Aperture blades and look...

[Former Member 8144]

Doing some research on lenses I have come across the following from Cristopher Perez on the hevanet website:

"After performing this test, I believe I begin to understand why old photos can have those qualities. Small blade number apertures can produce rapidly changing dark to light transitions that can be perceived as "harshness". On the other hand, old lenses tend to have many many aperture blades that define a nearly round aperture shape. This is the single most important influence on out of focus area rendition. Further, I believe it is possible to duplicate this effect with modern optics by simply mounting them in multi-blade apertured shutters."

So a simple question.
As far as these visual affects are concerned what is the number required for the 'many aperture blade' effect?

7?
10?
etc,
And then how old a shutter would one need to put the lens in to achieve this.

Thanks,

Marc

[Frank Petronio]

It also matters if the shape of the blades are curved versus straight like modern Copals are, why they couldn't cut them with a curve is beyond me.... Modern Copals only have 5 blades -- as do cheap SLR lenses -- higher end SLR lenses have 7 or 9 blades, Leica and VC lenses usually have 9. If you look into a stopped down Leica lens they make a nice flower-shaped aperture hole. And Rolleis have rounder apertures than on the similar 80/2.8 Zeiss in a Hasselblad... which is why Rollei TLR shots usually a bit creamier.

Sometime in the 60s Synchro-Compurs went from 7 to 5 blades, so the earlier Linhof shuttered lenses can be found both ways... I think all the Copals have been 5 blades, as have been most of the post war American shutters. You simply have to look and ask before buying.

Of course you get perfectly round apertures if you shoot wide open. Oftentimes that's what I try to do. But if you try to shoot outdoors wide open you may want to find a shutter with a fast 1/400 or 1/500 speed, and/or some ND filters and use slower film.

I went through all this myself and currently use a later 5-bladed shutter because I want a nice, newer, reliable shutter (I really prefer Compurs) and I do try to shoot wide open when I think the background will benefit. But it is always a compromise... I doubt Copal will change their design at this late date, it is a pity they didn't care about such details but in the 1950s-60s when things changed over... I think the prevailing trend was to emphasize how darn sharp everything was over other lens qualities.

[Former Member 8144]

Cheers Frank.
I just ask mainly out of interest really as most of my work is +F22 etc and not close up so never that many out of focus areas anyway!

But it is always interesting to understand how to create different visual looks etc.

[Frank Petronio]

It matters at f/22 too though... but then you have to weigh the other differences between modern glass and whatever was good 40 years ago. If you're used to state of the art modern lenses like the 110XL then a vintage lens is an entirely different look.

[Paul Fitzgerald]

Marc,

Ilex #4 = 10 blades
Ilex #5 = 12 blades
Alphax #5 = 15 blades
Alphax #4 = 15 blades
Rapax #3 = 10 blades
Xenar 360/4.5 barrel = 26 blades

It actually seems to make a difference. At f/22 and smaller the jagged edge of the circle would also induce diffraction.

[Jim Noel]

Doing some research on lenses I have come across the following from Cristopher Perez on the hevanet website:

"After performing this test, I believe I begin to understand why old photos can have those qualities. Small blade number apertures can produce rapidly changing dark to light transitions that can be perceived as "harshness". On the other hand, old lenses tend to have many many aperture blades that define a nearly round aperture shape. This is the single most important influence on out of focus area rendition. Further, I believe it is possible to duplicate this effect with modern optics by simply mounting them in multi-blade apertured shutters."

So a simple question.
As far as these visual affects are concerned what is the number required for the 'many aperture blade' effect?

7?
10?
etc,
And then how old a shutter would one need to put the lens in to achieve this.

Thanks,

Marc

I don't believe mounting a modern lens i an old shutter will have any appreciable effect on the "harshness" of the image. I call it "cut & paste" look. The crown/flint lenses just have a less sharp cut off and along with internal reflections produce those nice smooth effects.

[Frank Petronio]

Wow I didn't realize the old American shutters had so many!

We made such nice stuff back then...

[Ole Tjugen]

Compound #5, as in old Xenar 300mm f:4.5, Heliar 300mm f:4.5 etcetera: 23 blades.

I haven't counted #3 and #4 Compounds (yet).

[Ken Lee]

I don't believe mounting a modern lens i an old shutter will have any appreciable effect on the "harshness" of the image. I call it "cut & paste" look. The crown/flint lenses just have a less sharp cut off and along with internal reflections produce those nice smooth effects.

You may be right, but my Fujinon A lenses have very nice blur, when shot wide open. When stopped down, the influence of the 5-bladed diaphragm cannot be overlooked.

