Maxwell Render | Official public forum

I was very happy to see that the new FP supported Maxwell natively. Did a quick test using FPP to distribute water drops. It's just a quick one, but it worked out pretty well I think.

It was also a test of my newer brushed metal material, which I think also worked out really well

/b



Higher res here: http://heavyartillery.com/wp-content/up ... Can_03.jpg

...and a detail from the full res (only 4K for this test):



And this is without some of the post treatments to see the material/details a bit cleaner:

so are these drops geo that FPP randomly places, scales and rotates? let me know how the process works cause i've never used it. I've started using a script in maya that randomly places water drop geo but it's kindof a pain in the ass and is more than a hack than actually working real well.

Yes, they are placed randomly by FPP.

There are still some tweaks needed to get it really "correct" with FPP so to do this one I did it with two different FPP objects. One scattering the larger drops, one scattering the medium/small drops. The really small stuff is done with a normal map.

I really like doing it with FPP because it let's me use way more droplets than I can with Max's native PFlow system and it's really a PITA trying to do them any other way - aside from displacement, which is another route I have used extensively.

Anyway, not sure how you can match it with Maya - not something I have used - but FPP is a great tool, I have used it quite a bit and the support is first class.

/b

Very nice indeed.

Very impressive!

thanks gadzooks and Tom. I know it's not a very exciting image, but it is *very* nice to have these two tools working so well together

/b

Amazing, Brett!

Awesome images Brett! FPP4 is addictive isn't it? Everything I work on can suddenly use a scatter of some kind

Thanks Eric. I know what you mean It's a very useful tool in a lot of ways once you start thinking outside the box a bit, and I can't speak highly enough of the support from the guys at Itoosoft. One of my better CG investments for sure.

/b

The metal material is great, I like it more without the post treatment....at least, not quite so yellow. Blue cast is always good for metals

Thanks Mihai. I agree that it looks better for the metal, at least up close, but I felt the overall shot just needed something else to give it a bit more life and going cool was feeling overly monochromatic to me.

P. S this material is a perfect example of how much we could benefit from a way to link parameters in the material editor (we had a discussion about this a while back). The are 8-10 maps being used at least twice each so every time you have to adjust something like tiling or UV offset it can take a long time and is very prone to user error.

It seems like the droplets on the metal arent refracting and possibly not casting a shadow? The drops on the glass look more real, which is odd becasue for the metal there is no issue with booleans etc, just simple intersected geo is fine.

The normal mapped smaller ones would look a lot better if you duplicate the bsdf and flip the x/y of the normal map on the duplicate, and give it less weight, this will give the inner highlight.

For the bigger drops a scattering method seems like the best bet, or a hair sytem with droplets instanced onto the hairs, If someone finds a vector disp map of droplets that might work too..... just a thought.

For whatever reason, this seems to be how maxwell renders water drops on solid surfaces...looking like they're lacking refraction and a lot of times inner shadow and caustics. Still haven't been able to figure out what's going on this that.

eric nixon wrote:It seems like the droplets on the metal arent refracting and possibly not casting a shadow? The drops on the glass look more real, which is odd becasue for the metal there is no issue with booleans etc, just simple intersected geo is fine.

The normal mapped smaller ones would look a lot better if you duplicate the bsdf and flip the x/y of the normal map on the duplicate, and give it less weight, this will give the inner highlight.

For the bigger drops a scattering method seems like the best bet, or a hair sytem with droplets instanced onto the hairs, If someone finds a vector disp map of droplets that might work too..... just a thought.

Thanks for the feedback Eric. Actually the whole thing is metal, just two different ones. I am not sure why some ghost of that way, but as Justin pointed out it happens a lot. I've tested it quite thoroughly over the years and there have always been issues with Maxwell and interpenetrating geometry.

Displacement is the way I have usually done droplets in Maxwell and it works fine but I have never liked the lack of individual control. It also requires some cheats but that is no big deal most of the time.

We really need maxwell to change how it deals with those intersections, but for this test it didn't matter that much so I just cheated in some highlights and called it a day

Good idea on the normal map thing. in this case the normal mapped ones are pretty small so not sure how well it would actually show. I'll definitely give that a try next time and find out though!