Ummm.....is AGS now doing something really different....or only better?

[Regular Guy]

So...for the time being, I'll just suspend judgment, and will be interested in the various expert's feedback on this.

I'm not sure if, as has recently been inferred, AGS has really done an incredible job of marketing....if they've maybe overstated the case....or maybe they've simply bootstrapped language, also asserting fundamentals by restating things in a new way.

Not sure what got me thinking about this, this morning. Recently, Storm noted that he'd take a clear set of Helium output any day over a certificate, for example.

To his credit, JohnQ, somewhere on this board, has been a clear spoken supporter of AGS, and their new way of doing things. Also, he's talked about the different ways that a diamond can be measured, including direct and indirect. Also, I believe he's associated AGS's new approach to evaluation with direct measurement...at least in principle. Also, I believe I've done this.

In the past, RockDoc has criticized proportion based measurement, saying it can only tell maybe 70% of the story for a round, 30% for a fancy. Also, later, he's made mention of the fact that with AGS's PGS, he can reliably tell the cut grade of a diamond not graded by the AGS.

To my shame, blame, or credit, I've supported all these ideas. But suddenly, I have questions. I'm curious now what the claims are, what they can mean, and what is being asserted by the representation that the new AGS0 is more than proportions based, and is now also performance based.

Really, what I'm wondering if AGS is merely only still just proportions based, but now...much more precise in their predictions from this, such that their level of confidence is so high in asserting proportions are really all now that you need to know, that they're saying they're now doing performance based analysis. They presumably ARE now requiring the stone to be in hand, so that it's particular measurements are being measured. But then, I'm thinking it's only analyzing then those abstracted measurements, and imputing performance, based on what those measurements should do for the diamond.

In the past, I'm not sure that's what Rock Doc has meant by the fact that he needs to see the specific stone....and that knowing just it's stats isn't enough. But, maybe knowing the stats is all that AGS PGS software does anything with.

I went to the AGS site to see its claims. They say:

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

"Current knowledge and technology unknown to Tolkowsky have shown that a diamond with proportions slightly outside the Tolkowsky parameters still can perform extremely well and be beautiful. It also has been discovered that some diamonds, which are cut to those strict standards(1), are not as efficient in how they manipulate light. There can be a number of reasons for this. The AGS Cut Grading System considers not only the proportions of a diamond, but also the craftsmanship of its overall symmetry and polish. Most importantly, it uses technology to analyze the cut’s impact on diamond’s overall performance. This results in three categories of the AGS Cut Grade — Light Performance, Proportions, and Finish.

(1) Remember: The listed proportions, angles, and percentages should be viewed only as an indicator of what has traditionally been considered Ideal cutting for a round brilliant diamond. Simply because a diamond’s cutting standards fall within this range of measurements does not prove or guarantee that it is beautiful. Also, it is important to be aware that diamonds cut to proportions slightly outside the parameters can be extremely beautiful."

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

So, what must be this "technology to analyze" AGS is talking about. They provide the user of their software...software. They do not purport to measure light pixels doing anything, right? There's no other equipment apart from software, it seems.

Though possibly RockDoc can speak best to this, any professional opinion is welcome.

 

[oldminer]

Pure proportion grading works pretty well. The more accurate the measures, the better it might work. It will never be even for rounds, and especially for fancy shapes. AGS has done excellent marketing and Peter Yantzer has a convincing delivery which shows a distinct openness. People really like this combination of approach and personality.

However, you are very right that proportion measures do no always address a particular diamond and its unique characteristics of transparency and inclusion placement. It sounds to me that AGS and GIA, for that matter, are going to ignore these issues when it comes to cut grading analysis and report such issues under clarity only. The public will surely be misled on some stones by this approach. Why else would one of AGS'' strongest supporters on Pricescope often remind people that he must see the stone to know how well it looks in spite of the cut grade. Dealers will admit the same happens to them.

The logic of directly measuring LIGHT exiting each individual diamond graded and reporting exactly the characteristics of that particular diamond make more sense and will ultimately lead to the best reporting of cut and performance. I have promoted proportion screening for Cut Class grading for nearly 20 years to an audience that mostly said it wouldn''t work, when in fact it works pretty well. Tens of thousands of AGA graded diamonds are in the market and I am not busy with any complaints. With better measuring tools we do get better predictions, but nothing is going to beat directly measuring light behavior. Also, one should not forget that certain features of diamonds such as Durability, Finish and Size (spread) need to be measured on machines like Sarin, Ogi, Helium or ImaGem and reported as a Craftsmanship grade. We call such a grade "DFS". and it does not require tons of data or large numbers of scans to create. 3D is cool, but what does it do in terms of defining cut quality?

