Using the Empirical Method to Identify Ravenswood Blue Swirls

[Plutonianfire]

As per the previous thread on using an empirical approach to identify the maker of a blue swirl, it was noted that RAVENSWOOD made blue swirls that looked a lot like ALLEY. Namely, thick bright blue swirl lines. However, the RAVENSWOOD swirl lines tended to end at the marble seams, whereas the ALLEY lines ended in the middle of nowhere on a white base. The two photos below are known RAVENSWOOD marbles from the Joe Marble web site linked to Marble Connection. Each marble shows thick, bright blue swirl lines that tend to end at the marble seams.

 

[Plutonianfire]

Using the criterion of blue swirls with thick, bright blue swirl lines that tend to end at the marble seams, the photos below are for two marbles in my collection that meet the RAVENSWOOD criterion.

[Steph]

The base glass on your marble is the spongy kind that makes me think of Heaton or Cairo Novelty.    A great contrast between the smooth and bright white of the Ravenswood photographs.  

[Steph]

*my observation about the texture of the base glass carries no weight as to whether your marble is Heaton, Cairo Novelty, Ravenswood, or soemthing else. 

Basically I'm just pointing out a different feature which comes into play in IDs and could point to another marble company, or could stop one from being able narrow the ID down to a single company.  

[cheese]

There are no seams or cut lines presented in the two photos of yours. I would want to see more views of each and sizes before making an ID on them. Also, not every marble on the Joemarbles site is 100% infallibly identified. It's a very useful tool but as with pretty much every publication about marbles, it's not without mistakes. Just wanted to point that out since in another post you mentioned one being 100% solid on ID. The site is probably more than 99% correct though. Just be aware that there are a few. Many have tried to make WV swirls follow a flow chart and come out the end with a correct ID. It's not that simple. Spend some time with someone who knows in person and you'll see why. You need to consider way more than a couple "criterion". Size, bubbles, cold rolls, orange peel, ribbon width uniformity, tunnelling, metallic, UV response, striations, cut marks, machinery marks, tracers, opalescence, ribbon behavior, blowholes, mottling, iridescence, frit, aventurine, shapes, skips, opacity, fractures, what color it is that fractured even, and on and on are all things to be considered when making an ID. Remember the Plinko machine from The Price is Right? Imagine a big wide one with 30 or 40 chips falling through it. That's what it can sometimes be like when trying to determine the maker of a swirl. And other times it's obvious right off the bat.

[Plutonianfire]

1 hour ago, cheese said:

There are no seams or cut lines presented in the two photos of yours. I would want to see more views of each and sizes before making an ID on them.

The seams tend to get camouflaged because the bright blue swirl lines end at the seams. Please see a couple more photos that may show seams more clearly.  First two photos show continuation of seam on same marble.  Seam is less clear on second marble. polished? Both marbles 0.59-0.60 in.

 

[cheese]

I see folds. Take a couple colors of playdough and roll out two long thin strips like spaghetti noodles, then wad them up into a ball and roll it smooth. You will see lots of places where the colors "end" because the strip changed direction and then got rolled smooth. So it doesn't really end, it just dove in or got underneath the other color. Same happens with swirls. On your last pic that may be a cut at 1 or 2:00 and in a little. Where the darker blue hooks around the lighter blue a little and ends in a straight line. 

[Plutonianfire]

“Size, bubbles, cold rolls, orange peel, ribbon width uniformity, tunnelling, metallic, UV response, striations, cut marks, machinery marks, tracers, opalescence, ribbon behavior, blowholes, mottling, iridescence, frit, aventurine, shapes, skips, opacity, fractures, what color it is that fractured even, and on and on are all things to be considered when making an ID.”

I agree up to a point.  However, if you could develop a multiple regression model that simultaneously considered all of the variables that you mentioned, and the model selected only the most significant , you would likely end up with no more than four or five variables that contributed independently.

 

[Tommy]

I 💯 % agree with cheese , imho without having dug at the sites and examined bags with providence.  The marble companies in that area that made swirls lots of them shared product cullet from my research,  and glass will do pretty much the same thing with a sheer and some rollers.  That's is why I think some swirls are gonna be almost impossible . I'm just a few years old so I could be off....

