Rubric 1.0.0 · column 4 of 13
Color depth
screen.colorDepth, in bits. Effectively a binary signal: 24, or something worth looking at.
Under rubric 1.0.0 this column is worth between 0.09 and 6.64 bits, and the likeliest outcome the table knows — 24-bit — costs 0.09. Those figures are assumptions printed before your run. The counted box beside them is not: it is how many of the devices these sites have read share your value.
How a browser produces colour depth
The browser reports the bit depth of the display buffer it is compositing into, which the operating system decides on behalf of the panel and the graphics driver. It has been 24 on essentially every mainstream configuration for two decades, which is why the specification-level pressure has been to freeze it rather than to enrich it. Some platforms report 30 for a wide-gamut or HDR pipeline, and some hardened browser configurations normalise it back to 24 on the way out.
How rubric 1.0.0 prices colour depth
Two outcomes and a steep step between them. 24 is assumed at 0.94, which costs 0.09 bits — the quietest reading this rubric can produce for anything. Anything else at all is assumed at 0.01 and costs 6.64 bits, which puts it above the operating system, above the pixel ratio, and above most of the capability group, on the strength of a single unusual integer.
The spread between the cheapest and the dearest outcome is 6.55 bits, and that spread is the whole of what this column can ever do to a total. Where the four band edges sit →
Every outcome this rule can produce
This is the printed envelope, written out. The pen cannot land outside it, because these are
all the answers rule colorDepth has.
| Outcome | Assumed p | −log₂p |
|---|---|---|
| 24-bit modal outcome — the printed line | 0.940 | 0.09 |
| 30-bit any depth other than 24 | 0.010 | 6.64 |
What changing colour depth costs
There is nothing to spend here: this column is nearly free at 24 and there is no lower reading. The only decision is whether to normalise a 30 down to 24, and that is a rewrite in the same class as any other declared value — cheap in bits, and consistent only if the rest of the display group agrees with the ordinary machine you are now claiming to be.
The rubric's own reasoning, verbatim
Almost universally 24. Anything else is notable on its own.
What moves it
Almost nothing. HDR and wide-gamut displays report 30 on some platforms, and some privacy tools normalise it to 24.
Where this prior is weakest
The 0.01 prior for non-24 values is a guess with no source behind it at all. If 30-bit reporting is more common on a platform than we assume, this rule hands out 6.6 bits it has not earned.
Correlation
Grouped with Screen resolution and Device pixel ratio under "display". Whichever of them carries the most surprisal in a given pass takes full weight; the others take 35%. The chart draws the short stroke rather than describing it.
How the correlation discount works →
Challenge it
If you have a source that contradicts this assumption, the honest thing is to say so in
public against this specific prior. Open a challenge
against colorDepth →
What people ask about colour depth
Answered from rubric 1.0.0 itself. Where an answer quotes a probability, that probability is a constant in the shipped scoring code and is printed in the table above.
- Why is my colour depth worth 6.64 bits?
- Because your browser reported something other than 24, and the rule assigns every non-24 value the same assumed probability of 0.01. It does not distinguish 30 from 32 or from any other value, which is one of the weaker joints in the table.
- Can a colour depth of 24 ever cost me anything?
- Barely. It is assumed at 0.94 and costs 0.09 bits, and it sits in the display group where it is almost always discounted to 35% of that against a louder groupmate.
- Do HDR displays report a different depth?
- Some platforms report 30 for a wide-gamut or HDR pipeline. The rubric treats that as the notable case, which is exactly the assumption most likely to be wrong if HDR reporting becomes ordinary.
Where to read next
- The prior for screen resolution — the other reading the display group discounts this one against.
- The prior for device pixel ratio — the other reading the display group discounts this one against.
- The complete prior table for all 13 columns — every rule, every outcome, and the rationale the rubric ships with each one.
- Assumed and counted: what a prior is, what the counted column beside it measures, and why the two are never added.
- Compare two passes for the delta on this column — change one setting and read what actually moved, instead of guessing from a total.
Where the pens move
Knowing which prior is weak is only useful if something can actually move the signal.
Engine level, not JS level. Donut Browser spoofs through CDP injection in its Wayfern Chromium fork, so the value a page reads is the value the engine reports — there is no patched accessor to catch.
It refuses to contradict itself. A profile whose proxy exit geography, time zone or language disagrees with its fingerprint does not launch. That is a pre-launch gate in the app, not advice in a manual.
Open, and quiet. AGPL-3.0, desktop app and sync server both in the repository, and no analytics SDK anywhere in the source. macOS, Windows, Linux.