Why did GIA included Steep Deep diamonds in ''Excellent''

[RockDoc]

RE: Hasps

Thanks for the reply Garry.

In that I was not all that excited with the first version and supposedly having to buy multiple copies of the software, what would happen if I tried it only to find the same results.

If an upgrade is $ 50.00 and I could upgrade it on a computer which doesn''t have my original version, I''d try that.

As to the hasps....my Sarin thingy has a hasp... I think it goes into a USB port? I guess this means that if I want to use it on a different computer I have to crawl around on the floor, move the computer where I could take it out and plug it in another?

Plus I am a bit concerned with compatibility with the other haspie things I have

or does the hasp version come with more than one hasp that you don''t have to be crawling around the floor to change?

I already have a ton of USB thingies connected to CPU''s, multiple added thingies and keyboards, I already have a rat''s nest of wires living on the floor.




___________________

As for my previous problems with the settings, when I tried it - even though I locked in the input setting thingy, it still made changes to the previous entries I put in. The measurements I used came from the Sarin, so I would assume my numbers were "pure" as to existing. But my reports have always had "all the numbers" rather than averages, and of course DC didn''t accomodate putting in the individual numbers that varied at that time.

Maybe with it taking the data from the Sarin Scan that has changed, but I guess I want to know I was looking at the results from accurate facet by facet measurements, rather than an average.

I have a way of seeing the light exit paths that aren''t estimated or calculated and actually are done seeing the actual light in light out type view.

I am curious to see if the calculations really do match the real world result. I have a long way to go with the development of this.

 

[oldminer]

This is sure a highly technical thread. For those less technical, it sounds as if you are saying as follows: Give us a diamond. We will measure a lot of diemensions, facet lengths and angles and calculate (using complex software) what we believe to be true about your diamond.

What seems apparent is that not every measurement is taken into account or accurately measured. Some measures are computed to the best ability of the measuring device and some are just not measured. It seems there is general agreement that a diamond may be somewhat different than what is measured due to machine error, or not requiring a certain measure to be taken.

What degree of accuracy is involved in the final estimation of light behavior via ray or beam tracing using the equipment recommended? How much higher degree of accuracy will there be when machine error is decreased?

You know my side of the argument is based on direct measurement of LIGHT from a diamond, not calculating it from modelling. How far apart are the end results as you see it?

Thanks.

 

[Yuri]

Subject: Color optimization of a colored CZ .

Re: Cutters have learned how to INTENSIFY the color in radiant cuts, to getthe lucky vivid color call in fancy yellows (Citation of Martin Haske).


Dear Martin,

Some times ago we have started to study possible ways to optimize color in fancy colored diamonds. Because it is hard to have plates of colored diamonds necessary for reliable spectra recording we decided to play with CZ material first. We have CZ plates of different color and thickness and we have recorded their spectra. Fig. 1 shows 5 spectra of different thickness (1 mm, 2mm, 4mm, 8 mm, 16 mm) orange CZ plates made of one CZ rough piece. Being normalized on their thickness all these spectra are still slightly different because of uneven color distribution in the CZ rough piece.


to be continued

 

[Yuri]

We use these absorption spectra to build DiamCalc models (Fig.2). DiamCalc allows choosing a light source spectrum (we choose A type because of photo pictures were made under wolfram lamp which spectrum is close to A type). And DiamCalc can implement eye color adaptation function that should be corresponding to A type of the light source.



to be continued

 

[Yuri]

Verification has been done by Canon digital camera. Under wolfram lamp we took pictures for all plates and obtain very close colors and color coordinates (Fig. 3):

to be continued

 

[Yuri]

DiamCalc can model two different shapes of one linear size, for example 6 mm. Fig. 4 is a picture of a round brilliant cut and a princess cut. Here we use a spectrum for 16 mm plate and a lighting configuration similar with one that GIA uses for color grading of fancy color diamonds and light sources type D 65


to be continued

 

[Yuri]

