Indexing Upper Half Facets by Peter Yantzer, AGS Lab Director

The Effects of Indexed Upper Half Facets by Peter Yantzer, Director AGS Labs

The article analyses effect of the upper girdle facets on visual appearance, light leakage, weight retention, girdle, etc of the round brilliant diamonds.


American Gem Society's ASET ( Angular Spectrum Evaluation Tool ) images are used in addition to Firescope, H&A, and brightness simulations

Thank you Peter for posting this wonderful easy to understand explanation.

And thank''s also to the Boards of AGS and AGS Lab. for allowing Peter to post it here.

Awesome,
Thank you Mr. Yantzer and AGS.
A quick read finds it to be both informative and interesting.
It deserves a more detailed study when I get the time :}

Although this is an interesting article, I''m not really sure that this really covers many bases. Every simple viewing system such as the firescope and brilliantscope as well as this very well thought out and thorough analysis are completely focused on the appearance of a diamond from a direction perpendicular to the table under very specific lighting conditions. The GIA study uses a similar viewing setup and yet I can''t help but think that this is not the way that most diamonds will be viewed when they are set into jewelry. These studies all seem to point to a particular cut type with fairly narrow angle and symmetry charachteristics as being superior, but I''m not sure that this can be supported since the model does not mimic the actual use and lighting conditions that the diamonds will be used in. Are there any studies of this nature that take into account such real world conditions as tilt brilliance or unequal lighting models and somehow add them to this rather narrow type of study ? Do the diamonds which are cut to maximize brilliance perpendicular to the table also perform in a superior manner when tilted, say 5 or 10 degrees ? What happens to dispersion when the cut is maximized as this article would suggest ? This raises some interesting questions and I hope that this article is followed by some articles in the future that can address the questions relating to dispersion and tilt brilliance. Many thanks to Peter for providing this !

Michale E, are you familiar with the calculations performed by DiamCalc and Gem Adviser?
There is a one eyed and a two eyed version of the light return calculation - and there is a fixed and rocking result (broken down to table and whole crown).

AGS have used this approach.
They have also incoporated a multi position fire analysis shown here.

Thank you for the article, Peter, and welcome as a contributor to the Pricescope Journal.

Your article clearly shows another important point in the new cut-grading of AGS. Even when a diamond is cut nicely within the ''cutting guidelines'' of an AGS-0, it can still get a lower grade.

Even more, it is impossible to judge from the measurements, whether a stone will actually be AGS-0. In order to be sure, the stone has to pass through the future AGS-software, and preferably through the lab.

Like I said before: ''It''s the end of the world as we know it''.

Live long,

HI Garry,
I am familiar with those calculations. I suppose that my real question comes down to the practicality of using cut grades, formulae and the like to determine a realistic and observable difference between two diamonds of different grades. Say that you have an AGS cut grade "0" and a stone with an AGS cut grade "2". Same color, clarity and size, not carat weight, but size in diameter, (this because the price differential would allow you to buy a lower cut grade of the same size). If these two stones are sitting side by side in the same lighting conditions will the AGS "0" be dramatically "better" looking ? If so, does this hold true when they are tilted ? Can a cut grade of "2" look better than a cut grade of "0"?
Obviously there needs to be some scale of quality so that everyone can make a determination about what they prefer, prior to seeing the stone, but there seems to be a point where it becomes like splitting hairs, with no perceivable difference in the final look of the stone. Kind of like the difference between an IF and a VS2, if you can''t see the difference, does it matter ?
Here''s an even better question. If YOU had a fixed dollar amount to spend on a diamond, would you always buy an AGS "0" or would you buy a larger diamond of cut grade "2" or "3" with the idea in mind that the larger stone at the same price would actually give you a better look for your money ?
My post is not meant to be critical of anything, but rather to get an idea of just how close together all of the better cut grades are in how people actually perceive them.

To answer your question - Lets not consider AGS cut grade reductions based on spread, culet or a host of other factors that will result in a down grade.

