Some interesting things from AGS class

[Serg]

re:My own belief is that you know when a diamond is "ideal" without measuring fire at all. All fine diamonds have some dispersion, some more, some less. It is something a person might choose to go for or not be worried over.

Dave,

Imagem can not grade Fire.( BTW one of weakness part of Imagem)
It does not mean Fire is not important( or all diamond with good Brilliancy will have good Fire and vice versa)

 

[oldminer]

Wink: I appreciate the excellent response and your always courteous attitude. You are a gentleman when it comes to the subject we are covering.

I would question only this part of what you offered in your last response to me:

"If dispersion is truly not important to them then they can choose, possibly, a slightly less expensive diamond without dispersion if they wish. Fortunately, a truly well cut diamond will normally come with dispersion included, no extra charge."

We suggest that all finely cut diamonds have dispersion to a greater or lesser extent. A lot of it depends on the lighting they are viewed in, but that one can select the large range of "ideal Cuts" that exist without measuring fire. I don''t see any market evidence of "fire" having a plus or minus price effect other than firm''s branding or marketing strategies. Regular diamond dealers do not differentiate or haggle over fire in the wholesale market. As you yourself said: "Fortunately, a truly well cut diamond will normally come with dispersion included, no extra charge." This is what I am saying, too. How much fire may be something some particular folks will base a final buying decision on, but most people will just select a diamond they like from well cut stones that perform well, fire included, no charge.....

Spending a lot of R&D money on the measurement of fire will not create a better definition of a fine diamond. We can judge the4 best range of beautiful diamonds without measuring it. We can freely makes personal buying decisions based on secondary attribute, such as fire, without a metric for it. A maximum amount of fire would not make a wonderful diamond. Maximizing this effect might have a negative impact on overall beauty. I think it is a mistaken path, but we can at least disagree in a friendly way on this minor point.

All of this sounds like you are very excited about the future of diamonds, as I am.

 

[strmrdr]

Dave,
The amount of fire a diamond produced is one of the variables that can cause someone to like one diamond over another therefore its important.
We have discussed this many many times and I dont want to rehash it but after carefull and long consideration of your view including going out and looking at diamonds I keep coming back to the opinion that it cant be dismissed.

I think the overall light return is an important criteria but im of the opinion that how its returned may be more important to people seeing diamonds and picking one over another.

Im my world there has to be one best and it has to be defined :}
But human preference isnt that easy to nail down because everyone is different so with diamonds that is not possbile.
Im slowly coming to terms with that.

 

[oldminer]

Sergey: You are just uninformed about ImaGem. Partly because of your great distance away and partly for other reasons, too. I wish you were closer to our office. We would have a wonderful time

You said: "Imagem can not grade Fire.( BTW one of weakness part of Imagem)"
It can grade fire, but we chose not to. We think we can define the world of faceted diamonds with the three metrics we measure and we don''t need to add fire to the mix. We see fire as a part of total light return.

"It does not mean Fire is not important( or all diamond with good Brilliancy will have good Fire and vice versa)"
We say fire may be important to some people, but more importantly they will want al brilliant, sparkly and intense diamond in all cases. A diamond with only high brilliancy may fail to have sufficient sparkle or intensity to be selected as an "ideal" cut. This is true. That stone probably will lack fire, as well. But, we don'';t need a fire metric to know if it beliongs in the excellent cut category.....Obviously, "Ideal Cut" or "Excellent Cut" is a broad category considering the old Tolkowsky standards. We agree that beauty is found in a wide range of cuts. Some may have more fire than others, but all are highly attractive. If someone wanted to brand a diamond as "High in Fire" ImaGem could measure colored light return with very little problem.

I would question how one would choose to deduct body color from colored light return, or how one would measure total light return, if one only looked for "white light" when a stone had a definite tint to it and nearly all its returned light was non-white.

Anyway, the thread is a good one and people are thinking.....