On the other hand, my vintage lenses, (albeit limited in coverage, and not apochromatic) with their virtually circular diaphragms, continue to delight the eye, even when stopped down.

With such lenses, one need not shoot wide open, to reap the benefits. Here (http://www.kenleegallery.com/html/flowers/index.html?13) is a photo taken with a 210mm Braunschweig Heliar (at least 50 years old), stopped down a few stops - to get what felt like just the right amount of blur.

[Former Member 8144]

Great info guys.
Its quite a hard one as for the project I am embarking on although it is important for me that the images have a great clarity of detail as some of the subject matter will contain engraved text, etc, the subjects/objects are quite monumental and older relics and so a certain smoothness and sensitivity of look is also required.
The images will be colour and both printed in book form and also upto 40 inch wide and over (drum scanned and lightjet printed) exhibition prints.
So although I am not looking for an old fashioned look per se I am also not after the sharpest, modern xl lens type look.

Doubftull but does anyone have examples of the different look of the same subject achieved by newer and older lenses?

Marc

[erie patsellis]

Compound #5, as in old Xenar 300mm f:4.5, Heliar 300mm f:4.5 etcetera: 23 blades.

I haven't counted #3 and #4 Compounds (yet).
the #3 has alot, seemed like hundreds when reassembling mine..


erie

[Dave Moeller]

I haven't counted #3 and #4 Compounds (yet).

The Compound #3 that happens to be sitting next to me has 15 blades.

[Frank Petronio]

I think the older lenses that were meant to be sharp would probably give you the look you're after -- not soft focus or anything, but just smoother than a modern lens. You can always increase contrast on the file/print side to suit.

The other guys know more than I do about this, but I suspect the "sharp" vintage lenses start with Kodak Ektars...

[Hiro]

I don't believe mounting a modern lens i an old shutter will have any appreciable effect on the "harshness" of the image. I call it "cut & paste" look. The crown/flint lenses just have a less sharp cut off and along with internal reflections produce those nice smooth effects.
I don't believe it, either, for the same reason mounting an old barrel lens in a modern shutter wouldn't make it a Super Symmar (or whatever). Also I don't believe the aperture shape has "the single most important influence on out of focus area rendition." Lens' aberration, for example, is also a big factor for OVERALL out-of-focus characteristics. I agree the aperture shape does affect the rendition of an image, but with a world of respect for what Mr. Perez has done, part of the quoted statements sounds too simplistic.

One instance where the aperture shape is very pronounced is the shape of bright spots (e.g. light bulbs, sun reflections). As a compromise, modern "polygon" apertures, at least in small formats, have curved edges for the first 1-2 stops from wide open (originally a Minolta patent, IIRC). For the curious, I've seen a Photoshop plug-in that can simulate DOF and even the number of aperture blades :), though probably only for bright spots.

[Ole Tjugen]

... The other guys know more than I do about this, but I suspect the "sharp" vintage lenses start with Kodak Ektars...

Not exactly. Petzvals are bitingly sharp within the design coverage, and the central 20 degrees of good Aplanats can give higher resolution than all but a very few modern lenses. One of my 18x24cm negatives shot with a 270mm Meyer Aristoplanat is limited by film resolution, not lens resolution...

[Former Member 8144]

I think the older lenses that were meant to be sharp would probably give you the look you're after -- not soft focus or anything, but just smoother than a modern lens. You can always increase contrast on the file/print side to suit.


So guys, any lenses that fit this description that is, and thanks Frank, exactly what I am after for this particular project.
Remember looking at 240mm and then one longer around 480mm + or thereabouts.
Enough coverage for some movements on 8x10
And fairly light and small.
Shooting on colour often at sunset / sunrise.

I don't ask for much!

Cheers.

[Ole Tjugen]

A 240mm Dynar-type Heliar would do nicely, at least mine does. minimal movements on 8x10", but it does cover.

If you can find a Zeiss Doppel-Amatar, that might be even better (more coverage), but even the 150mm is rarer than hen's teeth.

For a longer lens a Rapid Rectlinear might do the trick, but most Aplanats will be too sharp (yes, I know they are supposed to be equivalent. Believe me, they're not).

[Murray]

I was told by a physics prof that odd vs even number of sides/blades in an aperture has an effect on the way a lens responds to specular areas. He seemed to think 'even', giving an example of a 6-blade iris, was largely the reason for a starburst look. I don't know if he meant those internal reflections you see in movies or a single one at a bright spot. He said the Fourier Transform (which is only a measurement or characterization tool that optics design predates) gives the different behavior...analogous to different waveforms with odd or even harmonics or both.

