Help! Not sure how much/where to sell my diamond.

[John P]

RE: VG Cut Grade

I assume 4.5% girdle thickness?

You got it. It's precisely 0.1% thicker than the weight-ratio permitted for EX. The interesting thing is that, even if the girdle was reduced by 0.1% uniformly around the diamond, it would remain 1.16 cts.

 

[yssie]

RE: VG Cut Grade

I assume 4.5% girdle thickness?

You got it. It's precisely 0.1% thicker than the weight-ratio permitted for EX.

Thank you!!

So this summary that lists the acceptable range as 2.5%* to 4.5% means non-inclusive. That's... horribly documented.


http://www.diamondcut.gia.edu/pdfs/estimating-a-cut-grade-hires.pdf
(Published 2006, revised 2009)

The interesting thing is that, even if the girdle was reduced by 0.1% uniformly around the diamond, it would remain 1.16 cts.

Can you explain this bit? How do you know this?

 

[arkieb1]

I am trying to read the date on the certificate, is it 2009? Sorry if I am wrong it looks like that to me? I seem to remember having a interesting discussion about grading like this (ie how strict they used to be with the numbers and how they really got it wrong with many stones that were great cuts but bumped down a grade because of they were just outside what the then acceptable range was) with Garry a while ago and he was saying they have actually changed the specs a little now when they grade, so that I stone like this graded today might receive an ex where it got a vg, is that correct (John or anyone else feel free to step in here)? If so maybe she should just get it regraded..

 

[yssie]

I am trying to read the date on the certificate, is it 2009? Sorry if I am wrong it looks like that to me? I seem to remember having a interesting discussion about grading like this (ie how strict they used to be with the numbers and how they really got it wrong with many stones that were great cuts but bumped down a grade because of they were just outside what the then acceptable range was) with Garry a while ago and he was saying they have actually changed the specs a little now when they grade, so that I stone like this graded today might receive an ex where it got a vg, is that correct (John or anyone else feel free to step in here)? If so maybe she should just get it regraded..

Yup 2009 - here's the reportcheck link:
http://www.gia.edu/cs/Satellite?reportno=2101878626&c=Page&childpagename=GIA%2FPage%2FReportCheck&pagename=GIA%2FDispatcher&cid=1355954554547

 

[Rhino]

It might be John but GIA's data, to my knowledge includes 4.5% in the Ex grade. Once I get my Sarin HD back I'll be able to pin point what the deal is. If it's fine I'll consult with makhro (hi makhro! ) and we may contest the grade.

Kindest regards,
Rhino

 

[John P]

Thank you!!So this summary that lists the acceptable range as 2.5%* to 4.5% means non-inclusive. That's... horribly documented.

The title "Estimating A Cut Grade" is appropriate. That page was created to serve tradespeople who don't have non-contact scanners (that's most tradespeople btw) but want to get an idea of what to expect if they send a second-hand diamond in to the lab. As an example, it provides five virtual images in each of the five cut grades. Similarly, in GIA classes, we are taught to make estimates of brightness, fire and scintillation using their DiamondDock... While useful, none of that is happening in the lab; a non-contact scan of the diamond is used to average and round its measurements which receive a proportions grade per charted parameters established in their observational studies.

This paragraph from that page is appropriate for this case. << The displayed girdle thickness and/or total depth (when not entered by the user) is derived from other input proportions. In some cases, this may lead to girdle thickness and/or total depth values that are different than the actual values due to rounding and/or conversion from verbal descriptions (culet size and girdle thickness) >>

The interesting thing is that, even if the girdle was reduced by 0.1% uniformly around the diamond, it would remain 1.16 cts.

Can you explain this bit? How do you know this?

The math implies it. In a no rounding scenario its proportions would result in a depth of 62.7%. However the grading report states a depth of 62.1% (4.17mm) which we can see is shallower. If we go to 57.4T 34.3CA 40.9PA we arrive at 62.3% depth - which is closer to the actual 62.1% - but at those proportions a girdle of 4.5% increases the diamond to 1.17 carats. This implies that the girdle is not 4.5% but actually on the lower side, and most likely 4.3%. Reduce it by 0.01% and the report now shows 4.0%, even though the carat weight would not change one point.

Do you want to go deeper? Here's another rub, and the main question I would pose with GIA about that VG: They report only the measurement of the 16 hill positions on the diamond (AGSL reports the thinnest and thickest places on the entire girdle, for those who may wonder why the girdle % ranges seem wider on AGSL reports). This means there is an important secondary dynamic we don't know about as it relates to the girdle here, and the one I presume is keeping it out of EX.