 

[WinkHPD]

Date: 2/24/2006 11:39:16 AM
Author: Serg
re:My own belief is that you know when a diamond is ''ideal'' without measuring fire at all. All fine diamonds have some dispersion, some more, some less. It is something a person might choose to go for or not be worried over.

Dave,

Imagem can not grade Fire.( BTW one of weakness part of Imagem)
It does not mean Fire is not important( or all diamond with good Brilliancy will have good Fire and vice versa)

Hi Sergey!

I was blown away when I saw the hoops that Peter and Jim put your software through! Both are extemely complimentary to your product and the changes that you have made in the past few years and how you are willing to work with them to make so many things possible. I am looking forward to downloading and using your software as soon as I catch up with my work here, which means I will probably be here this weekend as there is no way I can do it all today that needs to be done!

Peter was VERY insistant that I should have this software and after about two minutes of watching him make diamonds do tricks, especially watching him estimate recut weight and dimensions for poorly cut stones, I was in complete agreement with him.

Wink

 

[WinkHPD]

Date: 2/24/2006 10:50:35 AM
Author: JohnQuixote

Date: 2/24/2006 9:53:08 AM
Author: Wink

...I am not well enough versed in the light sciences to know if you are correct in your statement of dispersion being less in the colored diamonds or if it is still there, simply masked by the body color, making it harder for us to perceive. I certainly have seen some incredible dispersion in the colored diamonds that I have sold over the years, especially in the EightStar colored diamonds that I have been fortunate enough to have seen...

Optical symmetry improves dispersion due to the nature of the DCLR. Martin Haske attributes it to purer spectral colors firing simultaneously. In well cut OS diamonds with close angular relationships on the crown (those with painted girdles in particular) the purer spectral colors result in broader flashes and the fire/sparkle can be seen from farther away. The world''s top cutters have known this for years without recognition from the labs. I''ve been excited for some time that AGS is presenting analytics that will recognize this.

Translations please. What is DCLR? What is OS? I know what AGS is, but I quit learning Acronyms when I got out of the military...

 

[JohnQuixote]

Date: 2/24/2006 12:32:18 PM
Author: Wink

Date: 2/24/2006 10:50:35 AM
Author: JohnQuixote

Date: 2/24/2006 9:53:08 AM
Author: Wink

...I am not well enough versed in the light sciences to know if you are correct in your statement of dispersion being less in the colored diamonds or if it is still there, simply masked by the body color, making it harder for us to perceive. I certainly have seen some incredible dispersion in the colored diamonds that I have sold over the years, especially in the EightStar colored diamonds that I have been fortunate enough to have seen...

Optical symmetry improves dispersion due to the nature of the DCLR. Martin Haske attributes it to purer spectral colors firing simultaneously. In well cut OS diamonds with close angular relationships on the crown (those with painted girdles in particular) the purer spectral colors result in broader flashes and the fire/sparkle can be seen from farther away. The world's top cutters have known this for years without recognition from the labs. I've been excited for some time that AGS is presenting analytics that will recognize this.

Translations please. What is DCLR? What is OS? I know what AGS is, but I quit learning Acronyms when I got out of the military...

DCLR is dispersed colored light return (fire), defined also as “the visible extent of light dispersed into spectral colors.” OS is optical symmetry. USMC is... oh wait, you'll know that one!

 

[JohnQuixote]

Recalling that definition above reminded me of an old article (many of us are familiar with it I suspect) from Professional Jeweler regarding the Fall 2001 G&G. As it turns out the def I used is the one GIA used to distinguish fire from dispersion.

http://www.professionaljeweler.com/archives/articles/2002/jan02/0102dg.html

Excerpt 1

What Role Does Fire Play? In the new study, “Modeling the Appearance of the Round Brilliant Cut Diamond: An Analysis About Fire and More About Brilliance,” GIA now makes a distinction between fire and dispersion, terms once used interchangeably. The study says the distinction is needed because the dispersion value of diamonds is constant, while fire is understood to be a result of dispersion. The new definition of fire is “the visible extent of light dispersed into spectral colors.”

