Is a 1.00ct that looks like a 0.95ct diamond acceptable?

[Texas Leaguer]

I was looking at an old article and was reminded of one of my favorite diamond sayings, and I thought readers of this particular thread would get a kick out of it. It comes from Henry Dutton Morse who was cutting what essentially were Tolkowski ideals in the late 1800's in the US, more 20 years before Tolkowski published his famous work in Europe. Morse was a true pioneer and major proponent of cutting diamonds for beauty over weight.

“Judging diamonds by the carat is like judging a racehorse by the pound!”

 

[gm89uk]

Thanks Garry looking forward to seeing it in action.

 

[gm89uk]

Out of curiosity, how did you reach the 6.44mm benchmark?

 

[Garry H (Cut Nut)]

Out of curiosity, how did you reach the 6.44mm benchmark?

Hi GM, great question.
most common girdle thicknesses are thicker than others have used. The result in the real world is 6.47 etc are very very uncommon, even in the CBI world. Not to mention thin girdles are prone to chipping in the long run.
Take a smallish 1ct that measures 6.36mm for example.
That measure means 38% of the diamonds on Pricescope listings (6,690 stone sample size) that weigh exactly 1ct are smaller by millimetre spread. But using my algorthim to adjust for the apparent size in a range of diferent lightings for clean diamonds 52% appear to be smaller than a well cut 6.36mm.

If you think that is terrible (I do) I did the same thing on all the exactly 1.00ct stones on RapNet - there were 13,001 1.00ct RBC D-J IF-SI2

Only 1623 were +6.44mm = 12.5%

5,531 <6.36mm (OctoNus standard for 0.95ct) 42.5% had a smaller than optimal spread. 10% more than on PS! So the good news is that Pricescope has less worse stones
The bad news would be that more of those stones would also have shocking apparent much smaller size (I could not calculate the apparent reduction because I do not have the data. I only used the stones on the PS database that had dimensions, table size, crown and pavilion angles).

 

[gm89uk]

Hi GM, great question.
most common girdle thicknesses are thicker than others have used. The result in the real world is 6.47 etc are very very uncommon, even in the CBI world. Not to mention thin girdles are prone to chipping in the long run.
Take a smallish 1ct that measures 6.36mm for example.
That measure means 38% of the diamonds on Pricescope listings (6,690 stone sample size) that weigh exactly 1ct are smaller by millimetre spread. But using my algorthim to adjust for the apparent size in a range of diferent lightings for clean diamonds 52% appear to be smaller than a well cut 6.36mm.

If you think that is terrible (I do) I did the same thing on all the exactly 1.00ct stones on RapNet - there were 13,001 1.00ct RBC D-J IF-SI2

Only 1623 were +6.44mm = 12.5%

5,531 <6.36mm (OctoNus standard for 0.95ct) 42.5% had a smaller than optimal spread. 10% more than on PS! So the good news is that Pricescope has less worse stones
The bad news would be that more of those stones would also have shocking apparent much smaller size (I could not calculate the apparent reduction because I do not have the data. I only used the stones on the PS database that had dimensions, table size, crown and pavilion angles).

Very interesting, thanks for the detailed explanation.

It would be very interesting if you standardise all diamonds to 1ct by dividing mm / carat ^(1/3) and run your algorithm on apparent size to greatly increase your sample size.

It would be cool to see if those shocking numbers are inherently more at the exact 1ct mark or across the board to the same degree. You could even make a graph similar to the one you showed with X axis being carat size and y Axis comparing mean apparent % size to your bench mark 6.44 which would then certainly prove if the situation really is that dire as you approach the magic ct numbers, as the cutting numbers seem to suggest

 

[Garry H (Cut Nut)]

I just found this old video I made 5 years ago that I think shows very clearly why cut is so important, and so under valued. It compares 4 1ct diamonds with very different spreads, cut quality colour and clarity. BUT all are of the same cost i.e. value as the market sees it.

