Article Over Grading of Blue Fluorescent Diamonds Revisited

[Texas Leaguer]

Re: Article: Over Grading of Blue Fluorescent Diamonds Revis

HI Sharon, if we can move away from attack mode, and into solution mode....
What would be your suggestion for how to select the stones for a study? hundreds would be a good start
How many graders should assess the diamonds to give a more balanced and objective grading process?

I would rather see machines grading color but the most critical step is going to be identifying the ones the machine can not grade accurately and qc checking.
I also feel that humans could grade clarity without being near the diamond with the same or better accuracy as today and use that to teach AI how to grade. After a few years you may have a pretty good clarity grading AI or you may not.
Again the key is kicking out the ones it can't grade for human grading.

Karl, your statement above (my bold) is interesting in light of the subject of this discussion. I posted the following over a year ago in another discussion based upon a conversation I had with someone in upper management at the GIA lab, after learning that some dossiers are being color graded exclusively by machine.

Very few diamonds are actually color graded entirely by machine, and as your post suggested it is confined to diamonds under 1 carat.
The color grading devices that GIA uses are proprietary, developed in-house, and employ spectroscopic analysis. They output more than just a grade on the scale. They will indicate when certain "modifiers" are present such as hues other than yellow and fluorescence. Those stones are referred for human grading. So only the subset of small stones with no modifiers can be color graded entirely by machine.

It was the opinion of the person I talked to that color grading will go more in the direction of device-only grading as the machines continue to get better at taking all the subtle variables into account.

This is a further indicator that fluorescence complicates color grading, and further evidence that special precautions need to be taken with this subset of diamonds to avoid activating the color masking effect and to ensure accurate grading.

 

[Rockdiamond]

Re: Article: Over Grading of Blue Fluorescent Diamonds Revis

HI Sharon, if we can move away from attack mode, and into solution mode....
What would be your suggestion for how to select the stones for a study? hundreds would be a good start
How many graders should assess the diamonds to give a more balanced and objective grading process?

I would rather see machines grading color but the most critical step is going to be identifying the ones the machine can not grade accurately and qc checking.
I also feel that humans could grade clarity without being near the diamond with the same or better accuracy as today and use that to teach AI how to grade. After a few years you may have a pretty good clarity grading AI or you may not.
Again the key is kicking out the ones it can't grade for human grading.

Karl, your statement above (my bold) is interesting in light of the subject of this discussion. I posted the following over a year ago in another discussion based upon a conversation I had with someone in upper management at the GIA lab, after learning that some dossiers are being color graded exclusively by machine.

Very few diamonds are actually color graded entirely by machine, and as your post suggested it is confined to diamonds under 1 carat.
The color grading devices that GIA uses are proprietary, developed in-house, and employ spectroscopic analysis. They output more than just a grade on the scale. They will indicate when certain "modifiers" are present such as hues other than yellow and fluorescence. Those stones are referred for human grading. So only the subset of small stones with no modifiers can be color graded entirely by machine.

It was the opinion of the person I talked to that color grading will go more in the direction of device-only grading as the machines continue to get better at taking all the subtle variables into account.

This is a further indicator that fluorescence complicates color grading, and further evidence that special precautions need to be taken with this subset of diamonds to avoid activating the color masking effect and to ensure accurate grading.

A few things about your post Bryan- and we do agree
1) GIA is not going to give us exact methods- only general practices. We're never going to get them to describe exactly how they are doing it. This leaves examining stones which have been grades as the only way to "test" if they are doing it correctly.
2) Fluorescence does indeed complicate grading.

I believe GIA is taking the proper precautions based on the results we see.
And isn't this at the crux of the disagreement?
Michael is claiming they are doing it incorrectly, but there's no way to check his findings

 

[Texas Leaguer]

Re: Article: Over Grading of Blue Fluorescent Diamonds Revis

HI Sharon, if we can move away from attack mode, and into solution mode....
What would be your suggestion for how to select the stones for a study? hundreds would be a good start
How many graders should assess the diamonds to give a more balanced and objective grading process?

I would rather see machines grading color but the most critical step is going to be identifying the ones the machine can not grade accurately and qc checking.
I also feel that humans could grade clarity without being near the diamond with the same or better accuracy as today and use that to teach AI how to grade. After a few years you may have a pretty good clarity grading AI or you may not.
Again the key is kicking out the ones it can't grade for human grading.

Karl, your statement above (my bold) is interesting in light of the subject of this discussion. I posted the following over a year ago in another discussion based upon a conversation I had with someone in upper management at the GIA lab, after learning that some dossiers are being color graded exclusively by machine.

Very few diamonds are actually color graded entirely by machine, and as your post suggested it is confined to diamonds under 1 carat.
The color grading devices that GIA uses are proprietary, developed in-house, and employ spectroscopic analysis. They output more than just a grade on the scale. They will indicate when certain "modifiers" are present such as hues other than yellow and fluorescence. Those stones are referred for human grading. So only the subset of small stones with no modifiers can be color graded entirely by machine.

It was the opinion of the person I talked to that color grading will go more in the direction of device-only grading as the machines continue to get better at taking all the subtle variables into account.

This is a further indicator that fluorescence complicates color grading, and further evidence that special precautions need to be taken with this subset of diamonds to avoid activating the color masking effect and to ensure accurate grading.

A few things about your post Bryan- and we do agree
1) GIA is not going to give us exact methods- only general practices. We're never going to get them to describe exactly how they are doing it. This leaves examining stones which have been grades as the only way to "test" if they are doing it correctly.
2) Fluorescence does indeed complicate grading.

I believe GIA is taking the proper precautions based on the results we see.
And isn't this at the crux of the disagreement?
Michael is claiming they are doing it incorrectly, but there's no way to check his findings

The labs are color grading in light environments where color masking fluorescence can be activated, allowing for the possibility (if not the likelihood) of overgrading. Tom Tashey's study demonstrated that in some cases the extent of the grading discrepancy can be dramatic. Michael's study corroborated those findings, explained the causal mechanism, and suggested simple solutions for correcting the problem. He explained how doing so would benefit the market for fluorescent diamonds, while tying the need for correcting current practice back to GIA's own historical philosophy and teachings on the subject.