People are beginning to analyze the problem and discover ee where we are today. Some of these folks will think about how to make the remaining issues disappear and find direct measure of the Light coming from each diamond a best solution to getting the information to the consumer and to distant dealers.

 

[Garry H (Cut Nut)]

Ira AGS use their pixel counting software that takes the input of a 3D stl model of a diamond, not the proportions per se''.

the models are used to generate 45 ASET images thru a range of tilts from 0 to 45 degrees in 1 degree steps.

This is done for the 3 colors, plus leakage, and also the blue compment in a 40 degree aset is counted too.

And they also have forward and back ray traced images for the potential spread of fire - they also count 45 of each of these.

Now this is probably best described as a good approximation.

 

[Regular Guy]

Date: 9/22/2006 5:09:33 PM
Author: Garry H (Cut Nut)
Ira AGS use their pixel counting software that takes the input of a 3D stl model of a diamond, not the proportions per se'.

the models are used to generate 45 ASET images thru a range of tilts from 0 to 45 degrees in 1 degree steps.

This is done for the 3 colors, plus leakage, and also the blue compment in a 40 degree aset is counted too.

And they also have forward and back ray traced images for the potential spread of fire - they also count 45 of each of these.

Now this is probably best described as a good approximation.

Garry, correct me if I'm wrong, but isn't the process, in substance, what they might call in CA..."as if," although the diamond really is right there?

Seems like, rather than coming up with zillions of hypothetical models, for a specific diamond, they create a model based on its specific measurements, and then apply appropriate principles of light behavior to the created model. The actual diamond, then, becomes less relevant, than the proportions that describe it.

Do I understand this correctly?

 

[Garry H (Cut Nut)]

What is light?

 

[Regular Guy]

OK then!

(Maybe you''ll help me connect the dots...which are only apparently there...)

Off to services.

for any Jews, good yontiff.

Muslims are upon a special period as well, I believe.

Cheers,

 

[Rhino]

Good questions and excellent point too Garry about being a good "approximation" as beauty still lies in the eyes of the beholder.

While Dave''s point has great merit and I''m positive he''s not undermining the value of seeing a stone live for an inspection, I fall more in line with RockDoc''s, Garry''s and strms comments.

On a personal level, if I''m going to make a judgment based on numbers I want all of them or as strm says, those from a Helium scan. I want to know upper half and lower half, stars, azimuth deviations, variances etc. GIA and AGS'' inclusion of lower half and star info on the newer reports are a HUGE plus IMO.

Here are 2 stones that formerly were AGS Ideal Cuts that are now AGS "Very Good" or a 2 on the current scale and that for "light performance" as both their polish and symmetry were graded by AGS as "ideal".

http://www.goodoldgold.com/diamond/1648/

http://www.goodoldgold.com/diamond/869

These 2 make for interesting learning stones because if you look at their basic proportion data or were to view IS images you''d think either was a safe bet, even being accompanied by the older AGS Reports. I''m not trying to undermine these stones, they are still pretty compared to alot of kaka on the market but in the new system of things NOT ideal (for those wanting true ideal cuts based on the new systems). It''s one reason I believe AGS should do away with the old lab reports already as that is still an option and will be until mid next year. Stones using the current DQR''s or older style DQD''s are not taking these factors into account.

Hope that helps.

 

[strmrdr]

Interesting topic and a hard one to cover without going into a 2000 word essay that is hard to understand.

Pluses of an AGS0
1: strict lab for color and clarity grading
2: limited rounding of numbers
3: ASET is a very kewl tool

Minuses of AGS0
1: They use virtual ASET and other things to set the grade using sarin data. Garbage in Garbage out.
2: while they have helium scanners they don''t use them except if someone happens to suspect the sarin numbers. (aren''t they all suspect when it comes to models)
3: The best Sarin is not accurate enough for good models, Id rate it as decent with helium getting a very good and ogi getting a fair to poor.
4: It is one organisations opinion on what a top diamond is. They include some I disagree with and leave out some kicken combos.

In some ways we get a false picture here of what a AGS0 diamond is because the most familar ones from whiteflash, infinity, gog, niceice and others are in my opinion some of the best of the best of the AGS0s.
In the same way that the GIA graded gog diamonds are in my opinion some of the best GIA EX diamonds.
Neither is representative of AGS0 or GIA EX they are the cream of the crop.

