Leica R - what am I missing?

[JParker]

@miha
@Film-Niko
@Mark J
@Henning Serger

Thank you very much for all the detailed info, links, technical explanations, background info and test results.
That's all very interesting and extremely helpful. I am considering upgrading some of my lenses, and you have helped me a lot with my decisions .

 

[Mark J]

Glad to be of service !

 

[Helge]

And here another example: Portrait lenses.
You probably all know Steve McCurry's most famous picture "Afghan Girl". It was shot with the Nikkor 2.5/105 in its latest optical version from the early seventies.
This lens was my first portrait lens. Nikonians know that this is indeed a very good lens with an excellent reputation, with good performance at open aperture, and very good performance in the f4 to f8 range (as with all 35mm format lenses from f11 on diffraction is reducing the performance visibly).

So now here a quick comparison at open aperture between this Nikkor 2.5/105 AI-S, and the current Zeiss Milvus Planar 1.4/85 ZF.2.
Very important:
Both lenses used at their max. open aperture: The Nikkor at f2.5, and the Zeiss at f1.4
The Zeiss at f1.4 is even better than the Nikkor at f2.5 and f2.8 (the performance difference of f2.5 to f2.8 with the Nikkor is minimal and negligible).
So you get a two-stop performance advantage with the Zeiss. On film this difference is significantly bigger and more pronounced.
The 12 MP sensor is the limiting factor here.

1) Nikkor AI-S 2.5/105 at f2.5
2) Zeiss Milvus Planar 1.4/85 ZF.2 at f.1.4

Best regards,
Henning

Don’t know if this is a well known article, but I’ll post it anyway since it seems pertinent to the overall subject.

Historical perspective on Minolta lens design philosophy​

by David Kilpatrick ©2007

I may have posted some earlier thoughts on Minolta, Leica and Zeiss. Most comments about Minolta glass refer to lenses made pre-mid-1990s, because they changed their philosophy considerably once the Malaysian plant was opened. When Minolta made all their own glass, from the mix to the coating, they did something which no other maker did at the time (mid-1970s to the end of the first generation of AF lenses). They used the lens coatings to balance colour and contrast, so that an entire set of lenses from 7.5mm to 1600mm (originally) would need no CC filters if tested critically on a single roll of film.
This involved allowing some lenses to have less effective multicoating, in order to keep their contrast lower than would have been theoretically possible. Generally, simpler lenses with fewer air to glass surfaces were given single and double coatings on some surfaces and not the total of up to 9 layers which became possible in the 1980s. Zooms were given more efficient coatings to combat their naturally lower contrast. All the coatings were balanced, with the glass types, to give a neutrally matched colour transmission (Minolta's polariser and ND filters were also extremely accurate). We used Minolta colour measurement systems in the mid-1980s, and again in the early 1990s, to check lenses and filters and found the consistency exceptional.
Leica, for whom Minolta made lenses, elements, prisms and focusing screens (nothing to do with the G series which came long after Leica and Minolta ceased to work together) had never attempted to match colour or contrast and you will find radical differences between (for example) a six-element and seven-element Summicron. What they attempted to do was balance microcontrast and overall contrast (boosting MTF figure finer than 60 cycles at the expense of the important 10-30 cycles range). They also taught this concept to Minolta. It tends to produce a 'liquid, three-dimensional' look because overall tones are quite soft, but textures and surfaces are rendered far better.
Zeiss went in a different direction and picked a cutoff point for MTF, using equipment able to measure up to 400 cycles per mm (beyond the resolving power of any film, and theoretically unusable). They would decide that a particular range of lenses should maintain 60 per cent contrast at 80 cycles - or whatever - and then work like hell on the glass, the design, the coatings to achieve this target and never fail. They also tested each individual lens (in Germany) and retained a certificate against its serial number stating the actual figure for that one lens. Then, if returned for repair, they could instantly spot whether elements had become decentered. I do not believe the Kyocera-Zeiss team ever did this! All Hasselblad lenses were tested. Any lens which fell short was sent back for reassembly, or in the worst case, scrapped. The cycles per mm depended on focal length - tele lenses were not expected to reach the same figure as standard or macro lenses. Since retrofocus lenses naturally have extremely high central resolution, their focus with these was maintaining the edge MTF.
In pursuit of this, Zeiss actually ended up with rather variable contrast, usually as high as the design would permit, and used the T* coating thoroughly. They did not get absolutely consistent colour transmission because they used the coating to maximimise certified performance (chart tests) and not to balance colour. But the coating was so effective they often got high microcontrast plus high overall contrast, when Leitz was claiming the two functions were traded against each other.
Since the mid-1990s all you have is the heritage of these policies. Minolta started using outside sources, set up new factories, introduced cheap kit lenses which don't entirely match the range; Leica started trying to copy Zeiss; Hasselblad discarded Zeiss as sole supplier and went to Fuji, whose 9-layer Electron Beam (Super EBC) is probably the best around along with Pentax and Zeiss. Digital has made it all different, forcing designers to multicoat even the rear surfaces and the glued surfaces of lenses, whether or not this changes the contrast and colour. Avoiding digital sensor flare is now the big challenge. Ideas like 'liquid colour' and 'enhanced textural rendering' are no longer relevant (actually such lenses do worse on digital, as Leica owners have found, and would have REALLY found if Leica had dared to use an AA filter).
One thing I'll bet - when my CZ 16-80 arrives, it will be very different in contrast, microcontrast and colour transmission to my Minolta 24-105mm, which it will replace in daily use.
Canon and Nikon, like Sigma and Tamron, had entirely different targets in mind with lens design (like 'can we make this specification?') and generally their lenses are a real mixture of different qualities. Nikon's six blade iris, like Pentax's five-blader, gave their 1970s lenses that wiry, enhanced sharpness look - precisely what Minolta avoided, if you've ever seen the wonderful circular iris of the 1966 100mm f2 for example, or the extreme of the manual SLR lenses - can't remember how many blades the 135mm f4 had, but it's in the teens.
Consequently Canon owners have no real idea why Minolta owners get so deeply into lens quality (Canon lenses don't have a 'look') while Nikon and Pentax owners often really didn't like the softer, flare-prone Minolta glass. Leica owners of course have always liked Minolta glass, Minolta copied Leica from 1958 onwards and won Leica subcontracting work because they did it so well. Leica in return got the Minolta M-mount rangefinder system, previewed in 1958, shelved for ever. Minolta got Leica input into the design of the SR reflex series.
But in a way it is all irrelevant now as lenses are not made the same way, or to the same targets, today. If you collect vintage Minolta glass - 1970-80s MD/MC, 1985-1990 AF (particularly) you can enjoy the colour and contrast matching which made Minolta unrivalled for audio-visual production (the XE-1 was one of the few camera designed to space exposures perfectly to fit a Wess mount without needing pin registration - if you own one, check it out). It also made Minolta first choice for film-stills shooting and stop-frame animation - no need to test each lens and carefully fit a Wratten CC/LB filter pack and apply an exposure compensation (etc).

