Part 4 is almost ready!
07/06/17 11:00
The text for part 4 of my series of Leica books is almost ready. There are a few printed books of part three (Leica Lens Saga) available. To give an impression of the content of this book, find below a section of the text.
Leica’s optical design team has several constraints to work with when designing lenses for the rangefinder camera. These were listed in the first part of the book. As long as the M-designs have to be compact, they will be mechanical. It is not the rangefinder concept that forces this mechanical design. The Live View option in the current FPA-equipped M-cameras indicates that the mechanical linkage between lens and rangefinder mechanism is not the only option. The inclusion of an autofocus and optical stabilisation function in the compact dimensions of the M-lenses would not be possible. An M-lens can be compared with a T-lens of similar specifications. The Summilux-M 1.4/50 mm ASPH has a length of 52.5 mm and a diameter of 53.5 mm with a weight of 335 grams (+ integrated lens hood). Compare these figures with the Leica Summilux-TL 1.4/35 mm that has a total of 12 elements, and is substantially larger (length: 77 mm; diameter: 70 mm; weight: 498 grams with lens hood).
More elements imply better correction possibilities. The M-lens is also limited in the number of elements, at least for normal (50 mm) and short-tele designs (75 to 90 mm). Possible increases in optical performance have to found in the use of more aspherical surfaces and even more exotic glass types. In a compact design these choices imply even narrower mechanical tolerances than the ones now accepted in the manufacturing and assembly stages.
The previous period during which the flagship lenses for the M-camera were manufactured (1960 to 1980) is characterised by a slow progression in the optical performance and a heavy emphasis on the mechanical qualities of the design. This was also the period when the Japanese manufacturers switched their effort to zoom lenses where weight issues were more prominent and needed to be solved. The Leica lenses were universally acknowledged as having the best mounts of all manufacturers. The Leica lenses of that period were possibly not the best in the imaging sense, but the balance between excellence in mounting and material choice on the one hand and manufacturing cost on the other hand was a very good one. The original Summilux 1.4/50 mm stayed in production for over forty years and the current Summicron 2/50 mm (without the Apo-prefix) is now more that 40 years old. The potential of the Double-Gauss design and its variants on which many Leica lenses from 1935 to 1995 were based was completely exhausted around 1980.
The most recent lens constructions for Leica cameras indicate a major change in de design approach with a shift from opto-mechanical to opto-mechatronic constructions. This shift is mainly inspired by the switch to solid-state FPA-imagers for all Leica cameras (with the exception of the classical cartridge-loading rangefinder body). In the mechatronic world the mechanical connections are replaced by microprocessors and compact drive motors. The focusing movement of the M lens is accomplished by the turning of the focusing ring, that moves the optical cell forwards and backwards. The focusing ring has a rotary movement that has to be transformed into a linear movement by means of a mechanical screw-thread mechanism that connects the focusing ring to the optical cell. The main requirement of the mechanism is the accuracy of the movement and the smoothness of the movement that must be the same over the full extent and must not exhibit any sloppiness. The automatic focusing mechanism requires that the resistance of the focusing mechanism has to reduced and the travel of the components must also be minimised. When using auto-focus mechanisms, the focusing accuracy is now the task of a small precision stepper motor that moves the focusing lens or lens group in very small steps over a 360 degree circular movement. In addition, there must be an electronic interface between camera body and lens mount. The contact strip on the R-lenses for the R8/9 and the black-white lens code on the recent M mounts are very different in function from the ten-pin interface on the new SL camera. The relative masses of the camera body and the lens body are changing with smaller bodies and larger lenses. The advantage of the rangefinder body is the ergonomically correct balance of the total package (lens plus body).
The mechanical tolerances and the accuracy of assembly are for the foreseeable future fixed. The world of the micron, a very small distance in the physical world, sets a limit to what machines can accomplish. Typical tolerances are 0.002 - 0.005 mm for distances and a few arc seconds for decentring. Such a level of precision is difficult to handle and when it is possible, one should be aware that such small dimensions make the design very sensitive.
