Trying to understand the HCA just a little better....

[Regular Guy]

Garry,

Hoping you can help me here...

Though I know, rock paper scissors, idealscope may have your favor over this utility....HCA remains a convenient starting place in the search process for a diamond...witness the successful development of the search by cut utility here.

Also, from the point of view of "rubber meeting the road," and the way diamond shopping is frequently actually done with Pricescope being used as a resource...a shopper will start, and most frequently end their shopping process with the selection of diamond #1 that they say they want to see. Of course, your helpful admonition that it be used to reject bad performers, and narrow the process...is a good thing to understand, and qualify the HCA's use. Still, the practical matters of how it is used in actual practice for ultimate selection is something I am very mindful of.

Specifically, wanting to avoid providing poor direction concerning recommendations to any readers here on Pricescope motivates this thread. Wanting a little better understanding of how the HCA can be most helpfully applied and understood is the main intention here.

I've had reference to three ways to consider the application of the HCA.

1) For Regular Guys, the admonition...keep it between 0 - 2, and use this simple strategy to guide your initial selection. No complaints, and the simplicity of the idea stands up against the relatively similar simplicity of concepts relating to: color, clarity, i.e. G is better than I is better than K. VS2 is better than SI2 is better than I anything. Likewise, 0 - 2 is better than 2 - 4, etc. Likewise, the utility of the HCA allows for the implementation of a database like the search by cut utility here, as well.

2) But, at your other site dedicated to the development of HCA...the punchline I thought I understood as just a bit more involved...that the 0 - 2 score represents a composite of 4 elements, really: brilliance, fire, scintillation, and spread... and so...although the composite nature of the scores also includes the communication about a weighting of the relationship of these 4...still...more or less....when comparing two diamonds where brilliance (double weighting) are both excellent...then...you can actually compare two diamonds on the basis of the evaluation of their elements. In other words, although per (1): 0 - 2 is the target, per (2) regardless of the numbers between 0 - 2, more excellents trumps more very goods.

3) If my understanding of (2) is correct, I don't understand the concept behind your post quality only."

It seems to me that scores of lower than 1 only increase the chance of getting more excellents. Also, when the suggestion was made in recent post, Aljdewey commented on the 0 - 1 score, suggesting it may imply less contrast than would otherwise be the case. This may be true; is it? And if so, how does that work, exactly. Is there a basis in this idea, from the foundation of how a diamond gets this score?

Seeing the display of diamonds under the "search by cut" database and how they fall out, I wonder if the main point is simply that -- given 100 diamonds that appear there -- the majority will be 1 - 2. Perhaps it is then only a comment on their normal distribution, rather than a comment on the nature of those diamonds that happen to fall between 0 - 1.

Since my thinking in (2) is sort of different than what's been suggested in (3), since under the scenario of 0 - 1, there's simply a greater likelihood of having more, rather than less excellents...any advice on how to understand the basis for (3) is greatly appreciated.

Thanks in advance!

 

[Maxine]

I might be wrong, but I thought I read in a recent thread that someone was recommending a stone because it had a 1.3 on the HCA vs. a 1.1................So I agree w/RG, that it is confusing to see more ex ratings not as good of a stone????
IS there a difference in the scores under 2?

 

[Garry H (Cut Nut)]

It would help if you could distil the ? into 2 lines Ira

I think the answer is that most people seem to like the safe C 34-35 P 40.5 to 40.8 T 55-56%.
They will mostly be 1-2 (I guess).
But I think this is wrong - it is because of pre conditioning.

If Tolkowsky had have added girdle and a few more decimal places to his calculation - and had a few more flatter stones he could have pre conditioned pricescopers to 33, 41

 

[Regular Guy]

OK, so let''s try that again...

1) If you have 10 diamonds...5 scoring 0 - 1 on the HCA, and another 5 scoring 1 - 2...

based on whatever characteristics you would expect each set to have...

which set do you think would generally be preferred...

2) and why

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

(incidentally...running a - the default parameters in the HCA, as well as b - those you just recommended above, I get a) .4 TIC and 4 excellents, and b) .8 and 4 excellents).

Thanks!

