Octane for Softimage 3.0x Tweet

[face]

Great, you would release before release octane 4

You can make it faster, do it

Ok, VDB from points is also included, with the same features as in the previous post.
Added also a filter node with 12 filters.
Use two of these filters to get the bottom result in the image below.
Source was a simple emFluid4 demoscene as input for the point position/size.

[wireframex]

Oh very cool to see emFluid compatibility

[face]

Oh very cool to see emFluid compatibility

More or less, the input can be every point cloud data.

[face]

What a mess, trying to transfer attributes, like heat/velocity, from a pointcloud to a grid.
It works with generated fog volumes, the problem is, i can´t apply filters like erode/dilate to fog volumes, the filters works only with sdf volumes.
So the result looks like the top result in the image from the previous post.

I think the best way is to make the VDB support like the emTopolizer from Eric Mootz.
You have one main ICE node and plug in different opperators like filters, loader/savers, converters, etc...

Ok, some things to make...

face

[wireframex]

Any news Face ?

[Elvissuperstar007]

I think particles are volumetric, octane users do not often use it, you can release a version without supporting these functions, version 2.0 is faster by 90% compared to version 1.0, users are compelled to suffer and can not even make elementary animation! The priority of the particle is not justified!

[face]

Particles/strands are working, only VDB´s needs time to implement.
Especially a vdb combine takes time to implement because you can put a combine in a combine, in a combine...

Here is a little overview how the tree can look.
You will have two generators (VDB from Point/Geometry) which can have multiple references. The result can combine they together ( Add / Subtract / Multiply / Divide / Maximum / Minimum / (1-A)*B / A+(1-A)*B / Union / Intersection / Difference )
Each generator can have its own filter, Each Combine and the main result ( VDB_Main_Core ) also.
If you have a tree like the image below, it is a little bit complicated to find the right generation flow of the tree, this is the thing on which im working now.

The output from the tree looks for now like this:

Code: Select all

0 INPUT_CLASS/NODE_COMBINE/V1.0
1 INPUT_CLASS/NODE_COMBINE/V1.0
2 INPUT_CLASS/NODE_FROMPOINTS/V1.0/1/0.1/3/1.5/1/1/1/1/1*REFERENCE_CLASS/NODE_POINTREFERENCE/V1.0/1/#/Size/PointPosition/1/1/1/0,0,0*REFERENCE_CLASS/NODE_POINTREFERENCE/V1.0/1/#/Size/PointPosition/1/1/1/0,0,0
2 INPUT_CLASS/NODE_FROMPOINTS/V1.0/1/0.1/3/1.5/1/1/1/1/1*REFERENCE_CLASS/NODE_POINTREFERENCE/V1.0/1/#/Size/PointPosition/1/1/1/0,0,0
1 INPUT_CLASS/NODE_COMBINE/V1.0
2 INPUT_CLASS/NODE_FROMPOINTS/V1.0/1/0.1/3/1.5/1/1/1/1/1*REFERENCE_CLASS/NODE_POINTREFERENCE/V1.0/1/#/Size/PointPosition/1/1/1/0,0,0
2 INPUT_CLASS/NODE_FROMPOINTS/V1.0/1/0.1/3/1.5/1/1/1/1/1*REFERENCE_CLASS/NODE_POINTREFERENCE/V1.0/1/#/Size/PointPosition/1/1/1/0,0,0

The 0 is the first combine which is in the main core, the first 1 is the second combine in which the next two 2 generators where plugged.
The next 1 is the third combine which is plugged in the first combine which will have the last 2 generators.

You see also, the parameters from each node are a simple string.

You will have a INPUT_CLASS, which includes the from point/geometry/file nodes.
A REFERENCE_CLASS, which includes the point/geometry reference parameters/objects.
A FILTER_CLASS, which includes some filters ( blur / median / ... )
A OPERATOR_CLASS, which includes noises
And at last a OUTPUT_CLASS, which includes a visualize/file node.

face.

[Elvissuperstar007]

Well, you can actually release a release without a function defined) you have the same version 3.0, volumetric we can wait

[wireframex]

Well, you can actually release a release without a function defined) you have the same version 3.0, volumetric we can wait

+1

[Elvissuperstar007]

2014 is the last release of octane integration, I feel the release of the test version will be released in 2020 ))))))))