AGSL Scintillation Studies

[Paul-Antwerp]

Date: 6/16/2007 4:57:21 PM
Author: whatmeworry
Here''s a question sparked by another thread. It''s been opined on this board that round brilliants are more sparkly than say princess cuts or emerald cuts. Has the AGSL scintillation research shown any insights into this?

No immediate response coming out of the current AGS-research.

However, I do have some remarks on your question.

When you are saying ''more sparkly'', you probably refer to ''more scintillation''. In this, there are however two aspects: the number of the observable scintillation events, and the size of the observable scintillation events.

Now, what is more sparkly:

a. a high number of small scintillation events
b. a high number of large scintillation events
c. a high number of small, medium-size and large scintillation events

The list of possibilities is way longer with small numbers and so on. The question is not only what is ''more sparkly'', but also what do you prefer?

Intuitively, I would say that option c is best, but that is strictly personal. Very well-cut rounds will have a lot of such scintillation events, in all possible sizes, while most princess-cuts tend to have mostly smaller size scintillation events. However, it is entirely possible to cut princess-cuts in such a way, that you get scintillation events of all sizes. Then, their light return starts to resemble that of a round, and it could well be possible to surpass the sparkle of a round, I think.

I hope that this was somewhat clear.

Live long,

 

[JohnQuixote]

Date: 6/18/2007 5:21:08 AM
Author: Paul-Antwerp

Very well-cut rounds will have a lot of such scintillation events, in all possible sizes, while most princess-cuts tend to have mostly smaller size scintillation events. However, it is entirely possible to cut princess-cuts in such a way, that you get scintillation events of all sizes. Then, their light return starts to resemble that of a round, and it could well be possible to surpass the sparkle of a round, I think.

Also, we can also keep in mind that the surface area of each facet gets larger as the diamond gets larger. The ''splintery'' qualities one sees in a small stone become less splintery when the same configuration is cut on a larger size.

 

[Garry H (Cut Nut)]

When I was in Antwerp a week or so ago I had a short chat with Sergey.

Only those facets with reflections often crossing the arbitrary boundary between pre selected zones will contribute to scintillation in the AGSL model.

In particular, a diamond with reflections along zones’ boundaries will have maximum scintillation, AGS might want to consider this.

 

[Garry H (Cut Nut)]

Date: 6/16/2007 4:57:21 PM
Author: whatmeworry
Here''s a question sparked by another thread. It''s been opined on this board that round brilliants are more sparkly than say princess cuts or emerald cuts. Has the AGSL scintillation research shown any insights into this?

Round brilliants - especially if they are not too steep deep - gather more light from steeper directly above directions and have very short average number of reflections inside the stone (in, bounce bounce and out) so they have more very bright intense flashes or sparkles. This intensity level could be an issue with much of the AGS work (for fire too).

A princess has more smaller less intense sparkles because it has more ''virtual facets'' and more bouncing all about ray paths. A princess also gathers much more of the face up light from much lower angles where lighting tends not to be as common or as bright.
Hope that helps

 

[strmrdr]

Date: 7/8/2007 8:01:22 AM
Author: Garry H (Cut Nut)
When I was in Antwerp a week or so ago I had a short chat with Sergey.

Only those facets with reflections often crossing the arbitrary boundary between pre selected zones will contribute to scintillation in the AGSL model.

In particular, a diamond with reflections along zones’ boundaries will have maximum scintillation, AGS might want to consider this.

thats a good point and opens a huge can of worms dont it.

 

[surfgirl]

Well, I don''t no nuth''in ''bout no scientific diamond evidence but...I did enjoy Dave''s analogy to Cindy Crawford and I agree with him that this may be a very lucrative marketing strategy. I know after seeing all those pretty images on the OP, I''d like to pay someone to take similar photos of my stone just because it would be cool to see what they looked like. Having said that, to me, the numbers game is irrelevant, the only thing I want to know is what does the diamond look like in real life, with my own eyes, in a variety of lighting conditions. I dont really get the obsession with all the scientific mumbo jumbo because at the end of the day, the diamond isn''t appreciated by passing around it''s cert numbers or the nice pretty photos of it''s cert process. The only way to really appreciate and love a stone is by
l

king at it with one''s own eyes. So while this all is nice and groovy, it wouldn''t make me want to buy a stone that had a new cert/grading element to it. Ultimately, I use my eyes when deciding what''s exceptional vs. what''s good vs. what''s below par.

