Hello guys, I am checking out fuse+ply reader, by getting the pointcloud pos / color in a cache region I was able to load huge pointclouds and stay at very high FPS, I am quite amazed with the power of gamma
I would like to transform / scale / rotate this point cloud, and latter on mess up with fuse curl forces and so on… but I am getting stuck on the first step
Any advice ?
Also, to make this work there is a change need to be done on the VL.PLY document, I attach the image
you can transform the final entity, it should just apply the world transformation as if it would be one object. you don’t need to do that on the point cloud level.
if you want to apply forces to the points you first have to make a particle system out of it. you can use the positions of the buffer as starting position for particles. have a look at the particle patches, they should give you an idea.
another improvement that will give you a huge performance boost (~30%) is changing the buffers to immutable buffers. you only need dynamic buffers if you actually dynamically upload data in every frame. The GPU can optimize access for immutable buffers that it cannot do for dynamic buffers.
using dynamic buffers is automatic for beta users, and we kind of have to mention that everywhere. maybe we should rename the buffers again… so it is easier to get the right one.
I don’t find any world transform nodes, still If I want to load 2 point clouds from 2 files I would like to be able to transform them individually
I will check the emitters patches
Is there a way to do this,
I mean to transform the vector4 buffer from position and the RGBA buffer from the colors of the original PLY reader, so I can share this document to someone without asking them to edit the already “ready” plyreader nugget?
FuseAndPly.vl (54.8 KB)