Results:
1) DiamCalc can model diamonds and some other facetable materials with their absorption spectrum.
2) Photorealistic images and color coordinates can be used for color optimization of rough.
3) Color appearance depends on size, shape, cut style and cut parameters.
Discussion:
1) We did not find any light source with known emission spectrum. Modeled light can be different from real light
2) Canon camera has its own function (how color intensity depends on camera exposure time). This function is rather complicated. Also the camera has its own spectral sensitivity function while DiamCalc uses a standard function for human eye spectral sensitivity
3) We do not know how GIA will grade any particular color because we do not know color boundaries of different color ranges like Fancy light, Fancy, Fancy intense, Fancy vivid by GIA.
4) This work was done in the DiamCalc 2.4 with gamma correction sRGB. Other gamma will give other colors.
We will be glad to receive your suggestions.

 

[Serg]

Date: 8/15/2005 8:08:43 AM
Author: oldminer
This is sure a highly technical thread. For those less technical, it sounds as if you are saying as follows: Give us a diamond. We will measure a lot of diemensions, facet lengths and angles and calculate (using complex software) what we believe to be true about your diamond.

What seems apparent is that not every measurement is taken into account or accurately measured. Some measures are computed to the best ability of the measuring device and some are just not measured. It seems there is general agreement that a diamond may be somewhat different than what is measured due to machine error, or not requiring a certain measure to be taken.

What degree of accuracy is involved in the final estimation of light behavior via ray or beam tracing using the equipment recommended? How much higher degree of accuracy will there be when machine error is decreased?

You know my side of the argument is based on direct measurement of LIGHT from a diamond, not calculating it from modelling. How far apart are the end results as you see it?

Thanks.

“Direct measurement of LIGHT from a diamond” is not the synonym of measurement without mistakes.
BS and Imagem are direct measurement tools. Are result same?

Direct measurement is decreasing one type error and increasing other type error.
• operating characteristics of camera are far away from operating characteristics human eye.
• Stereo vision
• Very limited set of light condition
• Start position of diamond

Dave, please send answer to my remarks.Several times I showed to you these problems but never received any answer from you or Imagem.

 

[oldminer]

Sergey:

I have copied your request for a technical response to those in charge at ImaGem. I would encourage them to join into this ongoing conversation, but that is up to them.


You asked me as follows:

“Direct measurement of LIGHT from a diamond” is not the synonym of measurement without mistakes.
BS and Imagem are direct measurement tools. Are result same?"

I would reply that we know the results are NOT the same because it is our belief after sufficient testing that BS is not doing something correctly which ImaGem is doing correctly. The end result is that ImaGem can discriminate performance as the trained human eye would and that the BS fails to do this.
You then state as follows:

"Direct measurement is decreasing one type error and increasing other type error.
• operating characteristics of camera are far away from operating characteristics human eye.
• Stereo vision
• Very limited set of light condition
• Start position of diamond"

I would reply that I asked what sort of error comes from the devices and software others are recommending for making estimated readings? No one seems interested in giving a response to that. Within the wording of your question one sees that you are aware of "error", and I just wondered how you or others might describe this error. I don''t think "error" is involved with ImaGem techniques in that the end result is not error dependent. There is very slight variation in measures within stated bounds, but the overall result is within a stable and repeatable situation. More clarification may come from ImaGem on this to share.

A camera is NOT a human eye and I never said it was. What ImaGem has done is to take views from a camera and elegantly combine this with a statistical database that give artificial intelligence to the final answers provided. The final grade is not so much based on a device, but on intelligent decision making programming. We always look at the performance of diamonds with both eyes and the statistical database has human perceptual results in it. Therefore the camera data combines with human derived data to form a result that is a synthesized stereo from monocular measures.

The light conditions are "limited" as you stated. How true. No one will ever agree on any particular lighting model as being the one, only and unique way to grade a diamond. Argue til the end of time, but no one is qualified to set that standard. It would be very arbitrary no matter what it was. Remember, diamonds are viewed by consumers in every conceivable lighting environment. What ImaGem has done is to create an internal light environment which works with their cameras and statistical databases of human grading results and GIA grading results to give reasonable and correct grades to diamonds that make sense to gemologists when they view these diamonds in what we''d call a regular lab grading environment, a retail store environment, or in a daylight-outdoor environment. In other words, a practical, not unusual environment. Again, the unique use of statistics plays a role in obtaining results. One cannot do this just with fancy camera equipment. There is more going on.