 

[Paul-Antwerp]

Date: 9/23/2006 1:22:43 PM
Author: strmrdr

In some ways we get a false picture here of what a AGS0 diamond is because the most familar ones from whiteflash, infinity, gog, niceice and others are in my opinion some of the best of the best of the AGS0s.
In the same way that the GIA graded gog diamonds are in my opinion some of the best GIA EX diamonds.
Neither is representative of AGS0 or GIA EX they are the cream of the crop.

While I think that the rest of your post is slightly exaggerated, I can only say ''amen'' to the words quoted above. Some brands and vendors here are answering, like Neil Beatty says, to a higher authority than AGS.

It is very important to note that not all AGS-0''s are equal.

Live long,

 

[Garry H (Cut Nut)]

Date: 9/23/2006 1:22:43 PM
Author: strmrdr

Minuses of AGS0
1: They use virtual ASET and other things to set the grade using sarin data. Garbage in Garbage out.
2: while they have helium scanners they don''t use them except if someone happens to suspect the sarin numbers. (aren''t they all suspect when it comes to models)
3: The best Sarin is not accurate enough for good models, Id rate it as decent with helium getting a very good and ogi getting a fair to poor.
4: It is one organisations opinion on what a top diamond is. They include some I disagree with and leave out some kicken combos.

Paul I think it is reasonable to ask why AGS conntinues to push Sarin?

What is in it for AGS?

Helium Polished costs less than Sarin.
Helium takes more time to scan a stone if you stand there with a stop watch - but only slightly less time to complete a scan in an operational sense. This is because much of the time involved is recording which stone you scanned etc.

Point in case as to what exactly is the relationship - Sarin recently announced some sort of "exclusive" agreement with AGS and the use of the AGS data for round and brilliant rough allocation scanners. exactly what that ''exclusivity'' entails I am not sure, since Helium, Pacor and DiamCalc had this work done and provided for free to cutters and all of us last year.

So Paul it seems that there is some reasoning at AGS for using a less than best process. If it is just time / cost / benefit - as I have been told is the case, then it is a decision that I think has its limitations:

for instance, I imagine the process might be "aha - a border line failure stone - lets do a helium scan and see how it fiars - we might be able to bump it up." alternatively the same process should apply to stones that just scrape in on their Sarin scan models - but people being people - I doubt this process would be adhered too.

 

[Regular Guy]

The asides about value added that specific vendors may or may not be adding to this thread are certainly helpful. And it is noted that vendors who list diamonds but don't stock them add value by reviewing its performance. But...since the conjecture in this thread is that they may doing even something that substantially AGS is not hardly doing (looking and evaluating), broad speculation about what metrics they will look for are relatedly welcome, because in this discussion, we are not resting on the laurels of AGS's metric, and trying to figure out what teeth that AGS metric really brings altogether.

So, can I drag it back?

To assert this thread is intended to be about something is saying a bit too much. I woke up scratching my head, and am just trying to clarify some disconnects.

Seems like it is becoming about:

- trying to remove what may have been hype
- clarifying what AGS is actually doing, getting the "differential" right
- giving the differential its due, whatever that may be

A few thoughts:

Politicians who wish to run for higher office, who have been in lower office, have the harder time of clarifying past statements, even though they have tried and stuck their neck out previously, where their competitors have not. In that vein, I would like to credit John Pollard on this board as the most stand out individual to try to clarify clearly dense topics, and make them clear. WF making him director of education was a coup for them.

However, that said, my strategy to get at what has been said here has mainly been to pick on him, see what's he's said, and identify the disconnect that way. Sorry John.

Thread #1, which is not the set of threads I had been looking for, where I think he frequently tries to tag proportions based indirect measurement with GIA, and direct measurement with AGS, but maybe it's better, where he comments:

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

"The AGS has also concluded that, while proportions and measurements are helpful for making broad predictive judgments, direct assessment of a diamond's actual light performance is more meaningful than numerical data and simulated predictions. They have now moved to performance-based grading rather than proportions-based grading and are using a reflector device called ASET (more info than you probably want, here.) Much like Dave Atlas' lab, AGS is committed to a judgement based on the actual diamond. This is because proportions sets do not account for patterning dynamics, girdle treatments, presence of facet yaw, etc.

This is not a new concept: For years many of us have been saying that an ideal-scope image always trumps an HCA estimation. This is the same premise.