As much as you want to characterize the discussion as a couple of alarmists making noise, let me remind you what Martin Rapaport, arguably the most prominent person in the diamond industry, had to say about discounting of fluorescent diamonds: "Traders were and still are disturbed by the correlation between over-graded colors and fluor. Nothing upsets a trader more than buying a G and finding out the stone is really an H because some lab made a mistake. Unfortunately, the probability of a lab over-grading a fluor stone is much greater than a non-fluor stone and a large percentage of high color mistakes turn out to be fluor."

 

[Rockdiamond]

Re: Article: Over Grading of Blue Fluorescent Diamonds Revis

Bryan, if Mr Tashley was participating here, I would have the very same questions about verification and peer grading review.
Also, it seems Tom Tashley has his own gem lab, meaning GIA is his competition.
This again would mean we'd need independent verification of claims GIA is making mistakes.

Aside from that- what about Pricescope itself?
We've seen countless cases of grading problems- from EGL.
Where's the threads showing how GIA is misgrading?

"Traders were and still are disturbed by the correlation between over-graded colors and fluor. Nothing upsets a trader more than buying a G and finding out the stone is really an H because some lab made a mistake. Unfortunately, the probability of a lab over-grading a fluor stone is much greater than a non-fluor stone and a large percentage of high color mistakes turn out to be fluor."

About the Rappaort quote- He said "some lab"- there's no mention of GIA. I don't know when he said it, and I find it to be an over generalization of dealer's behavior. Traders ( the ones I know) do not buy blind- they look at the stone- so how exactly are they being tricked? In other words, how does their G become an H?
As I have pointed out repeatedly, cutters are taking a bath based on large discounts caused by fluoro.
You can simply look on your own site to see that you can buy an SB DVS1 for less than an inert F/VS1

Again- instead of discussion " the likelihood" or the "possibility" look at the actual market, and actual diamonds.

 

[Garry H (Cut Nut)]

Re: Article: Over Grading of Blue Fluorescent Diamonds Revis

Ladies and Gents, Tom was not considering diamonds graded with GIA's current practice.
Martin is a reporter not an analyst.
I too was unhappy with GIA's pre 2000 color grades of blue fluoro diamonds - but by their own round about statements they fixed it.
We do not know much about GIA's digital color grading.

What some of us do know is that GIA grades far more conservatively and my statement that no one has provided any evidence to refute is that some medium and most strong blue diamonds appear to be a better color than their GIA grade.

Sharon you are getting tired of the back and forward. Me too.
Bryan can you really tell me that you have never done a side by side set similar sized / cut comparison of diamonds and found that when there is enough light for a consumer to notice 1-2 grade differences that the fluoro stone appears to be a better grade than GIA gave it?
If you have not done it - please do it.
Approaching half the larger diamonds we sell are med-strong blue H to D and we manage to sell them because customers see this. They really do. And the motivation is rarely because of price differences. (I might add we screen out milky hazy and petrolish blue very strictly).

 

[Garry H (Cut Nut)]

Re: Article: Over Grading of Blue Fluorescent Diamonds Revis

Part of a 1998 trade article, before gIA changed the light source and moved further away - and way before digital grading.
http://www.professionaljeweler.com/archives/articles/1998/oct98/1098dg.html

To Filter or Not To Filter
In an informal experiment (attended by Professional Jeweler) conducted at the Gem Quality Institute's laboratory/booth at the JCK Show in Las Vegas in June, personnel color-graded a strongly fluorescent diamond twice – once with a UV-blocking filter, once without – using a Diamondlite with Verilux bulbs in a darkened room. The difference: two color grades.

Tom Tashey, owner of the Gem Quality Institute, says he has conducted many such experiments with moderately to strongly blue fluorescent diamonds and seen a difference of two to four grades in some cases.

 

[Rockdiamond]

Re: Article: Over Grading of Blue Fluorescent Diamonds Revis

A while back I mentioned that knowledge of color grading methodology is an important aspect of this discussion.
FL stones are a bitch to color grade- we all agree on that.
My position- GIA settles upon a fair compromise.
I took these quite informal shots, but maybe they will help.
Both stones graded G by GIA.
One is Strong Blue- the other no fluoro.
Let's start with a shot taken in a light box. I don't have a meter to test for UV, but I have noticed, over photographing literally thousands of diamonds in this box, the lighting is quite bright, yet does not activate fl color change.
To my eye , the SB stone is picking up more color than the non fluoro


Same two stones near a window,indirect sun. Now I can see a bit of that blue, which does make the SB stone look whiter.


Next office light totally shielded from sun- away from a window. In this lighting the inert stone again seems a bit whiter.



These are iphone pics, and sorry for the focus, but I believe they illustrate a bit of color change

Same two stones under the desk with UV penlight


Once we understand color grading- we can see that no matter which light we use, there's going to be other lighting environments where we will get different results. Therefore we have to select a compromise, and use it consistently.
Based on real world diamond results, it is clear that GIA's compromise works very well.

 

[michaelgem]

Re: Article: Over Grading of Blue Fluorescent Diamonds Revis

Part of a 1998 trade article, before gIA changed the light source and moved further away - and way before digital grading.
http://www.professionaljeweler.com/archives/articles/1998/oct98/1098dg.html

To Filter or Not To Filter
In an informal experiment (attended by Professional Jeweler) conducted at the Gem Quality Institute's laboratory/booth at the JCK Show in Las Vegas in June, personnel color-graded a strongly fluorescent diamond twice – once with a UV-blocking filter, once without – using a Diamondlite with Verilux bulbs in a darkened room. The difference: two color grades.

Tom Tashey, owner of the Gem Quality Institute, says he has conducted many such experiments with moderately to strongly blue fluorescent diamonds and seen a difference of two to four grades in some cases.

To be clear, Thom's 2009 discoveries were at a time before the switch to the DiamondDock when the DiamondLite was the lighting standard.