(I posted it on the colour discussion thread too, because it shows colour clarity cost trade off too)

 

[Texas Leaguer]

Hi GM, great question.
most common girdle thicknesses are thicker than others have used. The result in the real world is 6.47 etc are very very uncommon, even in the CBI world. Not to mention thin girdles are prone to chipping in the long run.
Take a smallish 1ct that measures 6.36mm for example.
That measure means 38% of the diamonds on Pricescope listings (6,690 stone sample size) that weigh exactly 1ct are smaller by millimetre spread. But using my algorthim to adjust for the apparent size in a range of diferent lightings for clean diamonds 52% appear to be smaller than a well cut 6.36mm.

If you think that is terrible (I do) I did the same thing on all the exactly 1.00ct stones on RapNet - there were 13,001 1.00ct RBC D-J IF-SI2

Only 1623 were +6.44mm = 12.5%

5,531 <6.36mm (OctoNus standard for 0.95ct) 42.5% had a smaller than optimal spread. 10% more than on PS! So the good news is that Pricescope has less worse stones
The bad news would be that more of those stones would also have shocking apparent much smaller size (I could not calculate the apparent reduction because I do not have the data. I only used the stones on the PS database that had dimensions, table size, crown and pavilion angles).

Garry,
Redirecting consumers attention to spread (dimensions rather than weight) is clearly a worthwhile cause. But the real power in your proposed upgrade to HCA is presenting an 'apparent' size using an algorithm to factor that in from the data.

Regarding the line in your post above which I have bolded, can you elaborate on your methodology for determining apparent size, in particular factoring in different lighting environments into that calculation?

 

[Garry H (Cut Nut)]

Garry,
Redirecting consumers attention to spread (dimensions rather than weight) is clearly a worthwhile cause. But the real power in your proposed upgrade to HCA is presenting an 'apparent' size using an algorithm to factor that in from the data.

Regarding the line in your post above which I have bolded, can you elaborate on your methodology for determining apparent size, in particular factoring in different lighting environments into that calculation?

Thanks Bryan. The question in bold blue above is an excellent one. Thank you. I was worried no one would ask. In low level diffuse light the size differences for peripheral leakage are much greater than what I set as a benchmark penalty. See after 20 seconds into the video.
In bright daylight or a store where seeing the actual edges of the diamond is easier, the difference is less obvious. I have used a penalty that is less than I think is fair, because the controversy and challenges will surely come.
In store purchasing decisions I will be proven to have been too harsh.
But in consumers and everyday diamond enjoyment environs, which have always counted more to me, my choice will be vindicated. I never agreed with AGS or GIA's close observation standards - they may be applicable for buying but not for enjoyment.

This 1.37min video really shows the differences that you actually see. I have done this demo many thousands of times and everyone is gob-smacked. The bad 1ct 6.25mm CZ looks like 0.50ct.

 

[Paul-Antwerp]

Hi Garry,

As you probably know, we are using the same CZ-set in presentations, and teaching sales-staff in stores to use it. It very often is an eye-opener.

Now, the set has a 6.3 mm. bad CZ and a 6.0 mm. nice CZ. I agree, in store-lighting, the apparent size is approximately the same, while in everyday life (with many low-light conditions), the bad CZ suddenly seems to shrink to half.

Thank you for explaining how you try to factor in various light-conditions into your metric. We will probably have to accept it not being perfect, but defensible and definitely better than simply taking weight as the indicator of apparent diameter or surface.

My suggestions to using your metric (technology?) are going further than including them in HCA.

- Why, when making it part of HCA, reduce it to 4 or 5 categories (Good, Less good, Bad, Horrible, ...)? Why not give the calculated number of apparent diameter (surface?), possibly compared to a standard of 100?
- Your metric is more than a sub-part of Cut Quality. It clearly undermines the value of the C of Carat Weight. Why not use your metric for better defense of the consumer?
- Is there a way to license your metric? I definitely would not mind posting apparent diameter or surface of all our diamonds, referring to your 3rd-party-metric?
- As far as education of consumers goes, is there a way to make this a better component of PS' educational content?
- As far as protection of consumers goes, can this metric become incorporated in the diamond-listing on PS, such that consumers can search for and compare apparent diameters, rather than carat weight only?