But thanks for the cool photos and thanks also for including an old cut in that comparison! I''ve always wondered what my stone might look like through an IS or an ASET and now I have a better idea...thanks for that!

 

[tanuki]

If requests are allowed, I would like to request a comparison between proportionate facet sizes, say a 7.8 mm round in 57 facet tolkowsky cut vs a 9 mm round with 66 facets such as the Leo in your example. Just to see how the additional facets change the picture of scintillation as the stone gets proportionately larger and the number of facets increases at about the same rate. I'm just curious as a Solasfera owner.

 

[Garry H (Cut Nut)]

Date: 7/8/2007 1:53:41 PM
Author: surfgirl
Well, I don''t no nuth''in ''bout no scientific diamond evidence but...I did enjoy Dave''s analogy to Cindy Crawford and I agree with him that this may be a very lucrative marketing strategy. I know after seeing all those pretty images on the OP, I''d like to pay someone to take similar photos of my stone just because it would be cool to see what they looked like. Having said that, to me, the numbers game is irrelevant, the only thing I want to know is what does the diamond look like in real life, with my own eyes, in a variety of lighting conditions. I dont really get the obsession with all the scientific mumbo jumbo because at the end of the day, the diamond isn''t appreciated by passing around it''s cert numbers or the nice pretty photos of it''s cert process. The only way to really appreciate and love a stone is by
l

king at it with one''s own eyes. So while this all is nice and groovy, it wouldn''t make me want to buy a stone that had a new cert/grading element to it. Ultimately, I use my eyes when deciding what''s exceptional vs. what''s good vs. what''s below par.

But thanks for the cool photos and thanks also for including an old cut in that comparison! I''ve always wondered what my stone might look like through an IS or an ASET and now I have a better idea...thanks for that!

Sure this nerdy stuff is not for everyone, but what it does result in is better ways to ensure that you have a better chance of walking into a store and finding nice sparkly diamonds for the places you want to wear them in.
That is what good research, coupled with good development does.

In some cases AGSL research has helped cutters, althugh their main goal is to assess the quality of already faceted diamonds - it has helped raise the efficiency / yeild / quality of currently cut diamonds.

The real future of vastly different cuts still awaits

 

[whatmeworry]

Date: 7/8/2007 8:09:05 AM
Author: Garry H (Cut Nut)

Round brilliants - especially if they are not too steep deep - gather more light from steeper directly above directions and have very short average number of reflections inside the stone (in, bounce bounce and out) so they have more very bright intense flashes or sparkles. This intensity level could be an issue with much of the AGS work (for fire too).

A princess has more smaller less intense sparkles because it has more ''virtual facets'' and more bouncing all about ray paths. A princess also gathers much more of the face up light from much lower angles where lighting tends not to be as common or as bright.
Hope that helps

Garry, thanks this makes sense to me. Lots of folks have written about places where their diamonds are the sparkliest, grocery stores, home depot, costco, stores with strong overhead lighting. Yet this is a very specific lighting model and I guess I still can''t see the application of of lighting independent metric when in the real world the are distinct lighting setups that make diamonds spakle.

 

[Garry H (Cut Nut)]

Date: 7/9/2007 1:31:27 AM
Author: whatmeworry

Garry, thanks this makes sense to me. Lots of folks have written about places where their diamonds are the sparkliest, grocery stores, home depot, costco, stores with strong overhead lighting. Yet this is a very specific lighting model and I guess I still can''t see the application of of lighting independent metric when in the real world the are distinct lighting setups that make diamonds spakle.

WMW some of the work that I am involved in has a primary goal to work for cutters because they are the market for grading (i.e. they pay the labs).

If we can create semiautomatic tools for the development of new cuts and their appraisal using Basic Light Return (BLR) and ETAS technology then it is possible to search for the best yielding combinations of parameters for existing cuts.

We can then use this data in rough diamonds software for designing the best marking technologies for princess and other existing fancy shapes.

And we can help cutters to use deviations from ideal symmetry to gain the maximum yield, and therefore lower priced diamonds, but still achieve Ideal Cut grades for any grading system.

Then we can progress to the designing of many different types of new cuts.


But! Firstly of course we need to know that laboratories will grade new good cuts as good, and new bad cuts as bad
If labs can accomplish that with or without the help of the Cut Group (I am one of the participants), then we will be satisfied. And AGSL is so far doing the best job.