 

[Garry H (Cut Nut)]

Ok
Then do this for me
Take a HCA chart - say 57% table size here, and put some eg numerals on it and list their HCA results and I will make some comments

 

[Regular Guy]

Date: 5/25/2005 11:34:10 PM
Author: Garry H (Cut Nut)
Ok
Then do this for me
Take a HCA chart - say 57% table size here, and put some eg numerals on it and list their HCA results and I will make some comments

Sorry for the ugly work...a) I don''t have experience working with drawing graphically, b) the options for HCA are based on actuals found in the search by cut option here, and c) as I''ve never plotted this before...I''ve come up with results you must be very familiar with.

Hopefully this will import below fine, and you can see the results as the clusters I''ve tried to represent (with the lower HCAs plotted more towards the center, and the higher HCAs above & below):

A - below 1

a1, hca .4, cr 34.7, pav 40.5
a2, hca .6, cr 34.3, pav 40.6
a3, hca .8, cr 34.2, pav 40.7
a4, hca .5, cr 33.6, pav 40.5
a5, hca .7 cr 35, pav 40.6

B - above 1

b1, hca 1.2, cr 34.5, pav 40.8
b2, hca 1.3, cr 34.8, pav 40.8
b3, hca 1.4, cr 34.5, pav 40.9
b4, hca 1.7, cr 34.9, pav 40.9
b5, hca 1.6, cr 33.9, pav 40

 

[lostdog]

"I might be wrong, but I thought I read in a recent thread that someone was recommending a stone because it had a 1.3 on the HCA vs. a 1.1 "

I didn't see that, but if someone was posting this, they probably shouldn't have been. My original point on the topic was that, simply as a matter of the mathematics, you can't split hairs with HCA score, because the number is interpolated 80% of the time (more or less). The resolution of the numbers doesn't tell you anything that precise. Possibly if you investigate the neighboring points on the contour map, you can decide overweight or underweight the HCA score for a given stone.

Generally, for 2, you have greater uncertainty as to whether it's a 2.2 or a 4.2.

So there are multiple issues. What you want to know is the most representative HCA score for the given stone (let's call it the "true" score). There's the HCA score that you get from entering your data (the reported score), which is approximately related to the HCA scores that the original evaluated images represented (the benchmark scores). If the stone you are evaluating is exactly on the benchmark intervals, then the benchmark and the reported numbers are the same (1 of every 5 cases), so it is a true score, but otherwise the reported score is an approximation of the true score.

Sometimes the variation from true to reported might be small, sometimes it might be large. Only rarely would it be large enough to mislead you by significantly understating a good performer, say a true 2.0 is reported to be a 4.5. (It is even less likely to overstate a poor perfomer). The only issue with the "small" variations is that you might be comparing, say, a reported .8 and a 1.6, and for all you know, but the true scores might actually be the reverse.

So look carefully before you buy a particular stone because it's "under 1.0". It might not really be that at all. If you really don't understand what you are doing with the math, then just accept that a reported score "under 2.0" covers the bases, and they are all equal after that. HCA is a valuable tool, but when you and I use it, it's a guide to whether a diamond might have good visual performance. You can't conclusively use it as a detailed comparison tool between similar diamonds, there are too many other things that can vary between actual stones.

Of course, you can get into what sort of performance differences there might be between a true .5 and a true 1.5 and so on, which is I think your actual question (and a good one).

Sorry, I don't mean to be a broken record on the subject. Just surprised people are trying to sweat out the tenths of HCA between two choices. There's no point. It's like picking the best basketball player by assuming the best is the tallest, and then using a photograph taken during the game to measure who is tallest. The underlying assumption (that absolute highest HCA is best regardless) is not necessarily true, and the method used to figure out the exact measure isn't precise enough to let you draw conclusions about height/score after a certain point.

Some people (present company in this thread not included) seem to want to use any diamond grading system not as a tool for finding diamonds, but as an emotional refuge tool regardless of how the actual diamond stacks up. 1.2 is lower than 1.8, therefore it's better, you're better, whatever. Enjoy it!

I think I'm done posting on this subject, probably for good.

 

[Garry H (Cut Nut)]

Storm I am preparing for a big trip
no time

Can you please put the LR, F, C and Spread results there please too.
And pick a couple from the far right and far left

 

[Regular Guy]

Garry,

If you''d said: Surely you''re joking, I''d have had to say: Don''t call me Shirley.

In this case, though I''m not storm (though, you know what...it''s the drawing chicken scratch, isn''t it!), if I understand the request, here goes:

HCA under 1

left most: A5: HCA .5 ex ex ex ex (cr 33.6, pav 40.5)
right most: A4: HCA .7 ex ex ex ex (cr 35, pav 40.6)

HCA over 1

left most: B5: HCA 1.6 VG ex ex ex (cr 33.9 pav 40)
right most: B4: HCA 1.7 vg vg vg ex (cr 34.9, pav 40.9)

Final note a reminder concerning both the point of this post, and as well, a response to Lost Dog...

