Sankyo Kohki Shootout

Bosun Higgs

Senior Member
Messages
1,203
Solutions
2
Reaction score
1,284
This post was prompted by my purchase of a Sankyo Kohki 105mm f2 lens, I already had the 100mm f1.8, but the 105mm was for sale at such an attractive price, I just could not resist ;O)

When I first looked at Sankyo 100mm(ish) lenses I was torn between the 100mm and the 105mm, and could not decide, but in the end the decision was made for me when a defective 100mm lens came up at a very low price. As I knew I could repair the lens, it was a no-brainer.

Strangely, these two nearly identical lenses were produced side by side for quite a while. The 100mm f1.8 was part of SK's "standard" lens range, but the 105mm f2 was part of their "High Speed" lens range, which is odd as it is slower than the 100mm.

Both lenses were made in fixed mount, and later, with interchangeable mounts; Unidapters for the standard lenses such as the 100mm, and special "High Speed" adaptors for lenses like the 105mm. With my lenses the 100mm is a Unidapter (Nikon F in this case), and the 105mm has a fixed Canon FD mount.

Precise dating on these lenses is difficult, but these two are definitely 1960's vintage. The earliest mention of the 105mm known at present is a 1965 EPOI advert for all four of the High Speed Lenses.

Having both lenses, I thought that instead of just describing the new 105mm, I would also do a comparison with the 100mm, and my usual yardstick, the Canon FD 100mm f2.

The Tests.

The test shots were taken with each lens at maximum aperture using the same A7rII body with the same settings. Images were put through ACR with default settings, no correction of vignetting or fringing.

My main interest in lenses this old is the bokeh and rendering, so I took no stopped down shots.

The Lenses.

From L to R, SK 100mm f1.8, SK 105mm f2, Canon FD 100mm f2. All mounted up as used for the tests.
From L to R, SK 100mm f1.8, SK 105mm f2, Canon FD 100mm f2. All mounted up as used for the tests.

The Sankyos were probably made mid 1960's and the Canon was introduced in 1980, this particular lens was made in 1986. The Sankyos were definitely designed by traditional means, but there is a slight chance that the Canon could have benefited from early computer optimisation. Canon is unclear which nFD lens designs had CAD help, apart from the aspherics.

Optical diagrams of the three lenses to the same scale,
Optical diagrams of the three lenses to the same scale,

All three lenses are Ernostar derived designs, the similarities are obvious. This is one of my favourite lens designs and is capable of excellent bokeh results. The two SKs are 5:4 designs and the Canon is a 6:4.

As mentioned above, the 105mm SK is a fixed mount lens (ironically Canon FD in this case), and the 100mm is a Unidapter lens (Nikon F).

The Results.

Contrast levels look very similar in these untweaked images.
Contrast levels look very similar in these untweaked images.

Note the stronger edge emphasis in the 100mm SK's bokeh bubbles.
Note the stronger edge emphasis in the 100mm SK's bokeh bubbles.

Although the 100mm has more blur, the less textured background from the 105mm can sometimes look smoother.
Although the 100mm has more blur, the less textured background from the 105mm can sometimes look smoother.

Background bokeh differences are very apparent here. The texturing inserted in the 100mm bokeh is offsetting its greater blur.
Background bokeh differences are very apparent here. The texturing inserted in the 100mm bokeh is offsetting its greater blur.

Note the strong catseyeing in the Canon pane, the 100mm SK shows the least of all.
Note the strong catseyeing in the Canon pane, the 100mm SK shows the least of all.

The two SK lenses are about even in sharpness, with wide open shooting and the DOF implications of this, it was difficult to tell them apart. The Canon was much sharper, every time, which does make me wonder about the possibility of CAD playing a part there.

Contrast-wise there was little to choose between the three, which is pretty impressive considering the Canon has the benefit of multicoating. The 105mm backgrounds sometimes looked a little less contrasty, but I think that this was probably due the softer bokeh effects (more on this later).

Vignetting was slight on all three lenses and field curvature was not a problem.

The 105mm showed the most fringing, with some subjects this could almost resemble a slight "glow", the 100mm had much less, and in comparison, the Canon seemed to have none.

The Bokeh.

Bokeh bubbles were predictably largest from the 100mm with its aperture advantage. Once again, the Canon had the smallest bubbles, noticeably smaller than the 105mm despite them both being f2 lenses.

