Fewer elements - lesser Bokeh?

Gordon Moat wrote in message ...
brian wrote:

Gordon Moat wrote in message ...

There is a mention in the above article that a switch from a Sonnar type design to a Xenotar type design improved
aberration correction. While this makes sense to me, is that what effected the reduction in Coma? Would that reduction
in Coma directly relate to improvements in defocus rendition? Thanks!

Ciao!

Gordon:
Coma correction for a single object distance is a trivial task with
either the Sonnar or double-Gauss (Xenotar) design forms. However,
since the double-Gauss has a greater degree of symmmetry the coma
correction remains stable over a wider magnification range.

Thanks for the awesome reply, I think I am finally understanding these lens types more. Since the Planar type seems to be
quite common in medium format choices, how does that fit in?

The double-Gauss design type is a natural choice when you want high
image quality and reasonable production economy in a lens with medium
to large aperture and moderate field angle. It is markedly superior
to Tessars in virtually all respects. Sonnars tend to be preferred
for longer focal lengths because they are more compact, and they lend
themselves more conveniently to apochromatic correction.


The old
Nikon 135mm f/2 AI/AIS is an example of a Sonnar type design which is
degraded by coma at close focusing distances, but which is very well
corrected for all aberrations at small magnifications.

Brian
www.caldwellphotographic.com

Since you bring up the close focusing issue, it makes me wonder why Nikon ?
did not apply their close range correction (CRC)
idea to the 135 mm. If I remember correctly, one group between the aperture and film plane moved. While the construction of
the 135 mm you mentioned might not allow that to work, I notice that the 85 mm f1.4 did employ CRC. Perhaps cost was an
issue.

I suspect that the Nikon designers felt that the 135/2 AI design was
good enough at the time. It is a very nice lens, and edge performance
degradation due to coma even at full aperture and closest focus is
still not what I would call dreadful.

The later 135/2.0 DC Nikkor is a double-Gauss derivative which does
use CRC to control coma in a manner very similar to the 85/1.4. Note
that alot of the wide angle Nikkors have CRC, but here the motivation
is to keep astigmatism under control while focusing rather than coma.


Ciao!

Gordon Moat
Alliance Graphique Studio
http://www.allgstudio.com

Brian
www.caldwellphotographic.com

jjs wrote:

In article ,
wrote:

Beautiful lens! The engraved information says that it's an f/5 lens,
yet the aperture ring has an f/4 setting, or have I misread something?

Good eye, James. It is more mysterious than that. See this - it is the
right side of the dial: http://wind.winona.edu/~stafford/2.jpg (Sorry four
the lousy underexposed picture, but I just snapped it by the window as
the sun had about set.)

So the lens is marked "F5" in one place and the widest aperture "1,5".
(The barrel and lens are not mismated. They are as much a single part as
one can imagine.) I'm going to take a guess that the "1,5" is European
decimal notation.

No. We Europeans have always used the f-value as it is. I'm not sure
when it was adopted, but possibly already in the 17th century when the
first telescopes were built.

Could the lens be "1,5" under a the manufacturer's convention, but branded
"F5" to follow the American standards by the American distributor?

Stranger and stranger!

Yes...

-- Lassi

Stacey wrote:

Gordon Moat wrote:


Since you bring up the close focusing issue, it makes me wonder why Nikon
did not apply their close range correction (CRC) idea to the 135 mm. If I
remember correctly, one group between the aperture and film plane moved.
While the construction of the 135 mm you mentioned might not allow that to
work, I notice that the 85 mm f1.4 did employ CRC. Perhaps cost was an
issue.


Several high end lenses do have designs (floating elements) where multiple
elements are moved when focusing to help closeup performance. My zuiko 28mm
f2, 50mm f2 and their 90mm f2 macro come to mind. The were all cost over
$700 20 years ago. I'm sure there are others as well made by other
manufacturers.

Cost is exactly why they don't do this more as it makes the mechanical
design of the lens much more complex.

Olympus seems to have solved the cost issue. Their IS series cameras
have an integrated zoom lens that has several lens groups moving around
in all directions, and still the whole camera costs far less than $700.

-- Lassi

In article , Lassi
=?iso-8859-1?Q?Hippel=E4inen?= wrote:

[...] We Europeans have always used the f-value as it is. I'm not sure
when it was adopted, but possibly already in the 17th century when the
first telescopes were built.

Okay, if you say so, but there were other aperture metrics in the USA and
possibly elsewhere. I do not mean the EV marks on some lenses, but
something else entirely. For example, I find it hard to believe the lens
in question really has an F 1:1.5 aperture.

brian wrote:

Gordon Moat wrote in message ...
brian wrote:

Gordon Moat wrote in message ...

There is a mention in the above article that a switch from a Sonnar type design to a Xenotar type design improved
aberration correction. While this makes sense to me, is that what effected the reduction in Coma? Would that reduction
in Coma directly relate to improvements in defocus rendition? Thanks!

Ciao!

Gordon:
Coma correction for a single object distance is a trivial task with
either the Sonnar or double-Gauss (Xenotar) design forms. However,
since the double-Gauss has a greater degree of symmmetry the coma
correction remains stable over a wider magnification range.

Thanks for the awesome reply, I think I am finally understanding these lens types more. Since the Planar type seems to be
quite common in medium format choices, how does that fit in?

