Primary colour
Comparison colour
Set each colour’s scene luminance independently to simulate HDR content at different absolute brightness levels. X swaps.
HDR Viewing Conditions
Lw = reference white in nits. SDR monitor: 100. HDR10 peak: 1000. Cinema: 48.
Tone mapping — HDR peak
Sets the HDR peak for tone mapping curves. Operators map scene luminance to display range.
Shortcuts
X — swap primary ↔ comparison  ·  R — reset defaults  ·  C — copy JSON  ·  1–5 — primary luminance presets
Colours
#D3AF37 200 nits
ΔEz (Jzazbz)
#4A90D9 200 nits
CIEDE2000: ΔE₇₆: ΔEz@100nit:
Jzazbz Absolute Coordinates
Colour Jz az bz Cz hz° Iz
Primary
Comparison
Jz = perceptual lightness (PQ-encoded, absolute) · az = red–green · bz = yellow–blue · Cz = chroma · hz = hue° · Iz = raw PQ intensity.
Jzazbz — az × bz Plane
Gold = primary, blue = comparison. Dot size ∝ Jz lightness. Rings at Cz = 0.02, 0.05, 0.10, 0.20.
Jz Lightness Comparison
ZCAM Appearance Correlates (Safdar et al. 2021)
Colour J Qz Mz Cz hz° Sz
Primary
Comparison
J = ZCAM lightness (0–100) · Qz = absolute brightness · Mz = colorfulness · Cz = chroma · Sz = saturation.
ΔE Metric Comparison
Tone Mapping Operators — Transfer Curves
Reinhard ACES Hable BT.2446A Log Clamp
X axis: input luminance 0→peak (nits). Y axis: output (display 0–1). Vertical markers show primary (gold) and comparison (blue) scene luminance.
Tone-Mapped Output Values (Y channel)
Colour Reinhard ACES Hable BT.2446A Log
Primary
Comparison
Values computed on Y channel at each colour’s scene luminance, normalised to HDR peak.
CIE 1931 Chromaticity Diagram
Spectral locus (white outline) · sRGB gamut triangle (gold dashed) · D65 white point · Colour markers.
CIELAB L*a*b* (D65, relative)
Colour L* a* b* C*
Primary
Comparison
CIELAB uses relative colorimetry (no luminance level) and is provided as a reference.
Export and share
JSON includes full Jzazbz, ZCAM, CIELAB data with ΔE metrics. CSV includes per-channel coordinates. Share URL encodes all settings.
Multi-TMO Comparison

Compare all 5 tone mapping operators on the current colour pair. Shows tone-mapped output values and post-mapping ΔEz side-by-side.

Operator TM(A) TM(B) ΔEz post-TM Quality
Click “Compare all 5 TMOs” to run analysis.
Shows how each TMO preserves or distorts the perceptual difference between the two colours after mapping to display range.
HDR Colour Science Standards
ITU-R BT.2100 — HDR Television

BT.2100 defines Perceptual Quantizer (PQ, ST 2084) and Hybrid Log-Gamma (HLG) transfer functions for HDR television. PQ encodes absolute luminance from 0 to 10 000 cd/m² using a perceptual model derived from Barten’s contrast sensitivity function.

PQ reference: SMPTE ST 2084:2014, ITU-R BT.2100-2 (2018).

Jzazbz (Safdar, Hardeberg, Ronnier Luo 2017)

Jzazbz is a perceptually uniform colour space using PQ-encoded absolute luminance. It covers 0.001 to 10 000 cd/m² with near-constant JND steps. Designed as an HDR replacement for CIELAB’s cube-root compression which is only valid for relative colorimetry.

Key achievement: ΔEz thresholds are scale-invariant across SDR and HDR — a ΔEz of 1 is roughly a JND at both 1 nit and 1000 nits.

ZCAM (Safdar, Ciu, Kim, Luo 2021)

ZCAM extends Jzazbz with CAM-style appearance correlates incorporating viewing conditions: adapting luminance Lw, background Yb, and surround factor Fs. It is the first HDR colour appearance model built on an absolute uniform colour space.

Reference: Safdar et al. (2021), Optics Express 29(4), 6036-6052.

ITU-R BT.2446 — HDR to SDR Conversion

BT.2446 defines standardised methods for converting HDR content to SDR for backward-compatible distribution. Method A uses a power-law curve with peak-luminance mapping. Methods B and C address scene-referred and display-referred workflows respectively.

CIEDE2000 & CIE 15:2004

CIEDE2000 (CIE 142-2001) is the recommended SDR colour-difference formula with corrections for lightness, chroma, hue, and the blue rotation term. CIE 15:2004 defines CIELAB and the standard observer. For HDR work, ΔEz from Jzazbz is preferred.

ICC.2 HDR Colour Management

ICC.2 (iccMAX) extends the ICC profile format for HDR and wide colour gamut workflows. Jzazbz is used as the Profile Connection Space (PCS) for HDR profiles, replacing PCSLAB/PCSXYZ for wide-gamut targets. ZCAM may serve as the appearance model for cross-media viewing condition adaptation.