Your research shows that a huge majority of what is available to consumers today is scoring low on your metric in a horrific way. I feel that the way you are planning to make your metric a small sub-aspect of HCA will not create a major change in that supply-situation. Broadening the education and making your metric a direct competitor of Carat Weight will have far more impact, especially if good forces join together.

We wish to join that fight. In fact, we are already educating our retail-network in that way for years. Who else wants to join in?

Live long,

 

[Garry H (Cut Nut)]

Hi Garry,
I will answer some now Paul, rest when I have time. Excellent questions. thanks.
As you probably know, we are using the same CZ-set in presentations, and teaching sales-staff in stores to use it. It very often is an eye-opener.

Now, the set has a 6.3 mm. bad CZ and a 6.0 mm. nice CZ. I agree, in store-lighting, the apparent size is approximately the same, while in everyday life (with many low-light conditions), the bad CZ suddenly seems to shrink to half. They would be ones I supplied - they should be 6.5mm nice make and 6.25mm very deep cut.

Thank you for explaining how you try to factor in various light-conditions into your metric. We will probably have to accept it not being perfect, but defensible and definitely better than simply taking weight as the indicator of apparent diameter or surface. Yes, Exactly

My suggestions to using your metric (technology?) are going further than including them in HCA.

- Why, when making it part of HCA, reduce it to 4 or 5 categories (Good, Less good, Bad, Horrible, ...)? Why not give the calculated number of apparent diameter (surface?), possibly compared to a standard of 100? I did explain early on the 4 grades and someone made an excellent suggestion to include another mid grade of "Average" which is about the avaerage for stones just over magic Carat weights (which is about +80% of all stones GM88 BTW, so even though statistically your proposed experiment would be interesting, a look at the distribution chart shows the answer and the cause of the problem).
Regarding giving a number: YES. but right now the only data input I have is GIA badly rounded data - and no input comp's for minor facets, painting or digging.
In the future, I hope to gain the real stone scan data and run stone at planning stage and post polished production. Then real numbers will make sense
But before then I also need some more patent protection than just a provisional


Live long,

 

[Paul-Antwerp]

Thank you for the reply, Garry.

As for the CZ-set, we use a slightly different composition, with a 6 mm well-cut CZ, not the 6.5 mm. We do use the 6.25 - 6.30 bad cut. Comparing with a well cut stone of clearly lower diameter has various advantages, in letting reality sink in:

Step 1: Comparison in jewelry-light: Agreement generally is that they look similar in size. Even with a strong light-environment, the well-cut stone already has an apparent diameter somewhat equal to the one with 4 to 5% higher diameter.

Step 2: Move to low light environment: Agreement is that the bad cut CZ seems to shrink enormously. The difference in apparent diameter suddenly becomes huge.

Steo 3: We communicate that actual diameter is higher for the bad stone. 6.25 versus 6.00. Wow, simply comparing diameters is not sufficient.

Step 4: We communicate that the bad stone is like a 1Ct diamond, the other like a 0.80 Ct diamond. Wow, this blows up the C of Carat Weight.

Step 5: Now put this in dollar-terms. A well cut 0.80 Ct is far cheaper than a bad 1Ct.

Step 6: Now, let's compare real diamonds. For instance, this well-cut 1.25 versus a random 1.50.

Live long,

 

[Garry H (Cut Nut)]

Step 4: We communicate that the bad stone is like a 1Ct diamond, the other like a 0.80 Ct diamond. Wow, this blows up the C of Carat Weight.

Step 5: Now put this in dollar-terms. A well cut 0.80 Ct is far cheaper than a bad 1Ct.
Great methodology Paul
And BTW on rap 70-89 price list with a 10% uplift for +80 premium, the 80 pointer is exactly half the list price for a 1ct G VS2

Live long,

 

[OoohShiny]

This thread is an excellent discussion, thank you very much to everyone contributing

I feel I have little to add from my noob standpoint but I was wondering...

