Short answer
They are measured at several angles and toleranced at each, usually through a weighted colour difference so that one limit can be applied across angles and colours. A metallic finish needs at least three measurement angles and an interference finish more. Even then the tolerance is incomplete: two panels can agree at every angle and still look different, because sparkle, graininess and the orientation of the flakes are appearance properties that multi-angle colour readings average away.
A metallic finish is light when viewed near the mirror reflection and dark when viewed far from it, and that change, the flop, is the point of the finish. A reading at one geometry records a single slice of it. Multi-angle instruments illuminate from one direction and measure at a series of aspecular angles, counted from the mirror direction back towards the surface normal. Research summarised by one instrument maker puts the requirement at a minimum of three angles and ideally five, and ASTM's method for metal-flake materials is built on three. Interference pigments shift hue and chroma as well as lightness, over a wider range, which is why instruments for them add an angle on the far side of the gloss and why ASTM's metal-flake method excludes them.
A single limit applied equally at every angle would be wrong twice over. In practice the limits differ by angle: tolerances at the near-specular angles of 15° and 25° and at the flop angles of 75° and 110° are usually set wider than at 45°. And an unweighted CIELAB difference is as colour-dependent here as anywhere else. The German standard for automotive colour tolerances is the usual basis for the weighting, which gives a tolerance parameter that does not depend on the colour; it was reissued in 2019 as a single document for solid and effect coatings, replacing two earlier parts. Vehicle makers commonly write their own limits on top of it for three-angle or five-angle instruments. Its factors and limits sit in the standard and are not repeated here.
ASTM's multi-angle method carries its own warning: specimens that agree at all three angles may still look different if gloss, texture or flake orientation differ. An instrument maker's white paper gives a concrete case. A car body was coated electrostatically and its bumpers with a bell and pneumatic process. The mean colour difference by the German standard's equation, at eleven positions across body and bumpers, lay between 0.40 and 0.90 and was judged acceptable. Viewed, the body sparkled considerably more than the bumpers. A conventional multi-angle reading averages the reflection over the measured spot, so it cannot separate the colour of the basecoat from the glint of individual flakes; that takes camera-based sparkle and graininess measurements.
The demanding comparison on a vehicle is not batch to batch but part to part: a plastic bumper against a steel wing, painted in different shops, on different substrates, sometimes by different suppliers, and seen edge to edge in sunlight. Flake orientation depends on how the paint was atomised and how it flowed before it set, so an identical paint applied by two methods gives two flops. The automotive recommended practice for instrumental colour difference is scoped across topcoat paint, hard trim, soft trim and film for this reason: parts in different materials from different suppliers have to be judged on one footing. Colour harmony is accordingly checked on the pair as assembled, as well as on each part against its master panel.
The visual standards used for solid colours do not apply as written. ISO's visual comparison of paints excludes special-effect coatings unless the illumination and viewing details have been agreed, and ASTM's visual practice excludes metallic and pearlescent materials because they need directional light. In practice the panel is tilted under a directional source to run through the angles, moving it towards and away from the lamp to see the colour travel of an interference finish. Two lighting conditions are needed as well as several angles: sparkle appears under direct, sun-like light and changes with the illumination angle, while a coarse, salt-and-pepper graininess appears under diffuse, overcast-like light and hardly depends on viewing angle.
| Comparison | Standard or scheme | Method | Condition | Figure, as the source states it | Caveat |
|---|---|---|---|---|---|
| Paint batches, painted bodies and add-on parts, and refinish against the original, for solid and effect colours | DIN 6175:2019-07 (replacing DIN 6175-1:2009 and DIN 6175-2:2001) | Multi-angle colour difference with weighting factors | Directional illumination with measurement at several aspecular angles; three or five angles in common instruments | None verified; agreed by the parties | The standard's factors and limits are in its paid text and are not quoted. Vehicle makers set their own values on top of it. |
| Metal-flake (metallic) finishes on coatings and plastics | ASTM E2194-14(2025) | Multi-angle colour measurement | Three measurement angles; pearlescent and interference materials are outside the method | None verified; agreed by the parties | The standard notes that specimens agreeing at all three angles can still look different if gloss, texture or flake orientation differ. |
| Vehicle topcoat, interior soft trim, interior and exterior hard trim, and exterior film against a colour standard | SAE J1545 (reaffirmed December 2021; first issued 1986) | Instrumental colour difference as a buyer–seller acceptance practice | Opaque or nearly opaque parts only; transparent materials excluded | None verified; agreed by the parties | Only the public scope was read. The numerical tolerances belong to each vehicle maker's own specification. |
Why: Flake orientation differs between the two application processes, changing sparkle and flop in ways the colour readings average out.
Fix: Add sparkle and graininess limits, and approve the pair assembled under direct and diffuse light.
Why: The repair was matched at one angle; its flop differs from the original.
Fix: Match and check at near-specular, mid and flop angles, adjusting application as well as tint.
Why: Interference pigments change hue and chroma with angle beyond what three angles capture.
Fix: Use a method and geometry set intended for interference pigments, with an additional angle beyond the gloss.
Each statement is labelled by kind — established fact, a standard’s requirement, observed market data, a convention, or Colourwise’s own interpretation or analysis — with the strength of the evidence behind it.
StandardStrong evidence
ASTM E2194-14(2025) requires three measurement angles to characterise the colour of materials containing metal flake, does not cover pearlescent and interference materials, and notes that specimens matching at three angles may still appear different if gloss, texture or flake orientation vary.
ConventionModerate evidence
In multi-angle colour quality control, tolerances are usually higher at the near-specular angles (15°, 25°) and the flop angles (75°, 110°) than at 45°, and automotive companies often set specifications on a weighted colour difference based on DIN 6175 using three-angle or five-angle instruments.
Caveat: An instrument maker's description of practice; individual vehicle makers' limits are not public.
StandardStrong evidence· Germany
DIN 6175:2019-07 sets out colour tolerances for automotive coatings covering both solid and effect coatings in one document, replacing DIN 6175-1:2009 and DIN 6175-2:2001.
Caveat: From the publisher's catalogue record; the standard's equations and limits were not read and are not quoted.
Source: DIN 6175:2019-07 Colour tolerances for automotive coatings — Solid and effect coatings
FactModerate evidence
In an instrument maker's example of a body coated electrostatically and bumpers coated by bell and pneumatic application, the mean DIN colour difference at eleven positions lay between 0.40 and 0.90 and was judged acceptable, yet the body visibly sparkled more than the bumpers.
Caveat: A single illustrative case in a manufacturer's white paper promoting sparkle measurement; the acceptance limit applied is not stated.
FactModerate evidence
Sparkle is seen under direct illumination and changes with the illumination angle, whereas graininess is seen under diffuse illumination and depends little on the observation angle; both are influenced by flake size, concentration and orientation.
StandardStrong evidence
ASTM D1729-22 excludes metallic and pearlescent materials that require directional illumination, and ISO 3668:2017 does not apply to special-effect coatings without prior agreement on all illuminating and viewing conditions.
Source: ASTM D1729-22 Standard Practice for Visual Appraisal of Colors and Color Differences of Diffusely-Illuminated Opaque Materials; ISO 3668:2017 Paints and varnishes — Visual comparison of colour of paints
Reviewed 6 October 2026. Colourwise summarises its sources in its own words and does not reproduce standards text or proprietary colour data. Spotted an error? Tell us.