Interestingly, after 10 years of experience with proportions-based cut grading AGS evolves to performance assessment - just as GIA is on the verge of introducing their first cut grading system based on (drum roll)... proportions."

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

In a more recent thread, we have suggested that GIA can be criticized for minimizing the constraints possible with its cut grade system by looking to the equipment available to its broad base of members, and setting the cut grade factors to be in line with this. Consider this recap from Garry::

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

"The point here about GIA rounding proportion data is that they set the rounding data based on what accuracy appraisers and other stake holders can measure to with reticules on microscopes and straight out visual estimation.

GIA's well looked after and callibrated Sarins can achieve beter accuracy - John the + - 0.2 degeees includes the crown angle which is the shortest and hardest to measure. Thier pavilion accuracy is much better - and this is the most critical angle for cut grading system that is parametric (proportion based). Rounding table to + - 1% is fine, and crown at .5 degrees - is almost OK - but pavilion counts 5 times more than crown - and =+ - 0.2 pavilion is like having + - 1.0 for crown - it is too much.

To run a system based on outside levels of accuracy seems stoopid"

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

What we should consider here, though, is a linearity between GIA and AGS...which is what has existed before, and now continues to exist. The difference maybe should be removed from the hyperbole. Also, removing the difference can add value.

It was asserted here (thanks John) concerning the light performance functionality that had been associated with Sarin equipment, that a proportions based approach can suck:

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

"The feature is available but not used much. "Light performance" evaluation is viewed as dubious even in machines like BS and Isee that do direct assessment. With Sarin’s given error any assessment is potentially off to begin with.

You know there is also a H&A function in DiaVision that works with a licensed DiamCalc component but the problem again is that the scans aren’t accurate enough (hence Helium being developed). If they were accurate enough we could rely less on live photos and use .srn files - although Sarin would have to upgrade their grading of the H&A... Brian spoke with them about the problem with their example of “perfect” hearts – the people at Sarin told him he was the first to comment on it not being perfect.

Additionally there are subtle factors currently being studied which may not yet be accounted for by any scanning devices. More on this as we get into '05."

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

But...is it clear, or possible, that AGS just ended up licensing this particular feature in its NEW approach to grading?

In the FAQ section here, we are brought forward, but perhaps with a footnote, on the nature of the new DQD...

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

"NEW DQD (performance-based)

Under this system each diamond is individually measured for face-up light performance as well as proportions and finish details.

Note 0 (Ideal) grades in:

Light Performance (evaluates angular spectrum via modeling and ray-tracing of each diamond)"

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

So...the detail is there, albeit in the parenthesis.

And...the new software does not require qualitative distinctions with respect to the class of machine being used to implement it. (eta, I think this touches on your point above, Garry, writen as I wrote..) Granted, you can't use a microscope to use it. But any old OGI will do. Also, although Helium is available, even to AGS, we are advised that even AGS doesn't consider it value added enough to implement it, and more routinely applies Sarin in the use of this software.

Also, there is the matter of the use of the hand held, and direct measuring devices, and where these fit into the AGS system. On the basis of their use, we can approach giving a nod to not only a different degree, but different direction in their comparative analysis. But...I read that the use of their ASET device only can begin the grading process. For their own purposes, they feel their PGS system's analysis includes consideration of ASET output, and super cedes it. Is there any checking one to the other? I don't understand this, but would like to. As I read here about real implementation...a user of ASET must not corroborate the findings of ASET resources with PGS...and that PGS, along with separate analysis of fit & finish, is enough to establish a grade.

And again, so, where does this bring us?

Perhaps AGS still trumps GIA, but I don't think in nature, but degree. As has been the case, it is tighter. But, it is still proportion based, basically. Also, it's development and implementation may have been sensitive to financial and equipment constraints available to get this better, by virtue of the technologies conveniently available, both for itself, and for its members.

On the fundamentals...is there any new understanding to be had? Has the research AGS has come forward with allowed us to understand something fundamental about the nature of proportion based measurement, making effectively obsolete the need to accompany it with some genuine form of direct measurement, not based on projections of a model? Has anything motivated a resource like Imagem, apart from the customer's need to compare this to that with a greater level of precision than is otherwise obvious?

Basically, if as had been conjectured about the coming advantages of AGS's system, if you really need direct measurement...then you still need it, even after AGS.

Alternately, if something has been understood enough about the nature of what proportions can tell us, such that direct measurement is effectively superfluous, then...that's something else again, and any output of data, whoever provides it to us, with sufficient rigor, may be just fine for ordinary evaluation. Perhaps this has always been the case. Seems like determining the level or rigor is still substantive. Depending on how thorough you want to be.

