Cut registry › Historic cuts
Historic cuts · six centuries, one protocol
What each step actually bought.
Five cuts rebuilt from Marcel Tolkowsky’s Diamond Design of 1919, from the eight-facet single cut to the old european, and put through the same ray census as everything else on this site. The answer is not the one the history books imply.
The short version
Three centuries of adding facets barely moved how much light a diamond returns. The single cut, the Mazarin and the Peruzzi are modelled on one shared proportion set, so facet count is the only thing that differs between them. Scintillation climbs from 43.89 to 60.68 to 80.69. Global goes 78.79, 74.88, 78.49. It goes nowhere.
The jump is the old european, and it is thinness, not facets. It carries the same 58 historic facets as the Peruzzi before it. Leak falls from 19.24% to 5.25%, and Global rises from 78.49 to 87.82, landing 0.78 short of Tolkowsky’s ideal. What changed was mechanical bruting: for the first time a stone could be made truly round.
And the Peruzzi beat the 1919 standard on sparkle, three hundred years early. 80.69 on scintillation against 71.82. It still finishes ten points behind on Global, which is the whole lesson in one line: facet count is one term of six, and the angles are worth more than all of it.
The line
In chronological order, in diamond. Historic facets is the conventional count, which omits the girdle; measured is the number of facets our model resolves as reaching the surface, and it is the count every scintillation figure is built on. The standard is T57, Tolkowsky’s 1919 ideal.
| Cut | Historic facets | Measured | Useful light | Leak | Scint. | Fire | Global | vs standard |
|---|---|---|---|---|---|---|---|---|
| Old Single from the 1300s | 18 | 26 | 72.11% | 19.30% | 43.89 | 84.19 | 78.79 | −9.82 |
| Mazarin mid-1600s | 34 | 42 | 64.56% | 17.27% | 60.68 | 62.94 | 74.88 | −13.73 |
| Peruzzi early 1700s | 58 | 74 | 74.99% | 19.24% | 80.69 | 68.40 | 78.49 | −10.11 |
| Old Mine c.1700 to 1890 | 58 | 74 | 58.82% | 31.96% | 80.69 | 72.42 | 74.31 | −14.29 |
| Old European c.1890 to 1930 | 58 | 74 | 90.26% | 5.25% | 80.69 | 73.54 | 87.82 | −0.78 |
| T57 1919 | 58 | 57 | 92.28% | 1.88% | 71.82 | 78.85 | 88.61 | 0.00 |
The single cut, the Mazarin and the Peruzzi deliberately share one proportion set. That is a modelling choice, and it is the only way to isolate facet count as a variable. The old mine and the old european each change proportions as well, which is why they move.
The lateral move
Tidying the Peruzzi’s facets into a more uniform, rounder stone lifted fire to 72.42. It also put a 60.3% table over a still-steep 44.4° pavilion, and that leaks: 31.96%, the worst of the six.
That is the nailhead an old mine actually shows, and we can put a number on it: the centre of the stone returns 54.3% against 66.9% at the edges, where a modern round brilliant is flat. The full page carries the caveat that its depths are an interpolation.
Outside the line
The Portuguese cut is a twentieth-century lapidary design, not a step in this lineage, and this section does not pretend otherwise. It is here because at 177 measured facets it is the extreme case.
It scores 98.04 on scintillation, the highest figure we publish anywhere, and still finishes 2.23 points behind a 57-facet stone from 1919. Fire falls to 65.37, because facets that small throw spectra too narrow to separate. More facets is not better. It saturates, and then it costs you.
All 6
Questions
The documented ancestry of the round brilliant runs: the single cut from the 1300s at 18 facets, the Mazarin double cut of the mid-1600s at 34, the Peruzzi triple cut of the early 1700s at 58, the old mine cut that dominated until the late 1800s, and the old european cut of roughly 1890 to 1930. The modern round brilliant follows from Tolkowsky's 1919 proportions. All five are rebuilt here from Diamond Design, pages 21 to 24.
The old european, at 87.82 Global in diamond against the 1919 standard's 88.61, a gap of only 0.78 points. That is closer to the modern ideal than any of the eleven jewellery shapes we publish.
Almost not at all, and this is the clearest finding in the section. Across the single cut, the Mazarin and the Peruzzi, which deliberately share one proportion set so that facet count is the only variable, scintillation climbs from 43.89 to 60.68 to 80.69 while Global stays flat at 78.79, 74.88 and 78.49. Facet count buys sparkle. It does not buy light return.
Being able to make it round. The old european carries the same 58 historic facets as the Peruzzi that preceded it, and the same 74 measured, yet leak falls from 19.24% to 5.25% and Global rises from 78.49 to 87.82. The change was mechanical bruting, which made stones truly circular in plan for the first time. One machine was worth more than three centuries of facet work.
Outline and culet. An old mine has a soft square or cushion outline with a large open culet; an old european is genuinely round with a smaller one. We measure the old european at 87.82 Global against the old mine's 74.31, and the old mine leaks 31.96% against 5.25%.
Because a wide table sits over a steep pavilion, so light entering the middle exits through the bottom rather than returning. The trade calls it a nailhead. Binning every returning ray by where it leaves the crown, our old mine returns 54.3% at the centre against 66.9% at the edges, where a modern round brilliant is flat.
Several do, on the terms facet count controls. The Peruzzi scores 80.69 on scintillation against the 1919 standard's 71.82, three centuries early. The old european beats it on both scintillation and fire. The single cut beats it on fire alone, at 84.19 against 78.85, because very large facets throw wide, well-separated spectra.
No, and the difference matters. The historic count omits the girdle, because a hand-bruted stone had no faceted girdle to count. Our models resolve it, so a cut counted at 58 facets historically measures 74 here, and one counted at 18 measures 26. Every scintillation figure on these pages is calculated from the measured count.
Not in the sense the other five are. It is a twentieth-century lapidary faceting design and it sits outside the brilliant's ancestry. It is published here because at 177 measured facets it is the extreme case for facet count, scoring 98.04 on scintillation, the highest figure on this site, while still finishing 2.23 points behind a 57-facet cut from 1919.
The five ancestry cuts are built from Marcel Tolkowsky's Diamond Design of 1919, pages 21 to 24, including his figures 9, 10 and 11. Each is modelled as a convex intersection of half-spaces, which is the honest model for a faceted gem and returns the plane set the engine measures directly. The Portuguese is built from a published faceting diagram.
Method
Marcel Tolkowsky, Diamond Design: A Study of the Reflection and Refraction of Light in a Diamond, 1919, pages 21 to 24, including figures 9, 10 and 11. The five ancestry cuts are built from that text and those drawings, not from photographs.
Each cut is a convex intersection of half-spaces, which is the honest model for a faceted gem: non-planar faces become impossible and it returns the plane set the engine measures directly.
Monte-Carlo on exact facet geometry: 3,200 rays face-up plus 2,200 at 20° tilt, cosine-weighted from a 7° near-vertical cone, at a fixed seed. Identical protocol on every cut on this site.
Engine estimates, ±2 points. These are reconstructions of documented designs, not surveys of surviving stones, which were cut by hand and by eye and vary enormously. Three of the six read at the tilt scale’s 100 ceiling, which on a light-starved cut means little was left to lose rather than that the stone is stable.
Read what Tolkowsky actually calculated, compare the modern shapes in the jewellery cuts, or turn any of these in the Index.