I''m trying to understand your affirmative statement, Garry:

"If my understanding.... is correct, I don''t understand the concept behind your post..., where you review that: "Most nice stones rate 1-2," and in the contemporary tutorial you reaffirm the same point for those buyers who are seeking "quality only."

Thanks again.

 

[valeria101]

RegularGuy and Strm...

If you like the topic, you might want to take a dozen examples of diamonds with both AGS and Sarin reports (too bad most of those will be super precise cuts, but that''s life), make the HCA scores based on either set of numbers and compare. Some time ago I did this and ended with HCA score variances up to 1. The difference between HCA scores for the same stone obtained using angles and % is a bit more on average. And there are outliers too

Of course I didn''t apply due diligence to take a large sample (or even define what a statistically large sample should be for this purpose). The range considered by Garry (0-2) stood the little test, of course. To double HCA precision (=to differentiate scores between 0-1 from those 1-2) you would need allot of diamonds.

Surely a mathematically precise model would construct differently looking glitter for slightly different main angles. So... you can compare GA models for HCA 0.5 and HCA 1.5 ceteris paribus. Measurement error apparently does not allow to translate the conclusion of such comparison into practice. Unless my sample was an unfortunate choice. Please feel free to check

 

[Regular Guy]

Date: 5/26/2005 7:13:01 AM
Author: valeria101

Surely a mathematically precise model would construct differently looking glitter for slightly different main angles. So... you can compare GA models for HCA 0.5 and HCA 1.5 ceteris paribus. Measurement error apparently does not allow to translate the conclusion of such comparison into practice. Unless my sample was an unfortunate choice. Please feel free to check

Ana,

Thanks for your input. Though related in substance, for the intention of this thread, the design is that measurement error is not the principle discussed here.

Also, although some effort was made to select options reflecting variance, the random walk was closer to the result, a variance of HCA scores above and below "1" was effected, and the question of "what is the character," if it can be discerned, of a diamond that has the descriptor "above or below 1," on the HCA is the nature of the question of this thread, itself. Surely, a better job of selection could have been done.

 

[JohnQuixote]

Date: 5/26/2005 3:28:26 AM
Author: lostdog

'I might be wrong, but I thought I read in a recent thread that someone was recommending a stone because it had a 1.3 on the HCA vs. a 1.1 '

I didn't see that, but if someone was posting this, they probably shouldn't have been. My original point on the topic was that, simply as a matter of the mathematics, you can't split hairs with HCA score, because the number is interpolated 80% of the time (more or less). The resolution of the numbers doesn't tell you anything that precise. Possibly if you investigate the neighboring points on the contour map, you can decide overweight or underweight the HCA score for a given stone.

Generally, for <2, you can have a high confidence the HCA is very low. For slightly >2, you have greater uncertainty as to whether it's a 2.2 or a 4.2.

So there are multiple issues. What you want to know is the most representative HCA score for the given stone (let's call it the 'true' score). There's the HCA score that you get from entering your data (the reported score), which is approximately related to the HCA scores that the original evaluated images represented (the benchmark scores). If the stone you are evaluating is exactly on the benchmark intervals, then the benchmark and the reported numbers are the same (1 of every 5 cases), so it is a true score, but otherwise the reported score is an approximation of the true score.

Sometimes the variation from true to reported might be small, sometimes it might be large. Only rarely would it be large enough to mislead you by significantly understating a good performer, say a true 2.0 is reported to be a 4.5. (It is even less likely to overstate a poor perfomer). The only issue with the 'small' variations is that you might be comparing, say, a reported .8 and a 1.6, and for all you know, but the true scores might actually be the reverse.

So look carefully before you buy a particular stone because it's 'under 1.0'. It might not really be that at all. If you really don't understand what you are doing with the math, then just accept that a reported score 'under 2.0' covers the bases, and they are all equal after that. HCA is a valuable tool, but when you and I use it, it's a guide to whether a diamond might have good visual performance. You can't conclusively use it as a detailed comparison tool between similar diamonds, there are too many other things that can vary between actual stones.

Of course, you can get into what sort of performance differences there might be between a true .5 and a true 1.5 and so on, which is I think your actual question (and a good one).

