Multiple Aperture Blades - Why?

I know from selling on the auction site that lenses with multiple blade "near circular" apertures are in some demand due to their perceived effect on the out of focus areas - the so called "Creamy Bokeh". The Pentacon 135 and 200mm Meyer Optic designed telephotos having 15 blade preset apertures are a good example. Pretty much all lenses with more than 9 blades were plain manual or preset, and non-automatic, although some semi-automatic lenses had more. For example the 300mm Tair, which came with the Photosniper set, had 16 blades which closed down automatically, but needed to be opened back up again manually

 

It occurred to me to wonder why they were made with so many blades, at a time when (as far as I remember) no-one cared a toss about Bokeh. They must have been very expensive and labour intensive to make. What was so special about a near circular aperture? And what about enlarging lenses, where out of focus rendering seems totally irrelevant? Some of those had multiple blades too.

 

Any thoughts?.

People did indeed care about the 'dot'-shape.

In barrel lenses, all sorts of shapes were used for the effect they had. The art was lost when the apertures changed from stops to iris diaphragms. Completely lost when they had to be automatic (and the number of blades had to be reduced).

And with it, the attention to what aperture shapes do was lost, the art (almost) forgotten.

I suspect friction increases with the number of blades, so modern designs with automatic everything tend to use fewer blades. I've seen some older lenses and some process lenses with a boatload of blades. I've read, but not confirmed, that just having a very round aperture doesn't guarantee pleasing bokeh. It also has much to do with the optical formula. The shape of the out of focus area does equate with the shape of the aperture.

I have (very) vague memories of reading, a long time ago, of lenses using, I think, 'T' stops which were discs or plates with the aperture required as a 'hole in the middle'. Can anyone else confirm this, and an associated 'memory' of some of these Stops having shaped 'holes', for a kind of vignetting effect on the image on the plate, and even some for plate cameras with full movements, to record less light on the sky area, so that cloud details were obtained. Or is this just the product of an already cloudy mind ?

On early cameras like the Kodak Six-20 Jiffy there was simply a bar/tab with holes in it for aperture settings. Many people were unaware of these "waterhouse stops" (LINK).

 

Generally, the holes were circular, although star shapes and such were used for special effects.

 

Here's the exposure guide from the Jiffy manual - the holes are simply ranked from large to small

Edited January 10, 2021 by JDMvW

Thanks, @JDMW - that is exactly what I was thinking of.

I have (very) vague memories of reading, a long time ago, of lenses using, I think, 'T' stops which were discs or plates with the aperture required as a 'hole in the middle'. Can anyone else confirm this, and an associated 'memory' of some of these Stops having shaped 'holes', for a kind of vignetting effect on the image on the plate, and even some for plate cameras with full movements, to record less light on the sky area, so that cloud details were obtained. Or is this just the product of an already cloudy mind ?

T stops were originally available on lenses intended for professional film making where precision was essential and generally possible under controlled studio conditions. I remember working with an Angenieux 9.5 -95 mm zoom lens that had f/stop markings for determining depth of field and T stop markings for determining exposure. This lens was f/2.2 but T 2.8 due to its complex design and the coating technology commercially available in the early 1970's.

Thanks, @AJG - thought I'd read the term somewhere. Thanks for the education and information.

I recently had the 'pleasure' of de-oiling and re-assembling the preset iris in a 300mn f/4 Meyer Pentacon lens.

I didn't actually bother to count the number of blades... but it was far too many!

 

As previously said. The reduction in number of blades used has everything to do with friction and stiction. It doesn't take a lot of oil to make a 19-blade iris pop apart under pressure or refuse to close.

 

The proximity of iris blades to greasy helicoids in medium format and miniature cameras played a part in the move to reduce iris-blade numbers as well.

 

Plus the advent of CAD has made the design of complex-curve blades much easier; such that an approximation to a circular aperture can be achieved with a smaller number of blades.