Bokeh bubbles cropped from one of the shots above, the size difference is obvious. Despitethe 105mm and Canon both being f2 lenses, the Canon bubbles are mcuh smaller.
Bokeh bubbles cropped from one of the shots above, the size difference is obvious. Despitethe 105mm and Canon both being f2 lenses, the Canon bubbles are mcuh smaller.

The bokeh bubbles from the 100mm were more strongly outlined than those from the 105mm, consequently they tended to stack, whereas the 105mm bubbles tended to blend. This was especially true with small specular light sources.

The bubble differences are reflected in the general bokeh of the lenses. Although the 105mm has smaller bubbles, the soft edges increase the blur effect, and for more diffuse lit areas the two lenses produce similar blur. In fact the edged bubbles of the 100mm can inject quite a bit of texture into the bokeh, and with some subjects, it can seem that it was producing less background blur than the slower 105mm because of this.

The Canon produced less background blur than either of the SKs. It also has quite strongly edged bokeh bubbles, and these inject far more texture into the bokeh than either of the SKs.

The Canon had the most catseyeing, and of the two SKs, the 100mm had the least. I saw no swirl, at any time, from any of these lenses.

The Bottom line.

The two SKs acquitted themselves very well against the Canon which is a 20 year later design and has multicoating. The main deficit was in sharpness, the Canon was clearly superior.

The sharpness difference is relative. These are all old lenses (the SKs are 60 years old!) and a modern computer optimised, 12-plus element, Sigma ART or Zeiss lens, packed with exotic glass types, will make just about any lens from this era look severely lacking.

Bokeh-wise the SKs produce more blur, the 100mm backgrounds show the least detail, but the 105mm can sometimes seem smoother. Background details were much more noticeable in the Canon's output and its rendering was much more textured.

As always, bokeh is largely down to personal taste, if you like smooth, then the SKs will probably suit you best. All of these lenses produce more pleasing bokeh to my eyes than the aforementioned modern glass, which is often described as having "clinical" rendering.

I have not had the two SKs long enough to get a good feel for their output, so I am not going to express a preference for either. If you have been thinking about getting one of these lenses for bokeh purposes, then I hope this small comparison will be of some aid to you in your choice.

Finally, I think that it has been a little unfair on the SK 105mm to have its coverage diluted in a group test, so I will leave you with a few stereos taken with that lens alone.

It can do this:-

9a683479d98a4269ba22fd3a5eeb4091.jpg

9cafd4eb0fc541b791ceb8336aea7994.jpg

a69cfe32655e425b95cced82efba2832.jpg

858d01f002c4465a9b801056cc9132fb.jpg
 

Attachments

  • ec31ed043bf94a87be96e79f6629e60f.jpg.gif
    ec31ed043bf94a87be96e79f6629e60f.jpg.gif
    60.5 KB · Views: 0
Last edited:
Contrast levels look very similar in these untweaked images.
Contrast levels look very similar in these untweaked images.



9a683479d98a4269ba22fd3a5eeb4091.jpg

9cafd4eb0fc541b791ceb8336aea7994.jpg

a69cfe32655e425b95cced82efba2832.jpg

858d01f002c4465a9b801056cc9132fb.jpg
Great looking shots - all really nice lenses for sure! In the direct comparisons I prefer the SK 100 mm, particularly because of the slightly lower contrast. I don't have any experience with any of those lenses, so it's hard to say if there would be any meaningful difference for how I'd use it.

--
Experimenting manual lens enthusiast.
 
These are all short distance shots and do not reflect the real quality of the lenses, in particular the Canon. While these 100-ish lenses are made for portrait (head-and-shoulder) shots and usually the border area is relatively unimportant, the image quality of the Komuras are usually OK to slightly above average and are usually sharp at the center. In general, the 100-ish lenses made by the big five camera makers are better to much better.



Komura 100mm 1:2 is a later version of their 100-ish lens, which is slightly better than the above-average level. I would rather use Canon, Minolta, Nikon, Olympus or Pentax. Even the Soligor may be at the same level or slightly better, if my recollection is right.

The failure of Komura is basically they did not aim high enough. People used to say Komura made cheap Leica lenses, which IMO is very true. So, Sigma, Tamron and even Tokina survived but Komura did not.

CK
 
These are all short distance shots and do not reflect the real quality of the lenses, in particular the Canon. While these 100-ish lenses are made for portrait (head-and-shoulder) shots and usually the border area is relatively unimportant, the image quality of the Komuras are usually OK to slightly above average and are usually sharp at the center. In general, the 100-ish lenses made by the big five camera makers are better to much better.