Of course, looking, comparing, picking that way...maybe makes things a lot easier. Then again, that's not largely what the long distance shopping associated with Pricescope has been about.

We could say it's about finding vendors you can trust. This post is about the system vendors can depend on, and what consumers can or cannot depend on, independent of any particular vendor. How helpful will Blue Nile's virtual AGS be? And on what basis will we understand that?

 

[Garry H (Cut Nut)]

Ira can you tell us / ask us what you just said in one sentence?

 

[Regular Guy]

Date: 9/23/2006 10:58:49 PM
Author: Garry H (Cut Nut)
Ira can you tell us / ask us what you just said in one sentence?

No.

but, I can say again what I wrote in my first post in this thread:

"Really, what I''m wondering if AGS is merely only still just proportions based, but now...much more precise in their predictions from this, such that their level of confidence is so high in asserting proportions are really all now that you need to know, that they''re saying they''re now doing performance based analysis."

 

[Regular Guy]

Hey Garry, I could try one more time.

You differentiate here...

Date: 9/22/2006 5:09:33 PM
Author: Garry H (Cut Nut)
Ira AGS use their pixel counting software that takes the input of a 3D stl model of a diamond, not the proportions per se''.

...and also, in the tutorial here, rather than contrasting between direct and indirect, you make the three fold distinction:

direct
proportions
3D

Do you have a basis for calling a 3D modelling approach different in substance, rather than just more thorough, than being proportions based?

 

[Garry H (Cut Nut)]

Date: 9/23/2006 11:39:52 PM
Author: Regular Guy
Hey Garry, I could try one more time.

You differentiate here...

Date: 9/22/2006 5:09:33 PM
Author: Garry H (Cut Nut)
Ira AGS use their pixel counting software that takes the input of a 3D stl model of a diamond, not the proportions per se''.

...and also, in the tutorial here, rather than contrasting between direct and indirect, you make the three fold distinction:

direct
proportions
3D

Do you have a basis for calling a 3D modelling approach different in substance, rather than just more thorough, than being proportions based?

As the first person to develop an effective proportion based cut system Ira, one that could be developed further after my patent is granted, I am the person on earth with the most to loose by claiming 3D approaches are better than parametric or proportion based systems(GIA''s patent application for the same thing was lodged about 4 or 5 years after mine).

There are some good articles about 3D benefits this from our Cut Conference in Moscow 2 years ago - but cant find the links
http://www.cutstudy.com/cut/english/conferens-article/7.htm this one is a little about the benefits.

It is a no brainer Ira.

But none of the other approaches is mutually exclusive of others - just 3D has many more benefits that others dont

 

[strmrdr]

RG:
The AGS system is computer simulation based with the input being the scanned proportions (usually sarin) of the diamond then the simulation is compared to AGS ideal computer simulated performance and given a grade.

Is that what your looking for?

Is it the ultimate answer to diamond performance grading.. NO .. that doesn''t exist .. is it a good one .. Maybe.
10 years from now it will likely be considered a joke as the technology and science moves on.

What is the AGS0 grade to me?
To me its a decent first round filter but not a requirement in rounds.

In princess cuts its a far more revolutionary because it was different from how they were being cut.
I think its safe to say that a lot of high performance princess diamonds on the market would not exist without AGS. They seem however to left some wiggle room some cutters are exploiting to save weight at the cost of performance. (Paul can expand on this better than anyone)

So far for asschers and emerald cuts it looks not as revolutionary as with princess cuts but more evolutionary than with rounds but it is too early to tell for sure.

So if your asking is the AGS system a huge deal Id say with rounds not really, with princess diamonds yes, and with asschers/emerald cuts its too early to tell for sure.

 

[Regular Guy]

Thanks Garry & Storm,

Garry, you''ve written here:

"Method of Analysis

Much of the development of HCA was done using observations of DiamCalc ‘virtual’, Firescope™ and Gilbertsonscope images, as well as other MSU website based virtual diamond tools. Comparing the effects of variations in diamond proportions using virtual diamond analysis eliminates problems with variation in color, clarity and the minor facet groups, all of which are encountered in studies using real diamonds."

I''m wondering, per my question above, why you consider the difference between proportion based modeling, and 3 D based modelling, different enough in substance to differentiate them. You''ve clearly used them in tandem, and the line between these two non-direct methods of analysis seems to be somewhat transparent to me. One is 2 D, one is 3 D. 3 D is better. Perhaps this is just semantics, and it is useful to maintain the distinction, so as not to confuse, since 2 D, prior to HCA and both newer systems from GIA & AGS was qualitatively inferior.

