Thinking a lot about “teachable” GL implementations this week. It seems to me that despite nostalgia for the good ol’ days of glBegin, GL ES 1.0 is actually closer to the sweet spot. It’s tiny, even compared to GL 1.2: https://www.khronos.org/registry/OpenGL/api/GLES/1.0/gl.h …https://twitter.com/BadMetaphor/status/1004424542520008704 …
-
-
My only real reservation on that is that shaders are kind of critical nowadays to most stuff, and shaders are also a good way to learn (I learned GLSL in quartz composer before I could code any C :)
-
Flipping that a bit: Learn basic shader stuff in shader toy or whatever first, then learn to put a quad behind it in GL. Expand from there into proper models. (Not saying you should go with that, just throwing it out there :)
-
I’m so torn on this. On the one hand, of course shaders are essential and ubiquitous. On the other hand, learning the fundamentals of transformations, rasterization, lighting models, etc. is already a lot of concepts for the beginner.
-
While I hesitate to teach a “toy” impl that’s further and further divorced from reality and practically, I’m very sensitive to overwhelming newcomers. I think shaders are separable, and I think fixed-function is still pedagogically useful, precisely because of what it abstracts.
-
Yep, agree with all of that… if only it was possible to used fixed function then just add a shader :) Dunno, ideally I’d like both approaches. Maybe fixed function first, then work back from a shader with ES2?
End of conversation
New conversation -
-
-
Reading the 2012 jwz xscreensaver post made me realise I was still angry about GL_QUADS.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.