The 1919 architecture solved a second time at n = 2.65, because the angles that produce the pattern in diamond stop producing it in moissanite.
Global score against T57, Tolkowsky’s 1919 ideal, measured the same way in the same material.
Diamondn = 2.417−0.22Behind
Moissaniten = 2.65+1.11Ahead
Cubic zirconian = 2.16−2.79Behind
−3T57 = 0+3
The short version
Every proportion standard in the trade is a diamond standard. Tolkowsky solved his angles for n = 2.417, and the whole of the modern round brilliant follows from that one number. Cut the same stone in moissanite at n = 2.65 and the geometry is unchanged but the physics is not: rays leave the pavilion at different angles, and the face-up pattern the design exists to produce comes apart. BL-Anteros keeps the 73-facet architecture and the 53% table, and re-solves the angles against the higher index.
It works, on its own terms. Against the diamond reference light map, moissanite cut to the reference angles agrees 55.4%; BL-Anteros agrees 80.2%, with the arrow budget back to 21.9% against the reference’s 21.8%. On Global, the registry’s own scoreboard, the result is more modest: +1.11 in moissanite, and behind the standard in diamond and cubic zirconia. Both readings are published here, because they are measuring different things.
The ledger
Where the points come from
Global is a weighted sum of six terms. Subtract the standard's figure from this cut's,
term by term, and the six differences add back up to the published gap. Nothing is hidden in the total,
so you can see exactly which part of the stone is doing the work.
In diamond n = 2.417
Behind
88.39 against the standard's 88.61, and the narrowest
deficit in the registry. The eased crown buys 2.3 points of brilliance and the 73-facet band
8.5 of scintillation, worth +0.78 and +0.85; fire gives back −0.92 and tilt −0.88.
The stone is not aimed at this material and very nearly holds the standard anyway.
TermThis cut / standardPoints
Brillianceweight 0.3494.692.3+0.78
Fireweight 0.2074.278.8−0.92
Tiltweight 0.1684.990.4−0.88
Scintillationweight 0.1080.371.8+0.85
Symmetryweight 0.10100.0100.0+0.00
Leakweight 0.10 ↓2.301.88−0.04
Sum of the six terms−0.21
Published Global gap −0.22. The six terms reconstruct it to
0.01 of a point. The difference is rounding in the source figures, not a second method.
In moissanite n = 2.65
Ahead
90.93 against the standard's 89.82. This is the
material it was drawn for: brilliance +6.6 and leakage cut from 7.76% to 2.76%, worth +2.24 and
+0.50 between them. Fire and tilt give back −1.28 and −1.20, which is what re-aiming
a pavilion cone away from maximum dispersion costs.
TermThis cut / standardPoints
Brillianceweight 0.3491.985.3+2.24
Fireweight 0.2091.197.5−1.28
Tiltweight 0.1685.793.2−1.20
Scintillationweight 0.1080.371.8+0.85
Symmetryweight 0.10100.0100.0+0.00
Leakweight 0.10 ↓2.767.76+0.50
Sum of the six terms+1.11
Published Global gap +1.11. The six terms reconstruct it to
the last hundredth of a point.
In cubic zirconia n = 2.16
Behind
88.98 against the standard's 91.77, the widest deficit on this page.
At n = 2.16 the critical angle opens to 27.6° and a pavilion cut for 22.2° stops holding
light: leakage goes from 0.51% to 6.55% and brilliance falls 5.7 points. Nothing in the design
was asked to survive this.
TermThis cut / standardPoints
Brillianceweight 0.3492.398.0−1.94
Fireweight 0.2081.786.8−1.02
Tiltweight 0.1686.987.1−0.03
Scintillationweight 0.1080.371.8+0.85
Symmetryweight 0.10100.0100.0+0.00
Leakweight 0.10 ↓6.550.51−0.60
Sum of the six terms−2.74
Published Global gap −2.79. The six terms reconstruct it to
0.05 of a point. The difference is rounding in the source figures, not a second method.