There can be a big difference between Basic Light Return (BLR) for cutters and a Cut Grade for customers.
A Cut Grade must appraise each real stones base optical characteristic’s that arise because of deviations from ideal symmetry and variations of parameters.
But Cutters need precise technological instructions on producing a stone of maximum size and, simultaneously, raise the probability of the highest Cut Grade (taking into consideration the accuracy of the process control cutting quality of cutting at the factory).

Date: 7/9/2007 1:31:27 AM
Author: whatmeworry

I still can''t see the application of of lighting independent metric when in the real world the are distinct lighting setups that make diamonds spakle.

A good diamond should work in a variety of different lightings. It is possible to design diamonds to work in only certain types of light too. There can be many new options, but first we need to know more. It seems strange, but defining sparkle and quantifying it is extremely difficult.

 

[strmrdr]

Date: 7/9/2007 1:55:22 AM
Author: Garry H (Cut Nut)


And we can help cutters to use deviations from ideal symmetry to gain the maximum yield, and therefore lower priced diamonds, but still achieve Ideal Cut grades for any grading system.

Teaching cutters to game the cut grades is a good thing?
BANG!

 

[Garry H (Cut Nut)]

Date: 7/9/2007 2:12:46 AM
Author: strmrdr

Date: 7/9/2007 1:55:22 AM
Author: Garry H (Cut Nut)



And we can help cutters to use deviations from ideal symmetry to gain the maximum yield, and therefore lower priced diamonds, but still achieve Ideal Cut grades for any grading system.

Teaching cutters to game the cut grades is a good thing?
BANG!

Yes it is a good thing if there are ways to adjust symmetry to acheive cut grades. We believe this is viable Storm. There are situations where perfect symmetry is not optimal for appearance. If we can predict this in advance of cutting, or we know where the limit is at a manufacturing level to achieve an optimal cost benefit, then it is a thing worth doing.

We are still probably a long way off optimal cut grading though.

 

[diagem]

Date: 7/9/2007 1:55:22 AM
Author: Garry H (Cut Nut)

But! Firstly of course we need to know that laboratories will grade new good cuts as good, and new bad cuts as bad


Garry, and base these grades on what parameters?

If these are new ''novelty'' type Diamond cuts? how will the Labs be able to judge a good ''new'' cut vs. a bad new cut?

 

[Garry H (Cut Nut)]

Date: 7/9/2007 4:48:29 AM
Author: DiaGem

Date: 7/9/2007 1:55:22 AM
Author: Garry H (Cut Nut)

But! Firstly of course we need to know that laboratories will grade new good cuts as good, and new bad cuts as bad



Garry, and base these grades on what parameters?

If these are new ''novelty'' type Diamond cuts? how will the Labs be able to judge a good ''new'' cut vs. a bad new cut?

AGS is getting closer to being able to grade any diamond. GIA say they are trying again for fancy cuts. I believe HRD have a plan. GCAL seem to think they know what they are doing.
If any labs want our help then they are welcome to make a proposal.
Second tier labs seem happy to sit and watch, or partner with Gemex etc

 

[diagem]

Date: 7/9/2007 5:40:28 AM
Author: Garry H (Cut Nut)

Date: 7/9/2007 4:48:29 AM
Author: DiaGem

Date: 7/9/2007 1:55:22 AM
Author: Garry H (Cut Nut)

But! Firstly of course we need to know that laboratories will grade new good cuts as good, and new bad cuts as bad




Garry, and base these grades on what parameters?

If these are new ''novelty'' type Diamond cuts? how will the Labs be able to judge a good ''new'' cut vs. a bad new cut?

AGS is getting closer to being able to grade any diamond. GIA say they are trying again for fancy cuts. I believe HRD have a plan. GCAL seem to think they know what they are doing.
If any labs want our help then they are welcome to make a proposal.
Second tier labs seem happy to sit and watch, or partner with Gemex etc

That is my question...
On what basis are these Labs ''claiming'' to be able to grade any shaped Diamond...
For example a Briolette type Diamond that its beauty is not based on BLR...

 

[Garry H (Cut Nut)]

Date: 7/9/2007 5:51:52 AM
Author: DiaGem

Date: 7/9/2007 5:40:28 AM
Author: Garry H (Cut Nut)

Date: 7/9/2007 4:48:29 AM
Author: DiaGem

Date: 7/9/2007 1:55:22 AM
Author: Garry H (Cut Nut)

But! Firstly of course we need to know that laboratories will grade new good cuts as good, and new bad cuts as bad





Garry, and base these grades on what parameters?