All diamonds are viewed in ImaGem from the table / crown when it comes to Light Behavior grading. This is just like in regular grading and examination of diamonds everywhere.

My own comment is that modeling is very useful to cutters as it provides them choices and recipes for cutting and re-cutting diamonds to perform better. There is a huge need for cutters to learn about these changes and what to do with rough.
However, when it comes to GRADING diamonds, it just seems better are more reasonable to measure the actual performance than to estimate it through a huge process of measuring lengths, angles, facet ratios, etc.

My thoughts are that both methods, direct measurement and calculation of performance are highly useful tools, but for different parts of the trade. If one asks what a diamond weighs it is logical to place it on a scale, but in a pinch, one can measure it and calculate the weight within a reasonable degree of accuracy without a scale nearby. No one would always calculate the weight if they had a scale and I don''t think it is logical to always calculate the light performance if one can measure it. It is very proper to calculate POTENTIAL light performance for instructing cutters on how to work rough or to recut diamonds. The end result should be measured, not calculated..........

I think my logic is way more solid than my ability to argue scientific details or technicalities with PHD''s and optical experts. I am just a guy with a deep interest in what is going on and have a degree of involvement in this ongoing process. Our industry will benefit the most from developing the best products. I have seen so many half-baked products come and go. This time, I sure would like to see our business make the best choices and use solid science, not smoke and mirrors.

Please don''t be offended. I admire all the great people providing their points of view. I sure hope my views don''t have the appearance of forcing "my way or else" on anyone. I think we all truly want the best to succeed.

 

[Serg]

Dave, Thanks for reply.

There is very slight variation in measures within stated bounds, but the overall result is within a stable and repeatable situation

I am not speaking about stable and repeatable result. It is easiest part of work. It could be achieved by different ways.

Re: What ImaGem has done is to take views from a camera and elegantly combine this with a statistical database that give artificial intelligence to the final answers provided. The final grade is not so much based on a device, but on intelligent decision making programming. We always look at the performance of diamonds with both eyes and the statistical database has human perceptual results in it. Therefore the camera data combines with human derived data to form a result that is a synthesized stereo from monocular measures.
Fine. May be it is best way.( we have similar stage in our grade) But If you use this way, “error”( mistake, wrong result, distortion) on this stage could be much bigger than on any other stages during direct or indirect methods.
And if you want to minimize this “error” you should to increasing number of parameters ( metrics) of inputs to your intelligent software. It is much more easy to do in indirect method.

 

[adamasgem]

Date: 8/15/2005 9:57:13 AM
Author: Yuri

Subject: Color optimization of a colored CZ .

Re: Cutters have learned how to INTENSIFY the color in radiant cuts, to getthe lucky vivid color call in fancy yellows (Citation of Martin Haske).



Dear Martin,


Some times ago we have started to study possible ways to optimize color in fancy colored diamonds. Because it is hard to have plates of colored diamonds necessary for reliable spectra recording we decided to play with CZ material first. We have CZ plates of different color and thickness and we have recorded their spectra. Fig. 1 shows 5 spectra of different thickness (1 mm, 2mm, 4mm, 8 mm, 16 mm) orange CZ plates made of one CZ rough piece. Being normalized on their thickness all these spectra are still slightly different because of uneven color distribution in the CZ rough piece.


to be continued

Good work.. The colored CZ spctrum appears to be that of Calcium stabilied CZ..

The different thicknesses you used can also be mathematically modeled from one thickness of plate, as a cross check on your data..

