[dk2079]

will do some research and look dev stuff here as time allows.


the maxwell engine has gone a long way by now, cool new features and it feels faster too.
starting here with nested dielectrics and some custom drop creation&scatter functions setup with grasshopper/rhino
kinda looks like sparkling water.. which wasn't my intention. ideally one would bake custom maps for roughness/condensation which leave out the wet drops (+offset)
still thinking how to get that done in a quick and precise way..

[dk2079]

little update along the way.. playing with baked occlusion maps.
looks like I have a faulty bump map in one of the layers which reveals the tesselation on the frosted part.

no water inside this time to render faster, anyway this is only SL 14

[rusteberg]

looks like I have a faulty bump map in one of the layers which reveals the tesselation on the frosted part.

can you attach the image map you are using here for bump? does it have very fine noise in it (the image you are using in the bump)?

[dk2079]

the map was left from a pretty old material I built, and had some bad settings on the b/w bump slot (brightness adjustment too high)
rather big file

making some fine grain noise is what I actually wanted to try out (for the condensation layer), and it seems to be rendering fine with a normal map instead.. using something like this atm:

[dk2079]

last nights endeavours on a procedural ice shader. cought some tiling there. SL20

[rusteberg]

It might be how you have the bottle uv mapped. Look HERE at what your original bump map is doing unseen to the naked eye. There's some weird scribble pattern centered in the middle of it which could have something to do with the weird pattern you are seeing... along with the noticeable vertical striations which could explain some of what you're seeing - i don't think it's a faceting issue

[dk2079]

It might be how you have the bottle uv mapped. Look HERE at what your original bump map is doing unseen to the naked eye. There's some weird scribble pattern centered in the middle of it which could have something to do with the weird pattern you are seeing... along with the noticeable vertical striations which could explain some of what you're seeing - i don't think it's a faceting issue

thanks for the check up! looks like a good example for a crap map.. not really sure where that stuff is coming from
I have noticed before though that B/W bump value is quite sensitive if the maps are very contrasty or bright, and that shading artefacts can come up depending on the mesh topology underneath.
to make a clear statement though I'd need to run some controlled tests.

continued with the icy stuff for now:




breakdown

left emitter - right emitter - ibl on low - ibl on high

[rusteberg]

here's a pretty good reference:

[dk2079]

hey thanks for that fine reference image! I am afraid the plants in the background are gonna make the render really slow, but I will think of something.

kind of hitting a wall with the on-surface mapping only though, guess one can make it work for a single shot but not so easy for several camera views or even an animation.

trying to go in layers .. inside ice only. increasing frequency and adding a voronoi on the marble brought the benchmark down from 500 to 50. interestingly, adding the bottle around and the drops brought it back up to 180.