AGS new cut grade system early 2005

[Garry H (Cut Nut)]

No one has attempted to claim the prize yet? It is still not Tuesday in USA!

This is over $300 of value. here it is again:

*PLease think about a 90 degree prism( or NailHead diamond), hemisphere light with head.
Could anybody publesh here before Tuesday Why WLR will give incorrect result for such condition?
The prize for best correct answer is Diamcalc and IDCC proceedings.



Sergey Sivovolenko
CEO OctoNus

 

[strmrdr]

----------------
On 10/25/2004 5:23:15 AM Serg wrote:

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




The brain 'sees' number of flashes bigger or equal to what every eye sees individually in the conditions of dark room (and correspondingly dark diamond).

The maximum number of flashes visible by brain for static picture is the sum of flashes seeing by every eye individually.

Bingo I agree 100% but thats not what the quote I posted is saying.
"but the average number will be two times as large as that in the case of single-eye viewing." is not true.

I read your stuff as scientific works
maybe I am expecting too much.
If you take a shortcut or overly simplify please say so.
Something along the the lines of a very simplified
explaination is .....
Also keep in mind that the more you dont clarify stuff the more fights you open yourself up too and they harder it is to get it accepted.

 

[Garry H (Cut Nut)]

No one has attempted to claim the prize yet? It is still not Tuesday in USA!

This is over $300 of value. here it is again:

*PLease think about a 90 degree prism( or NailHead diamond), hemisphere light with head.
Could anybody publesh here before Tuesday Why WLR will give incorrect result for such condition?
The prize for best correct answer is Diamcalc and IDCC proceedings.
See links to both on this page.


Sergey Sivovolenko
CEO OctoNus

 

[strmrdr]

----------------
On 10/25/2004 5:01:09 PM Garry H (Cut Nut) wrote:

No one has attempted to claim the prize yet? It is still not Tuesday in USA!


This is over $300 of value. here it is again:


*PLease think about a 90 degree prism( or NailHead diamond), hemisphere light with head.

Could anybody publesh here before Tuesday Why WLR will give incorrect result for such condition?

The prize for best correct answer is Diamcalc and IDCC proceedings.

See links to both on this page.


Because the light would not be returned to the top
but would be returned to the side.
Therefore you would never see the light return from the top if you tried to measure it the way you would with a non-nailhead RB. diamond.
All you would see would be surface reflections.

 

[Garry H (Cut Nut)]

.

 

[strmrdr]

hmmm looks like we are talking about 2 different things,,,
This is a 90 degree prism adapter...
http://www.scopestuff.com/ss_eda6.htm

 

[strmrdr]

What your showing im finding as a 90 degree optical square.
http://www.davidsonoptronics.com/catD-613.htm

 

[Rank Amateur]

I may be way off base here, but...

The prism reflects all light back to the hempisphere. The more light the observer blocks (directly overhead) , the lower the WLR. In non overhead positions, the viewer will see little light return. Much of the light is sent away as surface reflections, most of which don't intersect the viewer's eye.

 

[adamasgem]

----------------
On 10/25/2004 4:15:28 PM Garry H (Cut Nut) wrote:

No one has attempted to claim the prize yet? It is still not Tuesday in USA!

This is over $300 of value. here it is again:

*PLease think about a 90 degree prism( or NailHead diamond), hemisphere light with head.
Could anybody publesh here before Tuesday Why WLR will give incorrect result for such condition?
The prize for best correct answer is Diamcalc and IDCC proceedings.



Sergey Sivovolenko
CEO OctoNus
----------------

I already did, I believe, likening it to a retro-reflector, since there is no light returned to the observer because there is no light starting there

 

[Serg]

*Please think about 90 degree prism( or NailHead diamond), hemisphere light with head.
Could anybody publish here before Tuesday Why WLR will give incorrect result for such condition?
I found 3 attempts of answer:

Marty: "Sounds like you are talking about a retro reflector which, with a head, will give a good WLR but not return anything to the observer."

Strmrdr:"Because the light would not be returned to the top
but would be returned to the side.
Therefore you would never see the light return from the top if you tried to measure it the way you would with a non-nailhead RB. diamond.
All you would see would be surface reflections."

