split spots on Diamond Square mold from CPC? (2 Viewers)

Nice chips! I see potential influences from Le Noir Hundo, West World fiver...is that correct or coincidental?
Yeah the Noir $100 is my favorite chip. The inspiration for the 5 was that I wanted to use the 514 pattern that’s exclusive to the 44’s and the L2 spot pattern price kept the costs down
Nice chips! Congratulation for the set!! May i ask why there is no dollar sign on the green chip?
Good eye, it’s going to play high or low depending on the stakes. When low, the set has a 3, 4, 5 spot progression. When high, it’s a green chip, greens are worth $25.
 
You will find very few businesses that are as responsive as David and folks at CPC

I just want to emphasize this - I've harassed David endlessly with orders, questions, special requests, changes, and more changes, and he ALWAYS responds personally and quickly. You just cannot get his kind of service anywhere else.
 
Needs the tracking history also. Not to mention the change in artwork you didn't know about :)
Sounds interesting... do tell!
My package arrived at the Regional Facility DC Distribution Center on the 10th, where they sat for a week. It was supposed to be a two day package. On the 17th, for reasons unbeknownst to me, they were transferred to a 'local' post office with the same zip code as the regional facility (not the same as my zip code.) They stayed there for over 24 hours before returning to the Regional Facility DC Distribution Center, at which stage things finally started moving at a normal speed and direction and they were delivered this afternoon.

As for the change in artwork, since David brought it up I suppose it's not a faux pas to share -- I received this email at 5am on Jan 21st

"We ran into an issue at the start of the final pressing for your chips last night and there was no opportunity for the factory to contact you directly at that point. The extreme pressure and heat needed for 44mm chips was affecting the inlay print in some way and the first few out did not look great.
The good news is we got a new state of the art printer a couple of months ago which was used to reprint everything. The problem was that the tolerances are not as great until we are fully used to it and the factory got a much better result from using the original version background where the text lays slightly over the rocks. However if you look at the attached and compare to the original if you still have it you can see the colors are so much brighter and the finished product will show so much more detail that I don't think you will be disappointed. This was the only way we could salvage the order in the minimal time remaining. The chips should be completed in a couple of days now."

I replied by saying "no worries on swapping out the inlay graphic. Of course I can’t comment the printing tolerance issue until I see the final product, but I very much appreciate the detailed update" and David said "I think they will look better than you were originally expecting."

David was correct, the inlays look fantasic.
 
My package arrived at the Regional Facility DC Distribution Center on the 10th, where they sat for a week. It was supposed to be a two day package. On the 17th, for reasons unbeknownst to me, they were transferred to a 'local' post office with the same zip code as the regional facility (not the same as my zip code.) They stayed there for over 24 hours before returning to the Regional Facility DC Distribution Center, at which stage things finally started moving at a normal speed and direction and they were delivered this afternoon.

I’ve seen a lot of silliness in the DC area, and not just USPS. Just Wednesday, UPS shipped a package from a local center, to one further away, and then to Baltimore, before racing it back here the next day. Bonkers.
 
The replacement chips have arrived already! It looks like CPC sorted out the worst of the split spots already. With what is good I have enough to make my set complete as ordered. I am very happy with this set!

Thank you @David Spragg
B4D3EAF7-7DA2-4050-BAD4-27BC03174F06.jpeg
 
The replacement chips have arrived already! It looks like CPC sorted out the worst of the split spots already. With what is good I have enough to make my set complete as ordered. I am very happy with this set!

Thank you @David Spragg
View attachment 643643
Well just to clarify, the only way to fix this was to make a ton of them in order get 100 ok and 50 almost ok.
Obviously this is not sustainable on an ongoing basis so for problem combos there are going to be some price adjustments (some optional, so as not to penalise those who accept a higher proportion of splits).
When I have all that figured I will make a post accordingly.
 
A few observations here. First, this is a much larger problem to solve than most of us realize. If this is an issue that only occurs with particular color/spot/mold combos, then we're talking about 19 molds, 8 different spot sizes/types, & 38 different colors. This results in 19*8*38!/(38-2)! = 213,712 different permuations of spot color combos to "memorize" when trying to tackle this problem. It's simply not humanly possible for someone to remember all of those.