Komura 100mm 1:2 is a later version of their 100-ish lens, which is slightly better than the above-average level. I would rather use Canon, Minolta, Nikon, Olympus or Pentax. Even the Soligor may be at the same level or slightly better, if my recollection is right.

The failure of Komura is basically they did not aim high enough. People used to say Komura made cheap Leica lenses, which IMO is very true. So, Sigma, Tamron and even Tokina survived but Komura did not.

CK
Do you have any idea of when that Komura 100mm f2 was made?

I know that the 105mm f2 is probably no earlier than 1965 and you mention that the 100mm is later. I had thought that only the auto diaphragm lenses came after.

The shots in my post reflect how I use these lenses, hence the short focus distances. I only share them as there is so little info available on the Komuras, so they may be of help to someone.

I did like the "cheap Leica lenses" quote :O)
 
Thank you for taking the trouble to compare these lenses - the differences are a matter of taste and few of us would have had the lens material or the time to create this set.
 
These are all short distance shots and do not reflect the real quality of the lenses, in particular the Canon. While these 100-ish lenses are made for portrait (head-and-shoulder) shots and usually the border area is relatively unimportant, the image quality of the Komuras are usually OK to slightly above average and are usually sharp at the center. In general, the 100-ish lenses made by the big five camera makers are better to much better.

Komura 100mm 1:2 is a later version of their 100-ish lens, which is slightly better than the above-average level. I would rather use Canon, Minolta, Nikon, Olympus or Pentax. Even the Soligor may be at the same level or slightly better, if my recollection is right.

The failure of Komura is basically they did not aim high enough. People used to say Komura made cheap Leica lenses, which IMO is very true. So, Sigma, Tamron and even Tokina survived but Komura did not.
Looking at the way Komura chopped and changed their lens output from a professional accounting point of view I would suggest that they had more issues with output complexity* than just "cheap Leica".

* Large volume of types and most likely limited production capacity would have them twisting and turning trying to keep "in-stock" faith with their distributors.

Also hence their extreme efforts to make as many lenses possible compatible with as many mount systems as possible.
 
These are all short distance shots and do not reflect the real quality of the lenses, in particular the Canon. While these 100-ish lenses are made for portrait (head-and-shoulder) shots and usually the border area is relatively unimportant, the image quality of the Komuras are usually OK to slightly above average and are usually sharp at the center. In general, the 100-ish lenses made by the big five camera makers are better to much better.

Komura 100mm 1:2 is a later version of their 100-ish lens, which is slightly better than the above-average level. I would rather use Canon, Minolta, Nikon, Olympus or Pentax. Even the Soligor may be at the same level or slightly better, if my recollection is right.

The failure of Komura is basically they did not aim high enough. People used to say Komura made cheap Leica lenses, which IMO is very true. So, Sigma, Tamron and even Tokina survived but Komura did not.
Looking at the way Komura chopped and changed their lens output from a professional accounting point of view I would suggest that they had more issues with output complexity* than just "cheap Leica".

* Large volume of types and most likely limited production capacity would have them twisting and turning trying to keep "in-stock" faith with their distributors.

Also hence their extreme efforts to make as many lenses possible compatible with as many mount systems as possible.
Tom,

This was probably very true. On the other hand, the Tamron T-Mount began its life from the early 1960s until sometime in 1980s T3 appeared (no-Tamron, though), that simple T and T2 mount served millions of photographers. It is so simple, just a M42 x 0.75mm thread. The lenses were designed to have a standard and long enough back focal distance (i.e., the distance from the center of the last glass to the focal plan when focused at infinity) so that a standard size T-mount would fit nearly all lenses rather than different lenses require different mounts like what Komura did. Some of the Komura Uniadapter mount were M39, which suggested that Komura indeed had an intention to make cheap lenses for Leica and its compatible RF cameras.

So I think the optical design team within Komura did not have a long term view and goal, which caused Komura's demise at the beginning just like the Sigma YS mount. In this respect, Tamron's strengths are obvious: (1) the T and T2 mounts' survivability and adaptability, (2) the AdaptAll strategy worked well to provide automatic aperture that the T and T2 do not have, and (3) the aim is rather high but not unreachable. In fact, Tamron's first lens, 135mm 1:4.5 (?) is already good enough although its aperture is small and the size is small too. If my recollection is correct, people criticized Tamron's (and some Olympus) lenses for its high contrast to "hide" its lower resolution. This criticism holds to a large degree but not 100% true. I am satisfied with most early Tamron lenses, in particular the 400mm 1:6.x series, all of which have high enough contrast and also high enough resolutions. The Tamron 400mm 1:6.9 Nestar is an English translation of my original Chinese version, but Google translation sometimes looks funny. Komura had never been able to make a cheap but good lens like the Nestar. What a pity!