Then, in substance, where you say, Storm:

Date: 9/23/2006 11:57:38 PM
Author: strmrdr

Is that what your looking for?

Is it the ultimate answer to diamond performance grading.. NO .. that doesn''t exist .. is it a good one .. Maybe.

I do say...is the current claim that the eval done by AGS is now performance based, reasonable.

 

[Garry H (Cut Nut)]

3 d can and will soon be used to reverse engineer many new cuts.

HCA is for rounds only - very hard to appply to many other shapes.

HCA does not do symmetry and painting etc would be very hard to apply

etc etc

But 3D does not handle trasmittance issues -that needs to be solved too

 

[strmrdr]

Date: 9/24/2006 12:19:47 AM
Author: Regular Guy




I do say...is the current claim that the eval done by AGS is now performance based, reasonable.

***opinion***
I''m going to piss a lot of people off and say to me no it isn''t.
The AGS system is computer simulation based <--- My bottom line.
What they are simulating is performance under various tools that can be related to the real world but its not directly real world as the eye sees it performance based. <--- My bottom line.

Calling it a performance based system is marketing in my opinion.

 

[RockDoc]

Ira

You''ve brought out some very thought provoking questions...

I think most of your concerns will be addressed.

The exact "mechanics" of what AGS has done, I believe to a very advanced program as compared with what many others have done. Is it perfect, NO, can and will it be improved ? Yes.

AGS has done a lot more than what is disclosed, and given their multimillion dollar amount of research, not only by the lab, and those making comment and suggestion by the trade, they''ve also engaged the services of a significant scientific community in developing what they''ve done.

Some of this is proprietary, and the exact details of how it works will probably not be shared in full outside of the executive "arm" of AGS. AGS has been very forthcoming in their studies, and reasonably transparent in what they have disclosed and released.

But this is an ongoing devleopmental project, and will most likely be honed to a much sharper pointed analysis, in time to come.

AGS does have the challenge of living up to creating an industry standard, which is shared with GIA.

Although many others such as IMAGEM, GEMEX, ISEE2, have direct light analysis, and OGI /SARIN/ OCTANUS, measure proportions through scans, as far as becoming a "standard" I doubt regardless of how advanced they become, they will not have the consideration of being a viable standard, that will get as much "weight" that GIA and AGS will be expected to meet.

It is a very challenging job to take this on. It is a delicate, political balancing act where they have to "please" many different entitiies. The consumer, the dealers,the mining companies, DeBeers, the cutters, the retailers, and themselves all play a part in the mix. Sometimes, "pleasing" on category of the mix, will outrage other sectors, and draw criticism, sometime relevant, and sometimes with an underlying interest in its input.

At the moment I believe AGS holds most of the "trump" cards, and does significantly more in its analysis than is evident in the GIA cut grade system.

There''s politics and statesmanship that is part of this too, and AGS is doing it well, and because of WHO they are will be the most recognized method of grading, however, I think there''s a wide open future where things will be changed a bit.


Rockdoc

 

[Garry H (Cut Nut)]

Date: 9/24/2006 12:53:10 AM
Author: strmrdr

Date: 9/24/2006 12:19:47 AM
Author: Regular Guy




I do say...is the current claim that the eval done by AGS is now performance based, reasonable.

***opinion***
I''m going to piss a lot of people off and say to me no it isn''t.
The AGS system is computer simulation based <--- My bottom line.
What they are simulating is performance under various tools that can be related to the real world but its not directly real world as the eye sees it performance based. <--- My bottom line.

Calling it a performance based system is marketing in my opinion.

I dont disagree with you Storm.
AGS have never actually (publicly) tested their performance software against real people.

DiamCalc does not claim to have performance grading softwaree - although by AGS''s standards it has for several years.

But that is why we want to conduct a propper survey to find out for sure.

 

[Rhino]

Hi Garry,

Did you get a chance to look at the firemap during the poster session? I''m sure you''ve seen it before with the colors ranging from yellow to black with the blacks being the least potential for fire/dispersion. If you have, I''m curious to hear your thoughts on how you might think whether it does or doesn''t correllate to normal observation in spot lighting environments.

 

[Paul-Antwerp]

Date: 9/23/2006 9:35:26 PM
Author: Garry H (Cut Nut)

Paul I think it is reasonable to ask why AGS conntinues to push Sarin?