Which supports the notion that, though dispersion is a constant, different facet and angle configurations may result in visibly different perceptions of dispersion.

Excerpt 2:

No Single Best. Much as they discovered in the brilliance study, researchers found that no one set of proportions in a diamond best maximizes fire. In fact, they found a wide ranging set of proportions yielded similar results. Researchers also described an inverse relationship between the tests. Generally speaking, the set of proportions that yielded better weighted light return (brilliance), tended to yield poorer DCLR (fire), and vice-versa. But the researchers also found a happy medium where numerous diamond proportions yielded average or above-average results for WLR and DCLR.

Which supports the notion that fire is largely constant, and a futher notion about fire at the expense of brilliance & vice-versa... Neither idea has been wholly embraced.

Of course, symmetry deviations have been overlooked in all of these studies (despite GIA''s intentions in 1998 to include them).

 

[valeria101]

Date: 2/24/2006 10:04:56 AM
Author: Wink

How did you make this leap from the pictures that I shared?

Wink, you flatterer!

There was nothing brilliant: take two charts, put one over the other, see if features correlate - the overlapping points show coincidental characteristics for one set of proportions. Sure enough, the main splotches of color do... From your intro, it would appear that a small splotch around the Tolkowsky idea is (almost?) the only island on any chart that corresponds to ''good'' colors on all others. Etc.

If the charts do not have some input aside the deterministic models behind various bits of software, I would think that such optimization results could be spewed out of the machine in the form of un-impressive intervals of matching angles, set un-intuitively apart from each other by rifts of bad matches... Charts feel less out-of-nowhere than numbers - works for me, at least.

Staring at the keyboard, I wonder if the apparent complexity couldn''t be boiled down into something less mind boggling after all. What if the space art on the AGS charts is just a matter of picking up a weird facet pattern (the RBC) instead of making a rational choice for the purpose of the model ? The tangle seems hard to explain and manipulate. A simpler general form

Oh well...


Sorry for the messy post. And a quick question:


Do the fire metrics have anything to do with THIS.

The bit about trade off between fire & scintillation reminded me of this page.

 

[dyemonds2002]

Lets just find a diamond that comes at you though the table and also light performance all around girdle every diamond is diffrent how can you light apart a diamonds refractive if you want fire then higher color is the way

 

[JohnQuixote]

Date: 2/24/2006 1:30:33 PM
Author: valeria101

Date: 2/24/2006 10:04:56 AM
Author: Wink

How did you make this leap from the pictures that I shared?

Wink, you flatterer!

There was nothing brilliant: take two charts, put one over the other, see if features correlate - the overlapping points show coincidental characteristics for one set of proportions. Sure enough, the main splotches of color do... From your intro, it would appear that a small splotch around the Tolkowsky idea is (almost?) the only island on any chart that corresponds to 'good' colors on all others. Etc.

If the charts do not have some input aside the deterministic models behind various bits of software, I would think that such optimization results could be spewed out of the machine in the form of un-impressive intervals of matching angles, set un-intuitively apart from each other by rifts of bad matches... Charts feel less out-of-nowhere than numbers - works for me, at least.

Staring at the keyboard, I wonder if the apparent complexity couldn't be boiled down into something less mind boggling after all. What if the space art on the AGS charts is just a matter of picking up a weird facet pattern (the RBC) instead of making a rational choice for the purpose of the model ? The tangle seems hard to explain and manipulate. A simpler general form

Oh well...


Sorry for the messy post. And a quick question:


Do the fire metrics have anything to do with THIS.

The bit about trade off between fire & scintillation reminded me of this page.

Ana - are those juxtaposed charts something you can share? Or would we have to manipulate transparency manually in Photoshop to get the 'gist'?

Garry or Serg will likely answer your ETAS question more thoroughly, but those studies relate to all aspects of performance: "An observer sees light in a diamond if the source is placed within any cone coming to our eye through a diamond. All these cones form part of space called ETAS." This premise is the basis for a meaningful scintillation metric.