If these are new ''novelty'' type Diamond cuts? how will the Labs be able to judge a good ''new'' cut vs. a bad new cut?

AGS is getting closer to being able to grade any diamond. GIA say they are trying again for fancy cuts. I believe HRD have a plan. GCAL seem to think they know what they are doing.
If any labs want our help then they are welcome to make a proposal.
Second tier labs seem happy to sit and watch, or partner with Gemex etc

That is my question...
On what basis are these Labs ''claiming'' to be able to grade any shaped Diamond...
For example a Briolette type Diamond that its beauty is not based on BLR...

A briolette has its own set of BLR''s. For example it can have very good fire as well as very poor light return. But it is unlikely that anyone would want to submit a briolette for cut quality grading.

Imagine all diamonds would be considered by the same rules. That means a well cut princess might score 90% of a standard based on Tolkowsky as a yard stick.
A well cut round can get 100, an asscher maybe 80 and overall the briolette might score 50 (simplistic example: 20% for brilliance, 30% for scintillation and 100% for fire = 150/3 = 50%).

We believe labs will come to realise that all diamonds should be graded by the same scale. Ultimately this will lead to diamonds with better than 100 grades.

 

[diagem]

Date: 7/9/2007 8:12:27 AM
Author: Garry H (Cut Nut)

Date: 7/9/2007 5:51:52 AM
Author: DiaGem
That is my question...
On what basis are these Labs ''claiming'' to be able to grade any shaped Diamond...
For example a Briolette type Diamond that its beauty is not based on BLR...

A briolette has its own set of BLR''s. For example it can have very good fire as well as very poor light return. But it is unlikely that anyone would want to submit a briolette for cut quality grading.
Why.., A briolette would be extremely important in regards to cut quality..., unless you move symmetry away from cut quality grading.
Imagine all diamonds would be considered by the same rules. That means a well cut princess might score 90% of a standard based on Tolkowsky as a yard stick.
A well cut round can get 100, an asscher maybe 80 and overall the briolette might score 50 (simplistic example: 20% for brilliance, 30% for scintillation and 100% for fire = 150/3 = 50%).
Now another example..., a portrait cut (flat step-cut table-cut), which i assume will have a significant effect on its symmetry and less on its BLR.

We believe labs will come to realise that all diamonds should be graded by the same scale. Ultimately this will lead to diamonds with better than 100 grades.

What I am trying to point out is: as newer (innovative) cuts are being designed (and I am one who pushes that direction)..., and cuts that are moving away from the standard (step/brilliant) faceting arrangements that we are all used to..., that parameters to determine which is a ''good'' new cut vs. ''bad'' new cut will become completely alienated to any Laboratory..., am i making sense?

So how will the Labs be able to grade these type of new cuts on one scale as standard cuts? It seems to me these new cuts will have to be graded on an individual stone per stone basis.

 

[Garry H (Cut Nut)]

Date: 7/9/2007 8:43:54 AM
Author: DiaGem
What I am trying to point out is: as newer (innovative) cuts are being designed (and I am one who pushes that direction)..., and cuts that are moving away from the standard (step/brilliant) faceting arrangements that we are all used to..., that parameters to determine which is a 'good' new cut vs. 'bad' new cut will become completely alienated to any Laboratory..., am i making sense?

So how will the Labs be able to grade these type of new cuts on one scale as standard cuts? It seems to me these new cuts will have to be graded on an individual stone per stone basis.

If we face curved or convex facets then there could be some difficulties for some 3D model systems and an advantage to direct mesaurement DG.

But for regular flat faceted diamonds of any arrangement a system that can garade any stone is infact easier to develop than one where AGS (for example) gives its top grade to an AGS 0 oval, a AGS 0 round and a AGS 0 princess, when clearly they are not all the same.

But there is nothing to stop AGS (or Gemex) from applying their round stone grades to fancy shapes, and then there is a real challenge for creative cutters to optimize other existing cuts / shapes, just as we did over a couple of hundred years with the round, and we have been with the princess for 40 years.