Sorry, I don't mean to be a broken record on the subject. Just surprised people are trying to sweat out the tenths of HCA between two choices. There's no point. It's like picking the best basketball player by assuming the best is the tallest, and then using a photograph taken during the game to measure who is tallest. The underlying assumption (that absolute highest HCA is best regardless) is not necessarily true, and the method used to figure out the exact measure isn't precise enough to let you draw conclusions about height/score after a certain point.

Some people (present company in this thread not included) seem to want to use any diamond grading system not as a tool for finding diamonds, but as an emotional refuge tool regardless of how the actual diamond stacks up. 1.2 is lower than 1.8, therefore it's better, you're better, whatever. Enjoy it!

I think I'm done posting on this subject, probably for good.

I quoted this whole thing, hoping that people will read it again.

People should also be advised that some Sarin/Ogi error is possible, so what you're punching in might be a bit off in the first place. Not usually enough to make a drastic difference, but in tenths? Sure.

Garry has exclaimed many times that HCA is a tool for rejection.


ETA: Ira - not trying to derail. The above is just important consumer advisory that's not on the play list often enough.

 

[Garry H (Cut Nut)]

might get time in a few hours

 

[Regular Guy]

Date: 5/26/2005 9:29:42 PM
Author: JohnQuixote



People should also be advised that some Sarin/Ogi error is possible, so what you're punching in might be a bit off in the first place. Not usually enough to make a drastic difference, but in tenths? Sure.

Garry has exclaimed many times that HCA is a tool for rejection.

ETA: Ira - not trying to derail. The above is just important consumer advisory that's not on the play list often enough.

We could try methodically to account appropriately for measurement error, and in fact, could suggest yet again a 4th view on procedure, looking at the particulars of that is probably reasonable. For example, from the early post Lost Dog references, he -- as I recall -- keenly reviewed what Garry has elsewhere said -- that for certain sets of crown & pavilion angles, the critical points are much more sensitive, and so either getting an exact measurement is important, or understanding that the variance in actual measure is very sensitive, such that getting it "right" is going to be important for good performance.

But that is just a fourth view. And, for example, although Kenny, in his current post, is trying to get the best result by having a number of different measures for good cut synchronize, he can probably cut corners on some. For example, though he wants both AGS and sarin measurements to agree on a good HCA score, from what I understand, AGS's measurement can usually be counted on to be reliable, so I'm guessing just AGS scores alone will satisfy.

Like Kenny, where we are uncertain of which way to go, we try to get certain ideas to agree. In the thread he's initiated, he, Rhino, Mara, and Dancing Fire have all piped in on how to work with the HCA. John, you've suggested one view that Garry has suggested. From my reading, Garry himself has suggested not only this one, but he has either voiced or suggested the three different "angles" on reading the HCA that I began this thread with.

Since I personally find the dissonance at least slightly troublesome, (and with a quick review):

- Anything 0 - 2 is A OK (also John's view)
- given allowances for weighting, the normative meanings of Excellent trumping very good trumping good apply (Kenny's view)
- scores between 1 - 2 (where more excellents are less likely to appear) are preferred over scores of 0 - 1 (Rhino & Dancing Fire agree here)

with writers and readers here continuing to use the HCA to try to say different things...I am hoping Garry can put this item back on his agenda (as frankly...I am guessing there may yet be more to this than meets the eye).

...even if I'm the only one this is troubling.

Best,

 

[kenny]

HCA 0-2 is the best
HCA 2-4 is next best
HCA 4-6 is next, etc.
Then the only place you see 4 excellents is under 1.0.
I think it is understandable that all this led me to believe that the lower the number the better.
Who would have guessed that Ex Ex Ex VG is better than Ex Ex Ex Ex? - It's very non-intuitive.

Now I'm being told that too much under HCA= 1.0 is not good because of some other factors.

Forgive my curious nature but are you sure that the fact that much of the inventory for sale by PS vendors being Ex Ex Ex Vg (HCA= about 1) is not influencing the advice to simply buy the stones they have more of?
It would help if Garry explained why 1.0 is better than 0.5
Adding this exception to "lower is better" thinking" to the descriptions here would prevent my dumb question from coming up again too.

BTW a year ago I bought a hybrid car.
I drove the Honda and Toyota salesfolks crazy as I analyzed and comapred the cars to death.

