@Maywood I remember when certified coins in slabs of plastic was in its infancy and people swore resistance and said it wasn't a good idea. Look how many slabbed coins there are now.
Successful BST with ad4400, Kccoin, lablover, pointfivezero, koynekwest, jwitten, coin22lover, HalfDimeDude, erwindoc, jyzskowsi, COINS MAKE CENTS, AlanSki, BryceM
Working on segmenting Obverse vs Reverse scores, so there will be both a composite score and individual sides. Also may adopt an A+, A, A- etc scale instead of 1-70 as I think it is confusing.
Wanted to re-introduce myself, my name is Trey Benedict, I am an Economics & Data Science major at Berkeley who is working alongside engineers from Apple & AWS building Pure Technologies, an S23 Y-Combinator backed startup (Early Investors in Airbnb, Stripe, DoorDash, Instacart, etc).
Hi Trey! This is very exciting. Glad to see some Y-combinator interest in this!
Our goal is to use technology and software to create transparency within the Numismatics industry. As some of you have seen before, we have already created a pricing engine for all certified coins, but we have also been silently working on color/toning analysis, which I experimented with at the Smithsonian Institution as a visiting researcher earlier this year.
This is a very interesting goal. There's quite a bit of historical pricing data to use for supervised learning as well.
Some eye candy...
[...]
I found it interesting that the @AuroraBorealis / Sirna Moose rated so low and especially against those specific coins that rated higher. This would have me investigate the algorithms a bit more.
This is an interesting coin as this coin has had multiple photos taken, including multiple TrueViews and notable images by Brandon Kelley @poorguy below. It would be interesting to see how varied the results are for photos of the same coin taken with different lighting and equipment.
DISCLAIMER THIS IS A BETA TEST, TAKE RESULTS WITH A GRAIN OF SALT
Thanks for sharing this. It will be very interesting to see this evolve.
@Zoins its a Moose because it grades 68+ and has great toning. Is it toned on both sides? Not very many MS68 Morgans out there. The one I wonder about is the other Morgan with the tab toning. Its really nice, but also only an obverse pic.
Successful BST with ad4400, Kccoin, lablover, pointfivezero, koynekwest, jwitten, coin22lover, HalfDimeDude, erwindoc, jyzskowsi, COINS MAKE CENTS, AlanSki, BryceM
@gumby1234 said: @Zoins its a Moose because it grades 68+ and has great toning. Is it toned on both sides? Not very many MS68 Morgans out there. The one I wonder about is the other Morgan with the tab toning. Its really nice, but also only an obverse pic.
I'm not wondering why it's "The Moose".
I'm wondering why the "RGB colorfulness score" for this coin is 53 when some other coins score higher. For example, this coin has a "RGB-C" score of 61, but overall is much more uniform in color:
Wanted to re-introduce myself, my name is Trey Benedict, I am an Economics & Data Science major at Berkeley who is working alongside engineers from Apple & AWS building Pure Technologies, an S23 Y-Combinator backed startup (Early Investors in Airbnb, Stripe, DoorDash, Instacart, etc).
Hi Trey! This is very exciting. Glad to see some Y-combinator interest in this!
Our goal is to use technology and software to create transparency within the Numismatics industry. As some of you have seen before, we have already created a pricing engine for all certified coins, but we have also been silently working on color/toning analysis, which I experimented with at the Smithsonian Institution as a visiting researcher earlier this year.
This is a very interesting goal. There's quite a bit of historical pricing data to use for supervised learning as well.
Some eye candy...
[...]
I found it interesting that the @AuroraBorealis / Sirna Moose rated so low and especially against those specific coins that rated higher. This would have me investigate the algorithms a bit more.
This is an interesting coin as this coin has had multiple photos taken, including multiple TrueViews and notable images by Brandon Kelley @poorguy below. It would be interesting to see how varied the results are for photos of the same coin taken with different lighting and equipment.
DISCLAIMER THIS IS A BETA TEST, TAKE RESULTS WITH A GRAIN OF SALT
Thanks for sharing this. It will be very interesting to see this evolve.
Appreciate the insights! I do believe that the image I used for the AB coin did have an effect on the result of the score. This was run on a very early version (likely one of the first 1-2 coins to be run), so that could also have an effect.
