sRGB gamma exponent in apca-w3

[jengelh]

The current apca-w3 code at

apca-w3/src/apca-w3.js

Line 148 in c012257

mainTRC: 2.4, // 2.4 exponent for emulating actual monitor perception

employs a trivial 2.4 curve:

  function simpleExp (chan) { return Math.pow(chan/255.0, SA98G.mainTRC); };

Normally though, sRGB is described as having 2.2.
The value 2.4 is a consequence of sRGB reportedly defining a linear section for pixel values 0..10 that necessitates a harder (2.4) curvature for pixel values 11..255 to get the combined curve to approimxate a 2.2:

	return value255 <= 10
		? value255 / 3294.6
		: Math.pow((value255 + 14.025) / 269.025, 2.4);

Should apca-w3 switch from the trivial 2.4 curve to a trivial 2.2 curve, or use the {linear + 2.4 segment} composed curve? What's your take?

[Myndex]

Hi Jan @jengelh and thank you for your question.

Short Answer

The short answer is this is intentional pre-shaping, and possibly going to change before release.

Additional

This is answered more fully in the main documentation, "regardingexponents.md" at the documentation repo.

Also, the main discussion area regarding APCA and related color theory is at the SAPC-APCA repo:

github.com/Myndex/SAPC-APCA/discussions

If you'd like to discuss or explore this more, let me know and I'll re-open this thread and move it to the main area.