What's your thoughts on this Rolex Sea-Dweller, ref. 16600, calibre 3135 project?

[Nucejoe]

 Not sure of level of expertise at rwg.bz forum, yet every bit of info might help.

https://www.rwg.bz/

 

[VWatchie]

2 hours ago, rehajm said:

You’ve been duped by a bot!

That's what I'm hoping and believe, and why I wanted to discuss it with you.

2 hours ago, rehajm said:

OMG men I thought you were kidding

No, I'm being dead serious! What made you think I was kidding? It's a sincere question.

2 hours ago, rehajm said:

Rolex used those blue screws on the automatic movement to distinguish them from the plate screws.

What puzzles me is that only the top of the screw head is blued (see image in my previous post or below). Not even the screw slot is blued! In my opinion, it looks "Chinese-cheap" and not what I would associate with Rolex. Hence, it awakened my suspicion that something might be wrong.

1 hour ago, Kalanag said:

I‘ver never seen a fake Rolex with a free sprung balance and micro stella regulation.

Neither have I! I never paid much attention to all of those videos on YouTube explaining how to tell a fake and genuine Rolex movement apart, as I thought the escapement would be a dead giveaway. After all, who would be willing to pay for a fake free sprung balance with a Breguet hairspring and micro stella regulation, and having the insane precision that a genuine Rolex 3135 has? Looking at the timing machine results (my first post), I had no doubt whatsoever that I was dealing with the real McCoy.

So @rehajm, are you telling us we're looking at a fake Rolex free sprung balance with a Breguet hairspring, micro stella regulation and KIF shock protection? If so, we were wrong to assume that we would never see that in a Chinese clone. Ay caramba!  So, what's the giveaway?

1 hour ago, JohnR725 said:

But it would seem odd that you would make a generic screw blue if they were blue in the first place?

I get your point, John. It makes zero sense! But did you notice how both the original eBay screws and the generic screws are completely blued. The ones I found to hold the date indicator seating in place were only partly blued? Let me show you the picture again:

Makes you wonder if the Rolex screwmaker had a bad day or didn't understand the specification, as it was not written in Mandarin.

So, what's the verdict? Am I dealing with a genuine Rolex 3135 or a Chinese super clone? If you don't know or you're not sure, please feel free to take a guess! Here is a link to the first 100 pictures I've taken during the disassembly: https://1drv.ms/f/c/5643825a9f286b75/EpbF5n_qRCREvohfgbcZ4l4B0eUcw2h62xqmPYtdPzQ-QA?e=QIwcN0

Edited yesterday at 01:34 PM by VWatchie

[Kalanag]

See blued srews on a genuine Rolex (Fig. 12):

http://people.timezone.com/mdisher/andrewb/3135/3135_1.htm

and here:

https://www.chrono24.de/rolex/werk-komplett-rolex-3135---complete-movement--calibre-3135---genuine-rolex---fully-working--id29910928.htm

Edited yesterday at 02:12 PM by Kalanag

[Jon]

4 hours ago, VWatchie said:

So, what's the verdict? Am I dealing with a genuine Rolex 3135 or a Chinese super clone?

That is a legit Rolex. That is no clone.

I've got a friend who primarily services  the super clone Rolexes, so I'll get him to take some pics and post them

[VWatchie]

Thanks for the links @Kalanag!

If we assume the two examples you show are truly genuine, then we now know that sometimes Rolex use nice looking blued screws and sometimes screws where only the top of the screw head is blued and looks very cheap and not what I believe most would associate with such a prestigious brand. I guess I could have accepted it if the screw slot had been blued too. To be quite honest, I think they look horrible, but if they are genuine, I can live with them.

This is what the screws look like in the previous 3135 I took apart (my first Rolex). They look pretty, and so never sparked any suspicion in me. I guess I just have to try to deal with it. Had I found screws with only the top of the screw head blued in a Chinese or Russian movement I would have replaced them. Did I mention I don't like the partly blued Rolex screws? 

1 minute ago, Jon said:

That is a legit Rolex. That is no clone.

Big, big sigh of relief, Jon!  It's not my watch, but I care deeply not only for the watch but for its owner!

2 minutes ago, Jon said:

I've got a friend who primarily services  the super clone Rolexes, so I'll get him to take some pics and post them

The would be super interesting to see!

I must say, I'm so grateful we have you on WRT Jon!

[RichardHarris123]

The blue looks like some sort of coating not heat.

[VWatchie]

1 minute ago, RichardHarris123 said:

The blue looks like some sort of coating not heat.

I agree! It's ugly, and I wish Rolex would scrap them!

