Jump to content

merlino

Members
  • Content count

    103
  • Donations

    0.00 CAD 
  • Joined

  • Last visited

Community Reputation

9 Neutral

About merlino

  • Rank
    Initiate
  • Birthday 07/30/1978

Contact Methods

  • Skype
    marco.merletti

Personal Information

  • Name
    Marco
  • Location
    AR

Recent Profile Visitors

3,000 profile views
  1. Not sure, but have you tried putting a static solver for each static object and then to connect the two static solvers to a switch solver?
  2. Noise in Render

    Generally I tend to work that noise issue modifying "artistically" (read as "with no precise rule") the Noise Level and increasing the Max Ray Sample as the first step. Many times you can stick with a low Pixel Sample increasing the Max Ray Sample an lowering the Noise Level. Try to work in a small, noisy part of your render, so you can get fast feedback, and search for a good compromise between Render time and quality. If you can't manage to make the render less noisy that way, then start increasing Pixel Sample or Diffuse Quality or Global Quality. Hope it helps! Cheers!
  3. Happy 10th Birthday Od[force]

    Happy birthday!!! Congratz and thanks to all the community!!!
  4. Time-based paramater in /mat?

    Ok, I understand my bad! jeje ... Maybe creating the string as an attribute in sop and bind it in mat?
  5. Time-based paramater in /mat?

    Hi David, I'm not sure if it will works ... but have you already tried with "/my/disk/pointcloud.`$F`.bgeo" ?
  6. Birth particles on a shadow

    Buondí! If you're talking about self shadows I think you can make a dot product between normals and the difference from the points position and the light position. (if the light is a point light), something like this: @Cd = dot(@N,@P-point(1,"P",0))>0;
  7. I think the cheaper way, if you have already all approved and you just want to "merge" the effects, is to "point deform" the geometry. Just put the geo to deform (the dented, simulated one) in the first input, the bending box at rest into the second input (probably a timeshift at the first simulated frame will works) and the bending box simulated into the third input. I hope this helps!
  8. changing a parameter on everynode in my scene

    That's far easier and faster!!! Thanks!
  9. changing a parameter on everynode in my scene

    Surely with a python script that can look something like this: for object in hou.node("/obj").children(): if object.type().name() == "the type you're searching for": for node in object.children(): if ... //some other condition, if any node.set... //and here you set the parameter you want the value you want
  10. Mantra Render to Disk Stops Randomly

    Good idea the RFE Another option is to manage that with thinkbox deadline, for one (and maybe two) pc is free. And you can also manage dependencies and so on ... it's really useful for me
  11. Mantra Render to Disk Stops Randomly

    Sometimes you can have a read error loading the cache, you can try selecting "no geometry" instead of "report error" at "missing frame". That way it will be rendering your frame without your geo, so you must pay attention later.
  12. Blur Volume Using Volume Vops

    You can also convert the volume to vdb and use de smooth vdb node, where you can plug a mask for the blur effect (the mask should also be a vdb volume).
  13. Texture on flip mesh using rest, rest2

    Too late, really ... but I had the same problem working with rest and rest2 from flips and using procedural textures ... the only way it worked correctly was renaming rest as rest1 and then, in an attribute wrangle, using this expression f@rest = lerp(@rest2,@rest1,@rest_ratio); which is actually from this post by Mario Marengo
  14. Upcoming Build- future AMD vs Intel

    Hi @lukeiamyourfather! Just check here, there's some interesting info: https://www.sidefx.com/forum/topic/48975/?page=1#post-222234 It seems the Ryzen has a good performance on OpenCL, like a GTX 1070, but there you have all the system memory. It sounds interesting
  15. Upcoming Build- future AMD vs Intel

    Shouldn't be the internal OpenCL from the CPU? I used the i7 OpenCL capabilities in some test and worked well But now that you say that I go to check if the last CPU have OpenCL capabilities (Y)
×