 

[Garry H (Cut Nut)]

No bias Kenny

Need to get you guys thinking out of the box

try this
They are all 1ct stones

 

[Regular Guy]

Date: 5/29/2005 7:25:32 AM
Author: Garry H (Cut Nut)
No bias Kenny

Need to get you guys thinking out of the box

try this
They are all 1ct stones

...doesn't seem like changing the question to one that I understand you find more interesting, and of possible ultimate value, long term, to consumers (getting diamonds of very different looks available to consider, yielding BICs and FICs)...changes the need to understand better those existing options that do now fall "within" the box.

Maybe at this point you're merely being hounded by me, Garry. You've hopefully heard from others other than me the cacophony left by what seems to be three possibly valid views, yielding internally contradictory interpretations associated with the HCA, for at least normally conforming, readily available diamonds. Maybe you'd prefer this be left as is to clarify itself. Though I'm sure we'll do our best...the hope is you'll take one or more shots to clarify...or even...open up the box, which might be a better target (vs. thinking outside of it!).

With appreciation for the assistance the HCA provides already, warm regards,

Ira Z.

 

[strmrdr]

Date: 5/29/2005 1:04:33 AM
Author: kenny
HCA 0-2 is the best

HCA 2-4 is next best

HCA 4-6 is next, etc.

Then the only place you see 4 excellents is under 1.0.

I think it is understandable that all this led me to believe that the lower the number the better.

Who would have guessed that Ex Ex Ex VG is better than Ex Ex Ex Ex? - It''s very non-intuitive.


Now I''m being told that too much under HCA= 1.0 is not good because of some other factors.


Forgive my curious nature but are you sure that the fact that much of the inventory for sale by PS vendors being Ex Ex Ex Vg (HCA= about 1) is not influencing the advice to simply buy the stones they have more of?

It would help is Garry explained why 1.0 is better than 0.5

Adding this exception to ''lower is better'' thinking'' to the descriptions here would prevent my dumb question from coming up again too.


BTW a year ago I bought a hybrid car.

I drove the Honda and Toyota salesfolks crazy as I analyzed and comapred the cars to death.

hmmm dispite being mentioned a few times this is my first post in this thread
Someone been drinking again?

Now to answer your question about the VG on spread.
In order to get a well balanced diamond you need some height to the crown and depth to the pavilion.
Get either too shallow or deep and performance suffers.
The vast majority of high performing diamonds will come out as VG for spread.
That does not mean the spread is bad it just means that the weight is needed elsewhere to match up the angles to perform well.

The key to understanding the hca is that it tells you mainly if the pavilian angle average works with the crown angle average. Trying to read much more than that into it is a mistake imho.
If I was doing it Id make it a simple yes/no to the question is it a good combo.

 

[JohnQuixote]

Date: 5/28/2005 11:15:02 PM
Author: Regular Guy
- Anything 0 - 2 is A OK (also John's view)
- given allowances for weighting, the normative meanings of Excellent trumping very good trumping good apply (Kenny's view)
- scores between 1 - 2 (where more excellents are less likely to appear) are preferred over scores of 0 - 1 (Rhino & Dancing Fire agree here)

Ira, I think your endeavor to gather comments/opinion and sift out a static answer is admirable, but perhaps it is just not possible.

Short answer

Most people who find PS are using the HCA just to see if a diamond will be beautiful. A score <2.0 predicts it to be in the top 5% of known performers. After that there are too many potential measurement variances and unknowns to split hairs without further information; Direct and personal assessment of that particular diamond.

Think of it this way: The consumer is the king. HCA is his taster. After HCA determines the foods that are not poisoned (<2.0) the king must go forward to determine which entrée is his favorite.

Long answer

I am with Strm on this. I think HCA is the best tool available for non-direct assessment, but it should be thought of as providing a ‘yes/no’ answer. Splitting hairs via tenths on HCA may cloud the issue unnecessarily for the innocent consumer.

For a minute let’s pretend that Sarin/Ogi measurements are completely accurate. Even so, HCA does not take into account minor facets, which play a prominent role in the character of diamond beauty. Facet yaw influences character and performance, but is not a part of the HCA prediction. Details of optimum patterning and total facet symbiosis influence performance through a broad range of illumination environments in diamonds with supersymmetry, but these are not factored into HCA either. Should we begin dissertations on these aspects for anyone who comes to PS to make a simple purchase? These are things enthusiasts are absorbed by, but it's not necessary to force the casual shopper through these doors onto these winding, time-taking pathways which, on the whole, have some influence - but are still subject to personal taste beyond that.

