8 TIPS FROM OUR OPTICAL EXPERIMENTS

Over the past dozen posts, we tested a long list of traditional gem-cutting rules of thumb against 3D optical physics: how vertical girdle walls reshape light, what really drives rainbow fire, why some outlines hold their contrast in low-RI rough, and what happens when your faceting mast drifts by a fraction of a degree.

Detailed experiments are fun to read, but when you sit down at a faceting machine or open ProFacet Studio to design a stone, you just want clear, practical rules of thumb that hold up at the bench. Here is a quick cheat sheet of eight practical tips distilled from those studies, with a short explanation of why each one works and a link to the full post if you want to dive deeper.

1. Cut A 2% Girdle On Brilliants (and 4% To 5% On Step Cuts)#

Girdle thickness creates a tug-of-war between raw light return and visual pattern. Pure brightness is highest at a 0% razor knife-edge in almost every stone because a vertical wall does no work to bend light upward. Yet internal contrast, tonal richness, and sparkle climb to a peak between 2.5% and 4% because a thin vertical girdle acts like a ring of side mirrors that slices coarse reflections into a finer mosaic while giving rainbow fire room to spread out.

Combining both forces puts the all-around optical sweet spot right at 1.5% to 2.5% of stone width on almost all brilliant and fancy cuts (about 0.15 mm on an 8 mm stone). The one big exception is rectangular step cuts, which climb all the way to a 3% to 5% girdle sweet spot because vertical separation keeps the parallel crown and pavilion step ladders from blending into sleepy bands. Above 4% on a brilliant, every extra millimeter of girdle is just hidden belly weight that drains brightness and opens a tilted fishbowl ring around the rim.

Read the full experiment: The Girdle Height Paradox: Brightness, Contrast, Entropy, and Fire.

2. Shrink Large Flat Tables When You Want Rainbow Fire#

Why do classic stones with a wide, flat top facet often look purely white and silver when viewed straight on, only throwing rainbow sparks around their outer rim? A horizontal 0-degree table behaves just like a flat windowpane head-on: rays entering and exiting perpendicular to a flat top surface experience almost zero refraction and zero wavelength separation. Expanding a flat table past 65% of the stone's width crushes rainbow fire by 15% and cuts pinpoint sparkle nearly in half.

If you want rainbow fire right across the center of the crown without waiting for the wearer to tilt their hand by 20 degrees, shrink the flat table toward 25% to 40% of stone width (or replace it with a multi-facet dome) and surround it with two or three tiers of angled crown bevels between 28 and 36 degrees. Those sloped upper facets force exiting rays to cross near the crystal's critical angle, where Snell's law stretches red and violet light wide apart.

Read the full experiment: Playing with Prisms: What Our GPU Fire Detector Revealed About Gemstone Dispersion.

3. Use Pavilion Depth And Crown Height As An Internal Lever Arm For Fire#

Beginners often assume that rainbow fire is dictated strictly by mineral chemistry (which is why Cubic Zirconia at 0.060 dispersion out-fires Diamond at 0.044 by nearly 60% when cut to the exact same diagram). Yet cut geometry alone creates a three- to five-fold spread in fire within the exact same mineral.

The two strongest geometric drivers of fire are crown height and total depth. When white light enters an angled crown facet, the initial angular split between red and blue is tiny, often less than a single degree. If the stone is shallow, those rays exit before they have room to separate and blend right back into white light on your retina. Giving a cut a 60% to 72% total depth and a 16% to 19% crown height (especially in 3-fold trilliants, ovals, and polygons where rays cross-bounce diagonally across the belly) acts as an internal optical lever arm that turns a tiny angle split into saturated flashes of color without sacrificing face-up brightness.

Read the full experiment: Playing with Prisms: What Our GPU Fire Detector Revealed About Gemstone Dispersion.

4. Pick Odd-symmetry Outlines And Cushioned Walls For Low-ri Rough#

A gemstone's base outline shape governs its optical performance far more strongly than its facet count. Across everything from Quartz (RI 1.54) up to Moissanite (RI 2.67), three-, five-, and seven-fold outlines (triangles, shields, pentagons, and heptagons) achieve up to triple the elite optical performance of traditional four- and eight-fold shapes.

In an even-symmetry stone, pavilion facets sit 180 degrees opposite each other like parallel trough walls, which can leak under tilt in low-RI quartz or bounce your own dark head shadow straight back at you in a solid block. In an odd-numbered polygon, every flat wall faces a sharp corner across the culet, forcing light to corkscrew around non-opposing wedges as 3D skew rays. Better still, bowing flat sides outward into cushions and chamfering sharp corners starves leaky corner tips and multiplies elite optical performance up to sixfold on pentagons and nearly fourfold on heptagons.

Read the full experiments: Brightness, Contrast, and Base Shape and From Slide Rules to 3D Rays.