If AGS''s new software does what it should do (i.e does what DiamCalc and the un released parts of DiamCalc do), then the answer is Yes - you should easily be able to SEE a difference between AGS 0 and AGS 2.

One of my staff asked about a extremely steep FIC stone that I bought the other day. She had just finished ticketing the boxes of around 20 stones .40 to .50ct diamonds. I asked her to look in the box and find the brightest stone, and the firey''est stone. she did this easily.
When we examined them, one was about 36C, <55%T and 40P (firey), and the BIC was about 33C 60%T and 41P (no need to scan then as they all look great with the ideal-scope). So yes - people can see subtle differences.

If the AGS new grading system works, then all 2 grade variance should be easily seen from a close examination by anyone with good eyesight.

Michael,

As I read it, the GIA study is actually pretty similar to what you’re suggesting. They have recorded a large number of observations wherein an observer has ranked several superficially similar stones in the order of their perceived desirability. The study involved a rather small number stones (45) and a fairly large number of observations (70,000+). These observations were made in reasonably controlled environments by a few hundred different observers with varying degrees of gemological training. The conditions were varied and shuffled in order to get a pool of data that gave a reasonable chance of separating out the important parameters. Obviously there is room for error and there may be parameters that weren’t considered but number of eyes, viewing distance, number of stones presented simultaneously, ability of the viewer to manipulate the stone, viewing angle and nature of lighting were all elements that were adjusted, considered and recorded. There were quite a few more.

They have yet to make any announcements about their scale and how it will be applied but presumably they are going to take the data from the above study and distill it into something useful. I, for one, am eagerly waiting. It’s not out of the question that they’ll do a good job of it.

I''m nowhere near smart enought to tell if this approach is any better or worse than the AGS method but it''s certainly different and it addresses exactly the concern that you raise.

Neil Beaty
GG(GIA) ISA NAJA
Independent Appraisals in Denver

The big problem with all these aproaches Neil is that the poor old manufacturer needs to know what will make the grade and what will not.

Take the current discussion in the GIA symmetry thread for example. Do I make this stone lopsided, but make all the facets meet nicely and get a very good GIA grade? Will that stone get Ideal, VG, good or fair from AGS? etc etc

Date: 3/21/2005 2:28:38 PM
Author: Garry H (Cut Nut)
The big problem with all these aproaches Neil is that the poor old manufacturer needs to know what will make the grade and what will not.

Click to expand...

Is that really a problem, Garry?

The poor young manufacturer will find his way in the jungle, either ending up in Xanadu or completely missing it.

In a way, it also is an expedition into the unknown. And why should all be clear from the beginning? Like they say: ''No guts, no glory''.

Live long,

Garry,

I understand that it’s a difficult dance, but difficult is not the same as impossible. The AGS approach seems much more quantifiable and, as is illustrated by this very discussion, even that is going to cause some problems for the manufacturers. For either system to be useful, it must be possible for the manufacturers to train their workers to produce a particular result. ‘9 out of 10 people surveyed preferred this set of parameters’ must become a specific set of instructions in order for it to become a useful grading scale. This would account for the separation of the team involved in the scientific study from the creation of the grading scale. It’s a pretty reasonable approach to science. The same sets of problems are going to fall on appraisers and other gemological professionals as with the manufacturers. We are supposed to be able to look at a stone and assign a particular grade to it, hopefully by using existing or reasonably available tools. GIA & AGS are going to have to train future gemologists to apply their scales and they won’t be able to require them all to buy a Helium machine (sorry Sergey). It looks pretty clear that AGS is planning on selling a version of DiamCalc, or perhaps some knockoff, as part of their grading system and it may or may not be made available to non-AGS stores. They haven’t said yet. GIA’s plans are a closely held secret but at least it’s expected that they will make their system broadly available. This whole process still has a long way to go. The result will almost certainly be an improvement in the quality of cutting made available to consumers.