Originally posted on Minolta Users Discussion Group

Historical perspective on Minolta lens design philosophy

www.dyxum.com

What do you think about the above Henning?

 

[Hassasin]

@Helge Thanks for posting this, great insight.

 

[Pioneer]

Thanks Helge. Very interesting information.

 

[Henning Serger]

@miha
@Film-Niko
@Mark J
@Henning Serger

Thank you very much for all the detailed info, links, technical explanations, background info and test results.
That's all very interesting and extremely helpful. I am considering upgrading some of my lenses, and you have helped me a lot with my decisions .

You're welcome.

Best regards,
Henning

 

[Sirius Glass]

Don’t know if this is a well known article, but I’ll post it anyway since it seems pertinent to the overall subject.

Historical perspective on Minolta lens design philosophy​

by David Kilpatrick ©2007

I may have posted some earlier thoughts on Minolta, Leica and Zeiss. Most comments about Minolta glass refer to lenses made pre-mid-1990s, because they changed their philosophy considerably once the Malaysian plant was opened. When Minolta made all their own glass, from the mix to the coating, they did something which no other maker did at the time (mid-1970s to the end of the first generation of AF lenses). They used the lens coatings to balance colour and contrast, so that an entire set of lenses from 7.5mm to 1600mm (originally) would need no CC filters if tested critically on a single roll of film.
This involved allowing some lenses to have less effective multicoating, in order to keep their contrast lower than would have been theoretically possible. Generally, simpler lenses with fewer air to glass surfaces were given single and double coatings on some surfaces and not the total of up to 9 layers which became possible in the 1980s. Zooms were given more efficient coatings to combat their naturally lower contrast. All the coatings were balanced, with the glass types, to give a neutrally matched colour transmission (Minolta's polariser and ND filters were also extremely accurate). We used Minolta colour measurement systems in the mid-1980s, and again in the early 1990s, to check lenses and filters and found the consistency exceptional.
Leica, for whom Minolta made lenses, elements, prisms and focusing screens (nothing to do with the G series which came long after Leica and Minolta ceased to work together) had never attempted to match colour or contrast and you will find radical differences between (for example) a six-element and seven-element Summicron. What they attempted to do was balance microcontrast and overall contrast (boosting MTF figure finer than 60 cycles at the expense of the important 10-30 cycles range). They also taught this concept to Minolta. It tends to produce a 'liquid, three-dimensional' look because overall tones are quite soft, but textures and surfaces are rendered far better.
Zeiss went in a different direction and picked a cutoff point for MTF, using equipment able to measure up to 400 cycles per mm (beyond the resolving power of any film, and theoretically unusable). They would decide that a particular range of lenses should maintain 60 per cent contrast at 80 cycles - or whatever - and then work like hell on the glass, the design, the coatings to achieve this target and never fail. They also tested each individual lens (in Germany) and retained a certificate against its serial number stating the actual figure for that one lens. Then, if returned for repair, they could instantly spot whether elements had become decentered. I do not believe the Kyocera-Zeiss team ever did this! All Hasselblad lenses were tested. Any lens which fell short was sent back for reassembly, or in the worst case, scrapped. The cycles per mm depended on focal length - tele lenses were not expected to reach the same figure as standard or macro lenses. Since retrofocus lenses naturally have extremely high central resolution, their focus with these was maintaining the edge MTF.
In pursuit of this, Zeiss actually ended up with rather variable contrast, usually as high as the design would permit, and used the T* coating thoroughly. They did not get absolutely consistent colour transmission because they used the coating to maximimise certified performance (chart tests) and not to balance colour. But the coating was so effective they often got high microcontrast plus high overall contrast, when Leitz was claiming the two functions were traded against each other.
Since the mid-1990s all you have is the heritage of these policies. Minolta started using outside sources, set up new factories, introduced cheap kit lenses which don't entirely match the range; Leica started trying to copy Zeiss; Hasselblad discarded Zeiss as sole supplier and went to Fuji, whose 9-layer Electron Beam (Super EBC) is probably the best around along with Pentax and Zeiss. Digital has made it all different, forcing designers to multicoat even the rear surfaces and the glued surfaces of lenses, whether or not this changes the contrast and colour. Avoiding digital sensor flare is now the big challenge. Ideas like 'liquid colour' and 'enhanced textural rendering' are no longer relevant (actually such lenses do worse on digital, as Leica owners have found, and would have REALLY found if Leica had dared to use an AA filter).
One thing I'll bet - when my CZ 16-80 arrives, it will be very different in contrast, microcontrast and colour transmission to my Minolta 24-105mm, which it will replace in daily use.
Canon and Nikon, like Sigma and Tamron, had entirely different targets in mind with lens design (like 'can we make this specification?') and generally their lenses are a real mixture of different qualities. Nikon's six blade iris, like Pentax's five-blader, gave their 1970s lenses that wiry, enhanced sharpness look - precisely what Minolta avoided, if you've ever seen the wonderful circular iris of the 1966 100mm f2 for example, or the extreme of the manual SLR lenses - can't remember how many blades the 135mm f4 had, but it's in the teens.
Consequently Canon owners have no real idea why Minolta owners get so deeply into lens quality (Canon lenses don't have a 'look') while Nikon and Pentax owners often really didn't like the softer, flare-prone Minolta glass. Leica owners of course have always liked Minolta glass, Minolta copied Leica from 1958 onwards and won Leica subcontracting work because they did it so well. Leica in return got the Minolta M-mount rangefinder system, previewed in 1958, shelved for ever. Minolta got Leica input into the design of the SR reflex series.
But in a way it is all irrelevant now as lenses are not made the same way, or to the same targets, today. If you collect vintage Minolta glass - 1970-80s MD/MC, 1985-1990 AF (particularly) you can enjoy the colour and contrast matching which made Minolta unrivalled for audio-visual production (the XE-1 was one of the few camera designed to space exposures perfectly to fit a Wess mount without needing pin registration - if you own one, check it out). It also made Minolta first choice for film-stills shooting and stop-frame animation - no need to test each lens and carefully fit a Wratten CC/LB filter pack and apply an exposure compensation (etc).