Well I am off quite a bit , and missing a few marbles...🤪

[cheese]

How would it know which is most significant? It might see ribbons and a blowhole and say Pelt because that's the very logical conclusion given those two traits, but it might show a road and tunnel effect in at the seams and seem insignificant, but CACs can have blowholes too and Pelts don't do road and tunnel effects at the seam. So which would it choose? No rule is ever 100%, I can't say a pelt will never do a road and tunnel. The road and tunnel effect can vary widely to sometimes just a hint of it wanting to try to make that pattern, will parameters pick that up and if so, how will they discern that from the german type that is also similar and can give chevrons at the seam also? I use apps to identify mushrooms and they are wrong quite often. Marbles are harder than mushrooms to ID. Give it a try. You will need some very knowledgeable people to spend lots and lots of time and tens of thousands of marbles at the least to study and input from.

[Tommy]

But I hope you can come up with a formula to help that would be rad...

[Steph]

1 hour ago, Plutonianfire said:

“Size, bubbles, cold rolls, orange peel, ribbon width uniformity, tunnelling, metallic, UV response, striations, cut marks, machinery marks, tracers, opalescence, ribbon behavior, blowholes, mottling, iridescence, frit, aventurine, shapes, skips, opacity, fractures, what color it is that fractured even, and on and on are all things to be considered when making an ID.”

I agree up to a point.  However, if you could develop a multiple regression model that simultaneously considered all of the variables that you mentioned, and the model selected only the most significant , you would likely end up with no more than four or five variables that contributed independently.

 

 

Assuming you were able to identify and communicate all of the potentially relevant visual and tactile information and come up with a reliable model, reliability in this circumstance would  still give results in the general neighborhood of "there is a 75% chance that the marble was made by one of these  two or three companies."   
 

It has been a long time since my stats classes but I do remember how inexact the most practical results were.   


In my youth, from the security of the  generally well-definable notions I dealt with as a mathematician,  I glibly insulted some social scientists once because they had to make the best of some very vague answers, because, duh, it was social science, piffle.   I got schooled and eventually came to appreciate the work they did with the vague results they got.  But still ... vague.   

 

Being able to come up with as practical a model as you could given the physics and chemistry of glass, and the wide range of materials used, with standard batch formulas in early years and with common sources for glass cullet in later years, and given the intertwined histories of the companies and their machinery ... this would be impressive ... but would not mean you would have a high percentage of definitive IDs.  

 

[Fire1981]

I’ve always wondered if several WVS Factory’s used the same batch glass from one resource to make there own marbles🔥

RAR

[cheese]

Sure they did. Ohio River Valley sand being probably first and foremost. Houze made batch glass and sold it to the different companies. Vitrolite too. Some cullet WAS batch glass, just made elsewhere. 

[Plutonianfire]

4 hours ago, Steph said:

In my youth, from the security of the  generally well-definable notions I dealt with as a mathematician,  I glibly insulted some social scientists once because they had to make the best of some very vague answers, because, duh, it was social science, piffle.   I got schooled and eventually came to appreciate the work they did with the vague results they got.  But still ... vague.   

Totally understand. When you start looking at human behavior, statistical models explode, high order interaction effects abound and it’s often difficult to draw any conclusions at all.  Some 30 to 40 years ago, I attempted to answer questions like who might be willing to donate bone marrow, who was most likely to progress from asymptomatic HIV positive to full blown AIDS, and what novel risk factors might exist to facilitate prevention of cancers that science knew little about.  Those models offered insight but the issue was simply that of prolonging human life.  

https://news.cancerconnect.com/non-hodgkins-lymphoma/red-wine-consumption-associated-with-reduced-risk-of-non-hodgkin-s-lymphoma

https://onlinelibrary.wiley.com/doi/abs/10.1046/j.1537-2995.1986.26486262738.x

https://www.semanticscholar.org/paper/A-90-kDa-protein-serum-marker-for-the-prediction-of-Briggs-Natoli/06412fe329202d3b40d36a7e5526858cf6dcc2df

 

Nevertheless, with the collective expertise on this website, it should be possible to develop some relatively simple approaches that would make marble identification much easier.  I took the initiative to develop an admittedly crude paradigm for identifying blue swirls. But this could be refined by input from those who actually have a good understanding of the subject area.

 

[Nantucketdink]

comparing marbles to known examples is without equal.  Here are some known white base with blue swirl Ravenswoods.  Sure there are others not included in this group but its a more  relaible comparison to a number of Ravenswoods than you will be able to find anywhere else.

