Universal Geneve, Zenith and 36 (or perhaps: the significance of 12)

Posts
6,256
Likes
9,814
Sometimes I run into interesting stuff. Recenty this UG was posted on IG:
I always check the serials of watches like this against my mastertable of chronographs to see whether they fit in (actually: fit out) and this time it was a close call with a Zenith serial. This watch was sold in the past by L'Orologiese but no working links anymore unfortunately. Serial of the UG: 596.534, Zenith: 596.547. However: these numbers were not the only ones in the caseback. Both watches have a 2 digit number below the serial:



The UG 33 and the Zenith 10 (BTW: the 33 is not that well visible but the seller confirmed it). As I have written in the past some batches of UG appear to have been produced in 36, 72 of 144 pieces. I have concluded this based on the numbers below the serial and in this case this leads to a very interesting conclusion: the numbering of the UG batch connects seamlessly into the Zenith serials



The relationship between UG and Zenith in this period is well documented but this is quite a coincidence (albeit N=1 of course).

The interesting question is whether all batches were made in multiples of 36 and how long UG kept doing this. A recent example I ran into below.

Starting point was this:

Explanation:

A: reference (redacted because I don't want to give traders a reason to sell their watches with the '1 of 50 made argument'; if you know which references this table is about: don't mention it here please)
832.974: highest serial I know
832.961: lowest serial
13: 832.974 -/- 832.961
2: number of examples
38: tank number
3 numbers below: rough estimation of the batch
orange: batch

Column A/E (green) is basically the same principle but then with the numbers of the previous 5 columbs combined.

Recently I found an additional example of ref A: 832.997. This leads to the following table:


Bit of an odd pattern isn't it? Especially C with 90 is a bit off but 100 does not fit. However: when I tried 36 multiples it fits much better...



In this case probably a pure coincidence but still intruiging.
Edited:
 
Posts
3,468
Likes
8,060
Very cool stuff. There was some kind of logic applied on the ground back when production happened, they weren't just picking completely random sized batches (at least I wouldn't think so), make sense that you would be able to pick up on some of those patterns if you can get enough data
 
Posts
30
Likes
12
Sometimes I run into interesting stuff. Recenty this UG was posted on IG:
I always check the serials of watches like this against my mastertable of chronographs to see whether they fit in (actually: fit out) and this time it was a close call with a Zenith serial. This watch was sold in the past by L'Orologiese but no working links anymore unfortunately. Serial of the UG: 596.534, Zenith: 596.547. However: these numbers were not the only ones in the caseback. Both watches have a 2 digit number below the serial:



The UG 33 and the Zenith 10 (BTW: the 33 is not that well visible but the seller confirmed it). As I have written in the past some batches of UG appear to have been produced in 36, 72 of 144 pieces. I have concluded this based on the numbers below the serial and in this case this leads to a very interesting conclusion: the numbering of the UG batch connects seamlessly into the Zenith serials



The relationship between UG and Zenith in this period is well documented but this is quite a coincidence (albeit N=1 of course).

The interesting question is whether all batches were made in multiples of 36 and how long UG kept doing this. A recent example I ran into below.

Starting point was this:

Explanation:

A: reference (redacted because I don't want to give traders a reason to sell their watches with the '1 of 50 made argument'; if you know which references this table is about: don't mention it here please)
832.974: highest serial I know
832.961: lowest serial
13: 832.974 -/- 832.961
2: number of examples
38: tank number
3 numbers below: rough estimation of the batch
orange: batch

Column A/E (green) is basically the same principle but then with the numbers of the previous 5 columbs combined.

Recently I found an additional example of ref A: 832.997. This leads to the following table:


Bit of an odd pattern isn't it? Especially C with 90 is a bit off but 100 does not fit. However: when I tried 36 multiples it fits much better...



In this case probably a pure coincidence but still intruiging.
Fascinating. Thanks.
 
Posts
6,256
Likes
9,814
On an Italian forum somebody posted this overview. Perhaps 12 was kind of an industry standard
 
Posts
13,433
Likes
52,867
As I recall much of the work force back then farmed in the summer months and made watches in the winter. So a dozen would be a natural number on which to base a batch production process. I’m continually impressed at this growing data base of yours!
 
Posts
6,256
Likes
9,814
Thanks, much appreciated!

The only thing I can come up with is something my English teacher in high school pointed out: 12 is handier because it can be divided by 1, 2, 3, 4 and 6 - so 5 numbers - and 10 only by 3. This is reflected in English money.

In the Netherlands - where I live - the word ‘gros’ was used when ordering pencils or so. It means 144 pieces and was derived from the French word ‘grosse douzaine’ or big dozen e.g. 12x12. It may have a been customary number in (Western) Europe

Edit: AI says this



Another interesting post (from Reddit):




And also interesting: https://en.wikipedia.org/wiki/Finger-counting
Edited:
 
Posts
6,256
Likes
9,814
Just a quick update. This epiphany caused me a lot of work.... I need to update my whole table - which means hundreds of batches - but the 'fit' improves a lot