Have you looked at flat plates of colorless CZ, where sometimes you will find the rare earth spectrum due to Yttrium stabilization

Calcium stabilized colorless CZ is an fairly adequate representation of type IIa diamond spectra, HOWEVER, quite of bit of CZ is stabilized by Yttrium (stabilizers are needed so that crystalization occurs in the cubic form)

 

[adamasgem]

Date: 8/15/2005 10:01:55 AM
Author: Yuri

Results:
1) DiamCalc can model diamonds and some other facetable materials with their absorption spectrum.
2) Photorealistic images and color coordinates can be used for color optimization of rough.
3) Color appearance depends on size, shape, cut style and cut parameters.

Discussion:
1) We did not find any light source with known emission spectrum. Modeled light can be different from real light
2) Canon camera has its own function (how color intensity depends on camera exposure time). This function is rather complicated. Also the camera has its own spectral sensitivity function while DiamCalc uses a standard function for human eye spectral sensitivity
3) We do not know how GIA will grade any particular color because we do not know color boundaries of different color ranges like Fancy light, Fancy, Fancy intense, Fancy vivid by GIA.
4) This work was done in the DiamCalc 2.4 with gamma correction sRGB. Other gamma will give other colors.

We will be glad to receive your suggestions.

Yuri.. I noticed that you used 10nm spacing for your absorption spectra input, your photoreal representations might improve with finer spacing, rather than 30 points across the visible range.

Especially if you are modeling with fluorescent sources because of the Judge II booth..

10nm spacing is OK for computation of chromatic flare metrics but might be a little coarse for Photorealistic modeling purposes, I don''t know..

Were you using a Photopic ( or Scotopic) human efficiency function in the modeling?

 

[adamasgem]

Date: 8/15/2005 8:08:43 AM
Author: oldminer
This is sure a highly technical thread. For those less technical, it sounds as if you are saying as follows: Give us a diamond. We will measure a lot of diemensions, facet lengths and angles and calculate (using complex software) what we believe to be true about your diamond.

What seems apparent is that not every measurement is taken into account or accurately measured. Some measures are computed to the best ability of the measuring device and some are just not measured. It seems there is general agreement that a diamond may be somewhat different than what is measured due to machine error, or not requiring a certain measure to be taken.

What degree of accuracy is involved in the final estimation of light behavior via ray or beam tracing using the equipment recommended? How much higher degree of accuracy will there be when machine error is decreased?

You know my side of the argument is based on direct measurement of LIGHT from a diamond, not calculating it from modelling. How far apart are the end results as you see it?

Thanks.

Dave..
1) Dimensional "measurement errors" are more critical in modeling than in direct light measurements, HOWEVER

2) The relative stone to stone metrics one gets from one light source versus another WILL DIFFER, if one doesn''t DEFINE the source USED in EITHER case, then NO ONE has any idea of the VALIDITY

AND 3) In a direct light measurement system, the SIZE of the stone changes the INPUT angular light distribution relative to a fixed "source", AND then the resulting "measured metric", AND THAT IS A PROBLEM ALSO BECAUSE YOU ARE THEN COMPARING APPLES TO ORANGES, but the PROBLEM is more easily handled in the theoretical case than in the physical case, and, I believe STRONGLY, is as, or MORE important as errored "measurements" used in the theoretical case.

In either case, a single viewpoint Metric MAY BE MISLEADING, be it "measured" or calculated..

A calculated multi viewpoint metric MAY HAVE A CHANCE OF GIVING A BETTER representation of relative "goodness", from what I have seen (or not seen, as in the case of a couple of black boxes)..

 

[strmrdr]

Getting back to colored light for a minute.
Any opinions does tinted light mask contrast?
Could this have been an issue with the GIA study given what we know about "white" leds?

 

[adamasgem]

Date: 8/15/2005 11:33:32 AM
Author: oldminer



“Direct measurement of LIGHT from a diamond” is not the synonym of measurement without mistakes.
BS and Imagem are direct measurement tools. Are result same?'

I would reply that we know the results are NOT the same because it is our belief after sufficient testing that BS is not doing something correctly which ImaGem is doing correctly. The end result is that ImaGem can discriminate performance as the trained human eye would and that the BS fails to do this.Dave, I think that all one can say is that they are doing it differently, and get different results, as to which is "correct" or can discriminate better, is a technical issue NOT PROVEN, in any manner whatsoever. You can make statistics say anything you want to.