Edited January 10, 2021 by rodeo_joe|1

I have (very) vague memories of reading, a long time ago, of lenses using, I think, 'T' stops which were discs or plates with the aperture required as a 'hole in the middle'. Can anyone else confirm this

You're thinking of 'Waterhouse stops' Tony. Completely different from 'T-numbers' = T for Transmission. T numbers take the absorbtion/scatter of a lens into account and give the f-number of a perfect lens that would give the equivalent light transmission and exposure. IMO a bit OTT unless the shutter and film stock/sensor are equally well calibrated.

 

Confusingly, Zeiss marked some of its lenses with a red T or T*, which just indicated the lens was AR coated.

Thanks very much @rodeo_joe - yet another example of my mind concatenating two similar yet unrelated facts.

No problem Tony.

I sometimes think my head's too filled with that sort of useless information to remember really important stuff, like friends' birthdays, or where I left my spectacles and the like.:oops:

You're thinking of 'Waterhouse stops' Tony. Completely different from 'T-numbers' = T for Transmission. T numbers take the absorbtion/scatter of a lens into account and give the f-number of a perfect lens that would give the equivalent light transmission and exposure. IMO a bit OTT unless the shutter and film stock/sensor are equally well calibrated.

 

Confusingly, Zeiss marked some of its lenses with a red T or T*, which just indicated the lens was AR coated.

 

Not over the top, because the difference between transmission and what an f-number suggested about the same was great enough.

T-values were/are actual tested values of lenses that are not perfect. Or rather, of how much light actually passes through a lens. They give the effective f-stop rather than an focal length divided by diameter number.

 

Zeiss marked their coatings with a T precisely because antireflection coatings reduced the difference (almost) enough to make the difference between transmission and what an f-number suggests something you needn't worry about.

In motion pictures, images/scenes have to match and having to match in post was rather

Edited January 10, 2021 by q.g._de_bakker

Thanks for the comments, very interesting. I'm still not sure how the advantage of circular apertures was perceived.

 

People did indeed care about the 'dot'-shape.

 

I think this means the shape of an out of focus bright spot in the scene .assumes the shape of the aperture?

 

Stops having shaped 'holes', for a kind of vignetting effect on the image on the plate

 

I've seen these holes with different shapes, I think they were to create fancy vignettes - in which case they would have been in front of the lens, not at the aperture position.

 

Generally, the holes were circular, although star shapes and such were used for special effects.

 

Well, pretty much all low end cameras, when fitted with aperture control, did it with sliding circular holes. Although some I've seen had very oddly shaped apertures indeed, like the Bencini Koroll 35:

 

LINK --- The Bencini Koroll 35

Not over the top, because the difference between transmission and what an f-number suggested about the same was great enough.

Greater than any mechanical shutter's deviation from its marked value?

Or greater than the setting accuracy of a non-click-stopped aperture. Or even the repeat accuracy of an automatic stop-down mechanism?

 

Sometimes people worry about a dripping tap when there's a burst water-main two streets away.

I've seen these holes with different shapes, I think they were to create fancy vignettes - in which case they would have been in front of the lens, not at the aperture position.

Waterhouse stops remained in use in copying/process cameras for half-tone plate preparation. Right up until all pre-press work was computerised.

 

A suitable shape and orientation of a square aperture, for example, could improve the dot definition and shape.

Edited January 11, 2021 by rodeo_joe|1

As we wander as we wonder, I would only point out that one use of the DDR Meyer Domiplan lenses -- most of which have failed aperture mechanisms --is to take out the useless blades altogether and put in a waterhouse stop of some regular or exotic shape.

 

Meyer Domiplan with its original, non-functional aperture:

As noted already, reassembling aperture leaves is a 'delightful' occupation for a snowy winter night. Keep you busy, potentially, anyhow, for hours.

Edited January 11, 2021 by JDMvW

What are those variable adjusted diaphragms I've seen? What are they used for.

Greater than any mechanical shutter's deviation from its marked value?

Or greater than the setting accuracy of a non-click-stopped aperture. Or even the repeat accuracy of an automatic stop-down mechanism?