Just for the fun of it. I asked ChatGpt to read this thread and then ask if it affected its previous assessment. ChatGpt then lowered its "Confidence Level".

It's pretty clear that AI is still a lousy watchmaker. Looking at its generated images of watch movements can be pretty hilarious. It will be interesting to see where it is a year or so from now. When it comes to programming, solving complex programming tasks, and explaining and arguing for various types of software architecture patterns, it is mind-blowingly competent. I guess I made the mistake of believing it would be similarly competent in horology.

[Kalanag]

I asked ChatGPT if there did exist any Rolex clones with free sprung balance wheels and micro stella regulation. The KI said YES using reasoning! The provided internet links however showed nowhere any free sprung clones except one clone which (just for fun) was equipped with a genuine Rolex balance and balance bridge.

Don‘t trust KI… 

Edited 23 hours ago by Kalanag

[tomh207]

@VWatchie

im going to throw my tuppence in, might not help but hopefully gets you back down to earth.

current AI (LLM) are essentially web scraping and aggregator bots, they work across for those of us who have spent most of their working life’s building the internet think as the most untrustworthy source of anything. Look at Wikipedia as a prime example. Our issue here is as a society we have been conditioned over years to trust everything that the computer tells us. Garbage in, garbage out as I am sure you know is as valid here as it has always been.

AI could be useful in specific contexts, say we had watchgpt and it got its data by scraping the watch manufacturer data and only that then it’s erroneous output would be much reduced however if it referenced two tech docs from the same manufacturer and the same movement it would still depend on whether it’s algorithm could make a decision on which is best.

now out here in the real world the AI bot will scrape all and sundry data based on your request and it cannot validate any of that information at all but does seem to be able to present whatever it finds in a way that is fairly convincing at times.

i think that we are starting to see AI referencing AI generated output to validate what it’s put out. Not 100 % sure about the veracity but there is apparently a paper out in the wild that AI was used as an aid to compile, more papers have appeared citing the first in support of there thesis. The problem with these papers is they referenced a material from the first paper that doesn’t exist but was bits of words mashed together to sound good.

 

it can have its place and uses but should be treated with extreme caution.

sorry for the rant it’s hard enough on this planet to find Actual Intelligence as it is without introduction of Artificial Intelligence which is even more stupid.

 

Tom

[GuyMontag]

7 hours ago, Kalanag said:

I‘ver never seen a fake Rolex with a free sprung balance and micro stella regulation.

I believe they are out there (at least the free sprung balance):
 

 

Edited 23 hours ago by GuyMontag

[JohnR725]

6 hours ago, VWatchie said:

Makes you wonder if the Rolex screwmaker

I was thinking what makes you think that Rolex made the screws in their watch?

Here's an interesting link talks about the balance wheel and the  Parachrom Blue hairpspring. So a very clear indication of a counterfeit Rolex is the lack of this blue hairspring. Except watch was made for 34 years wonder if they ever changed anything or did every single watch be made to an exact specification all those years? Oh and the nifty blue hairspring came into existence around 2011 they actually have two types of materials for hairsprings so they blue hairspring just indicates it's the newer hairspring.

https://seikoparts.wordpress.com/2023/07/03/rolex-3130-3135-balance/

Here's an interesting site

https://calibercorner.com/rolex-caliber-3135/

Watchmaking has such an interesting definition doesn't it? Like for instance your company making movements do you make 100% of all the parts or could you outsource things like screws for instance? Then conceivably if you made a movement over a long time span as I mentioned before can things change? Or in the case of Rolex I always find this story interesting.

https://timefiles.ch/post/how-it-took-rolex-a-century-to-own-its-movement-manufacture/

 

 

[Kalanag]

@GuyMontag good find! Even the micro stellas are cloned in the Daytona fake!

Around $2600 is mentioned in the thread.

https://www.reddit.com/media?url=https%3A%2F%2Fpreview.redd.it%2Fnewest-development-from-ace-v0-61r4zxq1q2xd1.jpeg%3Fwidth%3D1080%26format%3Dpjpg%26auto%3Dwebp%26s%3Dd5dd3a29a3d1668077d4a4c894a09c76f6f66de9

https://forum.replica-watch.info/threads/ace-4131-new-movement.10992281/

Edited 10 hours ago by Kalanag

[fellerts]

15 hours ago, VWatchie said:

Just for the fun of it. I asked ChatGpt to read this thread

I asked ChatGPT's deep research mode about blue screws in Rolex without pointing to any particular source. After thinking for 10 minutes (deep research mode), it quoted your earlier conversation with ChatGPT:

Quote

You're absolutely right that Rolex does not use blued screws in the 3135 movement or any modern movement.