Originally posted on Minolta Users Discussion Group

Historical perspective on Minolta lens design philosophy

www.dyxum.com

What do you think about the above Henning?

Thank you for the detailed history.

 

[Mark J]

Thanks, very interesting.
David Kilpatrick is an excellent man with lots of industry knowledge, I bought his Pro magazine for several years in the 90's 2000's.
The info about the Minolta coatings confirms what I'd suspected from owning a few MC & early MD lenses. I had noticed that their coatings were more disparate colours than other lenses of the time and wondered if this was to achieve a certain colour balance. So in this regard they are like cine lenses.

The other discussion I'd like to read through thoroughly again. Some of it seems correct, but some of the comment about how Zeiss and Leica approached their designs in terms of MTF seems incorrect, in fact opposite to what I believe.

 

[Henning Serger]

Don’t know if this is a well known article, but I’ll post it anyway since it seems pertinent to the overall subject.

.........​

What do you think about the above Henning?

Hello Helge

my assessment ist that it is a bit too simplifying, and too generalized (and partly not correct).
The topic is much more complex.
As someone who is working scientifically for more than 30 years, and has worked as a researcher at University for several years, I prefer to differentiate.
And my experience in lens testing over the decades has taught me to avoid simplifying statements as "lenses from brand X behave generally like that, and those from brand Y generally like that". Because in real life it is more complex. In most lens lines from different manufacturers you will find lenses with quite different performances and characters.

Nevertheless there were / are of course some design philosophies of certain lens manufacturers:
E.g. Zeiss and Leica have always tried to make top-class, premium lenses. As good as technically possible. But also at a (much) higher price.
The potential customer base was / is logically much smaller.
In contrast to that approach Canon, Nikon, Minolta, Pentax, Yashica, Ricoh etc. had / have the mass market as main target market. So their lenses have to be mostly affordable for the "average" customer. That had / has significant implications concerning lens design options. You are much more limited because of cost reasons.
But despite that these mainly mass market manufacturers also had and have absolut top-quality / premium lenses in their lines: Mainly the super-telephoto lenses in the 300-800mm range. As these lenses were mostly bought by either professionals, or wealthy enthusiasts, the manufacturers were able to offer these niche products in low(er) production numbers and at the needed very high prices.

I won't get into all the details of that article. But just one point:
Quote:
"In pursuit of this, Zeiss actually ended up with rather variable contrast, usually as high as the design would permit, and used the T* coating thoroughly. They did not get absolutely consistent colour transmission because they used the coating to maximimise certified performance (chart tests) and not to balance colour. But the coating was so effective they often got high microcontrast plus high overall contrast, when Leitz was claiming the two functions were traded against each other."

Concerning the coating of Zeiss lenses: The current Otus and Milvus lenses offer the same, identical colour transmission and colour rendition over the complete line. So if you change from one Zeiss lens to a different one during the same roll of colour reversal film, you will not see a difference in colour rendition. I really like that feature.
It is absolutely correct that the combination of high overall contrast and high microcontrast is possible. All current Zeiss Otus and Milvus lenses demonstrate that perfectly. And meanwhile other manufacturers offer a similar performance with some of their lenses, too. There has been a real, very visible progress in the last years in that regard.

Concerning influences on lens design and lens design history I want to recommend the very detailed, long-term running series "Nikkor - The Thousand and One Nights":

NIKKOR - The Thousand and One Nights | Information | Nikon Consumer

Then you also will realize that the lens designer as a person - an individual with a certain approach as a photographer - has also a very great influence on the final result of a new lens.
Just recently in an interview the owner and CEO of Sigma also talked about exactly that, the great influence and importance of the designer as a person.

And we should not forget the influence of technology and market demands on lens design:
During the last 30-35 years we've seen very significant and often huge improvements in lens design. Besides the progress in computers and lens calculation methods, production improvements (e.g. for aspherical lenses), progress in glass, progress in coating technology etc. there are mainly two catalysts, two paradigma changes which pushed lens design significantly forward by wishes / demand of photographers:

1. Great improvements in film technology from second half of the eigthies on: T-Max, Technical Pan, Delta 100, APX 25, Ektar 25, Velvia, Provia set new, very high quality standards. And the better the lens, the more these outstanding films can shine.

