Cut-Grading Systems and Their Limitations

The word Excellent on two different laboratory reports does not necessarily mean exactly the same thing.

Different systems may evaluate:

• different sets of shapes;
• different ranges of color or clarity;
• different proportions;
• different optical metrics;
• finish;
• visual assessment;
• ray tracing;
• proprietary algorithms.

A cut grade should therefore never be mapped from one system to another solely by the category name.

First Separate the Meanings of the Word Cut

In everyday language, the word cut is used for at least three different things:

1. shape—round, oval, pear, cushion;
2. faceting style—brilliant, step, mixed, modified brilliant;
3. cut quality—how successfully the geometry and finish were executed and what optical result they produce.

This chapter addresses the third meaning, but the terminological confusion must remain visible.

Current terminology note—August 8, 2026.
On July 16, 2026, the CIBJO Diamond Commission published a pre-congress Special Report identifying proposed amendments to the Diamond Blue Book, including a redefinition of the relationship between the terms cut and cut grade. The congress at which the proposed changes are to be considered opens on September 4, 2026, in Vicenza. As of the date of this book, these are therefore proposals within a standardization process, not an already adopted universal definition. The controlling text for existing CIBJO nomenclature remains the Diamond Blue Book then in force.

Why There Is No Single Universal System

Cut is complex because it attempts to summarize several different dimensions:

• geometry;
• optical performance;
• finish;
• durability;
• weight distribution;
• visual preferences.

Systems can therefore be based primarily on:

• proportions;
• human observation;
• 3D geometry;
• ray tracing;
• structured lighting;
• a combination of multiple methods.

Different methodologies do not automatically mean that one system is “right” and another “wrong.” They mean that we must first know what each system actually evaluates.

GIA Overall Cut Grade

For comparison among systems, it is sufficient here to retain the boundary of scope: GIA’s overall Cut Grade applies to the standard round brilliant in the D–Z range, while the detailed logic of its grading is discussed in Chapter 21.

What follows matters more for the present topic: GIA categories, including Excellent, belong to a specific methodology and cannot automatically be mapped to identically named categories from other laboratories. The system does not reduce quality to one “ideal” number, nor does any category represent a universal unit of cut measurement.

Evidence note—internal GIA source conflict, August 8, 2026.
A GIA research review dated June 24, 2026, states in one passage that the overall grade is determined by the lowest result among all seven components. This is inconsistent with current official GIA cut-grade documentation and published grading boundaries, according to which the overall grade follows the limiting result among the five appearance/design components—brightness, fire, scintillation, weight ratio, and durability—while polish and symmetry are graded separately. Very Good polish and/or symmetry can be compatible with an overall Excellent Cut Grade. In this book, the current grading documentation is controlling; the disputed sentence in the research review is recorded as a source conflict, not as a change in methodology.

The AGS Legacy and the AGS Ideal® Report by GIA

AGS Laboratories historically developed a performance-based approach and a scale on which 0 represented Ideal. After AGS Laboratories’ technology and intellectual property were transferred to GIA in late 2022, the AGS Ideal® Report by GIA remained a separate supplemental light-performance layer for qualifying stones.

Detailed current eligibility requirements and their relationship to the GIA round-brilliant grade are discussed in Chapter 21. For comparison among systems, it is enough to remember that an AGS Ideal result is not synonymous with ordinary GIA Excellent, nor is it a universal grade that can be mapped onto other laboratories.

IGI: Round and Fancy Grading

IGI currently uses a different model.

For a round brilliant, it evaluates:

• polish;
• symmetry;
• proportions;
• with additional options for light-performance reporting.

IGI identifies Excellent-Ideal as its highest cut grade for a round brilliant.

For fancy shapes, IGI offers a separate optional overall cut grade from Excellent to Poor, based on a four-stage process:

1. polish and symmetry;
2. qualification of proportions;
3. shape-specific craftsmanship requirements;
4. visual assessment of overall light return.

This is an important update: it is no longer correct to say that major laboratories do not offer any overall cut grade for fancy shapes. The accurate statement is that no single universally accepted fancy-cut grading system exists.

HRD Antwerp: Separate Elements of Cut

In its current 4Cs documentation, HRD Antwerp separates the cut description into three lines:

• proportions;
• polish;
• symmetry.

Each is evaluated as Excellent, Very Good, Good, or Fair.

This is not the same as one unified overall cut grade in the GIA sense. When comparing HRD and GIA reports, it is therefore not valid simply to look for the word Excellent and assume identical meaning.

GCAL by Sarine and 8X

GCAL by Sarine uses the proprietary 8X® Ultimate Cut Grade. The system considers eight areas that the issuer groups as:

• polish;
• external symmetry;
• proportions;
• optical brilliance;
• fire;
• scintillation;
• optical symmetry;
• an additional shape/precision criterion depending on the program.

In the round 8X program, the final criterion has historically included Hearts & Arrows; fancy programs use shape-specific logic.

Factual cutoff—August 8, 2026.
In May 2026, GCAL by Sarine expanded the 8X program to cushion shapes, following earlier programs for round, princess, oval, pear, marquise, and radiant cuts.

This is a useful example of rapid development in cut grading. However, figures such as “less than 1% qualifies for 8X” should be treated as a claim by the system’s issuer, not as independently confirmed universal market statistics.

ISO and CIBJO Are Not the Same as a Laboratory Cut Grade

ISO 24016:2020, confirmed in a systematic review in 2026, establishes terminology, classification, and methods for individual natural, unmounted, polished diamonds over 0.25 ct within its scope.