I am adopting a letter grade system from here on out and will share updated results with more training applied.
@gumby1234 said: @Zoins its a Moose because it grades 68+ and has great toning. Is it toned on both sides? Not very many MS68 Morgans out there. The one I wonder about is the other Morgan with the tab toning. Its really nice, but also only an obverse pic.
I'm not wondering why it's "the Moose".
I'm wondering why the "RGB colorfulness score" for this coin is 53 when some other coins score higher. For example, this coin has a "RGB-C" score of 61, but overall is much more uniform in color:
Also, another reason why I am segmenting obverse and reverse independent of each other. Blast white obverse with rainbow reverse will result in a suppressed composite score, and vise versa. Dual-sided toned coins fair well with the current versioning.
@treybenedict said:
Working on segmenting Obverse vs Reverse scores, so there will be both a composite score and individual sides. Also may adopt an A+, A, A- etc scale instead of 1-70 as I think it is confusing.
@treybenedict said:
Working on segmenting Obverse vs Reverse scores, so there will be both a composite score and individual sides. Also may adopt an A+, A, A- etc scale instead of 1-70 as I think it is confusing.
Why not a 1-10 score?
Binning items is easier and more accurate on an ANN than defined scoring (1-10, 1-70, 1-100). In the background it still creates a normalized score, however, bins can adapt and change over time which allows for some flexibility in the model.
@Zoins the coin you used as an example of a higher score may score 61 for color but it is AT. Do you know if the Moose or the tab toned dollars are one or two sided toners. As mentioned doubled sided toned coins will have higher scores. Like the monster ASE in the thread.
Successful BST with ad4400, Kccoin, lablover, pointfivezero, koynekwest, jwitten, coin22lover, HalfDimeDude, erwindoc, jyzskowsi, COINS MAKE CENTS, AlanSki, BryceM
@gumby1234 said: @Zoins the coin you used as an example of a higher score may score 61 for color but it is AT. Do you know if the Moose or the tab toned dollars are one or two sided toners. As mentioned doubled sided toned coins will have higher scores. Like the monster ASE in the thread.
In the production version, I have set AT/QC coins to not be scored. I do not want to cause confusion in the future, but was explaining the background of how it works.
@gumby1234 said: @Zoins the coin you used as an example of a higher score may score 61 for color but it is AT.
I was asking about the "RGB colorfulness score". Is this score affected by AT?
It's also worth understanding how is AT determined. The results here say "Abnormal amount of blue detected!" So is AT determined by the algorithm? And if so, was there a determination by a TPG, and if so, what was it?
Do you know if the Moose or the tab toned dollars are one or two sided toners. As mentioned doubled sided toned coins will have higher scores. Like the monster ASE in the thread.
The Sirna Moose is primarily an obverse toner. It does have some toning on the reverse but it is minimal.
An interesting thing to discuss is what the scores represent. If it's just trying to represent color, that's one thing, but from the original post it seems like it's trying to measure eye appeal via "Generative scores of color based on eye appeal (1-70)."
It could be that having color on both sides doesn't equal having twice, or even any more, eye appeal than a single-sided toner. Also, we know that one side, the obverse in the case of Morgans, can be worth more than the other, presumably indicating a difference in eye appeal. This is where the ML and supervised learning may become important in fine tuning the algorithm.
@gumby1234 said:
That a good idea. AT coins should get a zero.
The interesting thing about this is that PCGS QC coins can still sell for high premiums.
On the flip side, there may be a CAC "C" coin like scenario where instead of getting a zero, the grading would be hidden, for the situations where the TPG graded the coin as market acceptable.
It would be great to use these tools to track and identify raw coins.
So many immigrant groups have swept through our town that Brooklyn, like Atlantis, reaches mythological proportions in the mind of the world - RI Safir 1998
@Zoins I think obv toning on any series is more important than a reverse toner, but a coin thats toned like the ASE in this thread on both sides is amazing and should have a higher score than a single sided coin with similar toning.
Eye appeal on the coins I have seen the scans for in the thread seems to me that the coins with the "E" on the graph look like the better coins Like Trey said it is a work in progress.
