I am doing all kinds of experiments but been out of the loop for a bit on HISE so interested in others experiments and conclusions.

Have people been using GLSL or Webview for example and what are the implications in terms of CPU/GPU and dependencies?

Webview is absolutely amazing for dynamic graphics so what have people been using for dynamic graphics since the paint routines are on cpu and not that suitable for fancy moving graphics.

Is it viable to keep the fancy stuff as webview ( perfectly viable in terms of making the graphics ) but what is the practical experience? is it stable on all platforms?

Also Three.js , anyone been using that and are there different considerations than 2D ?

Also CSS, dont know much about that and have yet to really test that, what are the advantages of that?

I am doing some glsl experiments but its quite limited "usable but limited" It would be nice to be able to feed it a png like in shadertoy, but i have had some success making masks from png into js array.

Still experimenting with GLSL, making a pretty cool UI with a combo of png and GLSL for an everchanging UI. Again, wondering if it will be mostly conceptual or if its usable in projects and stable.

Are there any plans for a future graphic methods that would outdate the current ones or plans to strengthen the implementation of the current ones like GLSL or webview ?

I am also interested in paint routines in terms of delivered code, I can now do pretty fancy vector UI´s in paint routines but i am interested in the backend of the code delivered.
Currently I have all callbacks in a sep include as well as all text,optimisations ( flip between png and vector shadows, stuff like that) , color maps and styles.

Before I build out that delivery setup I would love to know how someone would prefer recieving a fully built paint routine UI.