Friday, December 29, 2023

A Technological Appendix to Postblogging August 1953: Aspheric Lenses


Nothing particularly complicated here: Aspherical lenses. What's up with that? And how about Lawrence of Arabia for a pop-culture reference?

It's the music that seems old-fashioned. 
Lawrence of Arabia was filmed in Super Panavision, meaning that it was shot with 70mm spherical lenses instead of the 65mm then becoming standard. The intent was to capture extreme wide angles, and distribution was complicated by the need to upgrade theatres with curved screens, which The Economist has been wringing its hands about. Fox's original Cinemascope, new in 1953, is illustrated opposite. Based on Henri Chretien's 1926 work,* which Bausch and Lomb quickly managed to patent troll into their own intellectual property inventory, but hopefully not before cutting a big cheque to Chretien to cover a long and arduous journey. It seems that they also did the actual lens designing in their "electronic computing department," which, Fortune notes, is the first in the industry (as of 1953.) In practice, it was Panavision's innovation of a "new lens set that included dual rotating anamorphic elements which were interlocked with the lens focus gearing" that made the running. 

None of this has anything to do with aspheric lenses, which are lenses with computed profiles rather than the much-easier-to-calculate spherical and cylindrical lenses in current use as of 1953.  The most interesting commercial application of aspherical lenses is in eyeglasses, 
Men don't make passes at girls who wear glasses. Unless they're aspherical! Rowrr! Model uncredited. 

where they can reduce astigmatism and, more importantly, give better results in the multipower lenses that we are all getting these days because we are old. However, as the Wiki article points ou t, asphericals are widely used in consumer optics because they are cheaper than noncomputed lenses. 

So what's a computed lens?

People will have probably played with simple cylindrical lenses in high school science class, since the math uses simple trigonometric expressions to explain lens magnification with such simplicity and power. I could go all woolly and philosophy-(or, indeed, sociology of)-science here, but that's not where this post needs to go in the moment of 1953. The above formula will be familiar as a series expansion of a function, although the causality goes backwards, in that we define a function as having the necessary slope in as many points along the "curve"  as we are up to specify, then use the series expansion to write a function that joins these points in a mathematically-valid way. 

This is why Bausch and Lomb has an "electronic computer" department. It is why James G. Baker gets his own inset, as the "world's fastest lens designer," able to do computations in months that other optics labs take years to do. (He is working on the camera for the U-2 right now.) There's nothing in this digression about aspheric lenses either, but I hope it does put the apocryphal quotes from assorted worthies about the world only needing fifty computers or so in 1950, 1955, or 1960, or 1985, into perspective. If anyone, from the president of IBM down to Bill Gates said that, it was because they had no idea how many fields were working with this whole "maths" stuff to make products that make a lot of money while still being apparently invisible. (Every movie coming out this year seems to have been shot with a new lens, but who cares? It's just movies.)

The interesting thing here about aspherics is that as of 1953, aspherical lenses are being considered for a much wider range of applications because there is an absolute limit on how precisely spherical and cylindrical lenses can be machined. As the Wiki notes, small aspherical lenses are relatively easily produced from moulds, which presumably can be made by stringing together micrometer settings. Larger ones are producing by traditional grinding and polishing. The process that I just described as "setting the micrometer" turns out to be  "point-contact contouring," but there's a great deal more technique,a nd many manufacturing options, for larger aspherical lenses, and the Wiki goes into some detail.  rougly 150 years after the city's last Seneca householders vanish into a curious lack of comment in the local histories of two generations later. Elgeet's first commercially available aspherical lens was produced by Elgeet of Rochester, based on its Navitar lens for the US Navy, but it was then introduced as a wide-angle camera lens in 1956, and became a household name, ,eventually leading "Elgeet" to decide that it had a stupid name, and change it to "Navitar." It is by this impressively science-fictiony sounding name it does business today. and, no, I have no idea what the Royal Navy was doing for its aspherical lens needs, at least before the founding of Knight Optical, "almos[/a little over] thirty years ago." (It  turns out to be 1987. Or 1991!  No, wait, that's another Knight Optical. Which is in the United States, but there is also a Knight Optical (USA). Someone's got to sue somebody sometime.) 

The best known Navitar today is the Cine Navitar 3" cine lens for home movie cameras, and isn't that a slice of 50s social history we've been skirting to avoid collective trauma! (And that was before I discovered the educational film, the 50s equivalent of the MOOC.)

Another concern here is that in 1953 there was no way to measure radial curvature to more than one part in 60,000, an inherent limit imposed on spherical grinding that did not apply to aspherics. So it is perhaps to an improvement in curvature measure that we can attribute aspherics' failure to become even more prominent. Maybe! I guess we'll find out as the postblogging marches on. 

Widescreen! "The substance is still insufficient for the vast spread of screen which CinemaScope throws across the front of the theatre, and the impression it leaves is that of nonsense from a few people in a great big hall."


 *Check out place and date of birth. I'm choosing to believe that Chretien died after three years of partying on Fox money after a lifetime of shabby academic propriety.

No comments:

Post a Comment