 

Sometimes people worry about a dripping tap when there's a burst water-main two streets away.

 

Waterhouse stops remained in use in copying/process cameras for half-tone plate preparation. Right up until all pre-press work was computerised.

 

A suitable shape and orientation of a square aperture, for example, could improve the dot definition and shape.

 

Reminds me of flawed Thatcherite logic: radiation leaking from a nuclear plant wasn't a problem, she once said, because the level of leaked radiation wasn't significantly higher than natural background radiation.

 

The flaw is that it adds up, Cowboy.

So yes, even if (!) the level of error would be below that of mechanical deviations in aperture and shutter, it makes matters worse.

 

Uncoated glass reflects upto 4%. That means upto 7.7% loss in a simple lens. In the much copied original Planar design, for instance, you lose about 33%. A significant amount. Even without additional effects introduced by who knows what else.

The 1896 Planar is a rather simple lens (6 elements in 4 groups, 10 surfaces for light to bounce off), compared to what we see in later lenses (for instance 17 elements in 13 groups in a 'kit' zoom lens. 29 surfaces for light to pass through). So yes, it's a justified concern.

Edited January 12, 2021 by q.g._de_bakker

Greater than any mechanical shutter's deviation from its marked value?

Absolutely. The Angenieux 8x8B (8-64/1.9) on my Beaulieu 4008ZM was t/3.3.

The flaw is that it adds up

Or cancels out!

Uncoated glass reflects upto 4%. That means up to 7.7% loss in a simple lens.

Totally irrelevant. We're not living in the 1930s any longer. All lenses are multicoated these days, and TTL metering takes lens losses into account automatically.

 

A modern f/2 lens has a T-stop rating of about T-2.2, and I'd like to see the mechanical shutter that stayed within +/- 21% of its marked speed across the whole range of speeds.

 

And all of this pointless argument is totally off topic. Pair of cowboys!

Edited January 12, 2021 by rodeo_joe|1

Interesting conversation. Some of this must still be... at the very least, something that somebody somewhere cares a teensy little bit about. Although perhaps not so much, considering the basement bargain sales prices on these (and other) shaped-aperture "gobo" style lens inserts from Lomo?

 

New Petzval 85mm Special Aperture Plates – Season 2

It occurred to me to wonder why they were made with so many blades, at a time when (as far as I remember) no-one cared a toss about Bokeh. They must have been very expensive and labour intensive to make.

 

It is hard to believe in our modern, micro-managed, marketing-led culture, but there was a time when "cost of mfr" and monetary exchange rates were not the driving forces behind design of everything. In the not-quite-distant past, all photographic equipment above box camera level was routinely very expensive: there wasn't any competitive advantage in cutting corners because nobody was offering any bargains. So it just didn't occur to European optics firms that "gee, we could shave costs off this lens by dropping back to a hexagonal iris" (it might even have been more costly to alter the production line for fewer blades). The tradition passed down from large format and motion picture lenses was a circular multi-blade iris, so that simply remained the standard across smaller formats until automatic stop down became a popular, then required, function.

 

The exploding popularity of sophisticated Japanese SLRs like the Nikon F and the groundbreaking medium format Hasselblad 500c upended the old "Keebler Elves" mfrg ethos. More mechanical automation necessitated fewer aperture blades (tho Hasselblad and Zeiss really should have rethought their stark drop to just 5 blades: even Nikon was still at 7 to 9 blades with the early F-mount Nikkors). By the late 1950s lens iris designs (below press camera format) scattered to the winds: you could encounter anything from 7 to 16 blades in just one common preset Zeiss lens for Exakta. Questionable shapes like sawtooth and asterisk went in and out of production.

 

There were always some photographers and art directors aware of what we now call "bokeh". They would choose lenses appropriate to different assignments, with some lenses developing a reputation for their distinctive "look". But it wasn't as discussed or obsessed over as it is now. Lucky for us, the all-metal, little to no plastic construction materials available back then allowed many excellent old optics to survive for our enjoyment today.

Edited January 20, 2021 by orsetto