Neil Beaty
GG(GIA) ISA NAJA
Independent Appraisals in Denver

Well reasoned assessments Gentlemen

RE: The poor young manufacturer will find his way in the jungle, either ending up in Xanadu or completely missing it.

--------------------------------------------------


The diamond cutter's are "artists" and will respond in a way that the grading labs may not like. For instance, do you remember when the GIA said a stone with a 60/60 would be evidenced by the stars making the "double box" look?
What did the cutter's do? Simple. We made the stars first. You can make the double box look regardless of the proportions. We changed the sequence of what we do to achieve the results the lab was looking for.

How does this apply to "performance"? Simple. If you say that a particular redfield view is what is desirable, then the cutter's will find a way to achieve this, which may or may not indicate good cutting.

In researching my patent, I have witnessed measurements of hearts and arrow stones that had a whole degree difference in the bottom main angle on the same stone. I have a sample BrayScored stone of 668 that a lab stated exhibits the H&A patter, one that's 868 that doesn't and one that's 922 that does.

Achieving "light patterning" in a diamond basically can be achieved by making corrections on the top of the stone to "mistakes" made on the bottom of the stone.

Basically you could arrive at the same point we were ten years ago where we have 4 types of certs/stones: 1. stones that look good to the eye and good on paper. 2. stones that look good to the eye and bad on paper 3. stones that look bad to the eye and bad on paper and 4. stones that look bad to the eye and good on paper.



Bill

Date: 3/27/2005 9:2:20 AM
Author: He Scores

... remember when the GIA said a stone with a 60/60 would be evidenced by the stars making the ''double box'' look?

What did the cutter''s do? Simple. [...] We changed the sequence of what we do to achieve the results the lab was looking for.

Click to expand...

How long does such adjustment take ? Out of curiosity... Not precisely, of course, a very approximate time scale would do: is it a matter of months ? years? A decade or more ?... for the "purpose" of some grading standard to be incorporated in manufacturing decissions for better or worse.

Is there still room for such adjustment after the new GIA or AGS standards are in place ?
I would imagine these two labs at least learned to give leeway for artistic improvement on their cut standards by now...

how long will it take? It depends. It may take a few hours of a cutters time, but it may take years to spread throughout the cutting industry.

I was in a lab recently and he showed me data and performance results of a manufacturer who had improved how a princess cut performs under his brilliance scope. I looked at the data and showed him what was improved and it was right back to what Tolkowsky had said.. you want brilliance in a diamond? Put the bottom mains on 40.75 degrees.

The "secrets" of Eightstar patterning may take considerably longer but eventually all cutters will "find" out how to make them if the market calls for it.


Bill

Date: 3/27/2005 9:12:52 AM
Author: valeria101

Date: 3/27/2005 9:2:20 AM
Author: He Scores


... remember when the GIA said a stone with a 60/60 would be evidenced by the stars making the ''double box'' look?

What did the cutter''s do? Simple. [...] We changed the sequence of what we do to achieve the results the lab was looking for.

Click to expand...

How long does such adjustment take ? Out of curiosity... Not precisely, of course, a very approximate time scale would do: is it a matter of months ? years? A decade or more ?... for the ''purpose'' of some grading standard to be incorporated in manufacturing decissions for better or worse.

Is there still room for such adjustment after the new GIA or AGS standards are in place ?
I would imagine these two labs at least learned to give leeway for artistic improvement on their cut standards by now...

Click to expand...

I know many manufacturers who now rountinely make 8* type indexing.

I am not sure that creativity is a bad thing Bill - so long as the creativity benefits all parties (ie, better yeild and better appearance.

Re the H&A''s with 1 degree variation in pavilion angle - I am sure I remeber we already had this arguement - about the tilted table and opposing pavilion angle averages? You say it is bad because you score symmetry deviations badly. I say it is good because if the cutter can chase out a VS inclusion from the corner of the table and make the stone IF, and still give great beauty to the diamond - then more power to her.

Date: 3/28/2005 12:13:13 AM
Author: Garry H (Cut Nut)
I know many manufacturers who now rountinely make 8* type indexing.