What is "correct" and "discrimination" are too different things. I have shown this clearly and theoretically, by showing that higher angle lighting "discriminates" one diamond from an other but looking at subsets of hemispherical lighting (and have presented it at the recent EightStar conference).

You can also READ, but PLEASE DO NOT DISTRIBUTE TO IMAGEM IN ANY MANNER WHATSOEVER (I would not be happy about that), my writings on 3D raytrace studies, available in my SAS2000 software, under the View menu of the 3D RayTracing Module.

 

[Serg]

Marty,

10 nm is more than enough for modeling diamond in any source light if you are using average spectrum. You can use even 20 nm

For some combination material and source light are better to use 1 nm of course.

We are using model eye for good illumination ( cone light model)

 

[belle]

wow, this thread got so suddenly technical..i can''t tell if the original question was answered yet or not.

after 4 pages are we any closer to figuring out why gia included steep/deep diamonds in ''excellent''?

 

[adamasgem]

Date: 8/15/2005 3:47:32 PM
Author: Serg

Marty,

10 nm is more than enough for modeling diamond in any source light if you are using average spectrum. You can use even 20 nm

Hi Sergey.. The published "theoretical" fluorescent spectra (Wyszecki) seem to have more energy in a wider band at the emmision points of the phosphors, than what I see using spectrophotometry, that is why I made the comment

For some combination material and source light are better to use 1 nm of course.

Yup.. The wide band averaging of absorption (or source) will effect the blue end of the spectrum more, especially in diamond


We are using model eye for good illumination ( cone light model)

 

[adamasgem]

Date: 8/15/2005 5:02:04 PM
Author: belle
wow, this thread got so suddenly technical..i can''t tell if the original question was answered yet or not.

after 4 pages are we any closer to figuring out why gia included steep/deep diamonds in ''excellent''?

Hi Belle.. Unfortunately, there isn''t a direct and/or easy answer to that..
We don''t "know" the relative weighting placed on "all" the factors going into their "grade" and the unrealistic uniform Hemispherical lighting model they originally used, now with a limited perspective viewpoint, smears the distinction between stones, at least when they used the WLR methodology. Also it appears that using a single "point" source of light, to evaluate chromatic flare or "fire" can also alias the results..

 

[Garry H (Cut Nut)]

Date: 8/15/2005 7:19:40 PM
Author: adamasgem

Date: 8/15/2005 5:02:04 PM
Author: belle
wow, this thread got so suddenly technical..i can''t tell if the original question was answered yet or not.

after 4 pages are we any closer to figuring out why gia included steep/deep diamonds in ''excellent''?

Hi Belle.. Unfortunately, there isn''t a direct and/or easy answer to that..
We don''t ''know'' the relative weighting placed on ''all'' the factors going into their ''grade'' and the unrealistic uniform Hemispherical lighting model they originally used, now with a limited perspective viewpoint, smears the distinction between stones, at least when they used the WLR methodology. Also it appears that using a single ''point'' source of light, to evaluate chromatic flare or ''fire'' can also alias the results..

Marty do you think GIA has actually used any of its computer aided WLR and DCLR in the formulation of its grading system?

I had come to the conclusion, based on things said publicly and privately, as well as the reading between the lines of the foundation article, that the grading system was entirely based on trade observer preference (with a few office gals chucked in for good measure).

I think they explored ideas like brightness and fire etc, but ultimately asked people to decide - "which is better, this dead firey stone, or that bright fireless one?"

Not that there is anything wrong with that.

THe questions we have in hand are:

1. tray / background
2. lighting type

The results of the GIA''s proportions that include steep / Deep''s are surprising, to say the least.