- Why, when making it part of HCA, reduce it to 4 or 5 categories (Good, Less good, Bad, Horrible, ...)? Why not give the calculated number of apparent diameter (surface?), possibly compared to a standard of 100?

Could/Should a metric such as this focus on the volume of light return? or the (relative?) brightness of the light return? or are both of those impossible to ascertain without a stone having an AGS-style scan undertaken?

I guess I'm pondering if (in terms of the volume of light aspect) such a metric would end up giving, say, 100% scores to stones with near-total light return but no contrast patterns, which is not what we would want to see (because boring!).

In terms of brightness, would a (relative) brightness metric it go some way to helping Serg's quest to illustrate the benefits of improved contrast and therefore improved (perceived) brightness, rather than having boring stones with low contrast?


I don't know... I'm not sure I've really thought this through and have no real conclusion to this post but I thought I'd mention it in case it prompted any useful discussion. If not, please carry on as you were!!

 

[Paul-Antwerp]

This thread is an excellent discussion, thank you very much to everyone contributing

I feel I have little to add from my noob standpoint but I was wondering...


Could/Should a metric such as this focus on the volume of light return? or the (relative?) brightness of the light return? or are both of those impossible to ascertain without a stone having an AGS-style scan undertaken?

I guess I'm pondering if (in terms of the volume of light aspect) such a metric would end up giving, say, 100% scores to stones with near-total light return but no contrast patterns, which is not what we would want to see (because boring!).

In terms of brightness, would a (relative) brightness metric it go some way to helping Serg's quest to illustrate the benefits of improved contrast and therefore improved (perceived) brightness, rather than having boring stones with low contrast?


I don't know... I'm not sure I've really thought this through and have no real conclusion to this post but I thought I'd mention it in case it prompted any useful discussion. If not, please carry on as you were!!

Hi OoohShiny,

I think that we are talking about something else or at least more than light return or brightness.

This is what happens with diamonds:
- A diamond with a physical diameter of 6.0 mm. may appear to have a diameter of 6.3 mm. in most light environments.
- Another diamond with a physical diameter of 6.3 mm. may appear to have a diameter of 6.2 mm in certain lighting, a diameter of 4.5 mm. in other lighting.

That is a well-known fact, and Garry years ago developed a CZ-set to demonstrate this.

Now, Garry has apparently developed this further into a metric, probably weighing certain light conditions to come to some kind of average.

Wouldn't it be cool if all diamonds would come with that calculated 'perceived diameter' (or surface), as per Garry's system?
Wouldn't it be cool if the PS-search offered the possibility to select or search on the basis of 'perceived diameter', much less on the basis of 'carat weight'?

I for one thinks it is a perfect case of consumer protection to do this.

Live long,

 

[Garry H (Cut Nut)]

Hi OoohShiny,

I think that we are talking about something else or at least more than light return or brightness.

This is what happens with diamonds:
- A diamond with a physical diameter of 6.0 mm. may appear to have a diameter of 6.3 mm. in most light environments.
- Another diamond with a physical diameter of 6.3 mm. may appear to have a diameter of 6.2 mm in certain lighting, a diameter of 4.5 mm. in other lighting.

That is a well-known fact, and Garry years ago developed a CZ-set to demonstrate this.

Now, Garry has apparently developed this further into a metric, probably weighing certain light conditions to come to some kind of average.

Wouldn't it be cool if all diamonds would come with that calculated 'perceived diameter' (or surface), as per Garry's system?
Wouldn't it be cool if the PS-search offered the possibility to select or search on the basis of 'perceived diameter', much less on the basis of 'carat weight'?

I for one thinks it is a perfect case of consumer protection to do this.