However, the vast majority of combinations do not cause issues, and I suspect that the mold itself actually has very little to do with it. It's much more likely to be a color combination and edge spot issue from the limited data I've seen so far. As an example, the color black as an edge spot appears to be by far the biggest offender based on the results from just this thread. Also, butterscotch spots seem to have been a problem a few times. But these will all need to be explored further to know for sure.

This is a problem that can be solved though, and it is one that is well suited for a machine learning / data science / artificial intelligence algorithm. Basically, you're just feeding a bunch of data to a computer and teaching the computer how to learn and predict which combinations would be most problematic. I write these algorithms for work every day, so I'd be happy to code this up. We'd just need to compile a dataset from various orders with everyone's chip design info (mold, spots, colors) and what percentage of them resulted in split spots and the machine learning model will tell you what to expect. It'll be extremely accurate too. (Sorry @DMack, but it can't tell you tomorrow's winning lottery numbers; it's science/mathematics, not magic).

If there is enough interest for me to solve this problem, I would be happy to do it, but I'll need everyone's help putting together the dataset. We'll need an unbiased sample though, so it can't just be data from complainers. We'll need good orders from people who've never had issues and who are unaware of this thread.
 
I would think CPC would be the best source of that information. They see every chip that comes out of the factory. I’m sure we make up a decent segment of the orders placed, but we are far from the majority of the whole.
 
A few observations here. First, this is a much larger problem to solve than most of us realize. If this is an issue that only occurs with particular color/spot/mold combos, then we're talking about 19 molds, 8 different spot sizes/types, & 38 different colors. This results in 19*8*38!/(38-2)! = 213,712 different permuations of spot color combos to "memorize" when trying to tackle this problem. It's simply not humanly possible for someone to remember all of those.

However, the vast majority of combinations do not cause issues, and I suspect that the mold itself actually has very little to do with it. It's much more likely to be a color combination and edge spot issue from the limited data I've seen so far. As an example, the color black as an edge spot appears to be by far the biggest offender based on the results from just this thread. Also, butterscotch spots seem to have been a problem a few times. But these will all need to be explored further to know for sure.

This is a problem that can be solved though, and it is one that is well suited for a machine learning / data science / artificial intelligence algorithm. Basically, you're just feeding a bunch of data to a computer and teaching the computer how to learn and predict which combinations would be most problematic. I write these algorithms for work every day, so I'd be happy to code this up. We'd just need to compile a dataset from various orders with everyone's chip design info (mold, spots, colors) and what percentage of them resulted in split spots and the machine learning model will tell you what to expect. It'll be extremely accurate too. (Sorry @DMack, but it can't tell you tomorrow's winning lottery numbers; it's science/mathematics, not magic).

If there is enough interest for me to solve this problem, I would be happy to do it, but I'll need everyone's help putting together the dataset. We'll need an unbiased sample though, so it can't just be data from complainers. We'll need good orders from people who've never had issues and who are unaware of this thread.
There are many more permutations than that. Things change when the edge spots are multicolored, especially when they are not separated, like 3D, 4D, 3TA etc. I think you'll find nearer 4 million permutations.
There are other huge factors that there is no data for, which means you could not get anywhere near an accurate prediction:
Every mold has completely different pressing characteristics, and a different finishing diameter, but there is only one diameter of slug to start with.
Molds are heated by steam channels - you cannot assume that all cups in the mold have identical temperature at any one time and this is even affected by external temperature, humidity etc. Clay will cool much faster in winter than summer for example. Sometimes all the cups in the mold will produce identical results, sometimes they won't.
When an order is started, many more slugs are made than the order calls for. How many more is a judgement call at the factory based on past experience. Sometimes it might be 15% more, sometimes 50%. You don't know the % that may be rejected for reasons other than split spots.
The 'problem' orders referred to here only account for about 5% of orders shipped during that same period. Probably only 1/3 of orders are to customers even on PCF.
 
And, we do already use past stats to back up the 'judgement call' above, it's not guesswork. What you are seeing overall though, is that with the many new patterns and colors that have been introduced over the past 5 years, there are combinations being made that have never been made before so there is no past data.
 