Example for the reaction to the improved films by Nikon at that time: In one of my posts above I've talked about the famous Nikkor 2.5/105 AI / AI-S.
In 1988 Nikon introduced the Nikkor AF 1.8/85. Which already slightly surpassed the Nikkor 2.5/105 at f.2.5, f.2.8 and f4.
Not a huge, but a visible improvement.
And in 1993 Nikon introduced the AF-D Nikkor 2.0/105 DC. A dream lens, one of the best lenses Nikon has ever built. It significantly surpassed both above mentioned Nikkors at all apertures. And with the DC feature also a very nice artistic tool was offered in addition.

2. Different behaviour of digital amateur photographers: With the "digital revolution" "pixel-peeping" became not only a trend, but a kind of standard behaviour: People are sitting at their computer screens, zooming in the pictures at 100% or more, and want top-sharp results even in the far corners of the frame.
The lens manufacturers needed some years to realize the changed behaviour, but then reacted with further significantly improved lenses (from about 2008 on). Also because around that time 20+ MP sensors were introduced, which was a kind of "invitation" for even more "zooming in" and pixel-peeping. And with the 36 MP, 45, 50 and 60 MP FF sensors the technical need for improved lenses increased mainly for the parameters open aperture performance and more even performance to the edges.

And from these latest improvements we film photographers benefit as well, and often even more than the digital photographers. Because now we can use slower, higher quality film in much more situations (as described in this thread).

Best regards,
Henning

 

[Helge]

Hello Helge

my assessment ist that it is a bit too simplifying, and too generalized (and partly not correct).
The topic is much more complex.
As someone who is working scientifically for more than 30 years, and has worked as a researcher at University for several years, I prefer to differentiate.
And my experience in lens testing over the decades has taught me to avoid simplifying statements as "lenses from brand X behave generally like that, and those from brand Y generally like that". Because in real life it is more complex. In most lens lines from different manufacturers you will find lenses with quite different performances and characters.

Nevertheless there were / are of course some design philosophies of certain lens manufacturers:
E.g. Zeiss and Leica have always tried to make top-class, premium lenses. As good as technically possible. But also at a (much) higher price.
The potential customer base was / is logically much smaller.
In contrast to that approach Canon, Nikon, Minolta, Pentax, Yashica, Ricoh etc. had / have the mass market as main target market. So their lenses have to be mostly affordable for the "average" customer. That had / has significant implications concerning lens design options. You are much more limited because of cost reasons.
But despite that these mainly mass market manufacturers also had and have absolut top-quality / premium lenses in their lines: Mainly the super-telephoto lenses in the 300-800mm range. As these lenses were mostly bought by either professionals, or wealthy enthusiasts, the manufacturers were able to offer these niche products in low(er) production numbers and at the needed very high prices.

I won't get into all the details of that article. But just one point:
Quote:
"In pursuit of this, Zeiss actually ended up with rather variable contrast, usually as high as the design would permit, and used the T* coating thoroughly. They did not get absolutely consistent colour transmission because they used the coating to maximimise certified performance (chart tests) and not to balance colour. But the coating was so effective they often got high microcontrast plus high overall contrast, when Leitz was claiming the two functions were traded against each other."

Concerning the coating of Zeiss lenses: The current Otus and Milvus lenses offer the same, identical colour transmission and colour rendition over the complete line. So if you change from one Zeiss lens to a different one during the same roll of colour reversal film, you will not see a difference in colour rendition. I really like that feature.
It is absolutely correct that the combination of high overall contrast and high microcontrast is possible. All current Zeiss Otus and Milvus lenses demonstrate that perfectly. And meanwhile other manufacturers offer a similar performance with some of their lenses, too. There has been a real, very visible progress in the last years in that regard.

Concerning influences on lens design and lens design history I want to recommend the very detailed, long-term running series "Nikkor - The Thousand and One Nights":

NIKKOR - The Thousand and One Nights | Information | Nikon Consumer

Then you also will realize that the lens designer as a person - an individual with a certain approach as a photographer - has also a very great influence on the final result of a new lens.
Just recently in an interview the owner and CEO of Sigma also talked about exactly that, the great influence and importance of the designer as a person.

And we should not forget the influence of technology and market demands on lens design:
During the last 30-35 years we've seen very significant and often huge improvements in lens design. Besides the progress in computers and lens calculation methods, production improvements (e.g. for aspherical lenses), progress in glass, progress in coating technology etc. there are mainly two catalysts, two paradigma changes which pushed lens design significantly forward by wishes / demand of photographers:

1. Great improvements in film technology from second half of the eigthies on: T-Max, Technical Pan, Delta 100, APX 25, Ektar 25, Velvia, Provia set new, very high quality standards. And the better the lens, the more these outstanding films can shine.

Example for the reaction to the improved films by Nikon at that time: In one of my posts above I've talked about the famous Nikkor 2.5/105 AI / AI-S.
In 1988 Nikon introduced the Nikkor AF 1.8/85. Which already slightly surpassed the Nikkor 2.5/105 at f.2.5, f.2.8 and f4.
Not a huge, but a visible improvement.
And in 1993 Nikon introduced the AF-D Nikkor 2.0/105 DC. A dream lens, one of the best lenses Nikon has ever built. It significantly surpassed both above mentioned Nikkors at all apertures. And with the DC feature also a very nice artistic tool was offered in addition.