Everyone who is acquainted with measurement and control techniques knows about the margins of error inherent in all equipment, however accurate. A lens must be useful under conditions of physical stress and accept normal wear and tear. A very sensitive instrument requires frequent adjustments at even smaller intervals of time. The Apo-Summicron 50 mm lens is for the moment the design with the most critical dimensions which explains the high price and the low production volume.
The lenses for the Leica rangefinder cameras are unique in the current landscape of photographic lenses. Leica rangefinder cameras can be classified into two major groups: the screw mount models and the bayonet mount models. The last group has two subgroups: the film-loading cameras and the solid-state FPA-equipped models, also known as digital models. There are a few minor differences between the lenses designed in the film-loading (or analogue) era and the most recent lenses designed in the digital era. The most important differences are (1) the provision of the six-bit coding strip on the backside of the lens (this feature can be used to exchange information between the properties of the lens and the in-camera processing algorithms) and (2) the optical correction (or optimization) of the design to compensate for the small impact that the protective filter in front of the sensor surface has on the path of the light rays. Lenses for the screw mount cameras can be identified by the name of the lens, as was also the case with the earlier series for the bayonet mount cameras (the ‘M’-camera). After the introduction of the Leicaflex, Leitz used the suffix -R or -M to identify lenses for the two ranges of cameras. There is no special identification for the recently introduced or (slightly) redesigned lenses for the digital rangefinder models.
The premium characteristics of the M-lenses are the outstanding performance, the full metal mount, the manual operation and the compact size. The combination of compact size and high image quality is now an unusual one. Recent introductions by Zeiss and by Leica itself (for the SL- and T-cameras) indicate a trend to ever larger physical volumes. The small physical volume limits the optical evolution of the rangefinder designs.
The current designs have already a performance profile that most users have difficulty to fully exploit. At medium apertures the standard lenses are capable of a useful limiting resolution between 100 and 120 lp/mm (measured with the microscope on low speed microfilm, developed in Spur Nanospeed). This translates into a spot of about 0.0042 mm. A 24 x 36 mm sensor would need to have about 40 million pixels to record reliably this high resolution. Presumably 50 million pixels would be required because of the moiré effects at high resolutions. The Leica M-lens however should not be valued only for the high resolution potential, but for the clarity and smoothness of the reproduced details and gradation in the mid-range of spatial frequencies. Any audio aficionado knows that it makes no sense to look only for the high-frequency response of the equipment.
The Leica rangefinder user should however accept that higher levels of performance will be introduced more gradually for M-lenses in the future. The true advances in optical design will be introduced in the modern opto-mechatronic devices, like the Leica T- and SL- systems.
The complexity of photographic lenses has increased substantially in recent times. A high-speed-high-performance lens, like the Zeiss Otus has twelve elements in ten groups. A comparable lens for the Leica T (1.4/35 mm) has also twelve elements (including two aspherical surfaces) in eight groups. A zoom lens foo the Leica SL has 23 elements and seven moving groups. It is nearly impossible to handle this complexity with the help of the Seidel coefficients or the thin lens approach. The additional mechanical and electronic complexity (image stabilization, autofocus movements), including the analysis of the required accuracy and manufacturing tolerances can only be handled with the help of computer programs. The price for this added complexity is size! Lenses for the Leica M CRF are not as complex, but their physical constraints ask much of the designer to create high quality designs.
Leica’s optical design team has several constraints to work with when designing lenses for the rangefinder camera. These were listed in the first part of the book. As long as the M-designs have to be compact, they will be mechanical. It is not the rangefinder concept that forces this mechanical design. The Live View option in the current FPA-equipped M-cameras indicates that the mechanical linkage between lens and rangefinder mechanism is not the only option. The inclusion of an autofocus and optical stabilisation function in the compact dimensions of the M-lenses would not be possible. An M-lens can be compared with a T-lens of similar specifications. The Summilux-M 1.4/50 mm ASPH has a length of 52.5 mm and a diameter of 53.5 mm with a weight of 335 grams (+ integrated lens hood). Compare these figures with the Leica Summilux-TL 1.4/35 mm that has a total of 12 elements, and is substantially larger (length: 77 mm; diameter: 70 mm; weight: 498 grams with lens hood).