It does not apply as a universal grading standard to:

• fancy-color diamonds;
• synthetic/laboratory-grown diamonds;
• most treated diamonds;
• assembled stones.

ISO 6893:2024 applies to the inspection of lots of small, unmounted, colorless diamonds with an individual weight of up to 0.25 ct; it is not intended as a claim about one individual stone.

CIBJO primarily provides international terminology and trade/disclosure rules. A standard or Blue Book should not be presented as though it were itself a laboratory that issued a specific cut grade.

Fancy Shapes Remain a Methodological Problem

Fancy shapes show why a universal cut grade is difficult to create.

Two ovals with the same:

• length-to-width ratio;
• depth percentage;
• table percentage

can have very different:

• bow-tie patterns;
• contrast distributions;
• light return;
• ends;
• shoulders;
• windowing;
• behavior when tilted.

A table of proportions alone is therefore insufficient.

GIA currently has no standard overall Cut Grade for fancy shapes, but it conducts research and offers a supplemental AGS Ideal program for certain qualifying facet arrangements. IGI and GCAL by Sarine offer their own fancy-cut systems. These facts are not contradictory; they demonstrate that the market has multiple methodologies without a universal crosswalk.

2D Numbers Do Not Describe the Entire 3D Stone

A laboratory report often has to summarize complex facet geometry as:

• average crown angle;
• average pavilion angle;
• table percentage;
• depth percentage;
• girdle description;
• a few dimensions.

Two stones with the same reported average can have different individual facets.

A 3D scan can contain much more information, but it is not a complete description of beauty either. The result depends on:

• scan quality;
• the algorithm;
• the lighting model;
• the definition of the metric;
• stone transparency;
• surface condition;
• how different zones are weighted.

Modeled and Measured Optics

Ray tracing can model what light should do within a defined geometry and lighting environment.

Structured optical imaging can document the behavior of a specific stone more directly under particular conditions.

Human observation evaluates the visual experience in a real or standardized environment.

These three types of evidence are not identical.

A system that uses ray tracing is not automatically universal merely because it invokes “physics,” just as human assessment is not automatically worthless because it contains a subjective component. Validation of the complete system is what matters.

Proprietary Names Are Not Universal Units

Terms such as:

• Ideal;
• Super Ideal;
• 8X;
• Triple Zero;
• Hearts & Arrows

can be useful within a clearly defined system, but no single international mathematical scale makes all these expressions directly comparable.

A number in a marketing or proprietary name is not a physical unit of measurement.

An Estimator Is Not a Laboratory Report

An online calculator or estimation tool can be useful for:

• screening;
• planning;
• comparing proportions;
• rejecting obvious candidates.

But the result of an estimator should not be presented as a laboratory grade if the tool is not a laboratory system evaluating the specific stone.

GIA Facetware, for example, estimates a possible GIA round cut grade from entered parameters; its result is not a substitute for an official GIA grading report.

Crosswalk Tables Create False Certainty

A table such as:

GIA Excellent = IGI Excellent-Ideal = AGS Ideal 0 = GCAL 8X

is not professionally justified without serious comparative research.

A valid crosswalk would require:

• the same representative set of physical stones;
• current methodologies from every system;
• blind grading;
• controlled conditions;
• repeatability analysis;
• statistics on boundary cases;
• documented mapping of categories.

Without this, a category name is not a conversion formula.

The Highest Grade and Personal Preference

Cut grading attempts to objectify cut quality and its optical consequences. Within a high category, however, buyers may prefer different appearances:

• more fire;
• larger or smaller flashes;
• stronger contrast;
• a more open pattern;
• more or less of a crushed-ice appearance in fancy shapes.

A laboratory system can eliminate poorly executed combinations. It cannot prescribe personal taste down to the last nuance.

Protocol for Comparing Two Cut Reports

1. identify the laboratory;
2. read the exact service name;
3. check the report date;
4. check the shape and facet arrangement;
5. determine whether there is an overall cut grade or only proportions/polish/symmetry;
6. separate measured data from modeled data;
7. read the definition of the highest grade;
8. examine the light-performance section if present;
9. do not map proprietary designations into another system without validation;
10. observe the specific stone under the same conditions.

Chapter Summary

• The word cut can refer to shape, faceting style, or cut quality.
• No single universal cut-grading system exists for every diamond.
• GIA’s overall Cut Grade remains limited to the standard round brilliant in the D–Z range.
• The AGS Ideal® Report by GIA is a supplemental light-performance report for certain qualifying stones.
• IGI currently offers overall cut grading for fancy shapes according to its own methodology.
• HRD Antwerp reports proportions, polish, and symmetry separately.
• GCAL by Sarine uses the proprietary 8X system, expanded to additional fancy shapes in 2026.
• ISO and CIBJO have important standardization functions, but they are not the same as a proprietary laboratory grade.
• 2D averages and 3D models provide different layers of information.
• A crosswalk among the names Excellent, Ideal, Zero, or 8X must not be invented.
• An estimator is not an official laboratory report.
• The highest grade does not eliminate legitimate personal preference in appearance.

[VISUAL 29.1: Five approaches to cut grading—proportions, observation, 3D geometry, light performance, and a hybrid system]

[VISUAL 29.2: GIA, AGS Ideal by GIA, IGI, HRD, and GCAL—comparison of scope, not conversion of grades]

[VISUAL 29.3: Why the same name Excellent need not mean the same result]

[VISUAL 29.4: 2D report averages versus the actual 3D facet model]

[VISUAL 29.5: Warning against crosswalks—categories from different methodologies are not units of measurement]