In his 2009 Gems and Gemology letter to the editor, Thomas Tashey recounts: “I was shocked when I made the initial discovery, by placing a clear, UV filter, plastic film between the Verilux lamps in the DiamondLite and the diamonds to be graded, that stones with very strong blue fluorescence could change to a lower color by three or four letter grades.” He spoke of a 0.89ct marquise brilliant with Very Strong Blue fluorescence: “In the DiamondLite [Veriluxlamps, without UV filter] this stone was graded table down as a high “D”. ... When viewed table down, with the UV filter between the lamps and the diamond, the color grade of the diamond shifted to that of a low “H”.” Tashey also found that diamonds with “medium” to “strong” blue fluorescence generally shifted one to two grades when the filter was used. (The Professional Gemologist, 2000)

In recent conversations with Thom he told me from 2005 - 2008 he acted as Chief Gemological Officer to the World Gemological Institute and a professional consultant to the Israeli Diamond Industry (IDI). He ran studies of the recently arrived new lighting standard DiamondDocks. Using the UV filter with the DiamondDock, he and the Israeli graders did not find much difference or improvement over the DiamondLite. The improvement was less than the change the 2010 study found of from four grades of fluorescence enhancement in VST Blues in the DiamondLite to the two grades in DiamondDock lighting in 3 of the 5 VST Blues in the data base of GIA graded stones.

 

[Rockdiamond]

Re: Article: Over Grading of Blue Fluorescent Diamonds Revis

Michael: I appreciate that it's difficult having your work looked at with a critical eye, and again – I also appreciate that you're here discussing it. But in that light, would you please discuss it? I have raised considerations.
The questions are about stone selection, how it was done. Did you see the diamonds before the test?
Then of course the largest objection – you're claiming a four-grade difference, with no secondary opinion.
Instead of answering that, you're bringing in quotes from another individual.
I don't know Mr. Tashey, and he is not here to defend his opinions.
But looking at this objectively, how can we ignore the fact that Mr. Tashey owns a company that is in competition with GIA. Therefore his claims that GIAs doing this wrong have to be looked at in that light.
We all know that fluorescent stones undergo color change – so I don't see what the incredible discovery is.

 

[sharonp]

Re: Article: Over Grading of Blue Fluorescent Diamonds Revis

Recently their was a thread here on Pricescope https://www.pricescope.com/community/threads/color-vs-fluorescence.221665/#post-4010495#p4010495

"Like you said an I with no fluorescence will always be an I in any lighting, though I hoped that the G could be an F or an E in certain lighting environments. Meaning that the G with strong blue fluorescence would be considered undergraded in certain environments as opposed to being overgraded in those same environments. The stigma being that fluorescence having a bit price hit because of the potential of being overgraded is in the back of my mind."

I don't want the GIA-GTL, the market or some diamond dealer to decide what the true unenhanced body color of the diamond is.
That translates to poor information for the consumer, confusion, ambiguity, and a chance for exploitation and price premium.
This consumer is 'hoping' to a buy fluoro stone to 'game' the system and taking a gamble with the three outcomes where comparisons and comparables are difficult to make:

Win = The G cost the same as an non fluoro I, in most lighting it looks like an H or I, but consumer is paying the same price as an I all other things considered equal so if it is whiter than an I in some lighting that is just a bonus.

Draw= The G costs slightly more than an I, in most lighting it is still an I, except in strong UV/VV it becomes whiter so consumer paid slightly more but its whiter sometimes.

Lose = The G costs significantly more than an I but less than a non fluoro G. In most lighting it looks like an I sometimes improving to a G in strong UV/VV.

I'd rather GIA-GTL reports its true unenhanced body color and consumers can only 'Win', as certain lighting can only improve it, as can cut, but its no lower in any lighting than what is printed on the grading report.

 

[Rockdiamond]

Re: Article: Over Grading of Blue Fluorescent Diamonds Revis

Recently their was a thread here on Pricescope https://www.pricescope.com/community/threads/color-vs-fluorescence.221665/#post-4010495#p4010495

"Like you said an I with no fluorescence will always be an I in any lighting, though I hoped that the G could be an F or an E in certain lighting environments. Meaning that the G with strong blue fluorescence would be considered undergraded in certain environments as opposed to being overgraded in those same environments. The stigma being that fluorescence having a bit price hit because of the potential of being overgraded is in the back of my mind."

I don't want the GIA-GTL, the market or some diamond dealer to decide what the true unenhanced body color of the diamond is.
That translates to poor information for the consumer, confusion, ambiguity, and a chance for exploitation and price premium.
This consumer is 'hoping' to a buy fluoro stone to 'game' the system and taking a gamble with the three outcomes where comparisons and comparables are difficult to make:

Win = The G cost the same as an non fluoro I, in most lighting it looks like an H or I, but consumer is paying the same price as an I all other things considered equal so if it is whiter than an I in some lighting that is just a bonus.

Draw= The G costs slightly more than an I, in most lighting it is still an I, except in strong UV/VV it becomes whiter so consumer paid slightly more but its whiter sometimes.

Lose = The G costs significantly more than an I but less than a non fluoro G. In most lighting it looks like an I sometimes improving to a G in strong UV/VV.

I'd rather GIA-GTL reports its true unenhanced body color and consumers can only 'Win', as certain lighting can only improve it, as can cut, but its no lower in any lighting than what is printed on the grading report.

HI Sharon
My suggestion is that you call GIA and let them know your preferences. They never do squat when I complain but maybe you have better technique

Aside from that, I do sense that you are somehow invested in this discussion. We love interested consumers.
If that's what you are, please allow me to correct an important incorrect impression you're harboring
There's not a lot of stones that GIA grades G that look like an I in most lighting. I can't say such stones do not exist, but if they do they are exceptionally rare.

The G color examples I posted are "textbook"
Both the same color in whatever lighting GIA used to determine that they are both G ( and I agree) yet we can see a slight whitening ( maybe to an F) in indirect natural lighting.
Eliminating all reflected sunlight does cause the stone to loose two shades ( IMO)
BUT- trying to look at a diamond under non natural lighting bright enough to grade color is not in any way typical for a consumer.
Plus, we have to consider face up appearance if we're worried consumers are buying misgraded stones. In that case, it's far easier to see whitening than yellowing.
Add to this, the price of the SB stone is equivalent to an I without fl.
Win win.

I like the thread you linked to- our PS regulars ( and even you) gave good advice!

 

[michaelgem]

Re: Article: Over Grading of Blue Fluorescent Diamonds Revis

Recently their was a thread here on Pricescope https://www.pricescope.com/community/threads/color-vs-fluorescence.221665/#post-4010495#p4010495

"Like you said an I with no fluorescence will always be an I in any lighting, though I hoped that the G could be an F or an E in certain lighting environments. Meaning that the G with strong blue fluorescence would be considered undergraded in certain environments as opposed to being overgraded in those same environments. The stigma being that fluorescence having a bit price hit because of the potential of being overgraded is in the back of my mind."

