Don McCurdy

@Mugen87 I can't tell if you're saying we should wait to replace the constants until a different release or a different PR. I think a different PR is fine but...

Either `outputColorSpace` or `canvasColorSpace` are OK with me. The thing that confused me about the original name was that it is ignored if writing to canvas comes at the end...

The only part of "outputEncoding" I would change is "Encoding". Output is a common enough term for what it does, I think we can clear up any confusion around render...

> So at least with WebGLRenderer, we have to stick to the current approach. I think the design in `pmndrs/postprocessing` is perhaps a good alternative. But we can keep that...

Thanks for the comments @sciecode! The summary in https://github.com/mrdoob/three.js/issues/23614 covers where we are (tentatively) headed in the longer term, which is meant to be self-consistent in the end. We can...

It may not be too late to influence the designs for WebGL and WebGPU, if we feel the need. But personally — knowing that a color space is a choice...

We could also make our (new) enums more specific, while keeping the `.colorSpace` terminology. For example: ```js // Transfer Function, Primaries, White Point renderer.outputColorSpace = THREE.SRGB_Rec709_D65; // sRGB renderer.outputColorSpace =...

tl;dr — this change looks good to me! *** Summarizing legacy vs. physical attenuation: - **legacy:** light attenuates from 100% to 0% linearly (`decay=1`) or by inverse square (`decay=2`) with...

~~One other idea. This is obviously not physically realistic, but — if our ultimate goal is to _delete_ the legacy mode entirely, the path of least migration resistance would be...

Thank you for working on this! If `scene.background` were a cubemap, could/should that be used instead? I might suggest naming this parameter `blur` instead of `blurriness`. Easier to remember, and...