More elements imply better correction possibilities. The M-lens is also limited in the number of elements, at least for normal (50 mm) and short-tele designs (75 to 90 mm). Possible increases in optical performance have to found in the use of more aspherical surfaces and even more exotic glass types. In a compact design these choices imply even narrower mechanical tolerances than the ones now accepted in the manufacturing and assembly stages.
The previous period during which the flagship lenses for the M-camera were manufactured (1960 to 1980) is characterised by a slow progression in the optical performance and a heavy emphasis on the mechanical qualities of the design. This was also the period when the Japanese manufacturers switched their effort to zoom lenses where weight issues were more prominent and needed to be solved. The Leica lenses were universally acknowledged as having the best mounts of all manufacturers. The Leica lenses of that period were possibly not the best in the imaging sense, but the balance between excellence in mounting and material choice on the one hand and manufacturing cost on the other hand was a very good one. The original Summilux 1.4/50 mm stayed in production for over forty years and the current Summicron 2/50 mm (without the Apo-prefix) is now more that 40 years old. The potential of the Double-Gauss design and its variants on which many Leica lenses from 1935 to 1995 were based was completely exhausted around 1980.
The most recent lens constructions for Leica cameras indicate a major change in de design approach with a shift from opto-mechanical to opto-mechatronic constructions. This shift is mainly inspired by the switch to solid-state FPA-imagers for all Leica cameras (with the exception of the classical cartridge-loading rangefinder body). In the mechatronic world the mechanical connections are replaced by microprocessors and compact drive motors. The focusing movement of the M lens is accomplished by the turning of the focusing ring, that moves the optical cell forwards and backwards. The focusing ring has a rotary movement that has to be transformed into a linear movement by means of a mechanical screw-thread mechanism that connects the focusing ring to the optical cell. The main requirement of the mechanism is the accuracy of the movement and the smoothness of the movement that must be the same over the full extent and must not exhibit any sloppiness. The automatic focusing mechanism requires that the resistance of the focusing mechanism has to reduced and the travel of the components must also be minimised. When using auto-focus mechanisms, the focusing accuracy is now the task of a small precision stepper motor that moves the focusing lens or lens group in very small steps over a 360 degree circular movement. In addition, there must be an electronic interface between camera body and lens mount. The contact strip on the R-lenses for the R8/9 and the black-white lens code on the recent M mounts are very different in function from the ten-pin interface on the new SL camera. The relative masses of the camera body and the lens body are changing with smaller bodies and larger lenses. The advantage of the rangefinder body is the ergonomically correct balance of the total package (lens plus body).
The mechanical tolerances and the accuracy of assembly are for the foreseeable future fixed. The world of the micron, a very small distance in the physical world, sets a limit to what machines can accomplish. Typical tolerances are 0.002 - 0.005 mm for distances and a few arc seconds for decentring. Such a level of precision is difficult to handle and when it is possible, one should be aware that such small dimensions make the design very sensitive.
Everyone who is acquainted with measurement and control techniques knows about the margins of error inherent in all equipment, however accurate. A lens must be useful under conditions of physical stress and accept normal wear and tear. A very sensitive instrument requires frequent adjustments at even smaller intervals of time. The Apo-Summicron 50 mm lens is for the moment the design with the most critical dimensions which explains the high price and the low production volume.
