I'm recently working on implementing deferred lightning into my engine. I got the basics so far ( rendering the worldpositions/viewspace-normals/color) into textures, and then apply them onto a screen filling quad. However, I ran into a few problems doing so.
First of all, I managed the bring one pointlight into the scene. That seems to work fine, but I wanted to be able to bring a lot of lights into the scene, and now I can't really see how I am suppost to move on from one light to multiple ones.
Should I just pass a big big buffer with all their positions and attributes into the shader and then loop through them? Where would be the benefit compared to pass all of them to a usual shader?
I was told that shadowmapping would be an easy thing to add when rendering lights with this method, but I can't see how this is supposed to work.
Is there an easy way to implement some kind of AA to this?
And one other thing, when I render things to simple quads, the textures seem to suffer a pretty big impact on their quality, looking very stepped from color to color, not really smooth. Anyone an idea what could cause this?
Thanks for helping me out and spending your time to do so