 

[JohnQuixote]

Date: 7/9/2007 1:31:27 AM
Author: whatmeworry

Garry, thanks this makes sense to me. Lots of folks have written about places where their diamonds are the sparkliest, grocery stores, home depot, costco, stores with strong overhead lighting. Yet this is a very specific lighting model and I guess I still can't see the application of of lighting independent metric when in the real world the are distinct lighting setups that make diamonds spakle.

Choosing only one, or two, or ten specific metrics may be fine for one person but it won't apply to all scenarios. We can design a 'costco' metric but the people who live where there are no Costcos will not be happy.

I think a combination of both types of analysis is where we're headed, similar to how players are selected for the NFL draft: Each has stats from their college career games to show how they are predicted to do (environmentally specific measure). There is also an annual scouting combine where they lift weights, run, pass & catch outside of a specific game scenario (independent measure). For the most analytic - coaches & investors - seeing how the players do at the combine is critical to determining performance. Most NFL fans pay no attention to the combine, they just want the player to perform in the games.

 

[whatmeworry]

Date: 7/9/2007 11:19:18 AM
Author: JohnQuixote

Choosing only one, or two, or ten specific metrics may be fine for one person but it won''t apply to all scenarios. We can design a ''costco'' metric but the people who live where there are no Costcos will not be happy.

I think a combination of both types of analysis is where we''re headed, similar to how players are selected for the NFL draft: Each has stats from their college career games to show how they are predicted to do (environmentally specific measure). There is also an annual scouting combine where they lift weights, run, pass & catch outside of a specific game scenario (independent measure). For the most analytic - coaches & investors - seeing how the players do at the combine is critical to determining performance. Most NFL fans pay no attention to the combine, they just want the player to perform in the games.

Yes but the NFL combine is always held at the RCA Dome in Indianapolis, a fixed standard environment. And scouts will discount 40 yard dash times that are run at places other than the combine. I suggest that scouts also could take 40 yard dash times in 4 inches of snow in the middle of winter or in a muddy field to see how they perform because those environments are possible too but yet the combine environment seems to be the standard on which they are graded.

For those who are lost on the football analogy, I''m just wondering alound if adding taking all possible lighting environments (and weighting them equally) is a better metric than using more likely lighting environments.

Two more months till football season !

 

[JohnQuixote]

Date: 7/9/2007 12:46:22 PM
Author: whatmeworry
Yes but the NFL combine is always held at the RCA Dome in Indianapolis, a fixed standard environment. And scouts will discount 40 yard dash times that are run at places other than the combine. I suggest that scouts also could take 40 yard dash times in 4 inches of snow in the middle of winter or in a muddy field to see how they perform because those environments are possible too but yet the combine environment seems to be the standard on which they are graded.

For those who are lost on the football analogy, I''m just wondering alound if adding taking all possible lighting environments (and weighting them equally) is a better metric than using more likely lighting environments.

Two more months till football season !

It''s cool that you''re familiar with the combine.

Sorry I wasn''t more clear. I consider in-game context as ''performance environment'' for the players (meaning other players on the field, specific down/ytg, etc). Running/lifting done at the combine to measure speed/strength are independent in that sense.

For the record, I think defensive linemen have charm even though they have chunkier performance. So many people like the flashiness of the QB (sigh).

 

[JohnQuixote]

Here is a comparison from my ''other'' world: Each year in Texas appx. 40,000 high school musicians begin the try-out process for All-State Band & Orchestra. There are 2 local rounds, an area round and final selection at state. The tryout package consists of 3 solo pieces (compositions) as well as fundamental exercises (scales or rudiments).

The scales/rudiments are more heavily weighted in the first rounds since they are ''independent'' measures of technical skill (no prescribed tempos, dynamics, articulations or stylistic interp). After the area round those advancing to state are issued a new tryout packet containing only music actually being used in the concert that year: By this time candidates have been whittled down to only those with top technical skill (from the composition-independent measures). Final placement is determined by how these top technicians interpret & master the actual pieces to be played (composition-dependent).

This is why I think there''s harmony

in light source independent and dependent evaluation: The best diamonds will perform well in any environment (light source independent), but the end-user will have specific tastes, whether that''s the ''look'' of the diamond, favorite lighting or both.

 

[surfgirl]

Hey Gary, just to clarify, I wasn''t dissing the research at all. Truthbe told, I''d love someone to give me all those cool photos of my own stone! But I''m not an online buyer because I need to see stones in real life (and as many as possible for making a decision) so online buying isn''t for me. I think the "numbers game" is more geared towards online buying simply because you cannot see an unending number of stones in person before buying. That said, I was wondering about your comment:

"...what it does result in is better ways to ensure that you have a better chance of walking into a store and finding nice sparkly diamonds for the places you want to wear them in."