CK
 
From L to R, SK 100mm f1.8, SK 105mm f2, Canon FD 100mm f2. All mounted up as used for the tests.

The Sankyos were probably made mid 1960's and the Canon was introduced in 1980, this particular lens was made in 1986. The Sankyos were definitely designed by traditional means, but there is a slight chance that the Canon could have benefited from early computer optimisation. Canon is unclear which nFD lens designs had CAD help, apart from the aspherics.
When the New FD 35-70mm f/4 came out in 1979 Canon made a big thing about how their computerised design processes had cut the time it took to design to six months compared to the several years it took them to design the original ground-breaking FD 35-70mm f/2.8-3.5 S S.C. released in 1973. As the New FD 100mm f/2 came out in 1980 and is 6 elements in 4 groups rather than the 5/5 of the 100mm f/2.8 I suspect computers were involved. They possibly weren't for the FD 55mm f/1.2 AL aspherical lens of 1973 though.
 
These are all short distance shots and do not reflect the real quality of the lenses, in particular the Canon. While these 100-ish lenses are made for portrait (head-and-shoulder) shots and usually the border area is relatively unimportant, the image quality of the Komuras are usually OK to slightly above average and are usually sharp at the center. In general, the 100-ish lenses made by the big five camera makers are better to much better.

Komura 100mm 1:2 is a later version of their 100-ish lens, which is slightly better than the above-average level. I would rather use Canon, Minolta, Nikon, Olympus or Pentax. Even the Soligor may be at the same level or slightly better, if my recollection is right.

The failure of Komura is basically they did not aim high enough. People used to say Komura made cheap Leica lenses, which IMO is very true. So, Sigma, Tamron and even Tokina survived but Komura did not.
Looking at the way Komura chopped and changed their lens output from a professional accounting point of view I would suggest that they had more issues with output complexity* than just "cheap Leica".

* Large volume of types and most likely limited production capacity would have them twisting and turning trying to keep "in-stock" faith with their distributors.

Also hence their extreme efforts to make as many lenses possible compatible with as many mount systems as possible.
Tom,

This was probably very true. On the other hand, the Tamron T-Mount began its life from the early 1960s until sometime in 1980s T3 appeared (no-Tamron, though), that simple T and T2 mount served millions of photographers. It is so simple, just a M42 x 0.75mm thread. The lenses were designed to have a standard and long enough back focal distance (i.e., the distance from the center of the last glass to the focal plan when focused at infinity) so that a standard size T-mount would fit nearly all lenses rather than different lenses require different mounts like what Komura did. Some of the Komura Uniadapter mount were M39, which suggested that Komura indeed had an intention to make cheap lenses for Leica and its compatible RF cameras.

So I think the optical design team within Komura did not have a long term view and goal, which caused Komura's demise at the beginning just like the Sigma YS mount. In this respect, Tamron's strengths are obvious: (1) the T and T2 mounts' survivability and adaptability, (2) the AdaptAll strategy worked well to provide automatic aperture that the T and T2 do not have, and (3) the aim is rather high but not unreachable. In fact, Tamron's first lens, 135mm 1:4.5 (?) is already good enough although its aperture is small and the size is small too. If my recollection is correct, people criticized Tamron's (and some Olympus) lenses for its high contrast to "hide" its lower resolution. This criticism holds to a large degree but not 100% true. I am satisfied with most early Tamron lenses, in particular the 400mm 1:6.x series, all of which have high enough contrast and also high enough resolutions. The Tamron 400mm 1:6.9 Nestar is an English translation of my original Chinese version, but Google translation sometimes looks funny. Komura had never been able to make a cheap but good lens like the Nestar. What a pity!

CK
I hear what you are saying CK but technical prowess does not always equate to business success. Komura management might well have been a muddled mess, let alone well financed, with well marketed product.

I have noticed that they dared to make longer telephoto lenses in LTM (than Leica dared?) with special finders. That they had a reflex housing of their own and that their relatively sparse shorter focal length LTM primes are presently listed for sale at quite impressive prices.

--
Tom Caldwell
 
Last edited:
These are all short distance shots and do not reflect the real quality of the lenses, in particular the Canon. While these 100-ish lenses are made for portrait (head-and-shoulder) shots and usually the border area is relatively unimportant, the image quality of the Komuras are usually OK to slightly above average and are usually sharp at the center. In general, the 100-ish lenses made by the big five camera makers are better to much better.