I can't vouch for whether history, collecting and usage support this or if there are innumerable exceptions. I see odd & even numbers listed in the postings in this thread.

murray

[Ole Tjugen]

Your physics prof is correct, as far as "starbursts" go. A six-sided aperture will give a six-ray starburst, a five-bladed one will give a ten-ray starburst where each "ray" is weaher since there's no doubling. That's why there is a strong preference for odd numbers of blades, even if a 23-blade aperture has to be a lot more difficult to calculate exactly than a 24.bladed one would be.

[Christopher Breitenstein]

This seems like the question "how many rocks make a pile?" or "how many grains of sand make a heap?" The answere is far to 'gray' to provide a definite answere. my sugestion would be to look at an old lens, that produces the desired effect, and count the number of blades.


yours;

[rippo]

seems to me you could just make your own waterhouse style stops and put them over the front of the lens. perfectly round, no pesky blades to make your bokeh pointy.

[Paul Fitzgerald]

"seems to me you could just make your own waterhouse style stops and put them over the front of the lens. perfectly round, no pesky blades to make your bokeh pointy."

Wish it was that easy but that would only vignette the image.

Hate me, hate yourself for missing it:

Doppel-Amatar 250mm (http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&rd=1&item=250234599881&ssPageName=STRK:MEWA:IT&ih=015) :eek:

[rippo]

i use stops in front of my homemade landscape lenses without vignetting. depends on the distance from the lens, and the focal length, among other things.

[mrladewig]

This seems like the question "how many rocks make a pile?" or "how many grains of sand make a heap?" The answere is far to 'gray' to provide a definite answere. my sugestion would be to look at an old lens, that produces the desired effect, and count the number of blades.


yours;

The engineer in me disagrees.

As one of the posters pointed out, odd blades produce double starburst points while even blades produce the same number of points as blades. And all of this can have an effect on the OOF rendering too.

But also part of the interest is that we're trying to achieve a round aperture opening in order to get smooth round specular highlights. Some new 35mm lenses are being produced with aperture blades that are designed make a round profile for this reason. I remember from some old classes that you could measure the deviation from a shape with straight segments. The higher number of segments the lower the deviation until eventually infinite segments would produce a circle. But the change in deviation from 4 segments to 5 segments is greater than the change in deviation from 5 segments to 6 segments. So at some number of segments, you can say it comes close enough. I know that 5 and 6 blades do not cut it. These result in visible pentagon of hexagon shaped specular highlights. So the question becomes what number of segments is needed so that the corners are obscured in the fuzziness of a specular highlight.

But as others have mentioned several factors will affect OOF rendering, most important being the lens itself. Some lenses tend to create double lined OOF areas, others create interference patterns and others create doughnut patterns. Often in the quest for ultimate edge to edge image sharpness the out of focus areas take on weird appearances. And since modern lenses are tested on MTF results and minimal CA rather than smooth out of focus rendering, newer lens designs typically tend to have less pleasant bokeh than older lenses.

[domenico Foschi]

What is a pleasant bokeh?
How do we define it?
SOmeone's garbage is other people's treasure.

[Turner Reich]

What about the number in Betax shutters and Ilex?

[mrladewig]

What is a pleasant bokeh?
How do we define it?
SOmeone's garbage is other people's treasure.

There is truth in that as there are several forms taken by the OOF.

Generally it is considered to be good when a highlight makes a bright point which diffuses out ward smoothly.

And some forms are considered to be extremely distracting such as doughnut shaped bokeh where the light source is expressed as a diffused ring, darker inside and outside. Generally the types of bokeh where the highlight is rendered as a smooth disc with sharp falloff is considered to be less than idea as well.

But there are also other forms taken that can be attractive in some circumstances such as the swirly bokeh.

[Murray]

I've been wanting to try out one of the old Polaroid folder lenses, the 'less desirable' ones, 95B, 150, 160, etc, a 130 mm coated glass triplet lens with round stops (except the wide open one has an edge lopped off). You have to take the whole front part, 'though.

The shutter is kind of cool too, crude and simple enough I can undertand it, not like a 'real' shutter. And big enough one can stick a finger inside to actually confirm (by feel) how it works for each speed. Inspiration for a homebrew behind-the-lens shutter someday.

Yeah, go ahead, laugh.