I don't believe in the weighted vs unweighted theory though. I think there is sufficient data to know that this isn't the culprit. I have a set of scrown tri-moons, each chip with 4 different colors on it, and half of the chips have unweighted base colors, so this set is pretty much the perfect petri dish to test out several theories with. Here are some pics of my set. As you can see, there are almost no problems with all of the unweighted chips (DG Peacock, DG Tiger, DG Yellow). The only chips that I got a number of split spots from are the dark green base and charcoal base chips, and it's only one color on each of those chips that have splits (DG Green on the dark green, and black on the charcoal). I guess I have a couple of the dark blue spots on light green chips that are close to being split, but not as bad as the black on charcoal chips. Here are pics of my random racks.

Also, I'm pretty sure that when the split spots occur, it's not the base clay color coming through the middle, between the spots. I believe it is caused by some colors contracting more than others at different temps. Black seems to contract the most (I would assume this is also what makes for different stack heights of different colors - and yes, even Paulson has this problem). The black spots contract when under heat and pressure more than the other colors, then when it contracts, that leaves a small gap on the outer edges of the chip, and that space gets filled in by the surrounding clay. So the split is actually the base color coming around the outside of the chip's edge, not through the middle. If we were to grind down the edges with a lathe, the splits would all disappear (and we'd have smaller chips). Perhaps worth noting is that Paulson chips are smaller in diameter than CPCs (usually about 39.1mm vs 39.5mm or so). Maybe they actually start off with more split spots at Paulson factory than we get to see in the end, but they grind them down to a smaller diameter with a lathe and that gets rid of most of them? I don't know.



IMG_20200114_212800.jpg

1614348896084.png

1614349089456.png

1614349706905.png

1614349784181.png

1614349869996.png

1614350086791.png
 
Man, we are missing out on seeing a lot of awesome custom sets then.
I don't know obviously, but I would reckon that CPC does a lot of business with people outside PCF through their stock line-ups (Rounders, Chesterfield, etc).

Then there's third party re-sellers like Key West selling their own lines/designs.

They have "Casino-only" molds too, but I have zero clue how many casino customers they'd have vs. private/individual customers. I would suspect the majority is to private individuals, but I could be way off base there.
 
Man, we are missing out on seeing a lot of awesome custom sets then.

I'll take the under on that bet. Almost every set of customs I've seen that came from outside this community looks like well, let's just say, they look like they came from outside this community lol. I'm sure there are a few gems we haven't seen and never will, but the number of ugly sets that get made with "Bob's Poker Chips" in Ariel font on a white background with a black '$1' in slightly larger Ariel font below is seemingly unending. Also, a ton of people seem to order blanks, or 312 sets where all chips are blanks with the same edge spot colors. I'd bet at least 90% of the good stuff comes from this community.
 
Every mold has completely different pressing characteristics, and a different finishing diameter, but there is only one diameter of slug to start with.

That's interesting. How are the finishing diameters determined? Are you using a lathe with chips, or something else? I seem to recall you saying before that it was possible to make smaller scrowns if someone wanted them (e.g., 38.5mm, similar to TRKs). Would that just require lathing them down further? If so, would that get rid of some of the split spots by grinding out the "split"?
 
That's interesting. How are the finishing diameters determined? Are you using a lathe with chips, or something else? I seem to recall you saying before that it was possible to make smaller scrowns if someone wanted them (e.g., 38.5mm, similar to TRKs). Would that just require lathing them down further? If so, would that get rid of some of the split spots by grinding out the "split"?

They use a centerless grinder.

Good question regarding the splits. Is the split coming from the inside to the outside (therefore passing through the whole spot insert) or is it from outside in?

Destructive testing on split spot chips, anyone? :)
 
They use a centerless grinder.

Good question regarding the splits. Is the split coming from the inside to the outside (therefore passing through the whole spot insert) or is it from outside in?

Destructive testing on split spot chips, anyone? :)

I may have some extras now that I can do that experiment with :oops:
 
I may have some extras now that I can do that experiment with :oops:

That has been long done.
See pics below. The body color 100% pushes its way through the spot and forces most of the spot clay out on the problem ones. If you ground this down a fraction more there would be no edge spot left at all, not a full one!
When chips are being pressed you can see the excess clay being forced out of the mold.

While it is not 100% weighted/unweighted causing it, it is down to a large degree to the mix of ingredients for each color. Obviously that is not, and can not ever be public knowledge for statutory reasons, but it does cause 90% of the issues.