I don't want the GIA-GTL, the market or some diamond dealer to decide what the true unenhanced body color of the diamond is.
That translates to poor information for the consumer, confusion, ambiguity, and a chance for exploitation and price premium.
This consumer is 'hoping' to a buy fluoro stone to 'game' the system and taking a gamble with the three outcomes where comparisons and comparables are difficult to make:

Win = The G cost the same as an non fluoro I, in most lighting it looks like an H or I, but consumer is paying the same price as an I all other things considered equal so if it is whiter than an I in some lighting that is just a bonus.

Draw= The G costs slightly more than an I, in most lighting it is still an I, except in strong UV/VV it becomes whiter so consumer paid slightly more but its whiter sometimes.

Lose = The G costs significantly more than an I but less than a non fluoro G. In most lighting it looks like an I sometimes improving to a G in strong UV/VV.

I'd rather GIA-GTL reports its true unenhanced body color and consumers can only 'Win', as certain lighting can only improve it, as can cut, but its no lower in any lighting than what is printed on the grading report.

I'd rather GIA-GTL reports its true unenhanced body color and consumers can only 'Win', as certain lighting can only improve it, as can cut, but it's no lower in any lighting than what is printed on the grading report.

Sharon,

Terrific one sentence summary of the take away message for the consumer from this 11 page interaction.

Robert Weldon's article referenced by Garry is a great snapshot in time of opinion and perception of this issue, and what its resolution should be in the interest of the consumer and the diamond industry.

It's an important read and shows you are in good company.

http://www.professionaljeweler.com/archives/articles/1998/oct98/1098dg.html

Here are a few key sentences from individuals dedicated to consumer knowledge and protection :

"The Gem Quality Institute (doing business as European Gemological Laboratories) in Los Angeles, CA, and the American Gem Society Laboratory in Las Vegas, NV, now grade diamond color using filters that block UV wavelengths. Otherwise, they say, color grades could end up higher than they should be.

GIA's own Diamond Lab Manual makes the following suggestion for color-grading diamonds: "Use cool white, filtered, fluorescent light (UV-free) in a darkened room."

The AGS Lab uses a permanent filter to screen UV light when color-grading diamonds. Director Peter Yantzer says a diamond's true color should never be called into question. He says: "We have a working relationship with the industry, but one of the tenets [of AGS] is to protect consumers."

Martin Haske, owner of Adamas Gemological Laboratory in Brookline, MA, agrees. "If UV in varying amounts is available under different lighting conditions, causing different grades, why not take it out completely?"

 

[Rockdiamond]

Re: Article: Over Grading of Blue Fluorescent Diamonds Revis

Michael,
Please compare the chart on page 9 of your own study, and this

The AGS Lab uses a permanent filter to screen UV light when color-grading diamonds. Director Peter Yantzer says a diamond's true color should never be called into question. He says: "We have a working relationship with the industry, but one of the tenets [of AGS] is to protect consumers."

There was one Strong Blue diamond with a two grade disparity between GIA and AGSL. All others were within one grade or the same. Using your data, GIA was slightly more lenient on the fl stones.
But AGSL was more lenient on an inert stone.
We have seen a lot of cases where AGSL is one color grade softer than GIA on inert stones.
So maybe the AGSL lighting could also be adjusted- and there's far more inert stones being traded than fluoros.

Also to again point out that you graded a diamond as J color, GIA graded it as F and AGSL graded it as G.
So which is it- AGSL is right, or they're with GIA, and also someplace out in left field in your opinion?

Of course I don;t expect an answer. The lack of response is telling though.

 

[pfunk]

Re: Article: Over Grading of Blue Fluorescent Diamonds Revis

Michael... Any chance you will respond to some of the direct questions asked about your study? I enjoyed reading it and it certainly got me thinking, but there were questions I would like to see answered before I deem the results definitive and generalizable. Again, I appreciate your work on the subject and understand the time and effort that goes into a publication, but you too must understand the scrutiny/review that will follow. I applaud and appreciate your work and do hope you feel free to openly discuss it here. I can't speak for any others but I certainly come in with an open mind on the subject. I have no skin in the game and prefer to approach the subject in an objective manner and hope you will answer some of these questions to help us all better understand your study and its findings. Thanks!

 

[Rockdiamond]

Re: Article: Over Grading of Blue Fluorescent Diamonds Revis

Thank you Pfunk.
Michael- since you are not a "regular" participant here- I want to point out that part of my frustration is that Bryan has been holding up your study as "scientific proof" that color improvement is impossible in consumer lighting.
I also recognize the effort you put in. Also that if there are errors in the conclusions or methodology it might be hard to admit.
I'm sorry that has brought us into conflict.
I would be a willing participant if you'd like to start a new study. I have plenty of sample stones and woul be willing to submit them to AGSL to compare to the GIA grades.
You're very welcome to stop in if you're ever on 47th st

 

[michaelgem]

Re: Article: Over Grading of Blue Fluorescent Diamonds Revis

Michael... Any chance you will respond to some of the direct questions asked about your study? I enjoyed reading it and it certainly got me thinking, but there were questions I would like to see answered before I deem the results definitive and generalizable. Again, I appreciate your work on the subject and understand the time and effort that goes into a publication, but you too must understand the scrutiny/review that will follow. I applaud and appreciate your work and do hope you feel free to openly discuss it here. I can't speak for any others but I certainly come in with an open mind on the subject. I have no skin in the game and prefer to approach the subject in an objective manner and hope you will answer some of these questions to help us all better understand your study and its findings. Thanks!

Pfunk,

Thank you for your continued interest.

After 25 posts answering questions and criticisms that now fill a 60 page, 20,000 word document, about all the questions have been adressed and answered, some multiple times. When the same questions are continually repeated it becomes apparent that the motives of the posters repeating these questions/criticisms are not a better understanding of the study and its findings.

Could you be specific about the questions you still have? If you have read the summary article posted here, along with the original study which is linked in the summary, and the commentary contained in these 25 posts, I think you will find most questions have been answered. Having said that, if I have missed something or have not been clear, I will do my best to clarify.