[Gordon Moat]

Oddly enough, I have one Nikon (35mm) lens with six aperture blades. This one gives me twelve points star look when used for urban night images. Interestingly, used in daylight (http://www.gordonmoat.com/life_24.html), the defocus rendition is nicely smooth, perhaps due to curved aperture blades more than having an even number.

Ciao!

Gordon Moat Photography (http://www.gordonmoat.com)

[Former Member 8144]

back again...so in my search into some the older lenses that will give both a smooth AND detailed look for my 8x10 requirements Ole has suggested a Dynar-type Heliar or a Zeiss Doppel-Amatar in the 240mm range and a Rapid Rectlinear for the longer lens..any other smooth and detailed gems out there?

Cheers,

Marc

[Frank Petronio]

I own about 854 lenses LESS than Ole ;-) but a readily available Kodak Ektar from the 1950s should do the trick and probably be better for color work as well. I bet various Schneider Xenars and Zeiss Tessars would be fine as well.

[Ole Tjugen]

All Heliars, except Universal-Heliars and pre-WWI Heliars, are really Dynars (late RF lenses named "Heliar" excepted - most are neither Heliars nor Dynars). These Heliars seem to have better coverage than "real" Heliars. Note that Voigtländer continued to use the "real Heliar" diagram in brouchures right until the end, 60 years after the design used was changed from Heliar to Dynar...

The only lens I own that is as smooth as the old Heliars is even more expensive - it's the Apo-Lanthar (which just happens to be an "enhanced Dynar").

The short-lived Doppel-Amatar really surprised me when I got the first films developed. I'll never understand why Zeiss stopped making them after a run of only 6 years or so!

Some triplets are nice and smooth too, but that design is so variable that I don't dare say anything general about them.

Tessars and -derivatives can be considered triplets for this purpose.

[Jim Noel]

I think the older lenses that were meant to be sharp would probably give you the look you're after -- not soft focus or anything, but just smoother than a modern lens. You can always increase contrast on the file/print side to suit.

The other guys know more than I do about this, but I suspect the "sharp" vintage lenses start with Kodak Ektars...

Frank,
To some of us Ektars are modern lenses, not vintage. I consider vintage lenses those made w/o coating, and up to about 1940.
Yesterday I made a few images with a 1903 Gundlach Rapid Rectilinear wide open and closed to f 16.
Sharp as a tack in the focused plane but with a roundness to the objects not evident when using coated, and multi-coated lenses.
I believe the real cut off point was when crown and flint glass were no longer the primes for lens making and the sands from one of the northern Japanese islands began to be prevalent.

[Dan Fromm]

Frank,
<snip>
Sharp as a tack in the focused plane but with a roundness to the objects not evident when using coated, and multi-coated lenses.
I believe the real cut off point was when crown and flint glass were no longer the primes for lens making and the sands from one of the northern Japanese islands began to be prevalent.Jim, this entire discussion has baffled me. I can understand the effect of the aperture's shape on rendition of out-of-focus highlights, but don't understand what other effect it has. Most of what people who talk about bokeh seem to discuss has to do with the effects of aberrations that I don't think are affected by the aperture's shape.

And then you come along and add to my confusion by asserting that coating, i.e., control of flare, affects rendition of out-of-focus areas.

Not only that, you blame the loss of roundness that you see in images produced by whatever you mean by modern lenses on the glass used in making them. Now, if I understand things correctly the glasses used in lens-making are characterized by two numbers, refractive index and Abbé number; modern glasses offer a wider range of both than ancient glasses, give the lens designer more ways of reducing aberrations. I'm sorry, but I don't see the connection between use of modern glass, as from Schott starting in the 1880s, and the roundness you value.

It just isn't fair. I try and try and try butI just can't understand what you (all of the posters so far in this discussion, not just you, Jim) are talking about, why you're writing what you do, and what you believe. Will someone, anyone, please post a clear explanation of the models behind all the confusing words and show why they should be taken seriously?

Yours in puzzlement,

Dan

[Don Hutton]

It also matters if the shape of the blades are curved versus straight like modern Copals are, why they couldn't cut them with a curve is beyond me.... Modern Copals only have 5 blades -- as do cheap SLR lenses -- higher end SLR lenses have 7 or 9 blades, Leica and VC lenses usually have 9. If you look into a stopped down Leica lens they make a nice flower-shaped aperture hole. And Rolleis have rounder apertures than on the similar 80/2.8 Zeiss in a Hasselblad... which is why Rollei TLR shots usually a bit creamier.

Sometime in the 60s Synchro-Compurs went from 7 to 5 blades, so the earlier Linhof shuttered lenses can be found both ways... I think all the Copals have been 5 blades, as have been most of the post war American shutters. You simply have to look and ask before buying.