I applaud experts and hobbyists who like to assess the HCA algorithm and observed behavior in order to debate it on PS: Those discussions are important to who we are as a group on a scientific level. However, such dialog is not relevant to the majority of "I am here-for-10-days" consumers who just want to know "is this a high performing diamond?"

HCA and its 2.0 threshold provide a yes or no about optimum angles. It places a diamond in or out of the top 5%. Any further discussion regarding character is best done with an expert who has the actual diamond in-hand. That avoids possibilities of measurement error and speculation about influential aspects of facet construction.

I would put it is thusly:

* A diamond <2.0 is predicted to be a very beautiful diamond – one in the top 5% of known performers. Beyond that, one should use images and direct assessment of the actual diamond, along with the help of a trusted vendor and-or appraiser, to understand nuances of its specific character/personality.

Summary

This is why diamonds will never be wholly commoditized: There are so many variables from diamond to diamond.

Poetically, each is like a snowflake.

Less poetically, we can identify optimum performers with accurate measure & proven combinations of angles/percentages via HCA, but we cannot tell what exact character a diamond has until we look at it. Throw personal taste into the mix and you can see why it is important to assess each diamond individually, and determine what the customer is seeking relative to that.

The beauty of PS is that we have experts, scientists and enthusiasts from every walk of diamond-life who can help each consumer find his/her comfort level in diamond education. I see no need to force consumers into thinking they must acquire a semi-GG or science degree to be happy with a purchase.

It is sad to watch the occasional consumer join PS, get overwhelmed with deep expostulations and go away as a result: The online purchase suddenly became far too complex for them, when it didn't need to be.

Some consumers just want to know 'is it going to be okay?'
Others want to know a bit more.
Others want to know a lot more.

Still others get all geeked-up about minor facets, patterning, leakage footprints, contrast, facet yaw and diamond parcel-paper-thingys.

We can serve them all elegantly - each on his/her own terms - without overwhelming them.

 

[Garry H (Cut Nut)]

Date: 5/29/2005 6:21:03 PM
Author: JohnQuixote

Most people who find PS are using the HCA just to see if a diamond will be beautiful.

Point of order:

Most people who find PS are using the HCA just to see if a diamond will be beautiful.


Most people who find PS are using the HCA just to see if a diamond CAN or could be beautiful.
Or even maybe, or might be beautiful.

also there are no ''algoritms - just look up charts (I am not that smart)

 

[Regular Guy]

Sorry, I know, never mind right now about deficits and liabilities...right now you guys have got ASETs to talk about...still....

- with all you experts in each other''s proximity
- with me contemplating the juxtaposition of
a) experts, and the degree to which consumers should want to need them,
b) DIY, and the extent that what we learn on Pricescope allows us to do this by ourselves,
c) Pricescope itself, as an environment that brings these two principles rubbing up against each other...
- and with Nerdbot (re) appearing here with his success story, reminding me of what seemed a potent recent discussion on what HCA can or cannot lead one to understand

I wanted to share that with you, too, should any further insight fall out from that.

with best wishes...whether you''re in Las Vegas, or not...

 

[JohnQuixote]

Date: 6/3/2005 11:51:25 PM
Author: Regular Guy

I wanted to share that with you, too, should any further insight fall out from that.
with best wishes...whether you're in Las Vegas, or not...

Ira, thanks.

People who are attracted to HCA may enjoy this: I understand GIA is developing something along the lines of HCA (similar interface). I don't know how exacting/elaborate it will be or what number of measurements will be required, but if a major lab is embracing the premise a lot of devils in the details will need to be addressed to arrive at definitive conclusions.

Garry may be able to elaborate since it's truly his premise they are adopting.

 

[hoorray]

Ira, you may not be able to get the specific answer to your question since the answer realy is more general (as JQ, Garry and others have said), but I think the discussion is useful -- especially for Newbies trying to figure out what to think of and how to use the HCA.

When I first found the HCA, I had no idea who Garry was (sorry Garry

) and whether the HCA was credible. I wanted it to be, but had no basis yet to say it was. More research and reading led me to believe that it was a great first cut (no pun intended) tool. I hope other Newbies will read some of these posts and realize that tHe HCA is trying to smplify a very complex issue into something useful, focusing on the variances in performance that different crown and pavillion angles produce.

