Computational optics studio · Founded 2026
Brilliani Labsthe mathematics of light
We work on one thing: the optical ceiling set by the Tolkowsky Ideal Cut (1919), still the benchmark the trade ships against a century later, and how to measure it, then move it.
We don't manufacture stones. We manufacture certainty, and publish it, clear.
What we are
Brilliani Labs sits at the intersection of applied physics, Monte-Carlo ray-tracing and precision facet geometry. The lab exists to answer a question the industry stopped asking: is the 1919 ideal actually the ceiling, or just the last place the mathematics was left?
Our answer is a dataset, not an opinion. Every cut we publish is traced from its exact facet geometry and scored against the same 1919 benchmark, in the same material, by the same engine, so the comparison holds whoever runs it.
The discipline
Pareto-frontier designers, and we mean it literally.
A Pareto frontier is the set of designs where you can't improve one property without spending another. Most of gemstone cutting lives inside that frontier, trading blindly. The discipline is knowing exactly what each cut trades, and refusing to hide it.
Some of our cuts buy fire and spend it in light leakage. One is solved for moissanite and gives ground in diamond. We show the trade every time, term by term, because a frontier you can't see isn't a result, it's a slogan.
And where the geometry allows, the frontier moves. BL-Miliano, the flagship cut, clears the 1919 standard on every measured dimension at once in diamond: more light returned, more fire, steadier under tilt, nothing traded away. That is the exception, not the rule. Which is exactly why we publish the rule.
The finding that matters
Every material now has a cut that beats 1919.
Measured on global light performance against the Tolkowsky Ideal Cut (1919), same protocol, same refractive index. Every figure is traced from published geometry: open the page and check it.
The record
The Cut Registry.
The lab's flagship dataset. A unified ray-census run across three refractive indices (diamond, moissanite and cubic zirconia) against reference stones that include the 1919 ideal itself, modelled from 57 facets up to 185.
Every figure is traced from published facet geometry rather than approximated. Every score is reproducible rather than graded by eye. The registry grows as the fleet does.
How we measure
Ray census
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. Useful light exits through the crown within 64° of vertical.
One benchmark
Every cut is scored against Tolkowsky's 1919 ideal, measured the same way in the same material. The comparison is the point; the benchmark never moves between stones.
Read with care
These are engine estimates, ± 2 points, not lab-certified grades. The certainty is in the method and its reproducibility, not in a decimal place.
Born Clear
Traced geometry, published terms, reconciled sums. The brand line means the same thing in the lab as it does on the box: clear origin, clear conduct.