Successful luxury product brands fulfill two aspects in consumers’ minds; functional and emotional. Diamond itself (and a great shopping experience) can fulfill emotive needs, but buyers to a greater or lesser extent, want the functional assurance of a quality diamond. Doubts in a consumers mind about a diamonds’ functional aspects will kill the sale.
Diamonds main functional aspects are the 4 C’s; 3 of those aspects, Carat, Colour and Clarity, can be guaranteed fairly convincingly. However consumers are often frustrated by diamond Cut Quality. Our Cut Group believe this functional weakness is the single greatest problem facing the diamond industry today.

Confusion: GIA Excellent = AGS 5? This will lead to worse functional support than the present cofusion.
President Bill Boyajian suggests that retailers will be in a position to better guide consumers in making purchasing decisions as a result of GIA''s broader range of diamonds than traditional ''Ideal'' type concepts. Is this believable? Will consumers really trust jewellers? Or just update the car, or take a trip to Europe.

 

[adamasgem]

Date: 8/15/2005 8:22:53 PM
Author: Garry H (Cut Nut)

Date: 8/15/2005 7:19:40 PM
Author: adamasgem

Date: 8/15/2005 5:02:04 PM
Author: belle
wow, this thread got so suddenly technical..i can''t tell if the original question was answered yet or not.

after 4 pages are we any closer to figuring out why gia included steep/deep diamonds in ''excellent''?

Hi Belle.. Unfortunately, there isn''t a direct and/or easy answer to that..
We don''t ''know'' the relative weighting placed on ''all'' the factors going into their ''grade'' and the unrealistic uniform Hemispherical lighting model they originally used, now with a limited perspective viewpoint, smears the distinction between stones, at least when they used the WLR methodology. Also it appears that using a single ''point'' source of light, to evaluate chromatic flare or ''fire'' can also alias the results..

Marty do you think GIA has actually used any of its computer aided WLR and DCLR in the formulation of its grading system?

It doesn''t appear so, does it?

I had come to the conclusion, based on things said publicly and privately, as well as the reading between the lines of the foundation article, that the grading system was entirely based on trade observer preference (with a few office gals chucked in for good measure).

Yah, and there may be biases in the "trade" group selected, based on what they typically see

I think they explored ideas like brightness and fire etc, but ultimately asked people to decide - ''which is better, this dead firey stone, or that bright fireless one?''

Good question when the environments used may mask one aspect or another..

Not that there is anything wrong with that. But there is something wrong if it is the tail wagging the dog

THe questions we have in hand are:

1. tray / background
2. lighting type

YUP
The results of the GIA''s proportions that include steep / Deep''s are surprising, to say the least. But it will sell more "paper"

Successful luxury product brands fulfill two aspects in consumers’ minds; functional and emotional. Diamond itself (and a great shopping experience) can fulfill emotive needs, but buyers to a greater or lesser extent, want the functional assurance of a quality diamond. Doubts in a consumers mind about a diamonds’ functional aspects will kill the sale. Well, on the retail brick and morter side, I doubt that MOST consumers get 10% of what "Truth" they get on the internet. Sales clerks are just that, clerks

Diamonds main functional aspects are the 4 C’s; 3 of those aspects, Carat, Colour and Clarity, can be guaranteed fairly convincingly. However consumers are often frustrated by diamond Cut Quality. Our Cut Group believe this functional weakness is the single greatest problem facing the diamond industry today. It has always been the flim flammers way to sell at a lower price, sell off make goods

Confusion: GIA Excellent = AGS 5? This will lead to worse functional support than the present cofusion.
Looks like "The world''s greatest authority in Gemology(TM)" may have caused it by bowing to the mass merchants and corporate big wigs who give them big bucks, look what happened with the QVC thing, GIA shilling their Diamond Dossiers for the "D-E-F collection". Or the change in the "standards" on fluorescence in diamonds. Loosen everything up, and make the "trade" happy, may have been the real motivation for management, but we will never know the ''truth", because of the close lipped society in NYC and Carlsbad. It is all about $$$$$$.