Successful BST with ad4400, Kccoin, lablover, pointfivezero, koynekwest, jwitten, coin22lover, HalfDimeDude, erwindoc, jyzskowsi, COINS MAKE CENTS, AlanSki, BryceM
@gumby1234 said: @Zoins I think obv toning on any series is more important than a reverse toner, but a coin thats toned like the ASE in this thread on both sides is amazing and should have a higher score than a single sided coin with similar toning.
I agree a two-sided toner should have a higher score than a single-sided toner where the toned sides are exactly the same. I'm more curious about pricing of actual coins where the sides are likely to to be exactly the same, but may have the same score.
This goes to what the color scale is trying to represent. By originally using a 1-70 scale, I was wondering if this had to do with pricing. With pricing, there's a growing amount of toned coin pricing data and it will be interesting to see how these come together, if they are intended to come together at all.
Eye appeal on the coins I have seen the scans for in the thread seems to me that the coins with the "E" on the graph look like the better coins
Interesting observation. The axis for the vertical on the E seems to be for Hue from 0-200. For an "E" to appear, it seems certain hues would be be absent or less represented. Perhaps those are less eye-appealing hues?
Like Trey said it is a work in progress.
Agreed. It's great to be sharing this as mentioned!
@gumby1234 said: @Zoins I think obv toning on any series is more important than a reverse toner, but a coin thats toned like the ASE in this thread on both sides is amazing and should have a higher score than a single sided coin with similar toning.
I agree a two-sided toner should have a higher score than a single-sided toner where the toned sides are exactly the same. I'm more curious about pricing of actual coins where the sides are likely to to be exactly the same, but may have the same score.
This goes to what the color scale is trying to represent. By originally using a 1-70 scale, I was wondering if this had to do with pricing. With pricing, there's a growing amount of toned coin pricing data and it will be interesting to see how these come together, if they are intended to come together at all.
Eye appeal on the coins I have seen the scans for in the thread seems to me that the coins with the "E" on the graph look like the better coins
Interesting observation. The axis for the vertical on the E seems to be for Hue from 0-200. For an "E" to appear, it seems certain hues would be be absent or less represented. Perhaps those are less eye-appealing hues?
Like Trey said it is a work in progress.
Agreed. It's great to be sharing this as mentioned!
Pricing will eventually end up integrating with the toning. This was a goal of ours. However, there are a couple more steps to get there. This will be covered in the next thread
@gumby1234 said:
That a good idea. AT coins should get a zero.
I'm more of a fan of not scoring them at all.
AT analysis and scoring will end up being two separate modules. The reason for this is that AT checks can go hand-in-hand with checking for PVC, corrosion, etc. Scoring should be for only naturally toned coins. Working on this segmentation now.
@gumby1234 said:
That a good idea. AT coins should get a zero.
I'm more of a fan of not scoring them at all.
AT analysis and scoring will end up being two separate modules. The reason for this is that AT checks can go hand-in-hand with checking for PVC, corrosion, etc. Scoring should be for only naturally toned coins. Working on this segmentation now.
@keyman64 said:
The obverse gets more weight in grading and for toning. This should be the same.
Do you know how much more weight the obverse gets for grading at say PCGS?
In theory, there are no fixed weights,
and the overall grade is the minimum of the obverse and reverse grades.
In practice, I don't know - maybe the reverse grade gets overlooked sometimes if it's close?
@gumby1234 said:
That a good idea. AT coins should get a zero.
I'm more of a fan of not scoring them at all.
AT analysis and scoring will end up being two separate modules. The reason for this is that AT checks can go hand-in-hand with checking for PVC, corrosion, etc. Scoring should be for only naturally toned coins. Working on this segmentation now.
What determines natural toning?
Serious question
Martin
The answer is that it’s purely subjective. A human is calling coins naturally or artificially toned based upon their MK1 eyeball analysis and experience. The computer model is learning from their decisions and turning it into objective logic. No matter what, it’s still an objective formalization of a subjective process. If the underlying subjective decisions are flawed, then the model will reflect those flaws.
@keyman64 said:
The obverse gets more weight in grading and for toning. This should be the same.