What is in it for AGS?

Helium Polished costs less than Sarin.
Helium takes more time to scan a stone if you stand there with a stop watch - but only slightly less time to complete a scan in an operational sense. This is because much of the time involved is recording which stone you scanned etc.

Point in case as to what exactly is the relationship - Sarin recently announced some sort of ''exclusive'' agreement with AGS and the use of the AGS data for round and brilliant rough allocation scanners. exactly what that ''exclusivity'' entails I am not sure, since Helium, Pacor and DiamCalc had this work done and provided for free to cutters and all of us last year.

So Paul it seems that there is some reasoning at AGS for using a less than best process. If it is just time / cost / benefit - as I have been told is the case, then it is a decision that I think has its limitations:

for instance, I imagine the process might be ''aha - a border line failure stone - lets do a helium scan and see how it fiars - we might be able to bump it up.'' alternatively the same process should apply to stones that just scrape in on their Sarin scan models - but people being people - I doubt this process would be adhered too.

I wonder why you ask me, Garry?

But I fear that there is an analogy with what you e-mailed me a few weeks ago.

You told me that my marketing problem was that we produce at a strictness beyond the boundaries of measuring systems, and that not always the best quality wins in the marketplace. Very often, the best marketing wins.

In the same way, Helium might have the same disadvantage. After all, remember the adagio of the 80''s among professional buyers: ''Nobody ever got fired for buying IBM-computers''. Of course, the adagio is now outdated, but there is definitely still some truth in there.

As for Sarin incorporating the AGS-system into their rough-planning, it sounds nice, but it will not do the trick. The pointers, which I sent you by e-mail, are still very valid, and by foreseeing this into rough-planning, one can really improve a scanner''s performance.

Live long,

 

[Paul-Antwerp]

Date: 9/23/2006 11:57:38 PM
Author: strmrdr
In princess cuts its a far more revolutionary because it was different from how they were being cut.
I think its safe to say that a lot of high performance princess diamonds on the market would not exist without AGS. They seem however to left some wiggle room some cutters are exploiting to save weight at the cost of performance. (Paul can expand on this better than anyone)

So far for asschers and emerald cuts it looks not as revolutionary as with princess cuts but more evolutionary than with rounds but it is too early to tell for sure.

The fact that the AGS-system produced ways of cutting a princess-cut, which were never used before, and that these stones clearly are better performing, is a proof by itself that the system works.

I still remember the beginning of last year, when AGS'' president Peter Yantzer happened to be in Antwerp, when we just finished our first attempts at AGS-0 princess-cuts. We showed these to him before sending them to the lab, and he visually checked them in detail. Not only was he amazed by the brilliance of the stone, but I most of all remember him repeating over and over again: ''Wow, a princess with such fire. I never saw this before.''

I now realise that not only was he happy with the beauty of the stone, but it must also have been a satisfaction of establishing a totally theoretical system, and then seeing that it indeed delivers extraordinary beauty in practice.

Unfortunately, all other cut-grading tools seem handicapped when used on any other shape than rounds. H&A of course, Idealscope has its limitations, and clearly Brilliancescope had a great database of rounds, and with the limited variation in rounds, it seemed to work, but it entirely misses the ball in fancy shapes. I do wonder if Imagem can take this hurdle.

Even more unfortunately, with only the AGS-system really working on fancy shapes, it is hard to differentiate within AGS-0.

Live long,

 

[Garry H (Cut Nut)]

Very insightful Paul

 

[Paul-Antwerp]

Date: 9/25/2006 8:46:20 AM
Author: Garry H (Cut Nut)
Very insightful Paul

You know, we all know the mountains and the highest peaks on this planet.

With the new AGS-system for fancy-shapes, AGS has put us on another planet. We have immediately explored the planet, and have gone for a mountain-range again. Others were satisfied to land on that other planet as such.

Now, we notice that extra tools are lacking, which clearly show that we are on a peak somewhere. In the meantime, we just sing a certain Randy Newman-song.

Live long,

 

[Garry H (Cut Nut)]

Date: 9/25/2006 1:32:18 AM
Author: Rhino
Hi Garry,

Did you get a chance to look at the firemap during the poster session? I'm sure you've seen it before with the colors ranging from yellow to black with the blacks being the least potential for fire/dispersion. If you have, I'm curious to hear your thoughts on how you might think whether it does or doesn't correllate to normal observation in spot lighting environments.

this keeps crashing.
Edited - it worked - my 3rd try !!!
Rhino the probleme with all the best methods that give a quantified numeric result for anything is that no one has done the wort to match them to all cuts and 'looks'. I mean things like DiamCalc light return, contrast etc, ImaGem, ISEE2, B'scope and AGS data.

eg you ran an unusual stone from the GIA set on your Bscope and found it did not work.