Rank Amateur:" The prism reflects all light back to the hempisphere. The more light the observer blocks (directly overhead) , the lower the WLR. In non overhead positions, the viewer will see little light return. Much of the light is sent away as surface reflections, most of which don't intersect the viewer's eye. "

Answer of Rank Amateur is not correct.( or I did not understand this answer)
Answer of Strmrdr is not answer to my question.

Answer of Marty is correct. But even Marty did not explain why WLR is not correct for "retro reflector". :

LR is zero for "retro reflector". It is minimum for LR. You should show at least that WLR for "retro reflector" is more than minimum of WLR. It is enough, easy but not wittily.

If anybody send smart clear and short explanation during 2 weeks :Why WLR for "retro reflector" is bigger than WLR for some good diamond. I will send second copy of prize.

Marty won first round.
Please send to me you post-address for IDCC proceedings.
Also you can send order for Diamcalc registration. See instruction
for Diamcalc registration. See instruction http://www.octonus.ru/oct/products/3dcalc/standard/howtoreg.phtml

 

[adamasgem]

Got a chance to run the 58 stones that GIA had defined WLR's for with their newer 23 degree head obscuration model hemisphere, just to see what difference it would make using the cosine squared weighting function.
ABH data (10000 rays),

There are 4 items on this plot, all normalized to the value(s) obtained for the Tolkowsky model, D65, 400 to 700nm

1) GIA's published WLR values (RED)(Full hemisphere)
2) ABH, my full hemisphere results (GREEN)
3) & 4) A hemisphere with the central 23 degree cone giving no light becuase of a "head", with and without the "glare" effect . Superposition holds so I add the weighted glare and renormalize.

One notes that the peak to peak differentiation between stones is decreased with the restricted illumination model, but am I surprised?, not really..

I'll try to run the restricted 1.5 degree FOV, but in the forward ray tracing technique, only less than one in a thousand rays make any contribution at all to the numerical metric (binary WLR, you either see it or you don't)

 

[Serg]

Marty,

I received your post-adress.

Could you add new stone to your plots? Please add "retro reflector".

 

[Serg]

----------------
On 10/26/2004 10:32:22 AM Serg wrote:

Marty,

I received your post-adress.

Could you add new stone to your plots? Please add 'retro reflector'. ----------------

Please use 'retro reflector' like pyramid( without parallel sides)

 

[Serg]

Marty,

I am sorry.

I can not publish my verification policy( system) yet.

I wrote, read it and I understood that proposals are bad for real world. These principles will not work in current diamond community.
I need to think more to propose something will be really used by more than 2-3 enthusiasts and at least 1-2 main laboratories will join.

 

[adamasgem]

----------------
On 10/26/2004 10:37:40 AM Serg wrote:

----------------
On 10/26/2004 10:32:22 AM Serg wrote:

Marty,

I received your post-adress.

Could you add new stone to your plots? Please add 'retro reflector'. ----------------

Please use 'retro reflector' like pyramid( without parallel sides)
----------------

Sergey I'll have to look at how I would have to change my stone definition software to do that, and how to handle the "corner cube" issue within the framework of my software, Table and four pavilion mains is what needs to be done I believe.

 

[adamasgem]

----------------
On 10/26/2004 12:38:10 PM Serg wrote:

Marty,

I am sorry.

I can not publish my verification policy( system) yet.

I wrote, read it and I understood that proposals are bad for real world. These principles will not work in current diamond community.
I need to think more to propose something will be really used by more than 2-3 enthusiasts and at least 1-2 main laboratories will join.----------------

I be happy to review it for you, under non-disclosure of course. Screw the rest of the world, be open about it. I understand your problem, and will help you in anyway I can with it.

On the "new" GIA metrics article, it was typical that devoted two pages in G&G to descibing the stones (table, crown, etc) they used, BUT THEY DID NOT PUBLISH the individual metrics (brightness, DCLR, etc) for them, and only broadly categorized the composite results on a subset of the stones.

 

[adamasgem]

----------------
On 10/26/2004 12:38:10 PM Serg wrote:

Marty,

I am sorry.

I can not publish my verification policy( system) yet.