IMO, the spread calculation is just added info. I would typically read the results as VG = the stone is within the norms for it''s weight. X= the stone will have a slightly larger diameter than other similarly weighted stones. The X''s are harder to find within the top performing stones, and in my mind are a bonus -- not something that is reasonable to require if your reason for using the HCA is to identify top performers.

 

[Garry H (Cut Nut)]

There have been some requests for something like this.
I decided to put it on this old thread.

 

[james101]

Just chiming in to say thanks. I''m one of the newbies, soaking up hours of information from this site and using HCA. Thank you for your discussion in this thread - I''m getting close to my final selection and the HCA has been handy in the process. I''m really glad I came across this thread and its comments on how HCA is really intended to be used in the decision-making process.

 

[aljdewey]

Date: 5/26/2005 3:28:26 AM
Author: lostdog

Sorry, I don't mean to be a broken record on the subject. Just surprised people are trying to sweat out the tenths of HCA between two choices. There's no point. It's like picking the best basketball player by assuming the best is the tallest, and then using a photograph taken during the game to measure who is tallest. The underlying assumption (that absolute highest HCA is best regardless) is not necessarily true, and the method used to figure out the exact measure isn't precise enough to let you draw conclusions about height/score after a certain point.

I'm in LOVE.......LostDog, WELL SAID.

Date: 5/26/2005 3:28:26 AM
Author: lostdog

I think I'm done posting on this subject, probably for good.

Gosh, I really hope not, because this contribution is VERY important and bears repeating.

The HCA is a rejection tool.....it's like a first audition. It's only purpose is to weed out KNOWN poor performers, leaving a smaller subset that merit more extensive consideration. People want to think of it incrementally, and it's just not that.

It would help if folks would think of it this way: Anything less than is a pass; anything over 2 is a fail. Those that pass move on to *further consideration* (and further elimination until you settle on THE ONE).

Making it into the "pass" group doesn't automatically mean it's a stellar stone; it just means it wasn't reject-worthy in the first round. If you have 4 stones that all make it by that first round, then you have to consider more carefully and more minutely to go from 4 stones down to 1, right?

 

[Regular Guy]

Al and All,

Let me ask you to close your eyes, click your heels together, and say as fast as you can: "I want to go to Houston right now" five times. If you cannot accomplish this in 7 seconds, do not pass go, and you do not collect your tokens.

OK, before being accused of being the most obscure guy on the planet, let''s just get real, and talk about why not only me, but anyone would be motivated to make as much of the system of HCA, or any evaluative system that can help us understand the relative beauty of a diamond before we actually see it. That the HCA system is for weeding, and not for picking, is not an indictment of the system in any way. Still, it is so obvious why we would like it to be for picking, and so, we are motivated to do what we can to make it so.

Since so many of us do use Pricescope to not only learn about diamonds, but also to buy them, so frequently, we seek to buy at a distance from one of these internet vendors. To the extent we use our eyes, let alone even ask the vendor to look at a group of diamonds on our behalf...we are rather limited in the number of diamonds we can consider. So, having selected 2 to....5 diamonds we would choose among and between, the weeding process begins there, on the basis of that small number of diamonds that we had chosen to begin with.

In consideration of this filtering process...hopefully the initial filter is pretty darn good, as in the typical buying situation, we do go from there, and in the buying environment we find ourselves in, we don''t have a lot of diamonds around us to compare them to.

So that''s all. I think the HCA is pretty good. Certainly, it is convenient to think that it is, as it makes the possibility of shopping with some rigor at a distance seem all the more reasonable. We may try to squeeze it for more than we can get from it, but the purpose of this note today is to say...I hope you can see why I''d rather not be blamed for looking for a lot from whatever resource I would use to help me with a narrowing decision from afar.

Perhaps the combination of an HCA and a trusted vendor, together, is pretty good.

I''ll drink to that.

 

[aljdewey]

Date: 5/28/2005 11:15:02 PM
Author: Regular Guy

Date: 5/26/2005 9:29:42 PM
Author: JohnQuixote

People should also be advised that some Sarin/Ogi error is possible, so what you''re punching in might be a bit off in the first place. Not usually enough to make a drastic difference, but in tenths? Sure.

We could try methodically to account appropriately for measurement error, and in fact, could suggest yet again a 4th view on procedure, looking at the particulars of that is probably reasonable. For example, from the early post Lost Dog references, he -- as I recall -- keenly reviewed what Garry has elsewhere said -- that for certain sets of crown & pavilion angles, the critical points are much more sensitive, and so either getting an exact measurement is important,

Ira, you are like a pit bull with this. I admire your tenacity, but you are trying to make EXACT determinations about something that isn''t EXACT! I think you are interpreting that as an unwillingness to help you, but it''s not.

