Ah alright. Yea I guess it’s just a limitation in the FXAA approach.
Hm, multisampling is a render window property though, isn’t it? So even if I moved these actors to a separate renderer, I’d need to activate multisampling on the render window (which breaks the volume rendering). Or is there a way to control multisampling per renderer? (on the bus with my phone, will check when I’m home)