Do you know how much more weight the obverse gets for grading at say PCGS?
I have heard a number second hand that has come from former graders but I won’t share in case it isn’t accurate these days. Most of us have heard the term “Money Side” of a coin which refers to the obverse. It is preferred to have the most attractive toning on the obverse and when we consider grading elements like cheek, portrait preservation and things like that refer to the obverse. I think it is safe to say the obverse is favored when grading.
"If it's not fun, it's not worth it." - KeyMan64 Looking for Top Pop Mercury Dime Varieties & High Grade Mercury Dime Toners.
@Martin said:
What determines natural toning?
Serious question
I think you have to go back to 1st principles, which is also why the whole NT/AT division fell apart.
Try this:
Toning is the creation of a thin film of sulfur oxide on the surface of a coin. The thin film refracts light causing the appearance of color. The color is pleasing to the human eye.
As the coin naturally continues to be exposed to contaminants, the film gets thicker, and the refracted colors change. The progression is a measurable change in color that can be quantified. At its ultimate progression, it's black. If you go back to Trey's first post, he shows the color progression and the five cycles.
This process can be accelerated by introducing contaminants beyond naturally occurring levels or increasing the reactability of the silver coin (heat), typically called artificial toning.
The progression is the same and so it is not definitively differentiable between natural and artificial.
Instead of NT/AT the classification has moved to MA/QC - Market Acceptable / Questionable Color. One sign of QC is that the coin exhibits an abnormal progression of color. Absent a blocking device like a paper tab, MA coins tend to have adjacent colors that are adjacent on the color charts. You would not expect to see a 2nd cycle gold adjacent to a 4th cycle purple without some help.
-----Burton ANA 50 year/Life Member (now "Emeritus")
Here are several I've always been curious of. The first coin I used to own and no longer do. The toning pattern always puzzled me, never seen one like it. There are 2 images of the 42 D, obviously taken under different conditions. Is it possible to run both and see how and if they differ?
I find this very interesting, thank you so much for sharing it with us. A very cool idea and I'm sure it will have use as you refine it. No rush, please take your time.
@keyman64 said:
The obverse gets more weight in grading and for toning. This should be the same.
@johnny010 said:
I’d propose a weighted score 65% for obverse vs reverse.
I think an interesting metric (that I will look at it) will be how a range of of solely obverse toned Morgan’s trade on the market versus solely reverse toned on the market.
It’s very difficult to say what the exact split is, even for grading coins. I agree that the obverse should be weighted heavier, but as to a definitive number it varies depending on the series and if numismatists are willing “overlook” other problems that are not on the obverse.
@ElmerFusterpuck said:
Still curious what kind of score my 1822 bust half would get.
I’ll try to get to some today. Have a huge backlog of emails, PM’s, and coins in the thread to get to. I’m not trying to ignore anyone’s post here, it’s just been crazy, and we’ve been developing other projects in the background.
@gumby1234 said:
That a good idea. AT coins should get a zero.
I'm more of a fan of not scoring them at all.
AT analysis and scoring will end up being two separate modules. The reason for this is that AT checks can go hand-in-hand with checking for PVC, corrosion, etc. Scoring should be for only naturally toned coins. Working on this segmentation now.
What determines natural toning?
Serious question
Martin
The answer is that it’s purely subjective. A human is calling coins naturally or artificially toned based upon their MK1 eyeball analysis and experience. The computer model is learning from their decisions and turning it into objective logic. No matter what, it’s still an objective formalization of a subjective process. If the underlying subjective decisions are flawed, then the model will reflect those flaws.
He’s not being subjective in approach h> @treybenedict said:
Sneak Peak.
Obverse & Reverse analysis will soon be segmented, and toggleable in-app to switch between. Defined Score and Letter Grade as well
My promise to you - will be out before Long Beach
Trey
Who is calling the shots behind the scene?
A- don’t like it
Why not the 70 point scale?
Don’t sell out to millennials.
@gumby1234 said:
That a good idea. AT coins should get a zero.
I'm more of a fan of not scoring them at all.
AT analysis and scoring will end up being two separate modules. The reason for this is that AT checks can go hand-in-hand with checking for PVC, corrosion, etc. Scoring should be for only naturally toned coins. Working on this segmentation now.