HDR Industry Applications
  • HDR10 / Dolby Vision QC: ΔEz allows comparing HDR master vs. compressed deliverable across the full 0–10 000 nit range.
  • Tone mapping evaluation: ZCAM correlates assess whether a TMO preserves appearance relative to the HDR master.
  • Display characterisation: ZCAM predicts appearance changes as display peak luminance varies (500 nit vs. 2000 nit).
  • Perceptual gradient generation: Jzazbz interpolation produces visually smooth transitions at any luminance level.
Mathematical Models and Formulas

Jzazbz Forward Model (Safdar et al. 2017):

Step 1 — Pre-adaptation crosstalk (b=1.15, g=0.66):
X′ = b·Xabs − (b−1)·Zabs
Y′ = g·Yabs − (g−1)·Xabs

Step 2 — M1 linear matrix → [L, M, S]
M1 = [[0.41479, 0.58000, 0.01465],
     [−0.20151, 1.12025, 0.07626],
     [−0.01660, 0.26480, 0.66848]]

Step 3 — PQ transfer (normalised to 10 000 cd/m²):
PQ(x) = ((c₁+c₂·xm₁)/(1+c₃·xm₁))m₂
c₁=0.836, c₂=18.852, c₃=18.688, m₁=0.1593, m₂=78.844

Step 4 — M2 matrix → [Iz, az, bz]
M2 = [[0.5, 0.5, 0],
     [3.524, −4.067, 0.543],
     [0.199, 1.097, −1.296]]

Step 5 — Jz compression:
Jz = (1+d)·Iz / (1+d·Iz)    (d = −0.56)

ΔEz = √(ΔJz² + Δaz² + Δbz²)
Research, Standards and Citations

HDR Colour Appearance

[1] Safdar, M., Cui, G., Kim, Y.J., Luo, M.R. (2017). Perceptually uniform color space for image signals including high dynamic range and wide color gamut. Optics Express, 25(13), 15131-15151. DOI: 10.1364/OE.25.015131

[2] Safdar, M., Cui, G., Kim, Y.J., Luo, M.R. (2021). ZCAM, a colour appearance model based on a high dynamic range uniform colour space. Optics Express, 29(4), 6036-6052. DOI: 10.1364/OE.413659

[3] Fairchild, M.D. (2013). Color Appearance Models, 3rd Ed. Wiley-Blackwell. ISBN: 978-1-119-96703-3

HDR Standards and Transfer Functions

[4] SMPTE ST 2084:2014. High Dynamic Range Electro-Optical Transfer Function of Mastering Reference Displays (Perceptual Quantizer).

[5] ITU-R BT.2100-2 (2018). Image parameter values for high dynamic range television for use in production and international programme exchange.

[6] ITU-R BT.2446-1 (2021). Methods for conversion of high dynamic range content to standard dynamic range content and vice versa.

Tone Mapping Operators

[7] Reinhard, E. et al. (2002). Photographic tone reproduction for digital images. ACM Trans. Graphics (SIGGRAPH), 21(3), 267-276.

[8] Narkowicz, K. (2015). ACES Filmic Tone Mapping Curve. Blog post.

[9] Hable, J. (2010). Filmic Tonemapping Operators. GDC presentation (Uncharted 2).

Colorimetry and Colour Difference

[10] CIE (2004). Colorimetry, 3rd Ed. CIE 15:2004.

[11] Sharma, G., Wu, W., Dalal, E.N. (2005). The CIEDE2000 color-difference formula. Color Res. App., 30(1), 21-30. DOI: 10.1002/col.20070

[12] IEC 61966-2-1:1999. sRGB colour space specification.

About this tool

This tool implements Jzazbz absolute colorimetry, ZCAM appearance correlates, 5 tone mapping operators, ΔEz/CIEDE2000/ΔE76 metrics, and CIE 1931 chromaticity — entirely client-side (zero network). Not a substitute for calibrated measurement or official CIE software.

Research Backend
Backend
Batch Pairwise ΔEz Analysis

Enter hex colours (one per line or comma-separated). Computes pairwise ΔEz at the current scene luminance settings, plus CIEDE2000 for SDR reference. Full statistics and histogram.

Click Run Batch to analyse colours.
ΔEz Distribution Histogram
Hue Constancy Analysis

Research note: Jzazbz exhibits improved hue linearity over CIELAB in the blue region (∼270°), where CIELAB’s a*b* plane shows significant hue non-linearity. This makes ΔEz more reliable for blue-heavy HDR content (e.g., sky gradients, neon lighting).

CIELAB hue error (blues): up to ~12° non-linearity
Jzazbz hue error (blues): <3° across 0.1–10000 cd/m²

For critical blue evaluation, always prefer ΔEz over CIEDE2000.
Spectral Reconstruction (Planned)

Future: reconstruct full spectral reflectance from sRGB, compute Jzazbz from spectral data via CIE 1931 observer integration, and compare with tristimulus-derived values for metamerism analysis.

Research backend provides: Jzazbz absolute colorimetry, ZCAM appearance model, 5 tone mapping operators, ΔEz/CIEDE2000/ΔE76 metrics, CIE 1931 chromaticity, multi-TMO comparison, batch pairwise analysis, full JSON/CSV export. All computation is on-device with zero network dependency.