I am not sure that creativity is a bad thing Bill - so long as the creativity benefits all parties (ie, better yeild and better appearance.

Re the H&A''s with 1 degree variation in pavilion angle - I am sure I remeber we already had this arguement - about the tilted table and opposing pavilion angle averages? You say it is bad because you score symmetry deviations badly. I say it is good because if the cutter can chase out a VS inclusion from the corner of the table and make the stone IF, and still give great beauty to the diamond - then more power to her.

Click to expand...

Gary,
I remember us talking about tipping the table. I thought you were doing it on a measuring device, not on the real stone. I''ll try to explain my position to everyone. Whenever any facet on a diamond is made, the shape of those surrounding it change also. The table is the one facet that so many others are measured from, so to "tip" it after the stone is finished will change the relationships of all the facets top and bottom. It''s ok to do...but then you have to go and recut the whole stone over again to make it right. It is done on recuts from time to time, because in the past, many cutters would "cut" the table to begin with, fashion the stone, and then "smooth" or "polish" the table as a final step. This was so there wouldn''t be any bruises, etc. from the dops and fingers left on the table. Often times when they would do this, the facet "pulls from one side" because of the influence in the grain and the table plane is changed.

What happens when the table plane is changed, or tipped as I call it?

The main facet angles from where the tipping facet is coming from are reduced, while the opposite side main facet angles are increased. The main facet angles on the bottom of the stone are affected in just the opposite way. The "crown heights, pavillion depths are changed on the same sides, reducing where the facet is coming from and increasing on the opposite side. Here''s the one thing that will happen that is virtually undetectable except by measuring but is one of the most important things in the premium sytle of cutting. The main facets top and bottom that are 90 degrees from where the "tipping" facet starts are made crooked.

That last sentence may be hard to understand, but if you think in terms of the table being a railroad track, when you tip if from one side, it''s like spreading railroad track from the side you tip from. In general, tipping the table has far more overall risks as a final measure in cutting than it does benefits, and I know few cutters who would relish doing it and then re-doing the consequences.

In addition to your comment of justifying tipping the table with the claim of improving the clarity of the stone, then the weight loss must be taken into account, and when removing flaws from a diamond, the rule of thumb is to remove them with a number of facets if possible rather than just one....let alone the largest one on the diamond (read the one that will lose the most weight during cutting), the table.

Again, it has to be understood that diamond cutting is three dimensional. I see errors in some cad type modelling programs that don''t really mirror actual cutting, as I tried to point out in my anaylysis of Peter Yanzer''s article.



Bill

Not quite what i meant Bill.

http://www.gemology.ru/cut/english/symmetry/6.htm the case study here explains it better.
The table is out of line with all the other Axes - but a sarin report never records a table out of line. The table is always 0 degrees according to Sarin.
Or maybe there was a tiny bit of dirt under one side of the stone when it was set in the pot. But everything else can be very tight - and still you have big pavilion and crown variances 9wrongly noted by the Saarin scan)

Gary,
I think I know what you''re trying to say in that article, but even as a cutter I find it confusing. Quick question: You mention the four axes. Of course the table axis is acomplished when you lay the diamond on it''s table. The platform of the Sarin machine. How is the axis of the girdle determined in your study?

Also, I agree with you that a stone can have it''s'' axis corrected in a viewer to achieve a better performing stone, but what would happen in a real life mounting to achieve this same visual effect? Wouldn''t the stone be mounted crooked?.


Bill

Bill a lot of calculations or a computer is required to calculate the average axis for each part of the stone.
It could be dome from the average of all the crown angles, or crowns and minor facets too etc. The girdle is the easisest - if the stone has a 1% girdle one side and 3% on the other side, evenly graduated, then you can visualize a plane that bisects the girdle plane with an axis being the perpendicular line running through the center.
The Optical Axis should be weighted because the pavilion has much more impact on the flow of light through the stone. Clever programmers can work such things out.