The lenses for the Leica rangefinder cameras are unique in the current landscape of photographic lenses. Leica rangefinder cameras can be classified into two major groups: the screw mount models and the bayonet mount models. The last group has two subgroups: the film-loading cameras and the solid-state FPA-equipped models, also known as digital models. There are a few minor differences between the lenses designed in the film-loading (or analogue) era and the most recent lenses designed in the digital era. The most important differences are (1) the provision of the six-bit coding strip on the backside of the lens (this feature can be used to exchange information between the properties of the lens and the in-camera processing algorithms) and (2) the optical correction (or optimization) of the design to compensate for the small impact that the protective filter in front of the sensor surface has on the path of the light rays. Lenses for the screw mount cameras can be identified by the name of the lens, as was also the case with the earlier series for the bayonet mount cameras (the ‘M’-camera). After the introduction of the Leicaflex, Leitz used the suffix -R or -M to identify lenses for the two ranges of cameras. There is no special identification for the recently introduced or (slightly) redesigned lenses for the digital rangefinder models.
The premium characteristics of the M-lenses are the outstanding performance, the full metal mount, the manual operation and the compact size. The combination of compact size and high image quality is now an unusual one. Recent introductions by Zeiss and by Leica itself (for the SL- and T-cameras) indicate a trend to ever larger physical volumes. The small physical volume limits the optical evolution of the rangefinder designs.
The current designs have already a performance profile that most users have difficulty to fully exploit. At medium apertures the standard lenses are capable of a useful limiting resolution between 100 and 120 lp/mm (measured with the microscope on low speed microfilm, developed in Spur Nanospeed). This translates into a spot of about 0.0042 mm. A 24 x 36 mm sensor would need to have about 40 million pixels to record reliably this high resolution. Presumably 50 million pixels would be required because of the moiré effects at high resolutions. The Leica M-lens however should not be valued only for the high resolution potential, but for the clarity and smoothness of the reproduced details and gradation in the mid-range of spatial frequencies. Any audio aficionado knows that it makes no sense to look only for the high-frequency response of the equipment.
The Leica rangefinder user should however accept that higher levels of performance will be introduced more gradually for M-lenses in the future. The true advances in optical design will be introduced in the modern opto-mechatronic devices, like the Leica T- and SL- systems.
The complexity of photographic lenses has increased substantially in recent times. A high-speed-high-performance lens, like the Zeiss Otus has twelve elements in ten groups. A comparable lens for the Leica T (1.4/35 mm) has also twelve elements (including two aspherical surfaces) in eight groups. A zoom lens foo the Leica SL has 23 elements and seven moving groups. It is nearly impossible to handle this complexity with the help of the Seidel coefficients or the thin lens approach. The additional mechanical and electronic complexity (image stabilization, autofocus movements), including the analysis of the required accuracy and manufacturing tolerances can only be handled with the help of computer programs. The price for this added complexity is size! Lenses for the Leica M CRF are not as complex, but their physical constraints ask much of the designer to create high quality designs.
December 2020
November 2020
October 2020
July 2020
June 2020
May 2020
April 2020
March 2020
February 2020
January 2020
December 2019
November 2019
October 2019
September 2019
May 2019
April 2019
March 2019
February 2019
January 2019
December 2018
November 2018
October 2018
September 2018
August 2018
July 2018
June 2018
May 2018
December 2017
September 2017
August 2017
July 2017
June 2017
May 2017
April 2017
January 2017
December 2016
November 2016
October 2016
September 2016
August 2016
May 2016
April 2016
March 2016
February 2016
January 2016
December 2015
November 2015
October 2015
September 2015
August 2015
July 2015
June 2015
November 2020
October 2020
September 2020
August 2020July 2020
June 2020
May 2020
April 2020
March 2020
February 2020
January 2020
December 2019
November 2019
October 2019
September 2019
August 2019
July 2019
June 2019May 2019
April 2019
March 2019
February 2019
January 2019
December 2018
November 2018
October 2018
September 2018
August 2018
July 2018
June 2018
May 2018
April 2018
March 2018February 2018
January 2018December 2017
November 2017
October 2017September 2017
August 2017
July 2017
June 2017
May 2017
April 2017
March 2017
February 2017January 2017
December 2016
November 2016
October 2016
September 2016
August 2016
July 2016
June 2016May 2016
April 2016
March 2016
February 2016
January 2016
December 2015
November 2015
October 2015
September 2015
August 2015
July 2015
June 2015
May 2015
April 2015March 2015
February 2015