What determines natural toning?
Serious question
Martin
The answer is that it’s purely subjective. A human is calling coins naturally or artificially toned based upon their MK1 eyeball analysis and experience. The computer model is learning from their decisions and turning it into objective logic. No matter what, it’s still an objective formalization of a subjective process. If the underlying subjective decisions are flawed, then the model will reflect those flaws.
He’s not being subjective in approach h> @treybenedict said:
Sneak Peak.
Obverse & Reverse analysis will soon be segmented, and toggleable in-app to switch between. Defined Score and Letter Grade as well
My promise to you - will be out before Long Beach
Trey
Who is calling the shots behind the scene?
A- don’t like it
Why not the 70 point scale?
Don’t sell out to millennials.
You can toggle between Letter and Number scores
The letter grades are “bins” which put alike toners into categories.
Comments
No, it should not affect previous results.
3 Whenever's cool with me. Thank you, sir.
@Maywood I remember when certified coins in slabs of plastic was in its infancy and people swore resistance and said it wasn't a good idea. Look how many slabbed coins there are now.
Successful BST with ad4400, Kccoin, lablover, pointfivezero, koynekwest, jwitten, coin22lover, HalfDimeDude, erwindoc, jyzskowsi, COINS MAKE CENTS, AlanSki, BryceM
I have been keeping up with this thread for a few days now, and very cool and interesting to say the least. Here are a few of mine for examination.
Chirp
Successful BST with ad4400, Kccoin, lablover, pointfivezero, koynekwest, jwitten, coin22lover, HalfDimeDude, erwindoc, jyzskowsi, COINS MAKE CENTS, AlanSki, BryceM
Working on segmenting Obverse vs Reverse scores, so there will be both a composite score and individual sides. Also may adopt an A+, A, A- etc scale instead of 1-70 as I think it is confusing.
Hi Trey! This is very exciting. Glad to see some Y-combinator interest in this!
This is a very interesting goal. There's quite a bit of historical pricing data to use for supervised learning as well.
I found it interesting that the @AuroraBorealis / Sirna Moose rated so low and especially against those specific coins that rated higher. This would have me investigate the algorithms a bit more.
This is an interesting coin as this coin has had multiple photos taken, including multiple TrueViews and notable images by Brandon Kelley @poorguy below. It would be interesting to see how varied the results are for photos of the same coin taken with different lighting and equipment.
Thanks for sharing this. It will be very interesting to see this evolve.
@Zoins its a Moose because it grades 68+ and has great toning. Is it toned on both sides? Not very many MS68 Morgans out there. The one I wonder about is the other Morgan with the tab toning. Its really nice, but also only an obverse pic.
Successful BST with ad4400, Kccoin, lablover, pointfivezero, koynekwest, jwitten, coin22lover, HalfDimeDude, erwindoc, jyzskowsi, COINS MAKE CENTS, AlanSki, BryceM
I'm not wondering why it's "The Moose".
I'm wondering why the "RGB colorfulness score" for this coin is 53 when some other coins score higher. For example, this coin has a "RGB-C" score of 61, but overall is much more uniform in color:
Appreciate the insights! I do believe that the image I used for the AB coin did have an effect on the result of the score. This was run on a very early version (likely one of the first 1-2 coins to be run), so that could also have an effect.
I am adopting a letter grade system from here on out and will share updated results with more training applied.
Also, another reason why I am segmenting obverse and reverse independent of each other. Blast white obverse with rainbow reverse will result in a suppressed composite score, and vise versa. Dual-sided toned coins fair well with the current versioning.
Not my coins. But for your gold rainbow.
Tysm. Was looking for those crazy gold toners!
How about a 1-10 score?
Binning items is easier and more accurate on an ANN than defined scoring (1-10, 1-70, 1-100). In the background it still creates a normalized score, however, bins can adapt and change over time which allows for some flexibility in the model.
@Zoins the coin you used as an example of a higher score may score 61 for color but it is AT. Do you know if the Moose or the tab toned dollars are one or two sided toners. As mentioned doubled sided toned coins will have higher scores. Like the monster ASE in the thread.