I wrote, read it and I understood that proposals are bad for real world. These principles will not work in current diamond community.
I need to think more to propose something will be really used by more than 2-3 enthusiasts and at least 1-2 main laboratories will join.----------------

Sergey.. Can you run the 58 GIA stones on DiamondCalc with the GIA "brightness" model and the restricted FOV (3 degree total angle 1.5 degree cone)with their 46 degree total head model (23 degree cone), and publish the normalized results like I did with the SAS2000 forward Monte Carlo WLR. It would take me eons to do it, as I only can raytrace about 2000 rays a minute using the GIA criteria for # of interactions, and with the very low (1 in a thousand returned rays or less) light return within the restricted FOV I would have to run 10's of millions of rays (instead of only 20000 rays) to get a convergence on the metric. DiamondCalc is the only way to go on that exercise.

 

[Serg]

re:"corner cube" issue within the framework of my software, Table and four pavilion mains is what needs to be done I believe.

Yes. It is right.

What is GIA "brightness" model ?

I see this term firstly.

 

[adamasgem]

----------------
On 10/26/2004 2:21:05 PM Serg wrote:



re:'corner cube' issue within the framework of my software, Table and four pavilion mains is what needs to be done I believe.

Yes. It is right.

I'm going to make it simple by making it a square stone, that way I don't have to model any girdle planes

What is GIA 'brightness' model ?

I see this term firstly.----------------

"Brightness" is GIA's new terminology for light return to a 3 degree total angle restricted cycloptic viewer using a hemisphere with a black hole at the girdle plane, with 46 degree total solid angle head light blockage. "Brightenss includes the effect of "glare".

 

[Garry H (Cut Nut)]

I assume they make so much blackness to compensate for a dealers body Marty?

But why not model the dealer?

BTW it looks like they are going tom make dealers all wear white shirts now

 

[Serg]

I read new GIA article just. I Hope I understand The new GIA Brightness definition correctly.( I think GIA do not use cos*cos for this metric)

See result.
If RD1=1, then RD39=1,15 and RD45=0,964

RD45 is nailhead. !!!

Even unnormal size of Head 46 degree can not help to grade RD45 in GIA metric correctly!
I should check result. The result is very strange.

also I do not know thresholds of 18 level GIA brightness( I use intensity. The threshold of minimum intensity is important for result. )

 

[adamasgem]

----------------
On 10/27/2004 11:03:35 AM Serg wrote:

I read new GIA article just. I Hope I understand The new GIA Brightness definition correctly.( I think GIA do not use cos*cos for this metric)
Your correct there, a binary (0,1) weighting depending on whether ray is in field of view or not

See result.
If RD1=1, then RD39=1,15 and RD45=0,964

RD45 is nailhead. !!!

Even unnormal size of Head 46 degree can not help to grade RD45 in GIA metric correctly!
I should check result. The result is very strange.
Remember Sergey, their general categorizations are a composite of brightness, fire, weight, sintilation, etc

also I do not know thresholds of 18 level GIA brightness( I use intensity. The threshold of minimum intensity is important for result. )----------------

There is a lot we don't know

 

[Serg]

re:There is a lot we don't know

Yes . But RD45 has very big rate in WLR too.( More than modern Tolkowsky diamond)


That is interesting new big GIA HEAD absorb o few light from RD45.
This is reason why rd45 has big rate in new metric too.

 

[adamasgem]

----------------
On 10/27/2004 4:22:17 AM Garry H (Cut Nut) wrote:

I assume they make so much blackness to compensate for a dealers body Marty?

Head model, or hairdo, or else "heads" from down under diamond dealers

But why not model the dealer?

If you are talking about a symmetrical modeling body shadow, you don't need to because of the symmetry, you are blocking light from let's say 30% of the stone in a symmetrical fashion, the absolute value of a metric will change but the relative value shouldn't, providing the "body" solid angle is a multiple of the stones symmetry(take away light from a quarter or an eighth of the stone)

BTW it looks like they are going tom make dealers all wear white shirts now

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

That was a very good attempt by GIA at a controlled experiment (I participated in Tuscon), note I say attempt but it was the best they could do; every "viewer" had a different head size, hair and facial color, and didn't look at the stone from the same perspective and light blockage because of their physical size (height in the chair), besides backgrond lighting issues, all of which contribute in some fashion to the experimental results they obtained. I might add that they might have placed too much "weight" on their correlation results, since I don't see a "modeled" environment in their list (G&G data repository Tables) that reflected the viewing environment I saw.