5. Do Not Judge Contrast By Counting Blue Ink In A 0-degree Aset Map#

On online faceting forums, a popular shortcut for grading contrast is counting the percentage of blue pixels (observer head shadow) inside a static 0-degree face-up ASET diagram. In practice, static 0-degree ASET Blue correlates slightly negatively (-0.17) with real 3D contrast across tilt.

Why does counting blue ink fail? Because area is not perimeter, and a frozen face-up snapshot is not a moving hand. Flooding a third of a step-cut table with one giant central puddle of blue ink produces a high blue percentage with almost zero bright-to-dark facet boundaries, reading to the human eye as dark central extinction. Conversely, braiding less than 10% ASET blue across dozens of small, non-opposing facets surrounds every dark reflection with bright sky-lit neighbors, matching Diamond Hearts and Arrows contrast even inside ordinary rock crystal quartz.

Read the full experiment: Making Contrast Computable: Sobel Gradients, Density vs. Intensity, and the ASET Blue Myth.

6. Stagger Crown Angles To Break Up Lockstep Ring Flashes#

Testing stones under a dome of discrete overhead spotlights across tilt answers two long-standing bench questions about sparkle. First, flat table glare is just window-pane glare that washes out the center of the stone, whereas tilted crown facets around the rim contribute about a tenth of real sparkle as crisp surface glints on top of the 90% coming from internal pavilion virtual facets.

Second, why do swirl ovals and mirror-trilliants out-sparkle a classic Hearts and Arrows round brilliant by 40%? In a traditional round brilliant, all eight main crown and pavilion facets sit at identical angles around a circle, so whole concentric rings wink on and off at the exact same moment. Staggering crown bevels across slightly different angles (for example, stepping gently from 19 to 35 degrees around a spiral crown, or pairing staggered roof tiers on a trilliant) breaks that lockstep so the slightest movement of your wrist sets off an independent cascade of pinpoint flashes.

Read the full experiment: From Pixel Blinks to Virtual Facets: How Scintillation V3 Measures Gemstone Sparkle.

7. Pair Edge Contrast With Tonal Entropy To Avoid The Barcode Trap#

Maximizing only brightness and sharp bright-to-dark edge contrast can produce crystals that look strangely metallic and flat, flashing strictly between pitch-black voids and blown-out white glare like a harsh QR code. Neither brightness nor edge contrast asks how many intermediate shades of grey exist inside the gem.

Measuring those intermediate shades with Claude Shannon's information entropy (across 32 perceptual luminance bins) reveals two practical rules: tonal richness costs zero brightness, and fine Contrast Density lifts entropy while extreme Contrast Intensity pulls it down. When you want a velvety, liquid-silver look in heptagons, hexagons, or cushions, raise Shannon Entropy alongside Contrast Density so two-thirds of the stone populates the middle pewter and dove-grey buckets (resolving 23 to 24 simultaneous shades). When you want bold, graphic piano-key flash in a compact trilliant or step cut, intentionally lower entropy and crank up Contrast Intensity.

Read the full experiment: All Thirty-Two Shades of Grey: Bringing Shannon Entropy to Gemstone Design.

8. Do Not Sweat 0.2-degree Drift For Brilliance (stay Off The Critical-angle Cliff Instead)#

Newcomers often worry that if their mast flexes or their protractor drifts by two or three tenths of a degree, the finished gemstone will be ruined, especially after hearing that a classic Hearts and Arrows diamond loses its eight viewer-scope arrows at just 0.15 degrees of error. In reality, "forgiveness" means two completely different things:

• Scope-pattern alignment is fragile, but 3D room-light brilliance is remarkably sturdy: Even when a 0.35 or 1.0 degree drift causes the eight dark head-shadow arrows in a viewer scope to shorten and detach from their arrowheads, classic Hearts and Arrows still keeps over 98% of its full 3D white brilliance and sparkle. Because its 40 to 41 degree pavilion sits more than fifteen degrees safely above diamond's 24.4 degree critical angle, a slightly shifted ray simply trades one total-internal-reflection bounce for another without ever leaking out the back.
• The real bench killer is the critical-angle cliff (especially under tilt): Where a 0.3 to 0.5 degree angle drift genuinely ruins a stone is when a diagram parks a low- or mid-RI pavilion (like Quartz at a 40.5 degree critical angle, or Beryl at 39.3 degrees) right on the edge of its escape cone to chase a flattering 0-degree face-up score. Drift half a degree shallow at the bench, or tilt the finished ring 7 degrees on a hand, and the center drops straight off the critical-angle cliff into a see-through window. Always pick a cut that sits on a wide, multi-angle optical plateau with a comfortable safety margin above the mineral's critical angle.

Read the full experiments: Optical Plateaus and Knife-Edges: What Our Sensitivity Study Reveals About Cut Forgiveness and From Slide Rules to 3D Rays.