Scoring Model

The score is a weighted sum of well-defined metrics. Contrast density and contrast intensity combine into a single contrast term because strong contrast requires both; the remaining terms are linear for clarity. The Window metric is inverted before scoring (lower window = better score). ASET color targets use a deviation model — the optimizer minimizes the difference between actual and desired color percentages.

What goes in

• Metrics — Average Brightness, Contrast Density, Contrast Intensity, Scintillation Score, Compactness, Shannon Entropy, Window, and three ASET color percentages (Blue, Red, Green). The first seven are compared to the Hearts & Arrows reference; ASET colors are compared to user-defined target percentages.
• Weights — The ten sliders (averageBrightness, contrastDensity, contrastIntensity, scintillationScore, compactness, shannonEntropy, window, asetBluePct, asetRedPct, asetGreenPct). Weights are always treated as zero or higher. ASET weights default to 0 (disabled).

How the score is built

1. Normalize the metrics
   Everything is turned into a ratio vs the reference. If a metric is better than the reference, we keep it above 1.0 but gently squash it so runaway values don't dominate.

2. Make contrast a single value
   Contrast density and intensity are blended with a geometric mix (think "both must be good, one can't carry the other"). If both contrast sliders are equal, this behaves like the square root of (density × intensity). If one slider is higher, that side is favored — but a weak partner still drags the contrast term down.

3. Set the overall weights
   The two contrast sliders also decide how important contrast is overall: their average becomes the contrast weight. Brightness, scintillation, compactness, and contrast weights are then scaled so they always sum to 1 (extras are included the same way).

4. Take the weighted sum
   Final score = brightness part + contrast part + scintillation part + compactness part + entropy part + window part. Higher numbers mean closer to, or better than, the reference cut.

The Window metric

Window measures how much light passes straight through the stone without internal reflection — a "hole" in the optical performance. It deserves special attention because it behaves differently from every other metric.

Why Window is inverted

For all other metrics, higher is better: you want more brightness, more contrast, more scintillation. Window is the opposite — you want less of it. A perfect stone has zero window.

To make Window compatible with the scoring system ("higher score = better stone"), the engine internally flips the value before scoring:

scored value = 1.0 − raw window fraction

A stone with 0% window (perfect) becomes 1.0 — exactly like a metric that matches the reference. A stone with 20% window becomes 0.8, which is scored as "only 80% of perfect." This lets the optimizer treat all seven metrics uniformly: push everything toward 1.0 or higher.

How to read the numbers

The optimizer normalizes every metric against the Hearts & Arrows reference before displaying it. For Window, this means:

A Window value of 1.000× in the optimizer table means "as good as H&A." Values below 1.0 mean the design leaks more light. A value of 0.606× means the design retains only about 60% of the optical integrity that the reference achieves.

Using Window with other metrics

Because the reference stone already has a near-perfect window (~1%), the normalized Window score saturates quickly — improvements above ~0.98× are very small in absolute terms. In practice this means:

• If your design already has a small window, the Window weight contributes close to 1.0 and there is little room to improve. Set the weight lower and focus on the metrics that still have headroom.
• If your design has a significant window (say 10–30%), increasing the Window weight will strongly guide the optimizer to close the leak, and you'll see large delta percentages.
• Balanced presets typically set Window to 0.5 alongside the other metrics. This is a good default unless you know your cut style inherently produces a window (e.g., some step cuts).

ASET Color Targets

The Angular Spectrum Evaluation Tool (ASET) analyzes how light enters a gemstone from different angles. ProFacet renders a top-down ASET view and classifies every pixel into one of four colors, each representing a different range of light angles measured from the vertical (the observer's line of sight):

How ASET targets work

Unlike the other seven metrics (where higher is always better, relative to a reference), ASET scoring uses a target percentage model. You specify how much of the stone's face-up area should show each color, and the optimizer tries to match those targets.

The optimizer panel shows three ASET weight sliders with target inputs:

1. ASET Blue % — editable target (default: 20%)
2. ASET Red % — editable target (default: 60%)
3. ASET Green % — auto-computed as 100 − Blue − Red (shown read-only)

The ASET Black percentage is not a target — it is simply reported. A well-cut stone should have minimal black.

ASET scoring formula

For each ASET color, the optimizer computes:

aset_score = 1.0 − |actual% − target%| ÷ 100

This means:

• A perfect match (actual = target) scores 1.0
• Every percentage point of deviation subtracts 0.01 from the score
• The score bottoms out at 0.0 (100% deviation)

The ASET score contribution is then weighted and added to the total score like any other metric:

total += (aset_weight ÷ weight_sum) × aset_score

Worked example

Suppose you set Blue target = 20%, Red target = 60% (so Green target = 20%), and all three ASET weights to 0.3. If the stone measures Blue = 18%, Red = 55%, Green = 22%:

Each contributes 0.3 / weight_sum × score to the final total. The optimizer will nudge facet angles to bring the actual percentages closer to the targets.

Typical ASET targets

Good brilliant-cut stones typically show roughly:

• Blue: 15–25% — enough obscuration for contrast
• Red: 50–65% — dominant brilliance
• Green: 15–25% — moderate environmental contribution
• Black: < 5% — minimal light leakage

These defaults match the industry standard for a well-performing round brilliant. Fancy cuts (emerald, cushion, etc.) may have different characteristic distributions.

When to use ASET weights

• Set ASET weights to 0 (the default) if you want the optimizer to focus purely on the optical metrics (brightness, contrast, scintillation).
• Set ASET weights > 0 when you want to steer the stone toward a specific light-performance profile — for example, maximizing red (brilliance) while keeping blue (contrast) in a specific range.
• ASET weights combine naturally with the other metric weights. The optimizer normalizes all weights together, so an ASET weight of 0.3 alongside brightness at 0.4 means ASET has a proportional influence.

Notes

• The contest verifier, Optimizer, and UI all share this exact formula, so what you see is what gets verified.