In their defense, GIA has done the world a service by introducing a cut grade. It is pretty cool to have clients arriving in showrooms with the expectation of some kind of cut-criteria (many of my peers will recall when polish & symmetry were being framed as "the cut grade")... So while we may frown and harumph and dissect this specimen with arms crossed let's remember GIA's system is less than 10 years old. It will continue to be refined. In fact, in the past few decades the major labs have launched, revised and re-revised their cut grading systems more than a dozen times collectively, while color and clarity have stood pat since the 1930s.

 

[John P]

It might be John but GIA's data, to my knowledge includes 4.5% in the Ex grade.

I've seen this before, where all-systems-seem-go, but GIA has some reason not reflected in the rounded numbers for doing what they did.

Once I get my Sarin HD back I'll be able to pin point what the deal is. If it's fine I'll consult with makhro (hi makhro! ) and we may contest the grade.

I just wrote a tome above. If makhro is cool with it, let me know what you find out. I'm curious.

Boy, my Saturday nights are different now than they were 15 years ago.

 

[makhro82]

Hi Everyone:

I didn't realize that trying to sell my measly little 1.16 ct. stone would result in so much conversation (much of which is way over my head, all I ever knew was that it sparkled), with so many of the Pricescopers I respect . Thanks everyone for your responses and I have no problem with Rhino coming back to share the results with you.

And thanks Yssie for inviting them into this conversation. It's interesting to read even though I don't understand much I hope that my baby gets a better report card because I need more upgrade funds lol!

 

[Christina...]

Are you sure that the GIA cut grade is accurate? I'm curious as to what prevented the EX cut grade, the proportions appear to be in range. Do you know if there are girdle enhancements?

I assume 4.5% girdle thickness?
Bump it down to 4 in Facetware and it gets the EX expected from everything else. But this doc (apparently revised 2009) states that EX range is 2.5-4.5%... Unless it is non-inclusive?
http://www.diamondcut.gia.edu/pdfs/estimating-a-cut-grade-hires.pdf

Ah! According to the link you provided it does appear as though it was knocked for girdle thickness. I'm still sort of confused though because I thought that GIA based it's girdle measurements on the thinnest and thickest points, so even if this particular stones girdle measured 4.5% at any point, regardless of amount, and the remainder was under 4% it would still have been knocked to a VG? Isn't it true that in some cases the girdle thickness is adjusted to make an inclusion less obvious...not always to increase weight in order to achieve the increased weight premium? If that is true, wouldn't this make it increasingly difficult to tell if an otherwise great stone has been downgraded for reasons that would otherwise be considered beneficial to the stones overall appearance or performance?

BTW I have updates for you. I'll try to get an email off in the next day or two.

I know that w/ GIA "girdle thickness" percentage is measured at the eight kite/main junctions and reported as a percent of diameter, whereas the "thin/med/slthk/etc." descriptions are descriptions of the thinnest and thickest valleys only, and that according to GIA's documentation the valleys are usually ~1.7% thinner than the reported "girdle thickness" percentage barring brillianteering... vs AGSL's "thin/med/etc." that describe the thinnest and thickest points anywhere around the girdle, not just at the valleys.
But I can't find anything that states definitively whether a 4.5% girdle thickness is a disqualifier for the EX grade. Facetware seems to indicate as much but the charts do not, and Facetware has disclaimers stating that results may not match GIA's official verdict, and yeah - the Culet and Girdle Assessment is not helpful!! I'm tempted to just call in on Monday. Or maybe we'll get lucky and one of our tradepeople will see this.
Curiouser and curiouser

ETA: Report makhro posted in an old thread:

Thanks yssie! I didn't realize until reading the assessment that GIA measured the kite/main junctions to determine diameter and that thin/med/etc was a descriptor of the valleys. Or I guess I did know, I remember you posting about it in past threads, but I hadn't actually spent much time considering it. I also just realized when I re-read my post above that I had confused the two labs and the different ways that they assess and determine girdle measurement. I'm going to have to go back and re-read the GIA article and AGSs cut guidelines again, because I think I'm more confused now then I was before. Prepare for an onslaught of questions! Apologies for the confusion Makhro!

I'm glad that you called in the big guns! I'm super interested in seeing how all this plays out.