Successful BST with ad4400, Kccoin, lablover, pointfivezero, koynekwest, jwitten, coin22lover, HalfDimeDude, erwindoc, jyzskowsi, COINS MAKE CENTS, AlanSki, BryceM
In the production version, I have set AT/QC coins to not be scored. I do not want to cause confusion in the future, but was explaining the background of how it works.
That a good idea. AT coins should get a zero.
Successful BST with ad4400, Kccoin, lablover, pointfivezero, koynekwest, jwitten, coin22lover, HalfDimeDude, erwindoc, jyzskowsi, COINS MAKE CENTS, AlanSki, BryceM
I was asking about the "RGB colorfulness score". Is this score affected by AT?
It's also worth understanding how is AT determined. The results here say "Abnormal amount of blue detected!" So is AT determined by the algorithm? And if so, was there a determination by a TPG, and if so, what was it?
The Sirna Moose is primarily an obverse toner. It does have some toning on the reverse but it is minimal.
An interesting thing to discuss is what the scores represent. If it's just trying to represent color, that's one thing, but from the original post it seems like it's trying to measure eye appeal via "Generative scores of color based on eye appeal (1-70)."
It could be that having color on both sides doesn't equal having twice, or even any more, eye appeal than a single-sided toner. Also, we know that one side, the obverse in the case of Morgans, can be worth more than the other, presumably indicating a difference in eye appeal. This is where the ML and supervised learning may become important in fine tuning the algorithm.
The interesting thing about this is that PCGS QC coins can still sell for high premiums.
On the flip side, there may be a CAC "C" coin like scenario where instead of getting a zero, the grading would be hidden, for the situations where the TPG graded the coin as market acceptable.
It would be great to use these tools to track and identify raw coins.
Agree. I wonder if one end goal is to create a computer algorithm-based TPG.
@Zoins I think obv toning on any series is more important than a reverse toner, but a coin thats toned like the ASE in this thread on both sides is amazing and should have a higher score than a single sided coin with similar toning.
Eye appeal on the coins I have seen the scans for in the thread seems to me that the coins with the "E" on the graph look like the better coins Like Trey said it is a work in progress.
Successful BST with ad4400, Kccoin, lablover, pointfivezero, koynekwest, jwitten, coin22lover, HalfDimeDude, erwindoc, jyzskowsi, COINS MAKE CENTS, AlanSki, BryceM
I agree a two-sided toner should have a higher score than a single-sided toner where the toned sides are exactly the same. I'm more curious about pricing of actual coins where the sides are likely to to be exactly the same, but may have the same score.
This goes to what the color scale is trying to represent. By originally using a 1-70 scale, I was wondering if this had to do with pricing. With pricing, there's a growing amount of toned coin pricing data and it will be interesting to see how these come together, if they are intended to come together at all.
Interesting observation. The axis for the vertical on the E seems to be for Hue from 0-200. For an "E" to appear, it seems certain hues would be be absent or less represented. Perhaps those are less eye-appealing hues?
Agreed. It's great to be sharing this as mentioned!
This is my new favorite thread.
Successful BST with ad4400, Kccoin, lablover, pointfivezero, koynekwest, jwitten, coin22lover, HalfDimeDude, erwindoc, jyzskowsi, COINS MAKE CENTS, AlanSki, BryceM
Pricing will eventually end up integrating with the toning. This was a goal of ours. However, there are a couple more steps to get there. This will be covered in the next thread
On the roadmap.
I'm more of a fan of not scoring them at all.
AT analysis and scoring will end up being two separate modules. The reason for this is that AT checks can go hand-in-hand with checking for PVC, corrosion, etc. Scoring should be for only naturally toned coins. Working on this segmentation now.
I’d propose a weighted score 65% for obverse vs reverse.
The obverse gets more weight in grading and for toning. This should be the same.
Looking for Top Pop Mercury Dime Varieties & High Grade Mercury Dime Toners.
Still curious what kind of score my 1822 bust half would get.
