I‘m currently working on a spacestation. At the moment I am rigging the lighting-setup (separate layer) and I got a surprise when I looked at a test render after a few spot-lights were set:
The result was like the samples-value had been scaled (is set to 250), the render is far more grainy than the test render without the spot-lights.
None of the spotlights do directly point to the rendered area.
The scene itself is lit by a sun-light.
Not sure if this applies to spotlights in particular, but I always disable MIS for my tiny and non significant light sources (and materials, and dark surfaces in general). I’m not sure, but I think sampling is more happy when only dominating sources have MIS applied, and then it will still divide/guide the samples between those. When I have interiors with 30-50 spotlights in addition to other lighting (generally we don’t do sun&sky for what we do), I have to aid the system and fake stuff, even with tons of samples.
thanks Daedalus_MDW, your had the right suggestion, my renders are back to normal.
It’s the first time in many years with Blender I use that functionality and I still don’t know when to use it…^^