 

[Karl_K]

Re: Article: Over Grading of Blue Fluorescent Diamonds Revis

Something that would be interesting is have flats cut to the same thickness, laser a number on each.
Some with v-strong florescence, some med, some with none and different color grades and even different tones if you wanted to. (agree on 3 experts to set the base color grades, ask the labs if they want to play too.)
They could be cut from tops or broken diamonds to keep costs down.
Then do a simple which has less color with different lighting and with different people.
That would help eliminate differences in cut from the equation and would be low value enough to send them around the world to different people.
Could even have consumers get in the game at the PS get together in Vegas.

 

[Rockdiamond]

Re: Article: Over Grading of Blue Fluorescent Diamonds Revis

Hi Karl, No need to get so complex!
There's a few basic questions raised that can be easily researched.
1)) Is GIA overgrading diamonds due to an error in methodology based on lighting?
Since we know AGSL is shielding for UV, we can simply compare. We can submit some MB/SB stones to GIA that have been AGSL graded, or vice versa. I know there's a study here someplace, but not focused specifically on FL diamonds.
2) Are consumers paying too much for FL diamonds? Easy to do a bit of research. We could limit the samples to RBC- either GIA XXX or AGSL 0 to take cut out of the equation ( as much as possible)
Then check average prices for inert, MB/SB stones on the market on any given day. We can do this in colorless, near colorless, and even down to the K-L range. If the first set of results shows a 5% difference in grades, and the market price difference is 20% we can deduce that consumers are not overpaying.
3) Do some stones show improvement in "normal" lighting? For this one I'd be happy to submit a stone which possesses this characteristic to send to Las Vegas ( regretfully we can't make it this year) and maybe Garry or others will be able to also bring samples.
Of course this third category will have the same basic faults as Micheal calling a stone J that GIA called F- we're talking about human perception, as opposed to "hard facts".
Still if we have observers testifying that would be positive.

 

[michaelgem]

Re: Article: Over Grading of Blue Fluorescent Diamonds Revis

Thank you Pfunk.
Michael- since you are not a "regular" participant here- I want to point out that part of my frustration is that Bryan has been holding up your study as "scientific proof" that color improvement is impossible in consumer lighting.
I also recognize the effort you put in. Also that if there are errors in the conclusions or methodology it might be hard to admit.
I'm sorry that has brought us into conflict.
I would be a willing participant if you'd like to start a new study. I have plenty of sample stones and woul be willing to submit them to AGSL to compare to the GIA grades.
You're very welcome to stop in if you're ever on 47th st

David,

Bryan can correct me if I'm wrong but it seems you are reading more into his words than is intended. The 2010 study supports the following :

At usual diamond viewing distances from most artificial overhead consumer lighting those with a UV radiometer and light meter including AGSL and the AGA lighting task force have found around 1uW or less of UV and a lot less than 400fc of visible light intensity. The study showed that in these circumstances there is insufficient UV and VV to stimulate any color improvement from blue fluorescence. Consequently, the true body color (as defined by GIA) is seen.

This is the color Sharon is asking GIA to grade when she says: I'd rather GIA-GTL reports its true unenhanced body color and consumers can only 'Win', as certain lighting can only improve it ...(the color would then never be) lower in any lighting than what is printed on the grading report.

For example, take the 1.13ct VST Blue pear brilliant (study diamond #4.) In the DiamondDock lighting GIA-GTL graded it an H, as did I and AGSL (Note that for compatibility with GIA grading, AGSL grades in a GIA unfiltered DiamondDock. see photo).

Absent the blue fluorescence stimulating UV and VV in the DiamondDock lighting the pear's color dropped two grades to a J, its true body color. Though she may not notice this drop of two grades in artificial light, the consumer of this diamond would have paid based upon a color grade that needs the boost of UV and VV to appear an H.

Had the consumer paid based on the unenhanced true color of J she might be delighted to find her diamond looking as white or whiter than her friend's H color stone when viewed at 7 inches from the jeweler's desk lamp. And she would be even more delighted if the jeweler is Garry, and he takes her near his store windows on a sunny day in Australia. There the diamond looks like its F color graded in the DiamondLite in 150uW of UV and 600fc of illumination. Around noon outside where there is over 1500 uW of UV the perceived color would rise past D and into the realm of blue-white.

I suspect we can all agree that blue fluorescent diamonds are easy to sell when they are shown to whiten from their graded color, and even appear blue-white in natural daylight.

This leads to the study conclusion #3: Promoting this advantage in comparison with non-fluorescent diamonds of similar grade would return the marketing advantage to blue fluorescent diamonds that they once enjoyed.

By grading in lighting that does not stimulate fluorescence, fairness and consistency can be achieved, restoring trust in and renewing desire for this outstanding gemstone.

 

[Karl_K]

Re: Article: Over Grading of Blue Fluorescent Diamonds Revis

Since we know AGSL is shielding for UV,

Do we know that? based on an experience with them I wonder about that.
Can someone ping them and ask? Since Peter left I don't have a good contact email addy for them.

 

[Texas Leaguer]

Re: Article: Over Grading of Blue Fluorescent Diamonds Revis

Thank you Pfunk.
Michael- since you are not a "regular" participant here- I want to point out that part of my frustration is that Bryan has been holding up your study as "scientific proof" that color improvement is impossible in consumer lighting.
I also recognize the effort you put in. Also that if there are errors in the conclusions or methodology it might be hard to admit.
I'm sorry that has brought us into conflict.
I would be a willing participant if you'd like to start a new study. I have plenty of sample stones and woul be willing to submit them to AGSL to compare to the GIA grades.
You're very welcome to stop in if you're ever on 47th st

David,

Bryan can correct me if I'm wrong but it seems you are reading more into his words than is intended. The 2010 study supports the following :

At usual diamond viewing distances from most artificial overhead consumer lighting those with a UV radiometer and light meter including AGSL and the AGA lighting task force have found around 1uW or less of UV and a lot less than 400fc of visible light intensity. The study showed that in these circumstances there is insufficient UV and VV to stimulate any color improvement from blue fluorescence. Consequently, the true body color (as defined by GIA) is seen.

This is the color Sharon is asking GIA to grade when she says: I'd rather GIA-GTL reports its true unenhanced body color and consumers can only 'Win', as certain lighting can only improve it ...(the color would then never be) lower in any lighting than what is printed on the grading report.