Makhro: I have no doubt that Jon will solve this mystery and guide you in the proper direction, and the end will result will be a beautiful new upgrade!

 

[diamondseeker2006]

Regardless of the grade, the images show it is an excellent cut stone, so it shouldn't be difficult to sell! This is the advantage of using a vendor who can prove that as opposed to buying stones from paper alone!

 

[yssie]

Thank you!!So this summary that lists the acceptable range as 2.5%* to 4.5% means non-inclusive. That's... horribly documented.

The title "Estimating A Cut Grade" is appropriate. That page was created to serve tradespeople who don't have non-contact scanners (that's most tradespeople btw) but want to get an idea of what to expect if they send a second-hand diamond in to the lab. As an example, it provides five virtual images in each of the five cut grades. Similarly, in GIA classes, we are taught to make estimates of brightness, fire and scintillation using their DiamondDock... While useful, none of that is happening in the lab; a non-contact scan of the diamond is used to average and round its measurements which receive a proportions grade per charted parameters established in their observational studies.

This paragraph from that page is appropriate for this case. << The displayed girdle thickness and/or total depth (when not entered by the user) is derived from other input proportions. In some cases, this may lead to girdle thickness and/or total depth values that are different than the actual values due to rounding and/or conversion from verbal descriptions (culet size and girdle thickness) >>

With you so far. The tools available to estimate a cut grade with don't necessarily match the methodologies used to actually assign a cut grade in the lab, and so the estimated cut grade may not match the assigned cut grade...

The interesting thing is that, even if the girdle was reduced by 0.1% uniformly around the diamond, it would remain 1.16 cts.

Can you explain this bit? How do you know this?

The math implies it. In a no rounding scenario its proportions would result in a depth of 62.7%. However the grading report states a depth of 62.1% (4.17mm) which we can see is shallower. If we go to 57.4T 34.3CA 40.9PA we arrive at 62.3% depth - which is closer to the actual 62.1% - but at those proportions a girdle of 4.5% increases the diamond to 1.17 carats. This implies that the girdle is not 4.5% but actually on the lower side, and most likely 4.3%. Reduce it by 0.01% and the report now shows 4.0%, even though the carat weight would not change one point.

It does!

Diameter avg FROM REPORT = 6.715mm
Depth FROM REPORT = 4.17mm
Table (GIA rounds to 1%) = 57.4% = 3.854mm
CA (GIA rounds to 0.5deg) = 34.3deg
CH = tan(34.3)[(6.715-3.854)/2] = 0.976mm
PA (GIA rounds up to 0.2deg) = 40.9deg
No culet, so
PH = tan(40.9)[6.715/2] = 2.908mm

CH + PH = 3.884
Girdle @ kites/mains 4.5% = 0.302mm; Depth = CH + PH + Girdle = 4.186mm
Girdle @ kites/mains 4.3% = 0.289mm; Depth = CH + PH + Girdle = 4.173mm Matches the GIA 4.17mm


I don't have DC or anything so my really (really, really) hokey calc by volume which ignores minor facets, turns continuous "faceted" arcs into circles...

V(PAV) = (pi)(3.3575^2)(2.90 = 34.329mm^3
V(CRN) = (pi/3)(3.3575^2)[tan(34.3)(3.3575)] - (pi/3)((3.854/2)^2)([tan(34.3)(3.3575)]-0.976) = 21.926mm^3
V(GIR @ 4.5%) = (pi)(3.3575^2)(0.302) = 10.702mm^3
V(GIR @ 4.3%) = (pi)(3.3575^2)(0.289) = 10.235mm^3
V(GIR @ 4.2%) = (pi)(3.3575^2)(0.282) = 9.987mm^3

Density = 3.51g/cm^3, 1ct = 0.2g, so
V(TOT, GIR @ 4.5%) = 0.066957cm^3 = 1.175ct (GIA reports 1.17ct)
V(TOT, GIR @ 4.3%) = 0.066490cm^3 = 1.167ct (GIA reports 1.16ct)
V(TOT, GIR @ 4.2%) = 0.066242cm^3 = 1.163ct (GIA reports 1.16ct)

...Super duper hokey though. Would you happen to have a real simulation for it?

A question: you said "Reduce it by 0.01%". Does this mean that GIA would note a girdle of 4.29% as 4% on the report, as opposed to rounding 4.29 -> 4.3 -> 4.5% on report? I know the Sarin scanners they use are classified as accurate to 0.01mm, so how do they treat uncertainty w/ significant digits?