IMG_3474.jpg
IMG_3475.jpg
 
@David Spragg thanks for the explanations. I find this whole process fascinating. It’s a shame there are trade secrets involved, as I’d love to watch a documentary on the process and the nuances of it all.:tup:
 
@David Spragg thanks for the explanations. I find this whole process fascinating. It’s a shame there are trade secrets involved, as I’d love to watch a documentary on the process and the nuances of it all.:tup:
It is not just 'trade secrets'. I hold a Nevada Gaming Licence, which involves regulations that are upheld by state and federal law, and they prohibit so much being disclosed that I can't explain a lot of things I would have no problem with.
 
Ah, forgot about that. Yeah, can’t mess with that.
 
That has been long done.
See pics below. The body color 100% pushes its way through the spot and forces most of the spot clay out on the problem ones. If you ground this down a fraction more there would be no edge spot left at all, not a full one!
When chips are being pressed you can see the excess clay being forced out of the mold.

While it is not 100% weighted/unweighted causing it, it is down to a large degree to the mix of ingredients for each color. Obviously that is not, and can not ever be public knowledge for statutory reasons, but it does cause 90% of the issues.

View attachment 644282View attachment 644283

Very interesting, thanks. I wonder, to what extent is this “squeezing out the inside of the spot” effect happening on all chips, even chips where the spots do not appear to be split?

I had always assumed that CPC edge spots are always solid through the whole chip.
 
Very interesting, thanks. I wonder, to what extent is this “squeezing out the inside of the spot” effect happening on all chips, even chips where the spots do not appear to be split?

I had always assumed that CPC edge spots are always solid through the whole chip.
I did bust a couple of non-problem ones at the same time and the spots were solid right through, no problem.
 
There are many more permutations than that. Things change when the edge spots are multicolored, especially when they are not separated, like 3D, 4D, 3TA etc. I think you'll find nearer 4 million permutations.
There are other huge factors that there is no data for, which means you could not get anywhere near an accurate prediction:
Every mold has completely different pressing characteristics, and a different finishing diameter, but there is only one diameter of slug to start with.
Molds are heated by steam channels - you cannot assume that all cups in the mold have identical temperature at any one time and this is even affected by external temperature, humidity etc. Clay will cool much faster in winter than summer for example. Sometimes all the cups in the mold will produce identical results, sometimes they won't.
When an order is started, many more slugs are made than the order calls for. How many more is a judgement call at the factory based on past experience. Sometimes it might be 15% more, sometimes 50%. You don't know the % that may be rejected for reasons other than split spots.
The 'problem' orders referred to here only account for about 5% of orders shipped during that same period. Probably only 1/3 of orders are to customers even on PCF.
I love these posts from @David Spragg . I understand that a lot of the process is a trade secret and for good reason, but info like this that can be put out there is fascinating. Can you do like a weekly fun facts?! :LOL: :laugh:
 
I would consider some of the chip edges posted here as problematic.
I however have to declare that I have never come to contact with a more responsive and accommodating business manager than @David Spragg , in any walk of life. I 'm sure he 'll continue to do his best, as he has always done.

To honour CPC, as a fully satisfied customer, here are some pics of non-problematic :) chips of "suspect" configurations:

DG base, Retro and DG spots:
View attachment 639022

Weighed base, DG and DG spots:
View attachment 639025

Bright White (unweighed) base, weighed spots
View attachment 639029

Retro (unweighed) base, weighed spots
View attachment 639032
Just seeing this set for the first time now, somehow. What a beautiful set! What does the "AOE" mean, is it a bounty or something like that? It is a very pretty chip. Gotta give love to that 50 too, the way those colors work together is great and a unique combo. And that 5, chefs kiss!
 
Just seeing this set for the first time now, somehow. What a beautiful set! What does the "AOE" mean, is it a bounty or something like that? It is a very pretty chip. Gotta give love to that 50 too, the way those colors work together is great and a unique combo. And that 5, chefs kiss!
Θ or Theta is a Greek letter pronounced Th.
The inscription ΑΘΕ was on ancient Athenian coins (especially the tetradrachm, of four-drachma coin), together with the Owl, the holy bird.
"Athe(e)" was just the first three letters for Athena the goddess, and her city, Athens, of for Athenians.

Edit: I f interested, here 's the initial pr0n:
https://www.pokerchipforum.com/threads/athenian-owl-club-pr0n.32945/
 
Last edited:

Create an account or login to comment

You must be a member in order to leave a comment

Create account

Create an account and join our community. It's easy!

Log in

Already have an account? Log in here.

Back
Top Bottom