2. Different behaviour of digital amateur photographers: With the "digital revolution" "pixel-peeping" became not only a trend, but a kind of standard behaviour: People are sitting at their computer screens, zooming in the pictures at 100% or more, and want top-sharp results even in the far corners of the frame.
The lens manufacturers needed some years to realize the changed behaviour, but then reacted with further significantly improved lenses (from about 2008 on). Also because around that time 20+ MP sensors were introduced, which was a kind of "invitation" for even more "zooming in" and pixel-peeping. And with the 36 MP, 45, 50 and 60 MP FF sensors the technical need for improved lenses increased mainly for the parameters open aperture performance and more even performance to the edges.

And from these latest improvements we film photographers benefit as well, and often even more than the digital photographers. Because now we can use slower, higher quality film in much more situations (as described in this thread).

Best regards,
Henning

Wow, thanks Henning! You provided a comprehensive and fascinating response—much appreciated!
I've thoroughly enjoyed the Thousand and One Nights series and highly recommend it.

It's intriguing how most photos remain digital, yet photographers prioritize resolution, while tonality, and colour is baked in and left to post editing.
Optimal resolution only truly shines when printing (or projection when we are talking film).

Modern lenses offer superior resolution and a flatter field but at the cost of size, weight and price. No two ways about it.
Yet lenses like Nikkor 50 1.8 and Rokkor 85 2.0 has a lot of elegance in their smallness and simple optical formula and still has remarkable image quality.

Indeed, simplifying concepts with a low pass filter, while acknowledging their complexity holds value, provided it's still overall truthful of course.

If anyone is interested, here's a link to more stories by D. Kilpatrick. Here only linked as to not overwhelm the discussion with quoted articles.
Be sure to check the comments too.

Minolta's Full Circle

The Friedman Archives Blog - photography wisdom and knowledge from the photographer at www.FriedmanArchives.com

friedmanarchives.blogspot.com

 

[brbo]

Modern lenses offer superior resolution and a flatter field but at the cost of size, weight and price. No two ways about it.

Yes. And those that don't want to go big and heavy almost as a rule compromise on distortion which is then fixed in firmware/software. Since we are in 100% analogue section, those lenses are pretty useless for (film) photography.

I'm glad that some are enjoying the "huge" advancement in modern lenses. I'm not one of them, sadly. Either the new lenses are too big or too heavy (Sigma, Zeiss) or won't work on film bodies (Nikon, Sigma) or their ergonomics are compromised (Voigtländer) because they need to (primarily) cater to the digital market... I think that Leica M line is an exception to the rule, but there is an obvious reason why I won't be enjoying their superb new lenses on my film bodies.

 

[Hassasin]

And then there was photography, making actual images, enjoying the process and results, taken with anything that projects an image on a receiving medium, made from the beginning of photography as history knows it.

In all this, the simple answer to OP's question is: chasing ghosts or enjoying photography? If you are not sure R-series can give you what you need, buy one and try it. Nobody can do it for you.

I like the bodies (R5 to R7), I don't particularly like the shutter release delay, but got used to it over time. Outside of that one detail, bodies have very solid feel, much more refined than Minolta XD line (I cherish everything Minolta made, not so much the XD). R lenses are well built, even if it comes at a price. But I see virtually no resulting difference to Canon FD or Minolta Rokkors or Pentax SMC. Of course I do not study MTF charts or shoot test targets.

The thing is that the more one spends the more one is going to see, or "see". It's hard to be objective when price difference cannot be forgotten.

I am not doubting technicalities, design philosophies, glass compositions having an impact on final rendering. I am doubting everyone who delves that deep into evaluating an actual final image. Blind viewing tests would prove that as not much more than a coffee time gossip.

 

[Helge]

Yes. And those that don't want to go big and heavy almost as a rule compromise on distortion which is then fixed in firmware/software. Since we are in 100% analogue section, those lenses are pretty useless for (film) photography.

I'm glad that some are enjoying the "huge" advancement in modern lenses. I'm not one of them, sadly. Either the new lenses are too big or too heavy (Sigma, Zeiss) or won't work on film bodies (Nikon, Sigma) or their ergonomics are compromised (Voigtländer) because they need to (primarily) cater to the digital market... I think that Leica M line is an exception to the rule, but there is an obvious reason why I won't be enjoying their superb new lenses on my film bodies.

As touched on in Kilpatricks piece, digital has some requirements film does not benefit from and that complicates lenses.
- Though not quite as big a problem as it used to be telecentricity/collimated light is still optimal for a CMOS or CCD sensor. Not so with film. Often SLR lenses are partway there, due to the mirror box, but it’s not film that is mandating it.
- As mentioned by Kilpatrick, sensors are mirror like in their reflectance and as such sets higher requirements for internal coating, baffling and flogging due to higher back reflection.
- As I recently learned, due to micro lens arrays and the geometric shape of the pixels, sensor equipped cameras run into diffraction effects before film. That is, you can stop down further with film before diffraction limits resolution.

While the last point does not affect lens size directly it can affect your need and want for compensating tech, like vibration compensation and use of ND filters.

 

[Helge]

And then there was photography, making actual images, enjoying the process and results, taken with anything that projects an image on a receiving medium, made from the beginning of photography as history knows it.

In all this, the simple answer to OP's question is: chasing ghosts or enjoying photography? If you are not sure R-series can give you what you need, buy one and try it. Nobody can do it for you.

I like the bodies (R5 to R7), I don't particularly like the shutter release delay, but got used to it over time. Outside of that one detail, bodies have very solid feel, much more refined than Minolta XD line (I cherish everything Minolta made, not so much the XD). R lenses are well built, even if it comes at a price. But I see virtually no resulting difference to Canon FD or Minolta Rokkors or Pentax SMC. Of course I do not study MTF charts or shoot test targets.

