in depth of field, so that the rendering of those unfocused airy disks looks the same throughout the f-stop range of a lens. It's the rare lens that does both well, and a lens might be smooth wide open but distracting two stops down. I think of those lenses as having fickle personalities, clear and light on first tasting but with a rough finish.

I rarely use selective focus with large format, but that's a statement about my choice of subject more than anything. I do use it a lot with smaller formats, where I'm less in control of the situation and have to manage backgrounds by whatever means might be at hand. There, lenses such as the classic Sonnars provide smooth rendering while others don't.

The patterns in a busy unfocused background are the sum of those airy-disk projections. The shape of the airy disk, which is controlled not only by the shape of the aperture but also seemingly by coma and physical vignetting, can cause all the airy disks representing the image to add up to strange swirly circular patterns. It can cause two unfocused edges to create a ripple that has more definition than the edges themselves. At the end of the day, some lenses are on my list of those to use with selective focus, and some are on my list of lenses to use when trying to achieve a sharp image from edge to edge.

I did recently see some 8x10 portraits (in another forum) of full faces. A full face on 8x10, of course, is nearly in the macro range. Even a f/9 or whatever, the depth of field was about the thickness of a sheet of paper. Those images profoundly drew me into the focused eyes of the subject. That represented selective focus at its best for that sort of portrait. I don't remember how the lens rendered the background--and that's exactly the desired result, it seems to me.

Rick "who conducted a bokeh comparison of a range of small-format lenses, and found an ancient B&L Tessar to be the worst of the lot" Denney