We are slowly breaking the Internet, post by post. Forums, populated with real people, are crucial!

[rehajm]

Sorry if anyone feels I reacted incorrectly, thinking is a timely, healthy discussion about the limitations of AI but less so about Rolex real vs fake. There’s good information on that, too…thank you…

Edited 6 hours ago by rehajm

[Geotex]

The more “niche” a topic is, the less likely the accuracy of an AI output. You can do a reality check before sending your prompt by doing a quick Google search. If relevant info is scant and you are already familiar with the sources, skip the AI. But if you get pages and pages of results, give it a try but verify what you can. Skilled users sometimes feed the responses from one AI model to another on an unrelated platform for fact-checking, and ask for specific sources when available.

[rehajm]

3 hours ago, fellerts said:

After thinking for 10 minutes (deep research mode), it quoted your earlier conversation with ChatGPT

…that happens to me quite often on a variety of topics. I think the AI bots tend to prioritize recent contributions…

[fellerts]

ChatGPT and others use search engines just like we do. I can peek under the hood of what happened in my case: ChatGPT searched for "Rolex 3135 blued screw clone" and this thread is the 4th result in Google at the moment. In a couple years we might get a different response. Search engines definitely do prioritize recent content.

32 minutes ago, Geotex said:

If relevant info is scant and you are already familiar with the sources, skip the AI

This is very good advice. AI models are very useful for wading through and organizing vast amounts of information, but tend to "wing it" when information is scarce. "I don't know" isn't really in their vocabulary. I'm sure that'll improve with newer models though.

[VWatchie]

2 hours ago, fellerts said:

AI models are very useful for wading through and organizing vast amounts of information, but tend to "wing it" when information is scarce.

And unfortunately, very convincingly so!

It reminds me of travelling in Asia. When asking for directions, everyone points you in a specific direction, whether they have any real idea or not. It's not about trying to deceive you, but rather about avoiding the risk of disappointing you or creating an awkward atmosphere. In other words, the social interaction is prioritised over presenting the facts, which, in my example, would simply be the honest answer: "I have no idea, ask someone else."

Either way, it's easy to be misled because AI is extremely competent in certain areas, which builds a strong sense of trust. That, in turn, leads to the mistaken assumption that its expertise is equally strong across all fields — which, as we can clearly see, it is not!

@rehajm

You never responded to the question I asked in this post. It would be very interesting and greatly appreciated if you could answer it or share your thoughts on it. Thank you!

On 5/4/2025 at 3:16 PM, VWatchie said:

So @rehajm, are you telling us we're looking at a fake Rolex free sprung balance with a Breguet hairspring, micro stella regulation and KIF shock protection? If so, we were wrong to assume that we would never see that in a Chinese clone. Ay caramba!  So, what's the giveaway?

 

19 hours ago, JohnR725 said:

I was thinking what makes you think that Rolex made the screws in their watch?

I'm not sure I do. Perhaps I should have written: "Makes you wonder if the screwmaker for Rolex..."  I mean, we have to assume that Rolex stands behind every component in their movements, no matter who actually makes them, right? If I were responsible at Rolex and management forced me to accept screws that looked like they came from a bargain-bin watch, I’d probably quit on the spot. At the end of the day, you’ve got to be able to live with yourself and sleep at night.

Anyway, isn't Rolex known to manufacture every single part of their watch, including producing the various alloys for them, in-house, at least since 2004?!

Looking at the serial number, the Sea-Dweller I'm working on was manufactured in 1997, so good point, who knows who manufactured the screws!? That's also consistent with the colour of the hairspring, which in my case is not blue.

So, reading the article you linked to "Rolex 3130 / 3135 MicroStella balance July 3, 2023", until recently, the super clones used a genuine balance assembly, and I would guess likely still do. Makes you wonder how they get hold of them?!

Then again, we have the new super clones with free-sprung Breguet balances. Soon, it might be nearly impossible to tell the difference between a genuine Rolex and a fake one, huh?

Couldn't the Swiss Air Force just bomb the Chinese super clone factories?! Problem solved!

 

Edited 2 hours ago by VWatchie

[Jon]

Here's pics from a friend of mine who services Super clone Rolex watches.

This is from a 3235.

He told me non of the clones use overcoils and the blued screws are completely blue.

Rolex make their own screws by the way. They don't outsource them

Edited 30 minutes ago by Jon

[RichardHarris123]

I have neither worked on Rolex, nor a super clone, but I have read that even experts struggle to tell them apart.  Wouldn't the lack of an over coil be obvious ?