schwungsau

go with this one : http://www.cgchannel.com/2019/05/review-boxx-apexx-t3-workstation/ its better then building yourself... if spend so much money, you don't want crashing machine because of some hardware glitches with each other....

a lots of ways.... simple and fast way at pointwrangle after the add note ithe code int prim_points[]; float umap; addattrib(0, "point", "umap", 0.0); prim_points = primpoints(0, @primnum); for ( int i = 0; i < len(prim_points); i++ ){ umap = float(i) / (len(prim_points) - 1); setattrib(0, "point", "curveu", prim_points[i], 0, umap, "set"); } then you can add a "color" node --> set it ramp from attribute with curveu as attr.

this is a much better explanation of L-systems : https://natureofcode.com/book/chapter-8-fractals/

xeon clocked at 2.2 ? your GPU's will wait long time to get data....

i think, you have to set your particle size (pscale / width) to small number like 0.001......

its that's easy. written stuff for GPU needs 2 things, a host program and a kernel. the host program is usually written mid/high level language like (C, python etc..) the host program handles data management, complies kernels etc... the host program detect hardware GPU and complies and send the kernels to the GPU and feed the kernels with Data. in Houdini the openCL wrangle is the host program. the openCL Kernels (code/snippets) can only handle the data from local GPU/Memory which it got send from the Host program (OpenCL wrangle node). current openCL wrangle Houdini does not support multi GPU's. you have write you own openCL wrangle node via houdini API to make it happen, but you have to dive deep into parallel processing and data management or the GPU's sit there waiting most of the time for Data to be moved around. If you rewrite nodes as openCL version, you need basic routines first. openCL is pretty low level. you need to write function like noise, PCopen, neighbours, ramp's etc... based on lastest test Houdini does not take full advantage of shared memory nor does it use multi GPU's. Pyro or Vellum uses only little part of the GPU. I could get self written Kernels to use 100% of the GPU for simpler Operation. I know only handful software packages, which run entirely Simulation's on GPU's. I am not sure, but I think only Cuda can take advantage of NVlink right now.

i using classic Julia Fractal to drive my spline : cool spline (Mantra) and black-and-white

go through first 3 chapters here: http://www.scratchapixel.com/lessons/mathematics-physics-for-computer-graphics/geometry/points-vectors-and-normals it good explansion about vector's, Normal's etc...

thanks for sharing !

look nice !

you can use lifespanPP = rand (0.5,2); you should go to maya forum, you will get not much feedback in a houdini forum about maya effects.

splines rendering with indigo renderer

tks, i did some volume renders with displacement, Lee Giggs style, but the rendertime was extreme slow. julia, manldebulb is quite data heavy and limits creative freedom. i am using only real geo / splines instead.

if you wanna learn more The physics of wind-blown sand and dust "

i am playing more with Vex right now: Vex Code: float resX = 600; float resY = 600; float deltaD = 0.5; int numVert = 3; float pi = 3.1415922653; float rotOffset = pi / 6.0; vector pos= {0,0,0}; float cx = (resX/2.0)*cos((2*pi/numVert) + rotOffset) + (resX /2); float cy = (resY/2.0)*sin((2*pi/numVert) + rotOffset) + (resY /2); int iter = chf("iteration"); for(int i=0; i<(iter+15); i++) { int num = int(((rand(i)*32767) % numVert) +1); float xVert = (resX/2) * cos((2*pi*num / numVert) + rotOffset) + (resX/2); float yVert = (resY/2) * sin((2*pi*num / numVert) + rotOffset) + (resY/2); cx += deltaD * (xVert - cx); cy += deltaD * (yVert - cy); if(i>15) { pos.x = cx; pos.y = cy; addpoint( 0, pos ); } }