Why the standard fails here
One ray, and the boundary it lands on
The 1919 pavilion is aimed so that light entering the table leaves the crown after two internal reflections at an elevation of 45.0°. That figure is not incidental. It sits exactly on the boundary an ASET scope draws between the steep light a stone gathers from overhead and the shallow light it gathers from the room, which is why a well cut diamond reads as a clean pattern rather than a wash.
What changes at n = 2.65
Snell’s law bends that exit ray further from vertical in the denser material. The same geometry now puts it at 39.2°, below the line, and the centre of the stone changes colour on the scope without a single facet having moved.
What that costs
Moissanite cut to the reference angles agrees with the diamond reference light map on 55.4% of pixels. Prime red falls 71.6 to 56.3, green rises 5.6 to 31.2, and the hearts-and-arrows shafts lose their black zone, 21.8% to 11.1%.
What the correction is
Ease the crown and shift the pavilion cone: bezel 34.5° to 33.0°, star 23.07° to 26.0°, upper girdle 40.58° to 43.0°, pavilion main 40.75° to 40.10°, lower half 41.94° to 41.20°, with the mains run in to 0.55 of the girdle radius.
What it recovers
Agreement returns to 80.2%, the blue arrow budget to 21.9% against the reference’s 21.8%, and the arrows agree 95.0%. Those four figures come from the design engine’s light-map comparison, not from the Global sweep above.
Where it departs from 1919
The crown moves. The pavilion barely does.
The registry’s standing finding is that the house cuts do not leave Tolkowsky’s pavilion: they depart on the crown, because the pavilion is pinned by the critical angle and the crown is not. BL-Anteros is the cleanest example of it, and the only one where the reason is a change of material rather than a change of facet plan.
Pavilion: −0.65°
The main goes 40.75° to 40.10°. Moissanite’s critical angle is 22.2° against diamond’s 24.4°, so there is more room to work in, and the design still spends almost none of it.
Crown: three tiers, all moved
Bezel −1.5°, star +2.93°, upper girdle +2.42°. The crown carries the correction, and the star is the largest single move on the stone.
Depth: 59.4%
Against the 73-facet reference’s 61.7%. A shallower stone at the same 53% table and the same 8-fold, 73-facet plan.
The table is untouched
53.0%, exactly the 1919 figure. Nothing was gained by moving it, and holding it keeps the comparison honest: the two stones differ in angle, not in plan.
What Global does not measure
T73 scores higher in moissanite than the cut drawn for it
T73 is the same architecture at the reference angles, and it finishes +1.22 in moissanite against BL-Anteros’s +1.11. That is not a defect in either stone. Global is a weighted sum of light return, fire, tilt, scintillation, symmetry and leakage; it does not contain a term for whether the returning light forms a pattern, and pattern is the whole brief here. A cut solved to a light map and a cut solved to a score will not agree, and this page publishes the disagreement rather than choosing the flattering half of it.
Within Global, what the re-solve actually buys in moissanite is brilliance and leakage: 91.9% useful against 85.3%, and 2.76% lost to leakage against 7.76%, the second lowest figure in the registry in this material. What it spends is fire and tilt, −6.4 and −7.5, which is the price of aiming a pavilion cone at a pattern instead of at maximum dispersion.
Known limits
What is still wrong with it
Green runs high
11.8% of the face-up map against the reference’s 5.3%. This is the chief remaining fidelity defect and it is not fixed in this version.
Blunt arrow tips
The reference’s arrows taper to points; these do not. An 8-fold single-cone pavilion is rotationally uniform near the apex, so the centre fills solid. Pointed tips need a stepped culet-cap tier and a generator rebuild.
Tilt
87% of face-up useful light is still returned at 20° and 74% at 30°. Tilt is the term this design spends, and it is the honest weakness of it.
Not modelled
Moissanite’s birefringence, 0.043, is not in the engine. It will soften real scope images off axis in a way none of these figures predict. Figures here are engine estimates, not lab-certified grades.
The rest of the registry
Ahead of the standard in moissanite. Behind it in diamond and cubic zirconia.