The thing is that the more one spends the more one is going to see, or "see". It's hard to be objective when price difference cannot be forgotten.

I am not doubting technicalities, design philosophies, glass compositions having an impact on final rendering. I am doubting everyone who delves that deep into evaluating an actual final image. Blind viewing tests would prove that as not much more than a coffee time gossip.

That is the relativism Henning warned against.

Double blind tests (ABX for those who want to Google along at home) are frought with their own set of difficulties and fundamental problems.
Ask audio guys about that.
They are not the end all be all test, that they are popularly perceived as.

 

[removedacct3]

... In all this, the simple answer to OP's question is: chasing ghosts or enjoying photography? If you are not sure R-series can give you what you need, buy one and try it. Nobody can do it for you ...

Thank you very much for referring to my original question. At the beginning of this thread I concluded for myself that R lenses are not worth it. Perhaps they are way, way better than my collection of first generation Pentax K mount lenses or my Hexanons, perhaps not. The alleged difference in optical qualities and the better build quality do not warrant to invest in new camera and pricey first R lens. I do not think I am missing out on R lenses.

However, I did enjoy the remainder of this thread. At times a bit too fierce for my liking, but nevertheless enjoyable and quite informative. Thank you all for your contributions!

 

[Henning Serger]

Wow, thanks Henning! You provided a comprehensive and fascinating response—much appreciated!

"Keine Ursache, gern geschehen", as we say in Germany .

I've thoroughly enjoyed the Thousand and One Nights series and highly recommend it.

Yes, it is really a very interesting and informative series.

It's intriguing how most photos remain digital, yet photographers prioritize resolution, while tonality, and colour is baked in and left to post editing.
Optimal resolution only truly shines when printing (or projection when we are talking film).

Yes, as film photographers we get by far the highest resolution and full detail rendition in optical printing and projection.
In comparison, scanners (and digital camera scanning) are not competitive at all. The difference is big.
I have presented the test results here on photrio some time ago.

Nevertheless you will also see lots of the advantages of the modern, improved lenses in scanned results: E.g. Better open aperture performance and slightly stopped down, more even performance across the frame, better flare control, nicer bokeh, less distortion, no focus-shift, improved colour transmission, 3d-Pop can also be seen in the results of a good scanner.

Modern lenses offer superior resolution and a flatter field but at the cost of size, weight and price. No two ways about it.

As I have explained in one of my posts above: If you want to have very small and light SLR lenses, you have to make compromises: Either in
optical quality, or in Lichtstärke / max. open aperture / speed of the lens.
But, and that is extremely important:
That the new, better lens isn't as very small and lightweight as the old very small one isn't a problem at all.
Examples:
1) My Zeiss Makro-Planar 2/50 ZF significantly surpasses my old Nikkor AI-S 1.8/50 long barrel in all relevant optical and mechanical parameters. Yes, the Zeiss is a bit bigger and heavier. But is extremely well balanced on all my Nikon bodies.
The difference in size and weight between these two lenses is absolutely negligible in my daily photography. I simply don't feel it when I am photographing, it does not play any role at all.
But the difference in results I get play a role, because it is clearly visible.

2) My Nikkor AF-D 2/105 DC and Zeiss Milvus Planar 1.4/85 ZF.2 are both bigger and heavier as my old, very compact Nikkor AI-S 2.5/105.
But both also significantly surpass the old Nikkor 2.5/105.
But here again the same is valid as explained under 1): The difference in size does not play a role when these lenses are attached to my F6, F5, F4s, F100 or F90X. It is all very well balanced and handles nicely.
I personally have even made the experience that the bit heavier lenses have a stabilization effect, and allow me to use 1/3 - 1/2 stop slower shutter speeds with still sharp results.
And again, what really counts is the result, the picture. And there I see the difference very clearly.

3) Same as explained under 1) and 2) is valid for my Nikkor AF-S 4/300 D ED, the Sigma Art 1.4/35, Zeiss Milvus 1.4/50, Milvus 2.8/21, Milvus 2/35, Milvus 2/50, Milvus 2/100, Milvus 2/135, Nikkor AF-S 1.8/24 / 1.8/20 / 1.8/28.....and lots of other modern lenses.
The difference in size and weight is negligible, but the results are far better compared to the older lens equivalents.

Yet lenses like Nikkor 50 1.8 and Rokkor 85 2.0 has a lot of elegance in their smallness and simple optical formula and still has remarkable image quality.

My old Nikkor AI-S 1.8/50 long barrel has excellent performance in the f4-f8 range, mainly in the center.
But it cannot compete anymore in most parameters with the newer Nikkor AF-S 1.8/50 G.
And especially it cannot compete anymore with the Zeiss Makro-Planar 2/50 and the Zeiss Distagon 1.4/50. Both are much better at all wider apertures and towards the edges (more even performance across the frame). The Zeiss' also have much better flare control, better colour rendition, nicer bokeh, better mechanics, wheather and dust sealing, the 3d-Pop.
I will never forget the moment when I first had transparencies of identical subjects from the old Nikkor and the Zeiss 2/50 on my lighttable: The better quality of the Zeiss results was visible even without a loupe, without magnification, with the 'naked' eye. I had not expected that.

Best regards,
Henning

 

[Henning Serger]

I am not doubting technicalities, design philosophies, glass compositions having an impact on final rendering. I am doubting everyone who delves that deep into evaluating an actual final image. Blind viewing tests would prove that as not much more than a coffee time gossip.

Let's for a moment ignore that I am running an independent photography test-lab as a permanent side-project , let's take the scientific test results aside.
Let's also ignore that I have only some hundred meters away from my home a camera and lens rental service.

