The Watch Site banner
41 - 59 of 59 Posts

·
Special Member
Joined
·
2,701 Posts
Like mine shown above. I am not really convinced of the dial swap theory yet. Would have happened to often in my opinion.
 

·
Registered
Joined
·
43 Posts
Discussion Starter · #42 ·
Ya, I am starting to think I agree. Which leaves me scratching my head over the lumed 6217-7010 version. It seems perhaps the Asian Games watch was an August 1966 only production and the standard 6217-7010 was a 1967 run - which might have both lumed and not lumed versions.
 

·
Registered
Joined
·
43 Posts
Discussion Starter · #43 ·
Like mine shown above. I am not really convinced of the dial swap theory yet. Would have happened to often in my opinion.
Interesting that both your non-lumed 7010, and the one in that listing are from June 1967 - maybe just coincidence. The lumed versions I have verified so far have been from May (two different serial no's), November and December - so if the non-lumed "standard" version was a "production watch" and not a dial and hands swap - it overlapped the lumed version production dates. Seems plausible to me at this point that it was a production version - not a hack.

I still find it odd (and frustrating) that I can't find a single catalog scan or ad with the 6217-7010 - Asian Games or Standard version, lumed or not...
 

·
Registered
Joined
·
43 Posts
Discussion Starter · #44 ·
Another interesting detail that I just stumbled on today. In the 6217-7000 there seems to be two GMT hand versions. One is thin, the other tapered (thicker at the center of the dial, thinner as it arrives at the arrow). Going through the ad scans, some do show the thin GMT hand on this first gen model, while most show the tapered version - so I think both are OEM and correct for the 6217-7000.

Watch Analog watch Silver Watch accessory Rectangle


Watch Analog watch White Light Product


I am not clear if the tapered version was used on the 6217-7010. I am also not clear where the red colored GMT (without lume) was used originally, but there do seem to be some of those floating around that look original ...
 

·
Registered
Joined
·
43 Posts
Discussion Starter · #45 ·
Also - there appears to be some 6217-7000's from 1964 which also have the "JAPAN | 6217-7000 TAD" wording printed at the bottom of the dial... instead of MADE IN | JAPAN - just like I am finding with the 6217-7010's. Perhaps different dial manufacturing facilities? They don't seem to line up with production dates / cut-offs.
 

·
Registered
Joined
·
43 Posts
Discussion Starter · #46 · (Edited)
I've started a Google spreadsheet documenting the largest serial number for each month/year that I can find on google image search, ebay, chrono24, in my (small) collection, and in my notes right now... If you own or have a photo of a bigger serial no. for a specific month/year - please let me know:

-- DELETED --

As I see new listings popup - I will check them against this list and update when I find bigger S/N's.

The goal here is to get an educated estimate as to the total number of each generation of watch produced - and also identify any months where watches were not produced. There are a lot of gaps still in the spreadsheet

A few assumptions:
  1. I am assuming that generations like the 6117-6010 and 6117-6019 - or the 6117-6400 and 6117-6409 share a common serial numbering - so for instance you won't find a S/N 1D0001 on the 6117-6400 and another 1D0001 on the 6117-6409. Please let me know if you know otherwise.
  2. I am assuming that serial numbers are not repeated for different dial colors or case metal (gold/gilt vs SS etc)
  3. I am assuming SEIKO likely did not go to 9999 or 99999 on each month / batch
  4. I am assuming SEIKO do NOT duplicate serial numbers in a given month year (I hear there are some modern examples of this)
And some interesting notes:
  1. It looks like the Gen 1 watches all had 7-digital serial numbers, allowing up to 99,999 watches per month - for both 1964 and 1967. This appears to also be true of the Gen 2 Asian Games watches (with the largest current production count at 87,365 in its one month run).
  2. I have also seen one Gen 2 lumed version with a 7-digit serial number, and another white dial lumed with a 6 digit - both in 1967 - still thinking about that one.
  3. all other gens and version appear to share a 6-digit serial number - so max 9,999 watches per month - so far only one month comes close to that max.
UPDATE: The Google sheet has been replaced with a full-fledged image study app, including lots more detail - to be found here: Seiko World Time Study - Casebacks
 

·
Registered
Joined
·
43 Posts
Discussion Starter · #48 · (Edited)
I have created a web page / app that takes a repository of case images collected online, and analyzes them - returning stats. It may change locations / web addresses, but is currently here:


I'm not sure how much usage this free Azure service will allow - but figured I would post it up and see what happens...

The cool thing is I can just keep dropping new files in (named correctly) and it will auto-update.
 

·
Registered
Joined
·
43 Posts
Discussion Starter · #50 ·
Well that free instance got used up quick. It is up and running again for now - on my dime. links in previous post
 

·
Registered
Joined
·
43 Posts
Discussion Starter · #53 ·
Ok, I've also updated everything to be a little easier to use on a mobile.

A few more things I am still pondering:

1. What's the deal with Asian Games S/N's? Of the 7 I have cataloged, 6 of them start with 688XXXX, and one starts with 681XXXX... I'd like to find another S/N or two between 10,000 and 80,000 to be sure that we are really looking at 80k+ Asian Games watches - it does seem that way right now.