People want to establish systems that work with ideal cuts and the best stones - but one of the things that prooves how effective a system is is if it can delineate between the extreme stones - extremely bad for each of the things you are seeking to measure.

As a basis - all systems must be able to grade all cuts, account for all lighting sizes and distances from light and intensity of lights and structure of lighting positions and quantity's. They should allow for different focal lengths, and different people's preferences.
Then they can work on the diamond - but only then.

This is an ETAS fire and scintillation view for a given lighting, viewer and diamond. It is all quantifyable. You can descibe all of the above in the inputs, and measure all the outputs.

ETAS is effective total anguular size that we can see from given light sources that have come from a given diamond.

Nothing less can quantify fire or scintillation, of light light return etc

 

[WinkHPD]

Date: 9/23/2006 10:12:12 PM
Author: Regular Guy

Perhaps AGS still trumps GIA, but I don''t think in nature, but degree. As has been the case, it is tighter. But, it is still proportion based, basically. Also, it''s development and implementation may have been sensitive to financial and equipment constraints available to get this better, by virtue of the technologies conveniently available, both for itself, and for its members.

This is where I think you are wrong. The AGS system, while acknowledging proportions, is using forward and backward ray tracing to determine the performance of a diamond and is performance based, NOT proportion based.

AGS uses the Sarin primarily because it is much faster than the Helium to create the models that they use to do the ray tracing. They also look at the stone in the desk top version of the ASET tool which is actually looking at the real stone, not just the model. I am told it takes about 3 minutes to do a Helium model and when you are doing the number of stones that AGS is doing on a daily basis, one or two minutes a stone can make a tremendous difference in time.

Wink

 

[WinkHPD]

Date: 9/24/2006 12:53:10 AM
Author: strmrdr

Calling it a performance based system is marketing in my opinion.

And you are wrong in my opinion.

Wink

 

[Regular Guy]

Three primary thoughts...brought to you by...

a) one side, who thinks only direction observation makes any reliable sense
b) the other side, who thinks modeling is where it's at
c) the guy in the middle, who is undecided

A)

Date: 9/25/2006 4:06:20 PM
Author: Wink
The AGS system, while acknowledging proportions, is using forward and backward ray tracing to determine the performance of a diamond and is performance based, NOT proportion based.

From the point of view of someone from Minnesota, this is no different than Dave Pilkey's analysis, and the critical use of Flip-O-Rama, found in each and every Captain Underpants book. See the ninth point in this link from Wikipeda. You'll see the direction of the pages, and that they really can go back and forth.

Anyway, and really...take princesses for example (seems on point)...for someone who could choose between an AGS0 with a more modest Gemex report, and an option from GIA, with a more robust Gemex report...even given the discussion here...who is dead certain, especially at a distance, which they'd prefer?

B)

From the point of view of someone who thinks that a diamond's proportions can be very telling.....even though I personally still only see the continuity of relationship between proportions vs. 3D...I would like someone to make me feel like they're passing a stone and just tell me that the representation of a diamond, based on 3 D modelling, is so good, with respect to the real thing, that you can just trust that that will tell you real well how it will perform. But, Wink, you can't just say it. I need to see that you've passed a stone (or give me some kind of dynamic by which to understand this with you.)

C)

Short of (A) or (B)...I am mindful that the language used in particular by AGS above...

"Most importantly, it uses technology to analyze the cut’s impact on diamond’s overall performance..."

...is at least a bit like the safe and smart guy who can always understand the truth, even when visiting an island where the natives are always and only of two types...those who always lie and those who always tell the truth. (Extra points if you can repeat the question you have to ask so you will always have the truth).

AGS certainly analyzes the cut's impact on the diamond's overall performance. How descriptively it analyzes the diamond's performance directly remains something of a question. Most of the experts here have only reinforced the question, from what I can see.

Further, I see that Garry and the cut group are more reserved with respect to their pronouncement making, and are not so ready to pass a stone.

Although it would appear I'm trying to be difficult...it's only partly so. We mostly agree AGS is better. How much is really sort of the question.

For someone deciding between options from AGS and other labs, the question is reasonably pertinent.