Here is a great page about it. Check out the difference in the ETAS of a standard RB with top symmetry (fig 1) and one with lesser symmetry (fig 3). The probability to see the light source grows with tilting - as the tilt graphic on that page shows.

 

[adamasgem]

Date: 2/23/2006 9:53:40 PM
Author: JohnQuixote

Date: 2/23/2006 9:05:45 PM
Author:Wink

AGS had fully developed the metrics for their dispersion grading as well as for the light proformance. This was NOT easy and required the use of FORWARD ray tracing, not backward ray tracing as while forward and back work for brilliance, backward ray tracing does NOT work for dispersion.

Thanks for the peeks so far and those to come, Wink.

The fire analytics are very exciting. Anyone who likes the graphic Wink posted will enjoy seeing the tilt effects composite Marty Haske placed on his site: It shows Tolkowsky 30 deg/40 deg/fire in 1 degree increments. You''ll need to save it and open it to zoom in and enjoy it (>400k).

http://www.adamasgem.org/pngs/tolk.png

As I noted in the original post, these tilt sequences were made available courtesy of AGS.

 

[adamasgem]

Date: 2/24/2006 10:30:07 AM
Author: Rhino

Greetings again Winkster,

I''m not 100% sure if what you were referring to with the darker arrow shafts vs lighter arrow heads but this pretty recent picture I snapped off came to mind which demonstrates in real world conditions this phenomena as seen with the eyes. Is this along the lines of what you were referring to?

Kind Regards,

Lots of non random glare there in your picture....

 

[Serg]

Re:
You are just uninformed about ImaGem. Partly because of your great distance away and partly for other reasons, too.
You said: "Imagem can not grade Fire.( BTW one of weakness part of Imagem)"
It can grade fire, but we chose not to. We think we can define the world of faceted diamonds with the three metrics we measure and we don't need to add fire to the mix. We see fire as a part of total light return.

Dave,
I am understanding Why Imagem has problem to seperate Fire from total light return. Are you understanding it(both)?

 

[WinkHPD]

Date: 2/24/2006 12:00:54 PM
Author: oldminer
Wink: I appreciate the excellent response and your always courteous attitude. You are a gentleman when it comes to the subject we are covering.

I would question only this part of what you offered in your last response to me:

''If dispersion is truly not important to them then they can choose, possibly, a slightly less expensive diamond without dispersion if they wish. Fortunately, a truly well cut diamond will normally come with dispersion included, no extra charge.''

We suggest that all finely cut diamonds have dispersion to a greater or lesser extent. A lot of it depends on the lighting they are viewed in, but that one can select the large range of ''ideal Cuts'' that exist without measuring fire. I don''t see any market evidence of ''fire'' having a plus or minus price effect other than firm''s branding or marketing strategies. Regular diamond dealers do not differentiate or haggle over fire in the wholesale market. As you yourself said: ''Fortunately, a truly well cut diamond will normally come with dispersion included, no extra charge.'' This is what I am saying, too. How much fire may be something some particular folks will base a final buying decision on, but most people will just select a diamond they like from well cut stones that perform well, fire included, no charge.....

Spending a lot of R&D money on the measurement of fire will not create a better definition of a fine diamond. We can judge the4 best range of beautiful diamonds without measuring it. We can freely makes personal buying decisions based on secondary attribute, such as fire, without a metric for it. A maximum amount of fire would not make a wonderful diamond. Maximizing this effect might have a negative impact on overall beauty. I think it is a mistaken path, but we can at least disagree in a friendly way on this minor point.

All of this sounds like you are very excited about the future of diamonds, as I am.

I am excited. My comment included the word, possibly, for a reason, which is if the stone was somehow to not have dispersion it would be an AGS 1 or 2 rather than a 0 so it would be slightly less expensive. If you look at the beautiful pictures above you can find some surprizing areas that have nice reds, but that do not show any dispersion, so it would not be possible to cut a diamond in those areas and have an AGS 0 cut grade, although they might be interesting and beautiful diamonds.