Font Screenshot Software Technology Electronic device


2. in 1967 we saw 3 watches...
  • The reissues of the 6217-7000 (current examples from May, June, July and December)
  • the non-lumed 6217-7010 in June only (so far)
  • and lumed 6217-7010's at least May and December

We also so at least 3 months were more than one model were produced concurrently.

The lumed 6217-7010's from May have a 7-digital serial number, with the same watches from December have a 6-digital serial number. Those May watches also have very high S/N's indicating over 30k watches in May, and another 8k+ in December.

This is perhaps the most odd production year for this series of watches. To add to the confusion there seems to be no ads or catalog scans.

3. The gold/gilt 6117-6409 - I had previously noted this watch as a Feb 1972 production, but have now found two examples from June 1972...


I think I will decommission the Google Sheet - this page is easier to keep up to date.
 

·
Registered
Joined
·
43 Posts
Discussion Starter · #54 · (Edited)
Ok big data update.

VIEW IT HERE:
  1. I've now cataloged 275 watches, including a lot more data points
  2. You can now click on a S/N on any watch as you scroll the (long) detail list - for a popup with more detail break down
  3. You can now click on the Model No in the summary to view break downs by dial color, code etc.
  4. You can now click on a Production Year in the summary to scroll to the start of the detail for that year
  5. There is a floating "TOP" button at the bottom right to quickly get back to the summaries. Click it twice to go all the way up
  6. I have added a "Data Explorer" to the top nav - this takes the photo data and puts it in a grid, where you can filter, sort, drag column order (or off grid to remove), etc. Do your own analysis a bit
It is much easier to NOT run the full image version - instead you can click on a specific S/N to check the images and data points. There is a NEXT and PREVIOUS option in the popup to toggle through the watches viewing the associated detail.

I am sure I will continue to revise this, but it is largely there. I would like add a better image zoom option and a few other bits - over time...

A few new observations based on the data
  • Gen 1 through 3 show about 75% of dials being silver, with 25% being black
  • Gen 4 is closer to a 50/50 split of black vs white dials
  • The majority of Gen 4 dials had WORLD TIME at 6 o-clock through Dec 1970. From there on the vast majority were at 12 o-clock. It seems plausible the that outliers are dial swaps
  • 6117-6409's seem to be manufactured only from 1970 - 1972, whereas the 6117-6400 version went from Dec 1969 through Dec 1976
  • The GMT hand on Gen 1 (6217-7000) watches seems to be mostly the tapered version through Nov 1964. This narrow version is spotted from July 1964 through the 1967 models, suggesting the tapered version was the original GMT hand style and the thin version was an update / change later in the first year of production
At some point soon, I will update the main Google Doc with better data-backed information from this study / tool.
 

·
Registered
Joined
·
43 Posts
Discussion Starter · #55 ·
Got to 300 watches cataloged today!

Also made the S/N's clickable on the Data Explorer grid - so you can see each watch details if needed.

I am pasting in screenshots of the summary data here - but the very latest (if and when I add more) will still be found here: Seiko World Time Study - Casebacks



Here's the summary by Model




Here's the break out for the 6217-7000 watches




Here's the break out for the 6217-7010 watches




Here's the break out for the 6117-601X watches




And here's the break out for the 6117-640X watches




Here's a summary by year




And finally top line summary

 

·
Registered
Joined
·
43 Posts
Discussion Starter · #56 ·
Ok - a few more changes:

1. Everything is a little faster now
2. We've got an actual domain now: Seiko World Time Study - Casebacks
3. You can now click the eyeball icon to drill down and see the data points used in each summary - in the data explorer... Might change the icon - if I find a more obvious one...

So for example if you click on the highlighted icon

Rectangle Font Material property Screenshot Technology


This will show you those items. NOTE: you can click the Full Serial to view each item

Rectangle Font Material property Parallel Screenshot
 

·
Registered
Joined
·
43 Posts
Discussion Starter · #57 ·
Ok - a few more updates

1. There are now 371 watches cataloged - that's +71 since 5 days ago.
2. The main page has been reformatted more like a blog entry, with splash image and a table of contents for easier navigation
3. The site should run faster - all images have be optimized and thumbnails have been generated and used where appropriate - so you are no longer downloading huge files (some of the main image files are over 6mb)
4. There is now an Thumbnail Gallery on the Data Explorer... Select the data (filter, sort etc) in the grid, then click the Thumbnail Gallery option to visualize the selection as case fronts or backs:

Watch Analog watch Photograph Light Product
 

·
Registered
Joined
·
43 Posts
Discussion Starter · #59 ·
Thank you for the kind words. Hopefully others find it helpful. If I get bored, I might consider applying the same engine to other vintage models ... at some point in the future.

In some ways as enthusiasts we exist in an echo chamber - when we read that someone said X, Y or Z about a watch, we then often repeat that to others without any real scrutiny or fact checking. With vintage watches, and especially Seiko with their high production quantities and an active "modder" culture and mentality, that can often lead to unintentional repeating of misinformation.
 
41 - 59 of 59 Posts
Top