Of course you get perfectly round apertures if you shoot wide open. Oftentimes that's what I try to do. But if you try to shoot outdoors wide open you may want to find a shutter with a fast 1/400 or 1/500 speed, and/or some ND filters and use slower film.

I went through all this myself and currently use a later 5-bladed shutter because I want a nice, newer, reliable shutter (I really prefer Compurs) and I do try to shoot wide open when I think the background will benefit. But it is always a compromise... I doubt Copal will change their design at this late date, it is a pity they didn't care about such details but in the 1950s-60s when things changed over... I think the prevailing trend was to emphasize how darn sharp everything was over other lens qualities.Only Copal 0 shutters have 5 blades - both newer (black ringed) and older (the silver ringed "coarse" tooth) Copal 1 and 3 shutters have 7 blades. Personally, I don't feel that the actual shape makes a huge difference to the Bokeh (and I know I am in the minority on that opinion...) - however, it does define the specular highlight shape and pentagonal ones do not look great. 7 sided highlights are close enough to round that I do not find them distracting.

Interestingly, the latest Rolleiflex TLRs (the FX) has a 5 bladed Copal shutter in it. It's a very nice camera, but I did find that single feature a drawback when I owned one.

[Paul Fitzgerald]

Marc,

"back again...so in my search into some the older lenses that will give both a smooth AND detailed look for my 8x10 requirements Ole has suggested a Dynar-type Heliar or a Zeiss Doppel-Amatar in the 240mm range and a Rapid Rectlinear for the longer lens..any other smooth and detailed gems out there?"

Goerz's answer to the 'Heliar', their 'Dogmar', very nice look to it and very well color corrected. The 420/5.5 just fits a 6" lens board.

Cooke portrait lenses but I have no idea of their series numbers.

Have fun with the hunt.

[Paul Fitzgerald]

Dan, you answered yourself,

"Now, if I understand things correctly the glasses used in lens-making are characterized by two numbers, refractive index and Abb&#233; number; modern glasses offer a wider range of both than ancient glasses, give the lens designer more ways of reducing aberrations."

"And then you come along and add to my confusion by asserting that coating, i.e., control of flare, affects rendition of out-of-focus areas."

Not only did the ability to control aberrations change, the taste of the buying public changed. The manufacturers could and did change their priorities. The blade count seems to have dropped from lack of competition and simple economics, less is cheaper.

As to blade count changing bokeh, the easy answer is Kodak Ektar 127/4.7 & 152/4.5. They did make them in Supermatic shutters (5 blades), Rapax shutters (10 blades) and barrel mount (12-15 blades). All anyone needs to do is buy all three of each and shot-em-up.

[Former Member 8144]

Thanks for these recommendations guys...the way I see it I am looking for heliar, dogmar, dagor, dynar, amatar, apo lanthars!

I'm going to spend a bit of time doing some research working all these names out and which goes with which and the exact models to look for...


Marc

[Christopher Perez]

Marc,

Take a look at "modern" (ie: post 1950) Compur shutters. Many of the shutters I have contain more aperture blades than their Copal counterparts.

Another "modern" shutter that I like is the Prontor Press. These tend to be very light and offer a pleasingly round aperture shape.

As for an absolute number of aperture blades, I happened to stumble upon a couple shutters (Prontor and Compur) that give more pleasing out of focus renderings than Copal shutters some of my modern lenses were originally mounted in. Then I had a chance to shoot a Compound shutter'd lens against it's Copal shutter'd equivalent. The difference in OOF rendition was equally obvious. It was really that simple.

A person can go really nuts and look for 1950's - 1970's lenses in Compur, Compound, or Prontor shutters. I love Schneider Xenars for their creamy OOF rendition. I think Ole mentioned a 300mm f/4.5 Xenar in #5 Compound - I have one and it's brilliant sharp where sharp is desired and lovely in the OOF regions.

There are many other lenses to look for too. In fact, a friend wrote me about a cheap 180 f/4.5 Xenar in Compur that he found. It's OOF rendition is actually nicer than a Compound mounted Heliar of similar focal length. Granted, optics design plays a role in OOF rendition. HOWEVER, aperture shape is easily verified as one of the most/more important factor in OOF rendition.

Thanks for reading/looking. :)




So a simple question.
As far as these visual affects are concerned what is the number required for the 'many aperture blade' effect?

7?
10?
etc,
And then how old a shutter would one need to put the lens in to achieve this.

Thanks,

Marc