5. Hexanon database

This section has to do with a 6-year research project whose initial purpose was to assess the approximate production volume of Konica’s line of Hexanon AR lenses, and how that volume had evolved over time. The project was spurred by my inability to find satisfactory answers to a number of technical questions, by the circulation of various theories about Hexanon lenses and, not least, by the general lack of verifiable information about them. I had been wondering about the production volume of Hexanon lenses and naturally assumed that their serial numbers and production codes would provide some indication.


The question I first sought to answer was how to find the date of manufacture of a specific Hexanon lens. Hexanon lenses usually have a two-digit code indicating their time of production stamped on the baffle surrounding the lens’ rear optical element. The first digit of this code is a number indicating the year of production, while the second digit is a letter that stands for a specific time interval during that year (see Section 3 – Hexanon lenses - Time of production). Unfortunately, I have never been able to locate any information about Konica’s production coding system in any of the company’s literature or anywhere else, and no Konica user or collector I asked was able to point to any evidence in support of either of two possible interpretations of how Konica's coding system worked or tell me where they had originated. 

Common wisdom in the Konica community has it that Konica divided the year into 26 successive two-week periods, assigning a letter of the alphabet to each in alphabetical order. This sounded neat at first, but soon struck me as exessively complex and just as impractical. Mathematically, the problem with this scheme becomes evident when you stop to consider that 4 weeks is 28 days, not 30 or 31. It follows from this that the year’s 9th production interval, for example, would be from April 23 to May 6, and the 24th from November 19 to December 2.  This entire structure presumably shifts forward one day with every leap year, and back the next. The more I thought about it, the more I became convinced that this scheme was highly improbable.

There is also a second, almost identical scheme in existence – one which divides the year into 24 half-month intervals, instead of 26 two-week ones. This scheme looks much neater mathematically, in that each interval was either the first or second half of any given month. In contrast to the first scheme, this one had the advantage of being simple and practical, but it had its own annoying little glitch: It called for only 24 letters, which meant that 2 letters of the 26-letter English alphabet had to be omitted from the scheme. Which two were they and how to find out?

As if the production code issue wasn't mysterious enough, that of serial numbers also proved to be a challenge. Here too, a rumor that struck me as a bit fanciful was doing the rounds in the Konica community. It held that Konica picked its serial numbers at random. Some even claimed that Konica had a deliberate policy of mixing up serial numbers, the argument being that Konica supposedly did this to keep the competition in the dark about its production volumes.


I have never had anything to do with manufacturing, so I am not well-placed to know just how desirable it is to hide production volumes per se, whether it is at all practical, or how effective it is to go about it in this fashion. I had never heard of any company going to such lengths to prevent the outside world from knowing what it was up to (unless it was active in a highly-sensitive branch of industry) and I felt that there had to be a way of getting an approximate idea of how many lenses Konica had made over nearly a quarter century.


It logically follows that since no sense could be made of the production codes, the serial numbers made no sense either. The more I thought about the production code and serial number theories, the plainer it became to me that they simply didn't stand up to scrutiny. In the end, I became convinced they were just convenient explanations bringing the appearance of rationality to a situation which defied comprehension.


Those are the reasons why, in early 2009, I began taking down the serial number, production code and a few other details of any Hexanon AR or Hexar AR lens I could lay my eyes on. The main source of this information has been Ebay and other similar online auction or trading sites (Yahoo.jp; Taobao.com; Allegro.pl; Leboncoin.fr; Priceminister.fr; Marktplaats.nl; Ricardo.ch; Tradera.se; Molotok.ru; Craigslist; Gumtree; Kijiji; and many others). My database now contains the serial numbers and production codes of 12,000 lenses.


In this section, I will try to present the information that emerges from this database. I have no training in statistical analysis, so this may prove a challenge. I welcome the input of anyone who is familiar with statistics and is interested in this subject.