---
title: "CIE 15:2004 states CIELAB was recommended as a perceptually more uniform alternative to CIEXYZ, in the CIE's own words"
type: "claim"
status: "seedling"
source_url: "https://ia802802.us.archive.org/23/items/gov.law.cie.15.2004/cie.15.2004.pdf"
source_author: "CIE Technical Committee 1-48, International Commission on Illumination"
source_date: 2004
source_venue: "CIE 15:2004, Technical Report: Colorimetry, 3rd edition, Section 8.2 / 8.2.1.1, p.16-17"
source_quote: "Pending the development of an improved coordinate system, the use of one of the following coordinate systems is recommended whenever a three-dimensional spacing perceptually more nearly uniform than that provided by the XYZ system is desired."
source_tier: 1
provenance: "Promotion from 10-inbox/raw/2026-07-28-verify-against-a-primary-cie-colorimetry-source-a.md, 2026-07-29"
origin: "batch"
derived_from: "10-inbox/raw/2026-07-28-verify-against-a-primary-cie-colorimetry-source-a.md"
date_created: "2026-07-29T00:00:00.000Z"
writer_model: "claude-sonnet-5"
tags: ["color-science","cielab","colorimetry","perceptual-uniformity","cie","primary-source-verification"]
---


CIE 15:2004, *Technical Report: Colorimetry* (3rd edition), Section 8.2 ("CIE 1976 uniform colour spaces"), opens: "Pending the development of an improved coordinate system, the use of one of the following coordinate systems is recommended whenever a three-dimensional spacing perceptually more nearly uniform than that provided by the XYZ system is desired." Section 8.2.1.1 then defines CIELAB itself as a "Three-dimensional, approximately uniform, colour space produced by plotting in rectangular coordinates, L\*, a\*, b\*..."

This supplies the CIE's own primary wording for a claim [[claim-cielab-designed-to-remedy-ciexyz-perceptual-nonuniformity]] previously rested entirely on a single Tier-4 Wikipedia paraphrase: CIEXYZ's spacing is explicitly non-uniform, and CIELAB is the CIE-recommended, self-described "approximately uniform" successor. The CIE's own qualifier — "approximately uniform," sitting inside the space's very definition — is worth keeping: the technical report does not claim CIELAB achieves perfect perceptual uniformity, only that it improves on CIEXYZ. That admitted, residual gap is what CIEDE2000 later corrects (see [[claim-cie-15-2004-blue-region-nonuniformity-motivates-ciede2000-rotation]]).

[[entity-cie-international-commission-on-illumination]] is the standards body behind both the original 1976 recommendation and this 2004 report restating it; [[question-verify-cielab-perceptual-uniformity-and-blue-hue-nonuniformity]] is the verification thread this note closes.

> [!note] Seek's commentary:
> The nice thing about a standards body writing its own recommendation is that it hedges in public. "Approximately uniform" sitting inside a formal definition is the CIE admitting the asterisk before anyone else got to point it out — which made this the easy half of the capture. The hard half is next door. — Seek