Do you want to go deeper? Here's another rub, and the main question I would pose with GIA about that VG: They report only the measurement of the 16 hill positions on the diamond (AGSL reports the thinnest and thickest places on the entire girdle, for those who may wonder why the girdle % ranges seem wider on AGSL reports). This means there is an important secondary dynamic we don't know about as it relates to the girdle here, and the one I presume is keeping it out of EX.

Another question - also related to your response to Jon re. where all-systems-seem-go, but GIA has some reason not reflected in the rounded numbers for doing what they did.:

I've always thought the GIA assigns a cut grade using the numbers finally presented on the report, but this must not be the case based on your statement that there must be something going on that's not reflected in the rounded averages on the report. So at what point in the grading process is a cut grade assigned? Do they use the averaged 57.4%/34.3/40.9/4.3%/etc. and internal proportions sets not available to the general public? Is "GIA grades strictly by the proportions" an oversimplication of how the final cut grade is determined - it's not such a binary "chart says jump, you jump" process IRL?

In their defense, GIA has done the world a service by introducing a cut grade. It is pretty cool to have clients arriving in showrooms with the expectation of some kind of cut-criteria (many of my peers will recall when polish & symmetry were being framed as "the cut grade")... So while we may frown and harumph and dissect this specimen with arms crossed let's remember GIA's system is less than 10 years old. It will continue to be refined. In fact, in the past few decades the major labs have launched, revised and re-revised their cut grading systems more than a dozen times collectively, while color and clarity have stood pat since the 1930s.

Okay, okay, I'll hold off on crossing my arms and harrumphing until Rhino's scanner comes home!!

 

[yssie]

Rhino - are we there yet?
How about now?


John - had a couple more questions for you when you have time!

 

[John P]

John - had a couple more questions for you when you have time!

Time always seems elusive Yssie, but ask-away and I'll gladly contribute as I can!

 

[yssie]

John - had a couple more questions for you when you have time!

Time always seems elusive Yssie, but ask-away and I'll gladly contribute as I can!

Gosh, don't I know it! But we appreciate your help on here

I'll spare you the novel and just copy them -

1. Re. rounding:
You said "Reduce it by 0.01%". Does this mean that GIA would note a girdle of 4.29% as 4% on the report, as opposed to rounding 4.29 -> 4.3 -> 4.5% on report? I know the Sarin scanners they use are classified as accurate to 0.01mm, so how do they treat uncertainty w/ significant digits?

2. Re. assigning the final cut grade:
I've always thought the GIA assigns a cut grade using the numbers finally presented on the report, but this must not be the case based on your statement that there must be something going on that's not reflected in the rounded averages on the report. So at what point in the grading process is a cut grade assigned? Do they use the averaged 57.4%/34.3/40.9/4.3%/etc. and internal proportions sets not available to the general public? Is "GIA grades strictly by the proportions" an oversimplication of how the final cut grade is determined - it's not such a binary "chart says jump, you jump" process IRL?

 

[John P]

1. Re. rounding:
You said "Reduce it by 0.01%". Does this mean that GIA would note a girdle of 4.29% as 4% on the report, as opposed to rounding 4.29 -> 4.3 -> 4.5% on report? I know the Sarin scanners they use are classified as accurate to 0.01mm, so how do they treat uncertainty w/ significant digits?

That should be 0.1%. I initially said that, but 0.01 seems to have appeared as a typo later...late night shenanigans. My point: If 4.3 becomes 4.2 the rounding goes from 4.5% all the way to 4.0%. The result is then EX, according to Facetware.

2. Re. assigning the final cut grade:
I've always thought the GIA assigns a cut grade using the numbers finally presented on the report, but this must not be the case based on your statement that there must be something going on that's not reflected in the rounded averages on the report.

Only in the sense that some human cognition is involved. In this diamond's case there are parameters that do not work with a 4.5% girdle thickness, whereas at some shallower proportion sets 4.5% will work just fine.

So at what point in the grading process is a cut grade assigned? Do they use the averaged 57.4%/34.3/40.9/4.3%/etc. and internal proportions sets not available to the general public? Is "GIA grades strictly by the proportions" an oversimplication of how the final cut grade is determined - it's not such a binary "chart says jump, you jump" process IRL?

