Mark Kellogg

Nothing comes to mind right now, I'll let you know if I think of anything :)

You definitely have to be careful with post-processing effects, especially when they depend on depth buffer data; they're not a one-size-fits-all solution. The issue with combining splats and post-processing results...

So I think I may understand what's happening. When you don't have the FXAA pass it's pretty straightforward -- the outline effect gets rendered first, but since there's no depth...

Well I'm just guessing, but I think the output of the FXAA pass will be rendered to `renderTarget` and not the screen, because it is designated as the output render...

It definitely makes sense that when you render the composer after the viewer renders, the splat is gone. I assume the composer clears the render buffer before performing its own...

So I've discovered a couple of things. Three.js' `EffectComposer` is meant to be used for all rendering, so you shouldn't be trying to call `viewer.render()` and then following that up...

Just make sure your code to add passes looks like: ``` const composer = new EffectComposer(renderer, renderTarget); const pass = new RenderPass(threeScene, camera); composer.addPass(pass); composer.addPass(outlinePass); composer.addPass(effectFXAA); effectFXAA.material.depthWrite = false; ```...

Awesome, that's really good to know!

I'm going to close this for now, but I think I might eventually add something to the README about doing stuff like this. Thanks again for digging into this!

I have also thought about ways of doing this kind of thing -- Another approach might be to include a universal scaling factor for all the gaussians because simply making...