I'm working on a project in wich as you can see in V1 it's doing a lot of drawcalls, 3 per instance of MyRenderer actually,

in the first place I would like your opinion on this pipeline, if I'd be rendering only a single instance of MyRender, as I need to apply different effects (in purple), could it have been more straightforward ?

enter image description here

Now I'm working on a V2, that has several subsidiary questions

enter image description here

I've done two things to reduce drawcalls & texture bindings, so :

  • is my understanding/use of GLSL imageStore good or should I split thoses steps with more drawcalls ?

  • is this "atlased" pipelline with uber shader a better solution & would playing with render regions be a simple operation for GPU ?

Finaly as you hopefully understood, my whole rendering is mostly a cloner for MyRenderer (with different attributes per instance), would a compute shader be even faster at rendering this ?

I realise now I could have made the diagrams horizontaly, would have been easier to read. Sorry for that !

Hope you understood my questions.

Thanks for reply


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.