The double-Gauss design type is a natural choice when you want high
image quality and reasonable production economy in a lens with medium
to large aperture and moderate field angle. It is markedly superior
to Tessars in virtually all respects. Sonnars tend to be preferred
for longer focal lengths because they are more compact, and they lend
themselves more conveniently to apochromatic correction.

Definitely makes sense then why I see so many of each type in their respective ranges.




The old
Nikon 135mm f/2 AI/AIS is an example of a Sonnar type design which is
degraded by coma at close focusing distances, but which is very well
corrected for all aberrations at small magnifications.

Brian
www.caldwellphotographic.com

Since you bring up the close focusing issue, it makes me wonder why Nikon ?
did not apply their close range correction (CRC)
idea to the 135 mm. If I remember correctly, one group between the aperture and film plane moved. While the construction of
the 135 mm you mentioned might not allow that to work, I notice that the 85 mm f1.4 did employ CRC. Perhaps cost was an
issue.

I suspect that the Nikon designers felt that the 135/2 AI design was
good enough at the time. It is a very nice lens, and edge performance
degradation due to coma even at full aperture and closest focus is
still not what I would call dreadful.

The later 135/2.0 DC Nikkor is a double-Gauss derivative which does
use CRC to control coma in a manner very similar to the 85/1.4.

Considering that even used, the DC version is quite a bit more than the more normal AI/AIS version, it makes one wonder whether
the added cost and complexity really hold many benefits. I have used the 105 mm DC lens once, and it was such a pain (slowing down
the pace of shots too much) that I stuck with the more normal 105 mm F2.5 AIS.

Note
that alot of the wide angle Nikkors have CRC, but here the motivation
is to keep astigmatism under control while focusing rather than coma.

Very interesting differences. Considering the uses to which the different focal lengths would often be used, I can see why there
would be a particular bias in correction. Of course, the more amazing (to me) aspect is that much of that correction work was done
several decades ago. Thanks for the information.

Ciao!

Gordon Moat
Alliance Graphique Studio
http://www.allgstudio.com

Lassi Hippeläinen wrote in message ...


Olympus seems to have solved the cost issue. Their IS series cameras
have an integrated zoom lens that has several lens groups moving around
in all directions, and still the whole camera costs far less than $700.

-- Lassi

Moving groups of elements around is really not very costly in high
volume production. I was once involved in a 3:1 zoom lens project in
which the complete opto-mechanical package cost less than $15.00. The
truth is that mechanical parts are quite cheap. The main difficulty
in my experience is getting a first-class mechanical design.

Brian
www.caldwellphotographic.com

jjs wrote:

In article , Lassi
=?iso-8859-1?Q?Hippel=E4inen?= wrote:

[...] We Europeans have always used the f-value as it is. I'm not sure
when it was adopted, but possibly already in the 17th century when the
first telescopes were built.

Okay, if you say so, but there were other aperture metrics in the USA and
possibly elsewhere. I do not mean the EV marks on some lenses, but
something else entirely. For example, I find it hard to believe the lens
in question really has an F 1:1.5 aperture.

Wasn't the lens made in Kansas :-)

Do you know for certain that the numbers are f-stops? Maybe the maker
had developed a system that indicates how long the exposure time is when
compared to some standard lens. If the reference lens is an f/4, this
lens (f/5) would then need 1.5 times that time when fully open, etc.
Should be testable with an exposure meter.

-- Lassi

nicholas wrote in message ...
brian wrote:
So, my conclusion is that all Xenotar-type lens designs
have this particular rendering...



I'm going to assume that Xenotar means double-Gauss.

Not quite... A Xenotar-type design is 5/4 and not symmetrical as you
have me believe...

_SNIP_

Nicholas:
Double-Gauss designs come in a very wide variety of subtypes, and it
wasn't clear to me whether you were referring to the general design
form or to the particular 5/4 subtype. So, if you want to restrict
the conversation to the 5/4 type, then its OK with me. Xenotars have
a degree of symmetry very similar to other double-Gauss designs.

The fact remains: a 5/4 Xenotar subtype need not have harsh bokeh.
The reason for this is that the design flexibility is nearly as great
as with the more complex double-Gauss forms. A competent lens
designer can vary the character of the defocused background highlights
at will. For example U.S. Patent 2,844,072 by Lowenthal discloses an
f/2.8 Xenotar type lens with moderate undercorrected spherical
aberration. This design will produce beautifully soft-edged defocused
background highlights.

Although Schneider adopted the Xenotar type fairly early, the design
was pioneered by Charles Wynne of Wray Optical back in the 1940's. At
the time it was hailed as a significant achievement since it was
apparently simpler than the similar 6/4 Gauss type. Oddly enough, the
reason that the 5/4 Xenotar design form isn't used more widely is that
it is more difficult to make and hence more expensive than the more
common 6/4 form. The thickness tolerance on the 4th element almost
always turns out to be unusually tight, and this element is also quite
difficult to center.

The formula for the Vivitar Series I 90mm macro lens is given in U.S.
Patent 3,942,875 (example 6 of 6). There are a couple of minor typos
in the prescription, and I can give you the correct version if you're
interested. As you guessed, this design is not a Xenotar derivative
since the rear shell is a cemented doublet. However, changing that
cemented doublet to a singlet is easier than you might imagine.

Brian
www.caldwellphtographic.com