The point was made the "exact" depends on the calibration of the various measurement tools, so there is already some margin of error built in there, right? Now you need to factor ANOTHER step away from "exactness" into the mix...the girdle. If you are manually plugging in numbers to the HCA, it automatically ASSUMES a medium girdle. if the girdle isn''t medium, tho, that will MARGINALLY change the numbers. A stone that scores a 1.0 based on a medium girdle might score a 0.9 with a thin girdle! NOW WHAT?

My point: the thing doesn''t have to meet NASA accuracy calibrations to provide a "starting" point for weeding out stones. That type of MICRO difference is more meaningful in the 2nd and 3rd round of elimination, but NOT the first....which is what the HCA amounts to: a first-round elimination.

Date: 5/28/2005 11:15:02 PM
Author: Regular Guy

....or understanding that the variance in actual measure is very sensitive, such that getting it ''right'' is going to be important for good performance.

No. Getting it right is possibly important to SUPERLATIVE or unmatched performance, but not to *good* performance. And the HCA isn''t supposed to be used to arrive at "superlative".

I know the engineer in you wants to fuss over infinitessimal alterations.....but I have to say that some differences are more significant/staggering than others. Judgment and common sense has to balance the engineer in you.

If you put one drop of red food coloring into an 4-oz glass of water, the color will change significantly....enough for me to notice. If you put one drop of red food coloring into *Lake Michigan*, there IS a measureable, infinitessimal alteration, but not one I can notice or appreciate. There is a difference between a "measureable difference" and an *appreciable difference*.

There''s been a lot of discussion recently about stones that score below 1 and how performance in such stones may not perform as well. HOW MUCH less well? Measureable or appreciable? If the difference means I''ll see 20% less light return in a majority of lighting conditions, that would potentially be meaningful to me....and likely appreciable to my eye.

If the difference means that I''ll only see 86.66667676767% light return instead of 86.66667676770% light return, should that matter? I don''t think so.

 

[Regular Guy]

Al,

Do you understand this post started in May, with a couple of posts today (that did not originate with me, although I did remind Garry of it)? I did make one post today. Is there anything in my post today you''re responding to?

 

[aljdewey]

Date: 11/9/2005 4:20:20 PM
Author: Regular Guy

OK, before being accused of being the most obscure guy on the planet, let's just get real, and talk about why not only me, but anyone would be motivated to make as much of the system of HCA, or any evaluative system that can help us understand the relative beauty of a diamond before we actually see it. That the HCA system is for weeding, and not for picking, is not an indictment of the system in any way. Still, it is so obvious why we would like it to be for picking, and so, we are motivated to do what we can to make it so.

Oh, we are getting somewhere, Ira. This is perhaps one of the most succinct and understandable posts you've ever made here. A joy to read.

Believe me, I can understand the desire to make it more, but desire cannot override reality. Diamonds are unique and individual things....no two are PRECISELY alike. That reality trumps our want to "standardize" this. That's why it will never capable of being used as *selection* tool, no matter how much we might wish it to be.

Further, it's impossible to standardize people's *preferences*.....the same look doesn't appeal to everyone equally. So, not only can we not standardize the diamonds, we cannot standardize the buyer's interpretation of "most desirable".

Simply put: what we WANT isn't possible unless it will fit within the limitations of reality. I fervently wish I had superpowers, but I'm a mere mortal, and nothing can be done to work around that.

I wish that I could use the plastic cape in my godson's Halloween costume to fly....but no matter how much I want that, it's not capable of being a viable flight instrument for me.

Date: 11/9/2005 4:20:20 PM
Author: Regular Guy

We may try to squeeze it for more than we can get from it, but the purpose of this note today is to say...I hope you can see why I'd rather not be blamed for looking for a lot from whatever resource I would use to help me with a narrowing decision from afar.

It already IS that, though....it already IS a pretty good narrowing tool. It eliminates 95% of poor performers. I'd say that's pretty damn good.

Date: 11/9/2005 4:20:20 PM
Author: Regular Guy

Perhaps the combination of an HCA and a trusted vendor, together, is pretty good.

I'll drink to that.

Amen.

That combination IS pretty good, and I'll drink to that with you.