 

[adamasgem]

----------------
On 10/27/2004 12:33:37 PM Serg wrote:

re:There is a lot we don't know

Yes . But RD45 has very big rate in WLR too.( More than modern Tolkowsky diamond)

That is interesting new big GIA HEAD absorb o few light from RD45.
This is reason why rd45 has big rate in new metric too. ----------------

I'll have to add the rest of the stones to my list (RD30-RD46) when I get time, to see what I come up with.. it will take me awhile have to add it to the todo list, unfortunately the generic girdle desciption might prove to be a problem for modeling. Additionally we don't have any published metrics to bounce our data against.

 

[Serg]

----------------
On 10/27/2004 1:00:40 PM adamasgem wrote:



I'll have to add the rest of the stones to my list (RD30-RD46) when I get time, to see what I come up with.. it will take me awhile have to add it to the todo list, unfortunately the generic girdle desciption might prove to be a problem for modeling. Additionally we don't have any published metrics to bounce our data against.
----------------

I took WLR RD45 from your plot. Is it not correct?

What is mean points 30-58 on your plot?
If your plot is for 29 stones only, what is mean other point? Why does your plot have 58 points?

 

[adamasgem]

----------------
On 10/27/2004 1:16:17 PM Serg wrote:

----------------
On 10/27/2004 1:00:40 PM adamasgem wrote:



I'll have to add the rest of the stones to my list (RD30-RD46) when I get time, to see what I come up with.. it will take me awhile have to add it to the todo list, unfortunately the generic girdle desciption might prove to be a problem for modeling. Additionally we don't have any published metrics to bounce our data against.
----------------

I took WLR RD45 from your plot. Is it not correct?

What is mean points 30-56 on your plot?
----------------

They are the GIA reference diamond and then RD01-RD29 (excluding RD2 I'll put up a list BUT HOLD EVERYTHING!!!!!!

IN WHAT MIGHT BE TYPICAL FOR THE "WORLD's FOREMOST AUTHORITY IN GEMOLOGY(TM)" WHAT GIA HAS NOW PUBLISHED FOR THE DEFINITIONS OF RD01-RD29 HAVE CHANGED FROM WHAT THEY PUBLISHED IN THEIR FIRE ARTICLE IN G&G. So what's a few tenth's of a degree in crown and pavilion angle definitions between friends

. Round up, round down, what difference does it make. Real good scientific technique!

(But there is a footnote explaining that the differences could be due to "recutting, measuring device tolerances, and/or application of rounding") But then again, they don't want to let anyone REALLY know what is or was going on do they.?

For example to compare the DCLR FIRE article with what they just published now
Crown Old Crown New Pavilion Old Pavillion New Table Old Table New
RD29 37.7 37.5 41.9 42.2 61 60

 

[adamasgem]

Sergey Here are the model definitions I used for my data plots

 

[adamasgem]

Interesting to note regarding the excuse for the change in the model definitions of "recutting", that not one of the stones in the RD01 to RD29 lost enough weight to change the international standard published weight that GIA uses.

In their original GIA brilliance study I had a couple of glaring discrepencies in comparing my normalized modeled results with GIA WLR, noting my Model #23. I tried and tried to get GIA to check whether there was errata in their published data in the Brilliance article, with zero success, only getting lip service from everyone (unfortunately the practice at GIA is that no one technical can talk to anyone), including a real stupid and intentionally misleading email from not one of the technical authors of the original brilliance and fire articles, but from one of GIA's double talking management mouthpieces (who is also a new author of the new article), who will remain unnamed, to protect the guilty.

One might take note that in the current G&G there was an errata sheet enclosed with the current issue, regarding the published RD01's values in that issue. Some progress at least!!!

 

[Serg]

Marty,
Please explain :why does your plot have 58 different points if you use 29 stones only?