For example, take the 1.13ct VST Blue pear brilliant (study diamond #4.) In the DiamondDock lighting GIA-GTL graded it an H, as did I and AGSL (Note that for compatibility with GIA grading, AGSL grades in a GIA unfiltered DiamondDock. see photo).

Absent the blue fluorescence stimulating UV and VV in the DiamondDock lighting the pear's color dropped two grades to a J, its true body color. Though she may not notice this drop of two grades in artificial light, the consumer of this diamond would have paid based upon a color grade that needs the boost of UV and VV to appear an H.

Had the consumer paid based on the unenhanced true color of J she might be delighted to find her diamond looking as white or whiter than her friend's H color stone when viewed at 7 inches from the jeweler's desk lamp. And she would be even more delighted if the jeweler is Garry, and he takes her near his store windows on a sunny day in Australia. There the diamond looks like its F color graded in the DiamondLite in 150uW of UV and 600fc of illumination. Around noon outside where there is over 1500 uW of UV the perceived color would rise past D and into the realm of blue-white.

I suspect we can all agree that blue fluorescent diamonds are easy to sell when they are shown to whiten from their graded color, and even appear blue-white in natural daylight.

This leads to the study conclusion #3: Promoting this advantage in comparison with non-fluorescent diamonds of similar grade would return the marketing advantage to blue fluorescent diamonds that they once enjoyed.

By grading in lighting that does not stimulate fluorescence, fairness and consistency can be achieved, restoring trust in and renewing desire for this outstanding gemstone.

You are not wrong about my my statements Michael, as anyone who has followed my posts here or on other fluorescence threads understands clearly. Unfortunately RockDiamond chooses to mis-characterize my statements on a regular basis. Either this is intentional misrepresentation or he is incapable of understanding my posts. In any case, it is bad for any discussion that is aimed at greater understanding because the actual words are important.

I have publicly asked him on other occasions to put quotes around those things he claims I have said, and to quote them accurately. He aparently has no interest in doing so. He actually imbeds his own quotes in statements that inaccurately paraphrase someone else, leaving the impression that they actually said what he claims. While that practice is academically dishonest and is intended more to obfuscate the subject matter rather than shine light on it, I think that most people reading here do understand my perspectives so I no longer feel the need to respond to this sort of thing. Since it is not likely to change, I have instead tried to adapt my response to it.

 

[Texas Leaguer]

Re: Article: Over Grading of Blue Fluorescent Diamonds Revis

Since we know AGSL is shielding for UV,

Do we know that? based on an experience with them I wonder about that.
Can someone ping them and ask? Since Peter left I don't have a good contact email addy for them.

My understanding is that AGSL emulates GIA color grading as closely as possible today. The statement about filtering came from the 1998 article. My guess is that when the DiamondDock came out they synchonized.

 

[pfunk]

Re: Article: Over Grading of Blue Fluorescent Diamonds Revis

Michael, as I read your paper I have a couple more questions for you. I am personally more experienced with reading medical journals (more specifically pharmaceutical research) and am accustomed to finding the "limitations" section at the end. I am curious as to whether you formulated a list of limitations and what that list may look like?

As a reader, the first question I am asking myself is what are your credentials and training as a diamond grader?
What sort of master set of stones did you use for your grading of these 25 stones?
Was the selection of diamonds to be used done at random and why only 25 stones?
Were the specified light sources the ONLY source of light in the room at the time of grading?

I hope you don't take these questions as an indication of me trying to poke holes in your study. They are very real questions I would like to know in order to evaluate the study. I happen to come into this discussion feeling like a UV filter should be used for diamond grading and don't see why the use of a filter would be detrimental to the consumer. If UV light is causing a chameleon effect on stones that is dependent upon the lighting it should be eliminated as a variable, especially considering that we don't know why certain fluorescing stones behave one way and others seem to behave another way.

Here is a post I made when I first got into the discussion Michael. Since you are the only one grading the stones and claiming the large (2-4 grade) discrepancies, I think it is imperative that you inform readers of your methodology in a very specific manner. What kind of master set? Had the stones been cleaned recently? Diamond masters or simulants? A description of your specific grading method would be helpful (i believe you touched on this to a degree in previous posts). Were stones chosen at random and were these stones ones that you had seen before and graded prior to your study? When using such a small sample set, it would be very important that these stones not be known to you or graded by you previously (IMO). If they were, you may introduce selection bias in which you chose stones you previously hypothesized may have been overgraded due to their fluoro. I appreciate your time here discussing these questions. Criticism can be a real pain, but the best researchers use that criticism to help strengthen their studies. Thanks!

 

[michaelgem]

Re: Article: Over Grading of Blue Fluorescent Diamonds Revis

Since we know AGSL is shielding for UV,

Do we know that? based on an experience with them I wonder about that.
Can someone ping them and ask? Since Peter left I don't have a good contact email addy for them.

My understanding is that AGSL emulates GIA color grading as closely as possible today. The statement about filtering came from the 1998 article. My guess is that when the DiamondDock came out they synchonized.

You are right Bryan.

from my previous post: "For example, take the 1.13ct VST Blue pear brilliant (study diamond #4.) In the DiamondDock lighting GIA-GTL graded it an H, as did I and AGSL (Note that for compatibility with GIA grading, AGSL grades in a GIA unfiltered DiamondDock. see photo).

The photo is of the AGSL grader using one of their DiamondDocks with (Inset) AGSL's master diamond set to grade the 25 diamond data-base.

 

[Rockdiamond]

Re: Article: Over Grading of Blue Fluorescent Diamonds Revis

Thank you Pfunk.
Michael- since you are not a "regular" participant here- I want to point out that part of my frustration is that Bryan has been holding up your study as "scientific proof" that color improvement is impossible in consumer lighting.
I also recognize the effort you put in. Also that if there are errors in the conclusions or methodology it might be hard to admit.
I'm sorry that has brought us into conflict.
I would be a willing participant if you'd like to start a new study. I have plenty of sample stones and woul be willing to submit them to AGSL to compare to the GIA grades.
You're very welcome to stop in if you're ever on 47th st

David,

Bryan can correct me if I'm wrong but it seems you are reading more into his words than is intended. The 2010 study supports the following :

At usual diamond viewing distances from most artificial overhead consumer lighting those with a UV radiometer and light meter including AGSL and the AGA lighting task force have found around 1uW or less of UV and a lot less than 400fc of visible light intensity. The study showed that in these circumstances there is insufficient UV and VV to stimulate any color improvement from blue fluorescence. Consequently, the true body color (as defined by GIA) is seen.
My issue with a statement like this is that you're quoting very scientific sounding measurments without qualifying the other vital aspect. What was considered "usual" and "most ariticial overhead consumer lighting"? Was there a window in the room? A large percentage of consumer lighting environments will contain some form of natural lighting- windows. If a consuner is admiring their diamond at night, or in a room with no windows, the liklihood of them being able to determine the difference between F and I color is very small.
So it seems very likely that you are taking measurements in non typical lighting environments. Bottom line is that the conclusion is incorrect.