Let's only have a look at my personal photography as a private person and photo enthusiast:
I am not a rich man.
I have to work hard and long before I can spend money on a new Nikon, Sigma or Zeiss lens for example. For me such an expenditure makes only sense / is only worth the hard work for it if
- I really see a clear improvement in the picture
- I have additional benefits like e.g weather-sealing, which increases my possibilities of taking pictures in worse situations, or an artistic feature like the DC function on my AF-D 2/105 Nikkor.

Because I have these significant improvements I have upgraded my lens portfolio over the years (process is still not completely finished yet).
Otherwise I would not have done it.
I am neither a GAS junkie, nor a collector. I only care for real advantages in my daily photography.

Best regards,
Henning

 

[Henning Serger]

Yes. And those that don't want to go big and heavy almost as a rule compromise on distortion which is then fixed in firmware/software. Since we are in 100% analogue section, those lenses are pretty useless for (film) photography.

It depends certainly on how sensible and tolerant the photographer is concerning distortion. As a young photographer I was quite tolerant about it.
But as my skills improved and my experience has grown over the years, that changed completely. Now I have a very low tolerance concerning distortion. Also because most of my film work is either printed in my darkroom or transparencies projected. Therefore no distortion reduction in post-processing possible.

That is also one reason why I prefer certain modern 35mm lenses from Nikon, Sigma and Zeiss: They have no or negligible distortion.

We as film shooters are currently in a very good position, as new lenses with such excellent performance are offererd to us.
Because some manufacturers of lenses for Interchangeable-Lens-Electronic-Viewfinder (EVIL / DSLM) cameras are meanwhile producing lenses with optically quite big distortion, which is then reduced in-camera via software (Sony e.g. is well-known for that approach).

I'm glad that some are enjoying the "huge" advancement in modern lenses. I'm not one of them, sadly. Either the new lenses are too big or too heavy (Sigma, Zeiss)

As explained above in the other post: There are lots of new modern lenses which are definitely not too big or too heavy.
I am only using modern lenses which are all very well balanced on my Nikon bodies (and there are enough options on the market).
And all of these 35mm camera-lens combinations are ligther and more compact than my equivalent medium format camera-lens combinations.
In an era in which hundreds of thousands of photographers are using very big 2.8/24-70 and 2.8/70-200 zooms as standards, I also cannot understand why for example the much more compact Sigma Art 1.4/35, Milvus 2/50, 1.4/50 or 1.4/85, Nikkor 2/105 DC or 1.8/24....etc. are considered as being (too) big.

or won't work on film bodies (Nikon, Sigma)

Fortunately there are enough which do work on film bodies. From ultra-wide to super-telephoto, there is a sufficient selection.

Best regards,
Henning

 

[Craig]

I will never forget the moment when I first had transparencies of identical subjects from the old Nikkor and the Zeiss 2/50 on my lighttable: The better quality of the Zeiss results was visible even without a loupe, without magnification, with the 'naked' eye. I had not expected that.

I find it interesting that you should say that, as the same was said about Leica M lenses in the 90's- "a Leica transparancy will stand out on a light table".

A good friend of mine bought a new M6 and 50 Summicron in the mid 1990's and I had just bought a Nikon F4 with the then current AF 50/ 1.8. So I gave the F4 to my friend along with a few rolls of Kodachrome 25 and asked him to use both cameras and lets compare. Then he did the same to me, and I shot a few rolls.

We each picked a selection of our slides from both cameras and the other person viewed our selection on the light table. Neither of us could tell which came from the Leica and which came from the Nikon.

 

[Henning Serger]

@Craig :

The difference between the Zeiss 2/50 and my old Nikkor 1.8/50 AI-S is very obvious at f2 and f2.8: The Zeiss is much sharper, higher resolving and delivers more contrast.
From f4 onwards the difference is mainly seen towards the edges, e.g. in architecture photography. The performance of the Zeiss is more even acros the whole frame. And when you look at the transparency under a 10x loupe, you also see better sharpness and resolution in the centre with the Zeiss.
What is also easily visible is the different colour rendition: The Nikkor is neutral leaning slightly towards blue, the Zeiss has very saturated colour rendition leaning towards a warmer colour balance.
Here we certainly enter the area of taste and personal preference. I really like the colour rendition of my Zeiss lenses, it matches films like Velvia 50 / 100, Provia 100F, Ektachrome E100 perfectly.
And the Zeiss has a very nice differentiation between the in-focus main subject, and the out-of-focus zones (the so-called "3D-Pop").

Best regards,
Henning

 

[Mark J]

(Kilpatrick)

What (Leica) attempted to do was balance microcontrast and overall contrast (boosting MTF figure finer than 60 cycles at the expense of the important 10-30 cycles range). They also taught this concept to Minolta. It tends to produce a 'liquid, three-dimensional' look because overall tones are quite soft, but textures and surfaces are rendered far better.

I believe that more or less the opposite is true. In the 70's and 80's when lenses were not 'fully corrected', it is clear that there were compromises which gave a 'house style'. I have read comment from Leica, and seen confirmed in many of the R designs, that Leica put more weight on achieving high contrast in the coarser 10cy/mm and 20cy/mm MTF across the field, and in getting good open-aperture performance at full aperture, than did Zeiss. They would often compromise resolution ( 40cy+ ) in the field - I'm talking most about the faster lenses here. The Zeiss correction was more aimed at getting a flatter field and high resolution across the field when stopped-down ; this at the expense of some contrast loss at full aperture. Also, overall contrast is not a function of the fine detail of lens design & MTF - it is determined by how efficient the lens coatings are, and the detail design of lens chamfers, lockrings and treatments - in which Leica were at the forefront till the 1990's, with perhaps Nikon being 2nd ?

