I posted a image of the network of the shading part of the ocean. In light blue, there is the 2 HOT I use. Theses nodes are linked the nodes used as deformer in vopsop. In dark blue, the texture used to “warp” the position of the surface. The texture are link to ramps ( red nodes) to control the amount of foam of the ocean. All the ramp are fused to obtain the final foammask. The foammask is used to mix 2 shaders (purple node). The surface model is use for the water part. At the top of the picture you can see the actual method use to obtain thickness ( the nice tips of alejandro is not implemented yet). The foam shader is a mix of lambert and constant to fade the dark parts of the lambert ( really really cheap sss fake).
the screen node is a subnetwork. I create it to keep the pixel maximum value of the mask to 1.0. In this network, the values of the pixels in the two layers are negated, multiplied, and then negated again.
I've been trying to get a grip on your setup but so far haven't been fully succesfull yet. Would you mind posting a hip file so we can have a more in depth look to see what is going on in the shader and disect it ? If you don't want to share the entire network, maybe a simplified version of it?
One of the things i don't get is why you are getting the HOTs in here and link them to the sops deformation hot. Wouldn't it be easier to create a variable for minEigenval in sops and import it here. Or is there something i'm missing here…
Sure, that's a piece of cake compared to your shop network. Check the attached file. In sops the mineigenval attribute that is generated by the HOT is used to calculate a new attribute shaderFoam. You could ofcourse get the attribute mineigval itself as well and do the calculations in shop on that attribute.
In shop you just use an parameter node to get the point attribute in. I made the interesting nodes red.
I check your hip file. Thanks for this. The thing is, you export POINT attribute. The mask you obtain is a like a vertex map. If you use new HOT node you obtain finer result. I jointed a hip file myself. It's a wip so … ther is some mess. But you can check it out.
A little thing i found in your file that i do in another way is the environment. I generally use a environment light for this and set the rendering mode to ray tracing background. I also don't understand why you use raytracing as a render engine instead of the micropolygon one. The micropoly here speeds up renders tremendously. There are a lot of settings in you mantra node i don't quite understand why you changed them to those values. For example the Ray tracing acceleration, the texture cache size and for the sampling the sample lock and Ray variance antialiasing. Setting all those back to default settings improved rendertimes a lot and i don't see an awfull lot of difference in the result. You might have good reasons for the settings you used though. Just curious.
no good reason, just tests … My mantra began to crash during the render of the anim so … But if the default value is good for you, it's ok. About the micropolygone, I think the default render look worse than the default raytrace. You need to increase the dicing to obtain something similar.
Here is one of my tests to create foam from boat. Particle age and point cloud fading is not properly set for the moment. Strange … I was sure to set the ramp parameter to fade out the particle's mask …