This is the color Sharon is asking GIA to grade when she says: I'd rather GIA-GTL reports its true unenhanced body color and consumers can only 'Win', as certain lighting can only improve it ...(the color would then never be) lower in any lighting than what is printed on the grading report.

For example, take the 1.13ct VST Blue pear brilliant (study diamond #4.) In the DiamondDock lighting GIA-GTL graded it an H, as did I and AGSL (Note that for compatibility with GIA grading, AGSL grades in a GIA unfiltered DiamondDock. see photo).

Absent the blue fluorescence stimulating UV and VV in the DiamondDock lighting the pear's color dropped two grades to a J, its true body color. Though she may not notice this drop of two grades in artificial light, the consumer of this diamond would have paid based upon a color grade that needs the boost of UV and VV to appear an H.
Michael- this is a crucial issue with the results. YOU thought it looked like a J. Without corroration any single graders opinion is meaningless. Every day I have cutters swearing GIA is giving their G color the grade of H.
Given that GIA is considered the industry leader, and you're claiming they are making a mistake, you would need a variety of other independant qualified graders to agree with your assesment . GIA themselves publish the fact multiple graders look at important stones. Add to this the fact that you've desribed how you assesed the diamond, at shoulder height 36 inches from a bulb. Using such unconvential methods while at the same time claiming GIA is 4 grades off is another consideration about your results.
Had the consumer paid based on the unenhanced true color of J she might be delighted to find her diamond looking as white or whiter than her friend's H color stone when viewed at 7 inches from the jeweler's desk lamp. And she would be even more delighted if the jeweler is Garry, and he takes her near his store windows on a sunny day in Australia. There the diamond looks like its F color graded in the DiamondLite in 150uW of UV and 600fc of illumination. Around noon outside where there is over 1500 uW of UV the perceived color would rise past D and into the realm of blue-white.

I suspect we can all agree that blue fluorescent diamonds are easy to sell when they are shown to whiten from their graded color, and even appear blue-white in natural daylight.

This leads to the study conclusion #3: Promoting this advantage in comparison with non-fluorescent diamonds of similar grade would return the marketing advantage to blue fluorescent diamonds that they once enjoyed.

By grading in lighting that does not stimulate fluorescence, fairness and consistency can be achieved, restoring trust in and renewing desire for this outstanding gemstone.

Hi Michael- I will add to my reply soon- but the two points I've brought up here are crucial

 

[Rockdiamond]

Re: Article: Over Grading of Blue Fluorescent Diamonds Revis

....snip

The AGS Lab uses a permanent filter to screen UV light when color-grading diamonds. Director Peter Yantzer says a diamond's true color should never be called into question. He says: "We have a working relationship with the industry, but one of the tenets [of AGS] is to protect consumers."

...snip

this was the post that caused confusion about AGSL grading. It seems as though this statment was made in 1998, but now AGSL confirms to GIA lighting- is that correct?

 

[Rockdiamond]

Re: Article: Over Grading of Blue Fluorescent Diamonds Revis

Thank you Pfunk.
Michael- since you are not a "regular" participant here- I want to point out that part of my frustration is that Bryan has been holding up your study as "scientific proof" that color improvement is impossible in consumer lighting.
I also recognize the effort you put in. Also that if there are errors in the conclusions or methodology it might be hard to admit.
I'm sorry that has brought us into conflict.
I would be a willing participant if you'd like to start a new study. I have plenty of sample stones and woul be willing to submit them to AGSL to compare to the GIA grades.
You're very welcome to stop in if you're ever on 47th st

David,

Bryan can correct me if I'm wrong but it seems you are reading more into his words than is intended. The 2010 study supports the following :

At usual diamond viewing distances from most artificial overhead consumer lighting those with a UV radiometer and light meter including AGSL and the AGA lighting task force have found around 1uW or less of UV and a lot less than 400fc of visible light intensity. The study showed that in these circumstances there is insufficient UV and VV to stimulate any color improvement from blue fluorescence. Consequently, the true body color (as defined by GIA) is seen.

This is the color Sharon is asking GIA to grade when she says: I'd rather GIA-GTL reports its true unenhanced body color and consumers can only 'Win', as certain lighting can only improve it ...(the color would then never be) lower in any lighting than what is printed on the grading report.

For example, take the 1.13ct VST Blue pear brilliant (study diamond #4.) In the DiamondDock lighting GIA-GTL graded it an H, as did I and AGSL (Note that for compatibility with GIA grading, AGSL grades in a GIA unfiltered DiamondDock. see photo).

Absent the blue fluorescence stimulating UV and VV in the DiamondDock lighting the pear's color dropped two grades to a J, its true body color. Though she may not notice this drop of two grades in artificial light, the consumer of this diamond would have paid based upon a color grade that needs the boost of UV and VV to appear an H.

Had the consumer paid based on the unenhanced true color of J she might be delighted to find her diamond looking as white or whiter than her friend's H color stone when viewed at 7 inches from the jeweler's desk lamp. And she would be even more delighted if the jeweler is Garry, and he takes her near his store windows on a sunny day in Australia. There the diamond looks like its F color graded in the DiamondLite in 150uW of UV and 600fc of illumination. Around noon outside where there is over 1500 uW of UV the perceived color would rise past D and into the realm of blue-white.

I suspect we can all agree that blue fluorescent diamonds are easy to sell when they are shown to whiten from their graded color, and even appear blue-white in natural daylight.

This leads to the study conclusion #3: Promoting this advantage in comparison with non-fluorescent diamonds of similar grade would return the marketing advantage to blue fluorescent diamonds that they once enjoyed.