Are you saying that different cuts will look better - or not - depending on the environments we''re planning on wearing them in? Like different lighting, for example? I ask because I wasn''t sure what you meant and I was thinking that I''d never buy a diamond solely for use in one type of place. Can you clarify? Thanks!

And John, in the future, I know there are plenty of us here that would love for you to continue to include the old cuts in your comparisons. That was nice to see!

 

[JohnQuixote]

Date: 7/9/2007 1:32:46 PM
Author: surfgirl
Hey Gary, just to clarify, I wasn't dissing the research at all. Truthbe told, I'd love someone to give me all those cool photos of my own stone! But I'm not an online buyer because I need to see stones in real life (and as many as possible for making a decision) so online buying isn't for me. I think the 'numbers game' is more geared towards online buying simply because you cannot see an unending number of stones in person before buying. That said, I was wondering about your comment:

'...what it does result in is better ways to ensure that you have a better chance of walking into a store and finding nice sparkly diamonds for the places you want to wear them in.'

Are you saying that different cuts will look better - or not - depending on the environments we're planning on wearing them in? Like different lighting, for example? I ask because I wasn't sure what you meant and I was thinking that I'd never buy a diamond solely for use in one type of place. Can you clarify? Thanks!

And John, in the future, I know there are plenty of us here that would love for you to continue to include the old cuts in your comparisons. That was nice to see!

Thanks SG. It's nice to see consumers commenting in this thread. I know it's waaay dry.

(clink)

I do want to make it clear that all these graphics come from AGSL by way of their researchers. We are not a part of their developments, we're just sharing it with you fine folks. All of it is very proprietary to their lab, so we can't do comparisons for you (sorry Tanuki). Peter Yantzer was generous enough to allow some of us into his home and provide a far-ahead look at what's brewing. Many places would keep it all under wraps, but AGSL has a history of being completely transparent so we've been permitted this look.

 

[whatmeworry]

John,
Let me jump on the AGSL transparency in case they are reading this, GIA has posted papers on their cut studies on their website. I can''t find anything comparable on the AGSL cut grading on the web exept for what excerpts gets posted here (Thanks BTW). One thing that has not been very clear to me is how AGSL drew the boundary between AGS0 and AGS1 for light performance. I know (from the GIA papers) that GIA used computer modeling and human observation to differentiate between Excellent and Very Good. Do you know if this is the case with AGSL? Certainly the computer modeling is there, but is there human observation as well?

 

[strmrdr]

Date: 7/9/2007 2:21:58 PM
Author: whatmeworry
John,
Let me jump on the AGSL transparency in case they are reading this, GIA has posted papers on their cut studies on their website. I can''t find anything comparable on the AGSL cut grading on the web exept for what excerpts gets posted here (Thanks BTW). One thing that has not been very clear to me is how AGSL drew the boundary between AGS0 and AGS1 for light performance. I know (from the GIA papers) that GIA used computer modeling and human observation to differentiate between Excellent and Very Good. Do you know if this is the case with AGSL? Certainly the computer modeling is there, but is there human observation as well?

AGS releases at a much higher level than GIA does.
You wont find much of this on AGS''s website because very few would understand it.
Peter thru the PS trade members has answered questions I have posted on here about their system we haven''t got anywhere close to that kind of attention from GIA.
It is very clear that AGS atleast listens and even some times acts on the points raised here.

 

[Serg]

Date: 7/9/2007 4:30:47 PM
Author: strmrdr

Date: 7/9/2007 2:21:58 PM
Author: whatmeworry
John,
Let me jump on the AGSL transparency in case they are reading this, GIA has posted papers on their cut studies on their website. I can''t find anything comparable on the AGSL cut grading on the web exept for what excerpts gets posted here (Thanks BTW). One thing that has not been very clear to me is how AGSL drew the boundary between AGS0 and AGS1 for light performance. I know (from the GIA papers) that GIA used computer modeling and human observation to differentiate between Excellent and Very Good. Do you know if this is the case with AGSL? Certainly the computer modeling is there, but is there human observation as well?