10-4,
My Instagram picturesErik
My registry sets
I think this whole thing is a 🥫of 🪱
Once again only my lack of knowledge talking
Martin
I am also still curious to see the score my Oregon Trail would get. I am excited for the website announcement
Collector of Capped Bust Halves, SLQ's, Commems, and random cool stuff! @davidv_numismatics on Instagram
Do you know how much more weight the obverse gets for grading at say PCGS?
What determines natural toning?
Serious question
Martin
In theory, there are no fixed weights,
and the overall grade is the minimum of the obverse and reverse grades.
In practice, I don't know - maybe the reverse grade gets overlooked sometimes if it's close?
The answer is that it’s purely subjective. A human is calling coins naturally or artificially toned based upon their MK1 eyeball analysis and experience. The computer model is learning from their decisions and turning it into objective logic. No matter what, it’s still an objective formalization of a subjective process. If the underlying subjective decisions are flawed, then the model will reflect those flaws.
I have heard a number second hand that has come from former graders but I won’t share in case it isn’t accurate these days. Most of us have heard the term “Money Side” of a coin which refers to the obverse. It is preferred to have the most attractive toning on the obverse and when we consider grading elements like cheek, portrait preservation and things like that refer to the obverse. I think it is safe to say the obverse is favored when grading.
Looking for Top Pop Mercury Dime Varieties & High Grade Mercury Dime Toners.
I think you have to go back to 1st principles, which is also why the whole NT/AT division fell apart.
Try this:
Toning is the creation of a thin film of sulfur oxide on the surface of a coin. The thin film refracts light causing the appearance of color. The color is pleasing to the human eye.
As the coin naturally continues to be exposed to contaminants, the film gets thicker, and the refracted colors change. The progression is a measurable change in color that can be quantified. At its ultimate progression, it's black. If you go back to Trey's first post, he shows the color progression and the five cycles.
This process can be accelerated by introducing contaminants beyond naturally occurring levels or increasing the reactability of the silver coin (heat), typically called artificial toning.
The progression is the same and so it is not definitively differentiable between natural and artificial.
Instead of NT/AT the classification has moved to MA/QC - Market Acceptable / Questionable Color. One sign of QC is that the coin exhibits an abnormal progression of color. Absent a blocking device like a paper tab, MA coins tend to have adjacent colors that are adjacent on the color charts. You would not expect to see a 2nd cycle gold adjacent to a 4th cycle purple without some help.
ANA 50 year/Life Member (now "Emeritus")
Here are several I've always been curious of. The first coin I used to own and no longer do. The toning pattern always puzzled me, never seen one like it. There are 2 images of the 42 D, obviously taken under different conditions. Is it possible to run both and see how and if they differ?
I find this very interesting, thank you so much for sharing it with us. A very cool idea and I'm sure it will have use as you refine it. No rush, please take your time.
I think an interesting metric (that I will look at it) will be how a range of of solely obverse toned Morgan’s trade on the market versus solely reverse toned on the market.
It’s very difficult to say what the exact split is, even for grading coins. I agree that the obverse should be weighted heavier, but as to a definitive number it varies depending on the series and if numismatists are willing “overlook” other problems that are not on the obverse.
Looking this moment that it’ll end up integrating in with our previous project.
There will be an app first.
I’ll try to get to some today. Have a huge backlog of emails, PM’s, and coins in the thread to get to. I’m not trying to ignore anyone’s post here, it’s just been crazy, and we’ve been developing other projects in the background.
Sounds fantastic. No worries, I am looking forward to the app!
Collector of Capped Bust Halves, SLQ's, Commems, and random cool stuff! @davidv_numismatics on Instagram
Also side note - I’ll be at Long Beach Thursday & Friday to talk about the model, theory and any other details that anyone wants to ask.
Also will have goodies for anyone who finds me
Sneak Peak.
Obverse & Reverse analysis will soon be segmented, and toggleable in-app to switch between. Defined Score and Letter Grade as well
My promise to you - will be out before Long Beach
Nice!
He’s not being subjective in approach h> @treybenedict said:
Trey
Who is calling the shots behind the scene?
A- don’t like it
Why not the 70 point scale?
Don’t sell out to millennials.
You can toggle between Letter and Number scores
The letter grades are “bins” which put alike toners into categories.
Very Interesting. I wanna see more!
https://smallcopperguy.com