President Bill Boyajian suggests that retailers will be in a position to better guide consumers in making purchasing decisions as a result of GIA''s broader range of diamonds than traditional ''Ideal'' type concepts. Is this believable? NO, NO, NO IT is not believable, but MOST jewelers will just parrot what GIA tells them, and happily so, since they just pass along the costs to the consumer imposed by the control so few have on the diamond market creating artificially limited supply. Just ask the cutters who got pushed aside with supplier of choice.

Will consumers really trust jewellers? OH sure they will, they will have more "paper" from GIA, and jewelers will probably laugh all the way to the bank selling off make goods like they have for years.

Or just update the car, or take a trip to Europe. I don''t know about that, but guys are "forced" into spending to "prove" their love, and get lucky, at least for one night

And Guys being guys, nuff said on that.

 

[adamasgem]

Date: 8/15/2005 9:57:13 AM
Author: Yuri

Subject: Color optimization of a colored CZ .

Re: Cutters have learned how to INTENSIFY the color in radiant cuts, to getthe lucky vivid color call in fancy yellows (Citation of Martin Haske).




Dear Martin,



Some times ago we have started to study possible ways to optimize color in fancy colored diamonds. Because it is hard to have plates of colored diamonds necessary for reliable spectra recording we decided to play with CZ material first. We have CZ plates of different color and thickness and we have recorded their spectra. Fig. 1 shows 5 spectra of different thickness (1 mm, 2mm, 4mm, 8 mm, 16 mm) orange CZ plates made of one CZ rough piece. Being normalized on their thickness all these spectra are still slightly different because of uneven color distribution in the CZ rough piece.


to be continued

Suggestion for DiamondCalc, ADD a reference PATHENGTH box (so you can type in 1.0mm or 2.2mm) to be associated with the input absorbace spectra, and then handle the scaling arithmatic internally. THAT IS HOW I DO IT.
It would make the function it more useable. You only have to do the rescaling once so it won't effect speed. You could also allow the data to be input in "transmission" rather than absorbance units..

 

[RockDoc]

RE: GIA''s position.


I have been quiet about it since the system is written in stone yet. GIA has certainly done some announcements of what they intend to do, but I think it is all subject to change.

GIA gets their funding and income from dealers primarilly. So if one thinks they don''t have to "please" their customers (dealers) they are looking at this scenario through rose colored glasses.

When Mr. Boyajian states it is a better system, I suppose that is a factual statement, but it only, I believe is relevant when comparing what GTL used to report, which was basically nothing. It is interesting to note that in the early 80''s they did come out with a 4 grade cut grading, but never adopted that to their lab reports.


GIA certainly knew about different cut qualities for years. It seems rather obvious that their recent research only stems from the AGS issuing more complete reports as to cut grade.

I wonder if GIA would have announced their currently published cut grading system if AGS hadn''t gone into the lab business?

I am also curious about Garry''s previous post about Mr. Boyajian saying that in Inida larger tables and spreadier stones are preferred.

This is certainly a challenge for all the major labs to adjust their cut grading results. I think it is going to be close to impossible to please everyone''s tastes and preferences.

But time will tell I suppose.

Rockdoc

 

[michaelgem]

I agree with David''s well spoken ideas repeated here:

"My own comment is that modeling is very useful to cutters as it provides them choices and recipes for cutting and re-cutting diamonds to perform better. There is a huge need for cutters to learn about these changes and what to do with rough.
However, when it comes to GRADING diamonds, it just seems better are more reasonable to measure the actual performance than to estimate it through a huge process of measuring lengths, angles, facet ratios, etc.

My thoughts are that both methods, direct measurement and calculation of performance are highly useful tools, but for different parts of the trade. I don''t think it is logical to always calculate the light performance if one can measure it. It is very proper to calculate POTENTIAL light performance for instructing cutters on how to work rough or to recut diamonds. The end result should be measured, not calculated..........