Komura 100mm 1:2 is a later version of their 100-ish lens, which is slightly better than the above-average level. I would rather use Canon, Minolta, Nikon, Olympus or Pentax. Even the Soligor may be at the same level or slightly better, if my recollection is right.

The failure of Komura is basically they did not aim high enough. People used to say Komura made cheap Leica lenses, which IMO is very true. So, Sigma, Tamron and even Tokina survived but Komura did not.
Looking at the way Komura chopped and changed their lens output from a professional accounting point of view I would suggest that they had more issues with output complexity* than just "cheap Leica".

* Large volume of types and most likely limited production capacity would have them twisting and turning trying to keep "in-stock" faith with their distributors.

Also hence their extreme efforts to make as many lenses possible compatible with as many mount systems as possible.
Tom,

This was probably very true. On the other hand, the Tamron T-Mount began its life from the early 1960s until sometime in 1980s T3 appeared (no-Tamron, though), that simple T and T2 mount served millions of photographers. It is so simple, just a M42 x 0.75mm thread. The lenses were designed to have a standard and long enough back focal distance (i.e., the distance from the center of the last glass to the focal plan when focused at infinity) so that a standard size T-mount would fit nearly all lenses rather than different lenses require different mounts like what Komura did. Some of the Komura Uniadapter mount were M39, which suggested that Komura indeed had an intention to make cheap lenses for Leica and its compatible RF cameras.

So I think the optical design team within Komura did not have a long term view and goal, which caused Komura's demise at the beginning just like the Sigma YS mount. In this respect, Tamron's strengths are obvious: (1) the T and T2 mounts' survivability and adaptability, (2) the AdaptAll strategy worked well to provide automatic aperture that the T and T2 do not have, and (3) the aim is rather high but not unreachable. In fact, Tamron's first lens, 135mm 1:4.5 (?) is already good enough although its aperture is small and the size is small too. If my recollection is correct, people criticized Tamron's (and some Olympus) lenses for its high contrast to "hide" its lower resolution. This criticism holds to a large degree but not 100% true. I am satisfied with most early Tamron lenses, in particular the 400mm 1:6.x series, all of which have high enough contrast and also high enough resolutions. The Tamron 400mm 1:6.9 Nestar is an English translation of my original Chinese version, but Google translation sometimes looks funny. Komura had never been able to make a cheap but good lens like the Nestar. What a pity!

CK
I hear what you are saying CK but technical prowess does not always equate to business success. Komura management might well have been a muddled mess, let alone well financed, with well marketed product.
Without the technical prowess, a company will end up junked very fast. Whether Komura had the technology during its time has a mystery to me because they never made good competitive lenses against Tamron, just name one three-party lens maker.
I have noticed that they dared to make longer telephoto lenses in LTM (than Leica dared?) with special finders. That they had a reflex housing of their own and that their relatively sparse shorter focal length LTM primes are presently listed for sale at quite impressive prices.
This was perhaps the reason for many people said: Komura made cheap Leica lenses. On the other hand, the image quality has always been an issue to me. Because some tele lenses made by Komura has M39 thread to mount a Uniadapter, perhaps without the Uniadapter maybe the lens can be used by Visoflex I. I don't know for sure, though.

In fact, any lens made for any SLR can be mounted to a Leica, which we all know well. Whether these lens can be used properly on a Leica is another matter, focusing, framing, etc. The flange distance of Visoflex II/III is about 68.8mm. If the lens is designed to have long back focal distance, it can be used used on Visoflex II/III. Rarely we saw third party lens makers made Visoflex II/III compatible lenses, perhaps due to the market being not large enough.

A similar company coming to my mind is Petri (Kuribayashi). Once into the SLR business, Petri demised so quickly. They started with M42 in one model and changed to Petri breach mount in the second model and on. Nearly at the end of Petri's existence, they switched back to M42. Isn't it odd?

CK
 
The oem Komura reflex housing that I have dubbed "Komuraflex" has a different registration distance from "Visoflex" (and all the complications associated with the latter).

Discussed here:

https://www.dpreview.com/forums/post/60214837

Petri? Not in my knowledge area. I presume that M42 was common and an easy choice. Brand hubris and perhaps a need to make a "technology point" may have caused the invention of their own mount system. Followed by a lack of understanding by their market and potential market and a need to provide a wide variety of mountable lenses and a last minute attempt to save the business by reverting to M42? Hard to get inside the heads of those that made the decisions.
 

Keyboard shortcuts

Back
Top