AGS releases at a much higher level than GIA does.
You wont find much of this on AGS''s website because very few would understand it.
Peter thru the PS trade members has answered questions I have posted on here about their system we haven''t got anywhere close to that kind of attention from GIA.
It is very clear that AGS atleast listens and even some times acts on the points raised here.

"..
From private exchange with Sergey.

Sergey sees the main problem of the AGSL approach to the scintillation grading is the use of ASET structured illumination.

If one takes reflections of the virtual facets on a sphere (ETAS), then it becomes apparent that facets moved within ASET zones won’t produce scintillations. Only those facets, which reflections often crossing boundary between red and green zones, contribute to scintillation in AGSL model.

In particular, a diamond with reflections along zones’ boundaries will have maximum scintillation, which is nonsense.

It is important for the facets reflections to move with maximum angular pace and direction is absolutely irrelevant. AGSL, however, doesn’t count movement in horizontal direction, while in vertical direction AGS counts only reflection crossing the abstract ASET zone borders. There is also no account of the reflection length… - Generally speaking: clear underdevelopment.

Hypothetical maximum of scintillation in this model could be reached when all the reflections are located along the zone boundary and jump back ad forth. If one can manage to cut a diamond in such manner, it would have no scintillation because probability of the light source position along the zone border is minimal while in other areas such diamond won’t reflect at all.

In reality, scintillation is high if facets’ reflections are distributed equally and move actively.

.."

 

[strmrdr]

Serg,
Yes the AGS scint research is very beta, that gives people time to help them adjust it as needed.
For dynamic contrast I think it will follow the ags ASET zones as it moves in RB''s but not willing to say it totaly does. The arrows[pavilion mains] flashing on and off is a large part.
But for dynamic fire its a different story.
That''s one reason I sceptical that scint will be nailed down, you have 2 very distinct components that at the same time interface with each other and act independent of each other.

 

[JohnQuixote]

Date: 7/9/2007 2:21:58 PM
Author: whatmeworry
John,
Let me jump on the AGSL transparency in case they are reading this, GIA has posted papers on their cut studies on their website. I can't find anything comparable on the AGSL cut grading on the web exept for what excerpts gets posted here (Thanks BTW). One thing that has not been very clear to me is how AGSL drew the boundary between AGS0 and AGS1 for light performance. I know (from the GIA papers) that GIA used computer modeling and human observation to differentiate between Excellent and Very Good. Do you know if this is the case with AGSL? Certainly the computer modeling is there, but is there human observation as well?

The AGS Diamond Grading Standards committee was established in 1955, originally a colorimetry committee, and developed the 0-10 scale based on research and human observation. The grading standards manual underwent many revisions over the next 30 years based on observation. When AGSL began grading the round brilliant in 1996 their system was based on a range of proportions humans recognized as producing good performance (like the GIA scale introduced in 2006). That fixed system had questionable sets at its 'outer limits' (which will arguably be a problem in any fixed system).

Since the introduction of the 1996 grading system tools and technology have improved. Today innovations like DiamCalc (Octonus) and the ability to predict diamond performance via scan/ray-tracing (AGSL) is not only possible, but becoming increasingly precise. They are already good enough to place a diamond within a grade-range and getting more accurate each year.

Also (forgive me if I'm being repetitive) please remember this thread has nothing to do with current AGSL grading. It's drawing-board info about scint, for which there is currently no agreed-upon metric. The AGSL studies and Sergey's ETAS are by far the most advanced concepts but it's not soup yet.

 

[JohnQuixote]

Date: 7/9/2007 4:30:47 PM
Author: strmrdr

AGS releases at a much higher level than GIA does.
You wont find much of this on AGS''s website because very few would understand it.
Peter thru the PS trade members has answered questions I have posted on here about their system we haven''t got anywhere close to that kind of attention from GIA.
It is very clear that AGS atleast listens and even some times acts on the points raised here.

For the record I find the researchers and teachers at GIA very communicative and always willing to answer questions when they''re contacted. I don''t believe they''re at liberty to disclose ongoing projects to the degree that AGSL can (which is to fling the doors open, tell you to grab some iced-tea and pull up a microscope) and Strm''s right - AGSL makes no secret that the discussions on this forum are a point of interest for them.

This has its risks, as it opens the doors for pre-judgment and criticism of beta concepts. In fact I used to worry when we''d share pre-info here, thinking the criticism might stop Peter from wanting to share it any more - but those guys embrace it. I suspect Peter has either read the cluetrain manifesto or it''s in his genes.