Since an AGS 1 or 2 would be slightly less expensive than an AGS 0 that is the reason I made my comment. The R&D has already been done and is part of the metrics that AGS uses to assign their grade, although the measurement is made entirely from the forward ray tracing capability of the soon to be released AGS software which uses the non-contact measuring capability of the above mentioned manufacturers.

So I do dissagree with you, slightly, and certainly in a friendly way. As you so correctly stated, most well cut diamonds will already display fire, and it is not necessary to charge extra for it. Some surprisingly good looking diamonds however will lack fire and will not be graded AGS 0 cut grades because of it.

Wink

 

[Scott Terence]

Here are the two charts (30 blue and dispersive) layered upon eachother, my first impulse would be to shoot for the areas that appear fluorescent orange (lots of red from ASET and the orange from the dispersive chart). Hope this helps.

Best,

Scott

 

[WinkHPD]

Wow! That is REALLY pretty, now when you add in the 40 blue you will see some dissapearing red areas. Interesting how you get some isolated areas that are completely different from what you might expect and how sharply some of them change from one 0.20 angle degree to the next.

Wink

 

[oldminer]

Sergey:

I said Imagem chose not to measure fire. This in no way should infer to you there is any problem or technical issue that evades us in this respect. We made what we hope is a knowledgeable choice not to do it. WE CAN MEASURE IT....We don''t, at the present time.

We are not saying fire is not something of interest to lots of people, but we are saying that we believe you can choose the rather wide family of top quality cut diamonds with metrics that include fire within total light return. Then, if one wants more or less fire, one can use their eyes to say which diamond they prefer. Or, if a brand is built with extra fire or less fire, they can go by the brand name for help in making the selection. One of the world''s finest retailers selects diamonds for their icy character with only limited fire. They are highly respected as merchants who sell top quality diamonds at large prices.
They know that many diamonds have equally high Light Behavior, but that there is a subset of their own preference, their brand''s look, that is more neutral and icy than color filled.

Having high fire or low fire, does not determine the quality of the cut. It is a secondary attribute which one might wish to measure, but it does not determine the cut quality......we believe.

Of course, some people have other opinions. Many have opinions, but no basis for their opinion. That''s an important difference. Before you jump to defend the metric of fire, be sure to examine how you know it is a measure that must be present in some quantity. There are many very well thought of dealers and retailers who strongly believe in only moderate to slight fire as the best stones. Its an opinion that varies and not useful to "grading".......

 

[JohnQuixote]

Scott - thanks for doing that.

 

[WinkHPD]

Okay, Stormy, this is for you. I had to ask pretty please for this, and I did it with you in mind. Special appreciation to Peter Yantzer for his kindness in allowing me to post this.

This is the universe for Square Emerald cuts with 45% tables. One of these was cut by an Indian company who have expressed great joy in the results.

First the 30 blue

 

[WinkHPD]

Now the 40 blue

 

[WinkHPD]

And finally the fire.

I hope you enjoy this as well as the rounds...

 

[JohnQuixote]

Hey Wink, the 3 graphics are titled 'princess-...' but have 45%T. Assume those were actually the square ems?

 

[WinkHPD]

One of the first things you will notice is that there is much less dispersion with the S.E. diamonds than with the rounds. Not a big surprize here.

I find the little dark ovals on the right hand side of the dispersion picture to be very interesting, little ovals of zero dispersion with wisps of dispersion around them, almost like something we would see in the Hubble Telescope pictures. By the way, one of the rocket scientists who worked on the light metrics with AGS was the one who figured out what was wrong with Hubble and helped design the fix. These are NOT lightweight scientists who figured this out!

Wink

P.S. I am resubmitting the metric description for the colors in the dispersion picture modified to show the average length of the flash for your informational purposes...

 

[WinkHPD]

Date: 2/24/2006 5:06:05 PM
Author: JohnQuixote
Hey Wink, the 3 graphics are titled ''princess-...'' but have 45%T. Assume those were actually the square ems?