Live long,

Thank Shiny and Paul,
I hope to live up to all those goals Paul.
Initially using the only available info - the rounded and incomplete data from GIA etc reports - there is no option than to do as we do with HCA.
That means provide a rejection tool.
But in the longer run I would love to provide a tool to manufacturers to run scans on any shaped diamond and provide a reliable perceived dimension.
Of course such a tool could be used in planning the cut to choose for an individual piece of rough!

This would also go some way to providing a rather effective back door cut quality tool for fancy shaped diamonds.

 

[OoohShiny]

Hi OoohShiny,

I think that we are talking about something else or at least more than light return or brightness.

This is what happens with diamonds:
- A diamond with a physical diameter of 6.0 mm. may appear to have a diameter of 6.3 mm. in most light environments.
- Another diamond with a physical diameter of 6.3 mm. may appear to have a diameter of 6.2 mm in certain lighting, a diameter of 4.5 mm. in other lighting.

That is a well-known fact, and Garry years ago developed a CZ-set to demonstrate this.

Now, Garry has apparently developed this further into a metric, probably weighing certain light conditions to come to some kind of average.

Wouldn't it be cool if all diamonds would come with that calculated 'perceived diameter' (or surface), as per Garry's system?
Wouldn't it be cool if the PS-search offered the possibility to select or search on the basis of 'perceived diameter', much less on the basis of 'carat weight'?

I for one thinks it is a perfect case of consumer protection to do this.

Live long,

Hi Paul,

Thanks for your reply

I think I was tilting at something but not describing it very well lol

To play devil's advocate, and being aware that 'gaming the system' is already in full flow with the GIA XXX 'golden standard' being hit by 60+% of GIA-graded stones, can the proposed perceived diameter metric take into account 'fish eye' MRB stones that have poor light performance in the centre but may look quite good in terms of outer diameter? Or are 'fish eye' stones always small diameter for their weight due to being steep/deep and have poor edge-brightness? (Does the 'fish eye' effect happen with shallow/shallow stones?)

I guess I'm wondering if MRBs can have the 'bright edges, dark centre' that Rose Cuts can have!


These are likely 'noob' questions but my mind is taken up with other things at this moment in time! lol

 

[Paul-Antwerp]

Thank Shiny and Paul,
I hope to live up to all those goals Paul.
Initially using the only available info - the rounded and incomplete data from GIA etc reports - there is no option than to do as we do with HCA.
That means provide a rejection tool.
But in the longer run I would love to provide a tool to manufacturers to run scans on any shaped diamond and provide a reliable perceived dimension.
Of course such a tool could be used in planning the cut to choose for an individual piece of rough!

This would also go some way to providing a rather effective back door cut quality tool for fancy shaped diamonds.

Hi Garry,

How about the following idea.

If the supplier includes a full scan of the diamond, calculate the perceived diameter or surface according to your system, and list it. Add a note that your calculation is based upon full data.
If the supplier does not give a full scan of the diamond, calculate the best possible perceived diameter or surface according to the GIA-data. Add a note that this calculation is based upon incomplete data, that it is the very best possible, but needs to be checked further.

Live long,

 

[Garry H (Cut Nut)]

Hi Garry,

How about the following idea.

If the supplier includes a full scan of the diamond, calculate the perceived diameter or surface according to your system, and list it. Add a note that your calculation is based upon full data.
If the supplier does not give a full scan of the diamond, calculate the best possible perceived diameter or surface according to the GIA-data. Add a note that this calculation is based upon incomplete data, that it is the very best possible, but needs to be checked further.

Live long,

That is exactly the plan Paul, but starting with just the GIA AGS IGI etc data.
Until there is some traction (and the patent is granted) it is too costly to develop the online software for probably very little use.

 

[Paul-Antwerp]

That is exactly the plan Paul, but starting with just the GIA AGS IGI etc data.
Until there is some traction (and the patent is granted) it is too costly to develop the online software for probably very little use.

And this will become part of the PS-search? Really looking forward to that.