[Nantucketdink]

Maybe it’s easier to see some other types separated in my tray

[Plutonianfire]

5 hours ago, Nantucketdink said:

Maybe it’s easier to see some other types separated in my tray

Wow!  Yeah, that’s what I’m talking about.

[Plutonianfire]

17 hours ago, cheese said:

How would it know which is most significant? It might see ribbons and a blowhole and say Pelt because that's the very logical conclusion given those two traits, but it might show a road and tunnel effect in at the seams and seem insignificant, but CACs can have blowholes too and Pelts don't do road and tunnel effects at the seam. So which would it choose? No rule is ever 100%, I can't say a pelt will never do a road and tunnel. The road and tunnel effect can vary widely to sometimes just a hint of it wanting to try to make that pattern, will parameters pick that up and if so, how will they discern that from the german type that is also similar and can give chevrons at the seam also? I use apps to identify mushrooms and they are wrong quite often. Marbles are harder than mushrooms to ID. Give it a try. You will need some very knowledgeable people to spend lots and lots of time and tens of thousands of marbles at the least to study and input from.

I'm not sure it would be so difficult.  There are some savants who lurk on the marble forums.   For example, over on the All About Marbles website someone who goes by the name of CHEESE showed how it was possible to make quick work of every marble house in the country using the single variable of marble size to identify the maker of a particular marble. I believe the title of the analysis was Marble sizes and how they relate to IDs.

[Chad G.]

15 minutes ago, Plutonianfire said:

I'm not sure it would be so difficult.  There are some savants who lurk on the marble forums.   For example, over on the All About Marbles website someone who goes by the name of CHEESE showed how it was possible to make quick work of every marble house in the country using the single variable of marble size to identify the maker of a particular marble. I believe the title of the analysis was Marble sizes and how they relate to IDs.

@cheese is also on this board as I to visit AAM, personally I don't separate one from the other, all commonly shared knowledge as we are all part of one marble community.

[Carowill]

4 minutes ago, Chad G. said:

@cheese is also on this board as I to visit AAM, personally I don't separate one from the other, all commonly shared knowledge as we are all part of one marble community.

👍👍👍

[akroorka]

13 minutes ago, Plutonianfire said:

There are some savants who lurk on the marble forums.

No doubt about it my marble collecting friend--no doubt at all.
All are welcomed here and all will have input to give.
Your choice of words has not been discouraged but researched to get the gist of them.
I welcome you and all to this board regardless of the activity behind the scenes of humanity and the many themes of “Humanity”.

We are here to judge marbles—not humans—we are all after all—quite crazy!
Well—I better speak for myself—Quite crazy indeed am I ( I do speak like Yoda sometimes)---as far as marbles go at least.
Lets just go crazy!!
https://www.youtube.com/watch?v=aXJhDltzYVQ
Marble—On!!

[cheese]

If that's what you came away with from that article, I suggest going back and re-reading it. I explicitly state that there are many factors and size is just one of many. I specifically stated that patterns, glass, and other tools to use as "hints" to come up with the maker. I wrote that article because members asked me to explain how size makes a difference. Twisting my words to say I claim to be able to pinpoint every marble in the country solely by size is inaccurate to say the very least, in fact it is an intentional gross misrepresentation of what I said. I also said that size was of virtually no help at all IDing marbles other than swirls. How you think that all means I claim to be able to tell every marble under the sun by size escapes me. I've spent a good deal of time humoring you and helping you see what's involved with sorting marbles. I have a lot going on in life with two handicapped family members I take care of full time plus 3 businesses, so my time helping with marbles is limited. If this game is what I have to play to help you then it's just not worth it. More important things...

[akroorka]

1 minute ago, cheese said:

If this game is what I have to play to help you then it's just not worth it. 

Cheese and I both posted within seconds of each other. 
This is not a response to me at all.
I suggest a re-think and better vibes from all involved.
Let’s just be nice to each other---please consider this!
We are all very passionate about this hobby/addiction. 
Let’s try to keep it on a level plain.
I prefer to educate—if I can.
Marble--On!!

    
 

[Plutonianfire]

30 minutes ago, Carowill said:

👍👍👍

 

21 minutes ago, cheese said:

. If this game is what I have to play to help you then it's just not worth it. More important things...

Please. You completely misinterpreted my overly dry sense of humor. Obviously, all of the qualifiers and caveats that you pointed out are highly relevant. Just to be able to do the analysis that you did is amazing! I couldn’t come up with something like that even if I spent a full year doing background research.