Internal decisions are certainly being made. It's why the introduction to the charts says: “The following tables provide ranges of individual limiting parameters for each GIA cut grade. However, the GIA Cut Grading System also considers a round brilliant diamond’s proportions together as well as individually.” Translation: While the tables cite 4.5% GT as the limit for EX, it doesn't mean that diamonds with 4.5% GT will always be EX... I consider this a more enlightened position than "chopping" blindly at this line or that, based on a chart.

GIA personnel I respect say Facetware is a good way to tell where boundaries are with different proportions sets. And...
- If you take the information on the GIA report and plug it into Facetware VG is indeed the resultant grade.
- If you use the same information but reduce GT to 4.0% EX is the resultant grade.
- This is consistent with their message that the charts are guides only. And it implies GT as the critical factor here.

I do wonder (as I believe you do) how actual data is considered vs rounded data. Looking above, Facetware gives a resultant depth of 62.7%. But the GIA Report says depth is 62.1%. To me this means the proportions are not as severe as FW seems to imply. So "IFF" the girdle is actually 4.26% (not 4.5%) but has been rounded to 4.5%...which could explain the depth discrepancy...should it still be VG? Or is there a case for EX? I'm interested in the scan Rhino will do, just to see where that micro-detail lands.

Ultimately it is GIA's system and they have every right to pronounce it as-they-will. I'm only pursuing these questions in the spirit of curiosity and better understanding.

 

[yssie]

Thanks for your thoughts John!!

1. Re. rounding:
You said "Reduce it by 0.01%". Does this mean that GIA would note a girdle of 4.29% as 4% on the report, as opposed to rounding 4.29 -> 4.3 -> 4.5% on report? I know the Sarin scanners they use are classified as accurate to 0.01mm, so how do they treat uncertainty w/ significant digits?

That should be 0.1%. I initially said that, but 0.01 seems to have appeared as a typo later...late night shenanigans. My point: If 4.3 becomes 4.2 the rounding goes from 4.5% all the way to 4.0%. The result is then EX, according to Facetware.

Okie dokie

2. Re. assigning the final cut grade:
I've always thought the GIA assigns a cut grade using the numbers finally presented on the report, but this must not be the case based on your statement that there must be something going on that's not reflected in the rounded averages on the report.

Only in the sense that some human cognition is involved. In this diamond's case there are parameters that do not work with a 4.5% girdle thickness, whereas at some shallower proportion sets 4.5% will work just fine.

So at what point in the grading process is a cut grade assigned? Do they use the averaged 57.4%/34.3/40.9/4.3%/etc. and internal proportions sets not available to the general public? Is "GIA grades strictly by the proportions" an oversimplication of how the final cut grade is determined - it's not such a binary "chart says jump, you jump" process IRL?

Internal decisions are certainly being made. It's why the introduction to the charts says: “The following tables provide ranges of individual limiting parameters for each GIA cut grade. However, the GIA Cut Grading System also considers a round brilliant diamond’s proportions together as well as individually.” Translation: While the tables cite 4.5% GT as the limit for EX, it doesn't mean that diamonds with 4.5% GT will always be EX... I consider this a more enlightened position than "chopping" blindly at this line or that, based on a chart.

GIA personnel I respect say Facetware is a good way to tell where boundaries are with different proportions sets. And...
- If you take the information on the GIA report and plug it into Facetware VG is indeed the resultant grade.
- If you use the same information but reduce GT to 4.0% EX is the resultant grade.
- This is consistent with their message that the charts are guides only. And it implies GT as the critical factor here.

I do wonder (as I believe you do) how actual data is considered vs rounded data. Looking above, Facetware gives a resultant depth of 62.7%. But the GIA Report says depth is 62.1%. To me this means the proportions are not as severe as FW seems to imply. So "IFF" the girdle is actually 4.26% (not 4.5%) but has been rounded to 4.5%...which could explain the depth discrepancy...should it still be VG? Or is there a case for EX? I'm interested in the scan Rhino will do, just to see where that micro-detail lands.

Ultimately it is GIA's system and they have every right to pronounce it as-they-will. I'm only pursuing these questions in the spirit of curiosity and better understanding.

The bolded is exactly what I wanted to hear more about. I had imagined a very linear, step-by-step process that did draw hard lines based on some parameters - much like the chopping block you speak of! So either human opinion (and therefore variability) has more influence than I had assumed, like with colour grading, or the lines are based on other/further parameters like the un-rounded data that suggests the thinner girdle - or both!

Whatever the story of this stone, it is, as you say, GIA's sandbox... these sorts of puzzles are fun to play with