Live long,

 

[SueAnne]

This whole thread is confusing, presumable Garry will provide an "Adjusted Diameter" or "Adjusted Surface Area" for round diamonds based on his opinion about certain CA/PA combos looking smaller or larger than actual Diameter.

The so called GIA XXX 'Steep Deeps' may be double penalized(they already are in Diameter), and the ones closer the 'ideal bullseye' may receive a bump in 'effective diameter'.

Am I understanding what you are attempting to do Garry?

If this is correct perhaps a table of 'modifiers' or 2d map of ranges and penalty/rewards would be instructive.

 

[Paul-Antwerp]

Hi SueAnne,

Let me try to clarify this. In essence, this goes back to articles Garry published way back, as far back as last century, and it is not merely an opinion, it is an observation which can be repeated by unprepared consumers.

So, there are three potential ways to look at the size-aspect of a diamond.
- Weight (which is the method, driving price)
- Diameter (as measured)
- Perceived diameter (the way a consumer perceives the diameter)

Like you already say, steep-deeps are classic examples where weight does not get translated into a physical diameter. But when it comes down to perceived diameter, it is even worse.

Garry's studies about this are proven by a simple CZ-set, with a poorly cut 1Ct-stone with 6.3 mm physical diameter. Next to that, we put a well cut 0.80 Ct-stone with 6.0 mm physical diameter. In store-lighting, unprepared observers see both having the same perceived diameter, thus the 0.3 mm diameter difference is not perceived. In a low light environment, the biggest and heaviest stone appears to suddenly become half the diameter.

This is an exercise we show on a daily basis in our stores, putting a 1.25 Ct next to a 1.50 Ct, or a 1.70 next to a 2.00.

The consequence of all this comes down to a conviction: The industry is organized in such a way that it sells weight. However, I (and probably Garry, and many others) are convinced that the consumer would really like to buy perceived diameter.

If our conviction is proven correct, Garry's attempts to make this information more clear are a classic example of consumer-education and consumer-protection. Every consumer being correctly informed about apparent size can save a lot of money in their purchase, while increasing the joy of their purchase.

I hope this kind of un-confuses this topic.

Live long,

 

[SueAnne]

Paul thank-you for the reply but my post was way beyond the general concept which I understand, and more an inquiry into the technical details.

The devil is in the details, the example you stated of 0.3 carats or 30% is an extreme one, and I am more interested in seeing an overview of the whole penalty/reward system. We don't often choose between awful cut and ideal cut, that is best used for marketing and sales. The choices usually contemplated on this forum are far less extreme.

For example GIA XXX 60/60 versus a Superideal, what would be Garry's calculated 'perceived spread' difference?

How about a 34.5/40.8 versus 35.5/40.6 versus 33.5/40.7 or even one of the more shallow diamonds Garry has an affinity for.

 

[Garry H (Cut Nut)]

Paul thank-you for the reply but my post was way beyond the general concept which I understand, and more an inquiry into the technical details.

The devil is in the details, the example you stated of 0.3 carats or 30% is an extreme one, and I am more interested in seeing an overview of the whole penalty/reward system. We don't often choose between awful cut and ideal cut, that is best used for marketing and sales. The choices usually contemplated on this forum are far less extreme.

For example GIA XXX 60/60 versus a Superideal, what would be Garry's calculated 'perceived spread' difference?

How about a 34.5/40.8 versus 35.5/40.6 versus 33.5/40.7 or even one of the more shallow diamonds Garry has an affinity for.

Hi Sue Anne, and thankyou Paul for excellent answers.
Firstly SueAnne, I have a provisional lodged patent, which I aim to finalise in the coming months. Once that is lodged and becomes public there will be loads of information available.
And before then, in the coming weeks, we will have the first working model up live on Pricescope as an addition to HCA. So you will be able to play and receive answers directly yourself.
Thirdly, your 60/60 example, assuming good crown and pavilion relationships, would most likely have a larger spread in mm's and because the larger table it could have a slightly larger perceived size than a Tolkowsky. But not by as much as the spread enhancement.