By grading in lighting that does not stimulate fluorescence, fairness and consistency can be achieved, restoring trust in and renewing desire for this outstanding gemstone.

You are not wrong about my my statements Michael, as anyone who has followed my posts here or on other fluorescence threads understands clearly. Unfortunately RockDiamond chooses to mis-characterize my statements on a regular basis. Either this is intentional misrepresentation or he is incapable of understanding my posts. In any case, it is bad for any discussion that is aimed at greater understanding because the actual words are important.

I have publicly asked him on other occasions to put quotes around those things he claims I have said, and to quote them accurately. He aparently has no interest in doing so. He actually imbeds his own quotes in statements that inaccurately paraphrase someone else, leaving the impression that they actually said what he claims. While that practice is academically dishonest and is intended more to obfuscate the subject matter rather than shine light on it, I think that most people reading here do understand my perspectives so I no longer feel the need to respond to this sort of thing. Since it is not likely to change, I have instead tried to adapt my response to it.

HI Bryan
Please do as you say I should and post the quoted places where I've misinerpted your writing.
You have been using the findings of this study to claim that MB SB stones can't be benefitting from whitening in many consumer enviroments, haven't you?
If my imperssion is not correct, please correct it.

 

[Garry H (Cut Nut)]

Re: Article: Over Grading of Blue Fluorescent Diamonds Revis

A while back I mentioned that knowledge of color grading methodology is an important aspect of this discussion.
FL stones are a bitch to color grade- we all agree on that.
My position- GIA settles upon a fair compromise.
I took these quite informal shots, but maybe they will help.
Both stones graded G by GIA.
One is Strong Blue- the other no fluoro.
Let's start with a shot taken in a light box. I don't have a meter to test for UV, but I have noticed, over photographing literally thousands of diamonds in this box, the lighting is quite bright, yet does not activate fl color change.
To my eye , the SB stone is picking up more color than the non fluoro


Same two stones near a window,indirect sun. Now I can see a bit of that blue, which does make the SB stone look whiter.


Next office light totally shielded from sun- away from a window. In this lighting the inert stone again seems a bit whiter.



These are iphone pics, and sorry for the focus, but I believe they illustrate a bit of color change

Same two stones under the desk with UV penlight


Once we understand color grading- we can see that no matter which light we use, there's going to be other lighting environments where we will get different results. Therefore we have to select a compromise, and use it consistently.
Based on real world diamond results, it is clear that GIA's compromise works very well.

I can not believe no one has been interested in this post David?

It shows what Michael and others are saying - that GIA have overgraded the Fluoro stone, but in a room with daylight lighting the stone has been undergraded.

Is it possible for you to show the position the and distance from the window, rough time of day and direction the sun was coming from, what other lighting was available and if you were blocking that with your body etc. Are the windows double glazed or have any films attached?
If photographs are a security issue then perhaps a simple floor plan sketch?

 

[Garry H (Cut Nut)]

Re: Article: Over Grading of Blue Fluorescent Diamonds Revis

Something that would be interesting is have flats cut to the same thickness, laser a number on each.
Some with v-strong florescence, some med, some with none and different color grades and even different tones if you wanted to. (agree on 3 experts to set the base color grades, ask the labs if they want to play too.)
They could be cut from tops or broken diamonds to keep costs down.
Then do a simple which has less color with different lighting and with different people.
That would help eliminate differences in cut from the equation and would be low value enough to send them around the world to different people.
Could even have consumers get in the game at the PS get together in Vegas.

Interesting Karl - not sure how you grade the color of a flat? Explain more please.

 

[Garry H (Cut Nut)]

Re: Article: Over Grading of Blue Fluorescent Diamonds Revis

Part of a 1998 trade article, before gIA changed the light source and moved further away - and way before digital grading.
http://www.professionaljeweler.com/archives/articles/1998/oct98/1098dg.html

To Filter or Not To Filter
In an informal experiment (attended by Professional Jeweler) conducted at the Gem Quality Institute's laboratory/booth at the JCK Show in Las Vegas in June, personnel color-graded a strongly fluorescent diamond twice – once with a UV-blocking filter, once without – using a Diamondlite with Verilux bulbs in a darkened room. The difference: two color grades.

Tom Tashey, owner of the Gem Quality Institute, says he has conducted many such experiments with moderately to strongly blue fluorescent diamonds and seen a difference of two to four grades in some cases.

To be clear, Thom's 2009 discoveries were at a time before the switch to the DiamondDock when the DiamondLite was the lighting standard.

In his 2009 Gems and Gemology letter to the editor, Thomas Tashey recounts: “I was shocked when I made the initial discovery, by placing a clear, UV filter, plastic film between the Verilux lamps in the DiamondLite and the diamonds to be graded, that stones with very strong blue fluorescence could change to a lower color by three or four letter grades.” He spoke of a 0.89ct marquise brilliant with Very Strong Blue fluorescence: “In the DiamondLite [Veriluxlamps, without UV filter] this stone was graded table down as a high “D”. ... When viewed table down, with the UV filter between the lamps and the diamond, the color grade of the diamond shifted to that of a low “H”.” Tashey also found that diamonds with “medium” to “strong” blue fluorescence generally shifted one to two grades when the filter was used. (The Professional Gemologist, 2000)

In recent conversations with Thom he told me from 2005 - 2008 he acted as Chief Gemological Officer to the World Gemological Institute and a professional consultant to the Israeli Diamond Industry (IDI). He ran studies of the recently arrived new lighting standard DiamondDocks. Using the UV filter with the DiamondDock, he and the Israeli graders did not find much difference or improvement over the DiamondLite. The improvement was less than the change the 2010 study found of from four grades of fluorescence enhancement in VST Blues in the DiamondLite to the two grades in DiamondDock lighting in 3 of the 5 VST Blues in the data base of GIA graded stones.

Michael do you see that when using the old GIA Diamondlite Tom (NOT THOM) said SOME diamonds. Do you see that you have quoted him wrongly and placed a higher emphasis on his interpretation.
If it was SOME back then, surely it is a small number today with Diamond Dock.

Michael do you agree or disagree that consumers (and us lot) often see grade improvement in Med and Strong blue diamonds (and please note I am not and have not been discussing VSTB's because they really must be on a case by case basis).