I think my logic is way more solid than my ability to argue scientific details or technicalities with PHD''s and optical experts. I am just a guy with a deep interest in what is going on and have a degree of involvement in this ongoing process. Our industry will benefit the most from developing the best products. I have seen so many half-baked products come and go. This time, I sure would like to see our business make the best choices and use solid science, not smoke and mirrors." David S. Atlas

I add the following from my post to Bruce Harding:

"We believe most aspects of optical performance or diamond beauty may be more accurately observed, (or predicted), and evaluated from stationary images of a diamond , (face-up and tilted), under a representation of typical illumination and viewing circumstances." Michael D. Cowing
Re: Michael:
. Thanks for compliments.
. It sounds like your last paragraph, highlighted, is a quote from something. If so, what? Bruce L. Harding

This is my wording of the concept that I have been promoting for 6 plus years beginning in print in my 2000 Journal article and, among other ideas, elaborated upon in the 2005 Journal article: "Describing diamond beauty - assessing the optical performance of a diamond". A four page digest version, which you would better appreciate for its brevity, appears in the latest NY Diamonds Magazine and Israel Diamonds Magazine.

I have found a number of people that promote or are sympathetic to this idea or concept. I believe we can adequately handle the objections that have been posed concerning ''direct assessment'' from diamond imaging. For example, I have made the camera lens adequately ''see'' what my asymmetrically placed eye sees in typical illumination (including the diamond''s fire).

I like direct assessment from images of diamonds under a representation of typical viewing and illumination circumstances, because it can clear away for the consumer the mystery, smoke and mirrors, and allows for clear and easy-to-understand explanations of what constitutes the best, (or do I dare say Ideal?), diamond cutting.

Michael Cowing

http://www.acagemlab.com

 

[Garry H (Cut Nut)]

Bump - the first 15 or so posts on this thread discuss the effect of stereoscopic vision.

I am considering ways to factor this into HCA.
It would make many very slightly steep deeps - probably 2.5 hCA''s - become under 2.0''s.

If any pro-sumer consumers want to review the first part of this thread and give opinins i would welcome it.

Some of the later stuff got a bit heavy - but i tried to introduce the idea early on - and Storm said why he thinks dominant eye''s take over. I am certain that is not true in this situation.

 

[valeria101]

Date: 3/21/2006 4:56:18 AM
Author: Garry H (Cut Nut)

Bump - the first 15 or so posts on this thread discuss the effect of stereoscopic vision.

I am considering ways to factor this into HCA.
It would make many very slightly steep deeps - probably 2.5 hCA's - become under 2.0's.

Not much to say - aside a little observation about the graphical presentation of the 'binocular vision' pictures suggest an analogy:

What is the difference between considering a combined binocular view of diamonds as a weighted addition of light return maps from 'single eye' views, and considering the overall visual impression of the same diamond seen by a cyclops while being tilted within an arc equivalent to the focal distance of the normal two-eye viewer?

As much as I can tell, both one eyed and two eyed viewers almost never look at diamonds sitting still - accounting for a bit of dynamic by some smart combination of snapshots at convenient angles makes at least as much sense to me as taking a stereo view of things.

And a quick note about the overlay itself: the picture shows a superimposition of transparent pictures which looks a tad muddy. It is probably a weighed addition. However, the description of what the stereo view account for suggests a different function for combining the two one-eye views - perhaps taking a the maximum brightness between corresponding points, or something else that doesn't 'muddle' the contrast.



Finally... is it possible to account for the famous GIA rounding in the HCA scores? As much as I can tell, this amount to assigning a small range of HCA scores to a stone (or the minimum of the range?

). Since as little as half a score point (from 2 to 2.5) is the object of intervention, now, this rounding thing may account for at least as much in many cases. Us

 

[adamasgem]

Date: 8/9/2005 2:37:55 AM
Author: Garry H (Cut Nut)

BTW this is the color of the CDAXVE or whatever they call it in the photo from the GIA website. A fairly dark gray. But it would be great if someone could give us the panetone or some other code # please?

Garry, I'll try to remember to bring my Munsell Matte Neutral Gray Set to Vegas To try to get a Comparison (If they let you or me near their booth), but you might really need to get a reflectance spectrum on the paint, because lighting will change the comparison perception.

Sorry for the repeat post , as I didn't notice the bump..