Zeiss went in a different direction and picked a cutoff point for MTF, using equipment able to measure up to 400 cycles per mm (beyond the resolving power of any film, and theoretically unusable). They would decide that a particular range of lenses should maintain 60 per cent contrast at 80 cycles - or whatever - and then work like hell on the glass, the design, the coatings to achieve this target and never fail.

This doesn't sound like a workable design philosophy to me and seems more like some nugget of truth that went through a series of Chinese whispers ! Bear in mind that the difference in a Leica MTF balance and a Zeiss MTF balance as I described above, could be merely the difference in half a dozen target values ( numbers ) in the lens optimisation file. The differences are however something that would come from the designer's or the 'house' philosophy on correction.

They also tested each individual lens (in Germany) and retained a certificate against its serial number stating the actual figure for that one lens. Then, if returned for repair, they could instantly spot whether elements had become decentered. I do not believe the Kyocera-Zeiss team ever did this!

This may or may not be true in detail, but you don't need the original performance of a lens example to check it and repair, if it comes back with poor performance. You always have the design MTF curves to refer to, and can do a test covering the whole field ( in the Zeiss machines the lens is rotated on a precision bearing ) to assess what character the errors have. Usually if something has been dropped and a lens element or group is decentred, it will show as increased variability of the higher-frequency MTF around the outer parts of the field.

All Hasselblad lenses were tested. Any lens which fell short was sent back for reassembly, or in the worst case, scrapped. The cycles per mm depended on focal length - tele lenses were not expected to reach the same figure as standard or macro lenses.

Again the reference would always be back to the design MTF, and Leica and Zeiss would be aiming to get the curves to within 5% of the design, or very rarely 10% down at 40cy/mm in the outer fields. In the old tele designs from the 70's, yes the MTF would be lower, but I can guarantee that there was no need to make any allowance like this on the 300mm f/2.8 TSA .... and many other telephotos that were around by the 90's and 2000's, when DK wrote this article.

Since retrofocus lenses naturally have extremely high central resolution, their focus with these was maintaining the edge MTF.

Correct-ish ; the same lens tolerances may be needed, but in a retrofocus lens, the ray pupils are so small that hardly any detrimental effect occurs in the centre of the picture, however the tolerance effects build up rapidly towards the edge of the field, where the rays pass through the steeply curved outer areas of each lens.

In pursuit of this, Zeiss actually ended up with rather variable contrast, usually as high as the design would permit, and used the T* coating thoroughly. They did not get absolutely consistent colour transmission because they used the coating to maximimise certified performance (chart tests) and not to balance colour. But the coating was so effective they often got high microcontrast plus high overall contrast, when Leitz was claiming the two functions were traded against each other.

In the modern era it is easier to get consistent colour balance without the Minolta 'tricks', because the glass ( esp. blue ) transmissions have improved, and the coatings are so much more efficient and flatter across the waveband. Again, I don't fully understand what Leica were saying about coatings determining microcontrast and the trades there. Perhaps there was more roughness in the layers with the 1970's deposition methods, and so the number of layers might be best limited ? I will check with our guys at work on that subject, if I can remember.

ps. sorry I couldn't hack in and replace 'Helge' with 'Kilpatrick' all the way through !

 

[Mark J]

And we should not forget the influence of technology and market demands on lens design:
During the last 30-35 years we've seen very significant and often huge improvements in lens design. Besides the progress in computers and lens calculation methods, production improvements (e.g. for aspherical lenses), progress in glass, progress in coating technology etc. there are mainly two catalysts, two paradigma changes which pushed lens design significantly forward by wishes / demand of photographers:

Thanks for all the excellent comment, Henning, much that I agree with.

Now, on this point above, I would like to say that AR coating design should be considered one of three main drivers that has spurred-on lens design over the years. This became very clear to me as I wrote and refined lens design training material for our department. In the pre-war years, without coatings, there became something of an impasse in lens complexity, where having more than six air/glass surfaces was a distinct trade-off with overall contrast. You can see that Dagors, triplets and Tessars were all still most favoured, and there were many attempts to get more correction using additional cemented surfaces or thick lenses - for examples the Hektor and Sonnar variants.
After the war, once we had Mag-Fluoride coatings, we saw new designs and acceptance of up to 5 or 6 elements - for instance in the double-gauss and plasmat designs. Remember that the double-Gauss had existed as the 'Planar' since before 1900, but could only now be widely used.
However this complexity reached another blockage, even though computer power started to become available, the designers still had to wait a few years until the late-60's and early 70's, when the first multicoatings started to be used, to have free-rein. At this point , we see another major jump in complexity, and in particular the explosion of new retrofocus wide-angle lenses that came through. Minolta were at the forefront at this time, for example with their excellent 17mm f/4.

 

[ph]

Interesting discussions: I am inspired to park my digital for a while and feed some tarnsparency film into my old R8 or R9 body in order to catch the autumn colours on a 6x6foot screen.

-As a postscript to the discussions above I might add that my R optics even though they have a useless ROMchip, remain much smaller and more robust without all manners of electronics for aperture and focussing (however their built in sunshades are too short for my tastes). Those who seek out aspheric elements as a quality sign may rest assured that at least the short R- zoom uses one-

- In order to spite the hunters for modern glass I will also use my ancient Novoflex/staeble made achromatic doublets to snap the same motive as with a "scale correcting" R- Leitz apomacro at the same aperture (light permitting). Newer Zeiss glass will not fit the R8.so that is not an option to compare the ancient 16mm Minolta construsted Leitz mount lens with something newer, but it performed well enough on my Z7 digital last week .

p.

 

[xkaes]

We each picked a selection of our slides from both cameras and the other person viewed our selection on the light table. Neither of us could tell which came from the Leica and which came from the Nikon.

Incoming!!!

 

[Pioneer]

Incoming!!!