Quote from: Dune on January 05, 2022, 02:55:36 AMIt's just one parameter that needs a zero, and I think it's pretty normal that one checks every shader and sets parameters like they're needed. You can't just add a shader and expect it to work as you need without adjustments. I also change the breakup shader, e.g. as I don't like the very warped lines created when greys are broken up. I could call that annoying, but I don't see it that way.
Isn't that ironic though because that's literally the point of the aforementioned suggested shader? And aforementioned options for the default shader? So it is more straightforward and to the point. Need a lit surface? Drop in a Luminous Shader and do simply that. Not extrenous options TG needs to check to see if it's doing this or that at runtime before rendering that were never part of your intended goal. In coding this would be bad logic as it just adds to the runtime execution. However miniscule it seems for one function, can add up for several, or dozens.
Again this seems more personal, and more just not wanting to be bothered with change. Not that it's actually valid beneficial points. Should be more open and not set in ways.
If you want to test this yourself. Use a default shader with no fresnel naturally, and a surface layer, use a 0.5 colour. Now benchmark both these shaders for just colour against a constant shader. The constant shader is faster in all tests, because naturally it is
not going through a series of checks to see what it's doing beyond the intended goal. I have no terrain in my scene, and default camera and resolution of 800x450.
Surface Layer: 11s
Default Shader: 12s
Constant Shader: 9s