Production codes

The Hexanon database has allowed me to establish three facts about Konica's production codes and all three came as a surprise. The first began to emerge quite early after I started gathering information on Hexanon lenses some lens versions simply have no production code: A and B-type lenses (the two lens versions that have an aluminum ring, made in 1965-70) don't have one; roughly half of the C-type lenses (all-black-all-metal lenses, made in 1970-74) have one; and the majority of D-type and all E-type lenses (lenses with a rubber-covered focusing ring made from 1973 on, and the compact lenses that followed them) have one. I seems Konica began stamping its production code on Hexanon lenses sometime in early 1972 (see below).

By the time I had about 500 lenses in my database a second fact also began to emerge the production codes are made up of only 12 letters: They are ‘A’, ‘B’, ‘C’, ‘E’, ‘I’, ‘K’, ‘M’, ‘N’, ‘O’, ‘S’, ‘T’ and ‘Y’. As there are twelve of them, I assumed they stood for the twelve months of the year. I have not encountered the letters ‘D’, ‘G’, ‘Q’, ‘R’, ‘U’, ‘X’, or ‘Z’ on any lens. A further 7 letters, ‘F’, ‘H’, ‘J’, ‘L’, ‘P’, ‘V’ and ‘W’, seem to appear on 11 lenses out of a sample of 12,000. Those 11 lenses represent 0.09% (not quite one tenth of one percent) of the entire lens sample. Statistically speaking, the number of codes with one of those 7 letters is so small, that it leads me to believe they are smudged renditions of the 12 letters first mentioned above. A factor that makes such an explanation plausible is that the production code on Hexanon lenses is not engraved but stamped and on some lenses its rendition is smudged. On such lenses, the code seems to have been stamped in such haste that it is necessary to examine it very carefully to determine whether the letter it contains is an 'H' or a 'K', or a 'V' or a 'Y'. In other words, the 7 letters are probably not the ones they appear to be.


In light of these findings, I think it safe to assume Konica’s coding system divided the year into 12 monthly intervals. This scheme is, to the best of my knowledge, the only one which has any evidence to support it. This evidence is circumstantial, but overwhelming. I first came to this conclusion when my database only contained a fraction of the number of lenses it contains now, and not a single lens since then has given me any grounds to reconsider. In contrast to the two schemes described in the introduction to this section, this scheme also has the advantage of being logical, straightforward and, therefore, practical. It is often said that the simplest of all possible explanations is usually the right one. I believe this is a case in point.


* * *


All this was most interesting and gratifying, but it hardly brought me any closer to getting an idea of Konica’s production volumes, and that was the purpose of the data collection exercise to begin with. By 2012, I had taken down the serial numbers and production codes of some 4,500 lenses and I had arranged them in the numerical and alphabetical order of their production codes (1A, 1B, 1C…, 2A, 2B, 2C…, 3A, 3B, 3C…, etc). Most disappointingly, the list of serial numbers I obtained in this manner looked without any rhyme or reason whatsoever. It was beginning to look like Konica had indeed sought to lead unduly inquisitive people like me astray.


Then, in December 2012, while I was browsing the net in search of Hexanon lenses which had inexplicably escaped my scrutiny, I came across the website of an Australian gentleman who had discovered that, for any given year, arranging the production codes of the Konica-made Koni-Omega medium format cameras in his collection in a certain order would produce a clear ascending pattern among the serial numbers associated with those specific production codes. What was especially encouraging was that the 12 letters used on the Koni-Omega cameras happened to be exactly the same as the ones I had found on Hexanon lenses. I contacted this gentleman to compare notes and he said that the first thing that had struck him with the twelve letters is that half of them spelled the word 'KONICA'. From there, finding the order for the remaining six that produced a clear ascending pattern among their corresponding serial numbers was easy. The resulting arrangement of the production code letters forms the expression 'MY BEST KONICA', with the letter 'M' corresponding to January, and the letter 'A' to December.


Incidentally, this elucidated the very first mystery I had noticed after embarking on this project: I had less than a few hundred lenses in my database when I noticed that the serial numbers associated with the letter ‘A’ were always much higher than those associated with the letter ‘Y’, while the opposite should theoretically have been the case. Naturally, I assumed that the 12 letters, standing for the 12 months of the year, were arranged in alphabetical order (in what other order are letters ever listed?). This inconsistency between the numerical order of serial numbers and alphabetical order of the letters was a clue to the probable existence of a key to the production codes, but the idea never crossed my mind.


Lo and behold, with each lens type for any given year in my sample now arranged in the 'MY BEST KONICA' sequence, the serial numbers of most Hexanon and Hexar lenses in my database fell neatly into place to form an clear ascending pattern from year to year and, on the whole, from month to month.


What the origin of the 'MY BEST KONICA' expression is I have no way of knowing. To my arguably untrained ear, it sounds like something an ambitious mid-level Konishiroku manager not entirely conversant in English might come up with, especially as the Japanese had, and continue to have, a liking for marketing slogans of this kind. There is perhaps an amusing story behind it.


 * * *

The database also reveals how and when the production codes were introduced. There are 3,320 lenses in my Hexanon database without a production code (PC) – about 27.6% of the total. Those lenses were produced from 1965, when the Konica AR-mount was introduced, to early 1973. The stamping of lenses with a PC seems to have begun in early 1972 (the earliest PC in the database is on a 52/1.8 lens from February 1972).

In the database, there is usually an overlap between the highest serial numbers of lenses with no PC, and the lowest serial numbers of the same lenses with a PC. It would seem that during the period of this overlap, lenses were produced with and without a PC. This overlap extends over 1 to 10 production months, depending on the lens. But in each case, by the end of this period it seems that no lenses were leaving the factory floor without a PC. For most of the lenses involved, the entire process took place in 1972, but for three of them it extended into 1973.

Let’s use the 21/4 as an example and use the following table, showing a very short segment of the 21/4 section in my database to illustrate what I mean (the lenses with a white dot under the SN are lenses without a PC):

To determine the approximate overlap of SN ranges, one begins with the highest SN for a 21/4 lens without a PC, and that is 7032188 (A). We then find the numerically closest SN to that one, but for a 21/4 with a PC, and that is 7032237 (B, at the far right of the table). We then look for the earliest SN for a 21/4 with a PC, and this is 7031162 (C). To apply this range to 21/4 lenses without a PC, we need to find the closest SN to that one among 21/4 lenses without a PC, and that is 7031117(D, at the far left of the table). This gives us two numerically overlapping SN ranges of approximately 1,100 lenses each, rounded off to the nearest hundred lenses. Given the relative orderliness of Hexanon serial numbers as a whole, it seems reasonable to assume that the two ranges of 1,100 lenses were produced during roughly the same period, in August 1972 for the most part.

The following table shows the above approach in relation to all Hexanon lens versions that exist with and without a PC:

It thus seems like the PC was phased in during a period of approximately one year. This is rather quick by Konica standards, considering that new lens versions were often produced side-by-side with the version they were meant to replace for 2, and in some cases even 3 years.

* * *

Now that we have a large sample of Hexanon serial numbers spread over a period of 23 years, I assume addressing the issue of production volumes, if only in approximate terms, should not be a terribly daunting proposition. I wouldn’t venture to make any suggestions at this stage, however, as I haven’t had the time to break down the sample into its various categories. I am also not entirely certain how to go about it, as I don’t have the necessary training to engage in statistical analysis. I suspect, however, that the size of the sample is more than sufficient to draw some valid conclusions.