Omigosh! What a bonehead error. I was so tired I got it in my brain that the second set were the SE cuts and they are in fact the princess cuts. Too bad I can not say I was confused and they really were the SE cuts, but now I remember, Pete and I looked at them and they did not have anywhere near the beauty of the princess cuts in the large pictures as there is so much less dispersion so I chose to get the princess cuts dispersion picture and this one I chose because it looked so beautiful and sort of like a deep space nebula.

Thank you for correcting me before my hole was completely dug and ready for being concreted... When I looked at the naming of the original files they are clearly the princess cut with 45% tables, not something you see very often, but potentially incredible...

Wink

 

[JohnQuixote]

Hey Wink, we're all human. You deserve a break - thanks for posting all of this.

Seeing those graphics just affirms that princess cuts have great potential in different configurations than we had become accustomed to. 'It's the end of the world as we know it...' right Paul?

 

[DiamondExpert]

Wink - thanks for all the info! Looks like the rest of us will have some serious catching up to do!...I hope AGS will give some repeat performances with that in mind.

By the way, once these "universes" get stacked, providing a 3D uber-universe, there may be "worm holes" in it representing cutting "no go" zones, a la your observation of the circular 2D events!

Thankful there is no need (at least so far!) for a time-dependent variable!

Keep it coming!

 

[WinkHPD]

Date: 2/24/2006 6:00:38 PM
Author: DiamondExpert
Wink - thanks for all the info! Looks like the rest of us will have some serious catching up to do!...I hope AGS will give some repeat performances with that in mind.

By the way, once these ''universes'' get stacked, providing a 3D uber-universe, there may be ''worm holes'' in it representing cutting ''no go'' zones, a la your observation of the circular 2D events!

Thankful there is no need (at least so far!) for a time-dependent variable!

Keep it coming!

Hey thanks! Neal the Denver Appraiser was there too. Interesting item. He just picked up his new J color master from the Lab at AGS. He had it graded by both GIA and AGS. It was one of the WhiteFlash diamonds that are cut to emulate the EightStars. He and I were sitting next to each other on the second day and he handed me the diamond to look at in the ASET. I went, "WOW! this stone has no leakage and incredible light return with a fabulous pattern, it looks almost like an EightStar."

He told me it was an AGS 0, but only a GIA very good. Wow! I was shocked.

By the way, there are way more "NO GO" zones than "GO" zones. Look at all the light green and blue zones, while they may look pretty on paper, the diamonds they represent will NOT be easy to sell! I like the idea of "worm holes" though, it gives such a finishing touch to the "rocket scientist" part of the light sciences!

Wink

 

[WinkHPD]

Date: 2/24/2006 6:00:38 PM
Author: DiamondExpert
Wink - thanks for all the info! Looks like the rest of us will have some serious catching up to do!...I hope AGS will give some repeat performances with that in mind.

Yes, it was a great class. I am sure that Pete will be sharing much of the information again at the Vegas show and you can contact the AGS Education Director Diane Flora at 702 255-6500 ext 1027 to ask her when the next Stars Class will be.

One of the truly incredible things to me was that there was no charge to attend two days of presentations by some of the brightest minds in Gemology. Truly an incredible membership benefit for AGS members and affiliates.

Wink

 

[WinkHPD]

I just got an email from Pete Yantzer with a small correction to a statement I made yesterday. It has to do with fire and my statement that the gray areas represented stones with no dispersion. I will let him state it in his own words:

Lastly, as a point of correctness, the gray areas on the very large, very fine FireMap matrixes are unproductive areas. They are not areas that don’t have any dispersion. You can see that in the FireMap key where it says “Obscured or low angle or leakage.” The gray areas on the ASET 30 and ASET 40 Blue VLVF matrixes are leakage only.

Heartily,

Peter

Thanks Peter! I will always try to bring to Pricescope correct information and as a non scientist there will be times when I do not state it completely correctly. When this happens I will always bring the corrections to you Pricescopers so that you can count on this information being correct.

Wink