Jump to content

markingleukc

Members
  • Content count

    34
  • Donations

    0.00 CAD 
  • Joined

  • Last visited

  • Days Won

    3

markingleukc last won the day on July 9

markingleukc had the most liked content!

Community Reputation

25 Excellent

1 Follower

About markingleukc

  • Rank
    Peon
  • Birthday 06/17/1993

Contact Methods

  • Website URL
    www.markinglevfx.com

Personal Information

  • Name
    Mark
  • Location
    New York

Recent Profile Visitors

1,225 profile views
  1. TOPs video mosaic for simulations

    Haven't used PDG yet myself, but from what I understand the generate_contact_sheet is an imagemagick node set to montage. And the gather_matching_frames would be a partitionbyframe node to make sure the mosaic renders the animations in sync. Check out Entagma's PDG series, part 1b specifically for creating mosaics.
  2. Floating Fountain

    A passion project re-using footage shot a few years ago. The water is split into two sims: the first covering the sphere and the downward stream, the other filling the basin and overflowing around the sides. A total of over 7 million particles make up the fluid sim. Software used: Houdini Redshift Nuke
  3. CacheIt -- HDA

    Nice! I have made a similar kind of thing for myself. Something I included was adding 'metadata' as detail attributes to help identify caches easier than just their version number. Definitely helps when your client asks you to revert back a few versions to one they liked earlier! Maybe you could add this in too?
  4. Floating Fountain WIP

    I've been remaking this old shot I had worked on 3.5 years ago, taking the old footage/files and completely reworking the sim, render and comp. Feedback welcome
  5. Fragrance FX spot

    The reassembly was a reverse particle sim. I didn’t use packed rbd because collisions weren’t a priority and I needed to keep things simple. Each particle represented a fractured piece and the transforms of the particles were applied to the pieces after the sim.
  6. Fragrance FX spot

    I'd like to share a project I have been on for the last 3 weeks as the sole creator. All CG work was done in Houdini, Redshift, and Nuke. Some minor work was also done in PS and AE. Shoutout to the Gridmarkets team for the support and helping me meet this tight deadline! Feedback/criticism is welcome. https://vimeo.com/346944545
  7. Whitewater spray- pop replicate on velocity

    Something like the file attached? You can do 'emit from attribute' in the point replicate. So all you would need is to define a 0 to 1 float attribute to drive that dop. I called that attribute 'birth_chance'. Your speed would be calculated by taking the length of you velocity vector: length(@v) And you can refit that speed value to isolate the faster points for replication: f@birth_chance = fit(@speed, 1, 6, 0, 1) You can also multiply this by a refit of the spray attribute to narrow it down further to just spray particles within a certain threshold. f@birth_chance = fit(@speed, 1, 6, 0, 1) * fit(@spray, .3, .5, 0, 1); I also isolated these new replicated spray particles in a pop group, so that they can be ignored by the pop replicate - otherwise you have cloned points creating cloned points which gets messy quick. If you need additional help, upload your .hip workfile so we can get a better sense of how you're working flip_whitewater_from_v.hipnc
  8. affect all except a group

    preface it with a '!' !group1
  9. Smoke Target Force

    I've attached a cleaner and updated (for H17.0+) version of this hip file. In the original file, the rubber toy was in there for initial tests. It serves no purpose. Load your alembic geo in as houdini geo or delayed load primitives. Put down an unpack sop, and a convert sop to manipulate it in Houdini. If you're in need of more help, upload your file and your alembic caches so I or anyone else can set you in the right direction. run_smoke_v003.hipnc
  10. You can try using a pointdeform sop instead of the transform, using the animated rings as the deformation lattice. curtain_fix2.hipnc
  11. I believe you need to add animation to the curtain for the pinned points to connect to. Even if you add animation to the whole geo and plug it into the vellum solver, the solver only takes the first frame of the geo - unless the points are in a pin group. curtain_fix.hipnc
  12. Move pivot to center of other node?

    Ah okay, then I would use expressions such as below to calculate that point without having to do a merge. Add these to the pivot translate of the 'rotate' xform node. And vice versa for 'rotate1', just edit the path of the second centroid function. (centroid(opinputpath(".", 0), D_X) + centroid("../rotate1/", D_X)) / 2 (centroid(opinputpath(".", 0), D_Y) + centroid("../rotate1/", D_Y)) / 2 (centroid(opinputpath(".", 0), D_Z) + centroid("../rotate1/", D_Z)) / 2
  13. Move pivot to center of other node?

    Looks like a dependency loop. In the 'rotate' xform node you're referencing the centroid of 'transformall' - which is downstream. edit: Looking at your original file, $CEX $CEY $CEZ in the pivot translate of 'transformall' seems to work fine for giving you a pivot between the two boxes. They only seem to evaluate to 0 using the old dollar sign variable syntax. Use the following expressions in the pivot translate of 'transformall' to get the equivalent in hscript which evaluates correctly. Then use the 'transformall' rotation controls to rotate the two boxes together. centroid(opinputpath(".", 0), D_X) centroid(opinputpath(".", 0), D_Y) centroid(opinputpath(".", 0), D_Z)
  14. Yep, you got it Although instead of using three separate attributes, you can use '@id+123', '@id+456', '@id+789' for the x, y and z random seeds (or any other values you see fit). The rand() function just uses the seed you give it to generate a value from 0 to 1.
  15. Creating a primitive out of a stream of particles

    Add an attribute to your source points to identify where each particle has come from. Your first popnet will by default give you an 'id' attribute already, so use add an 'attribute rename' sop to change this to something like 'parent_id' (otherwise your new popnet will overwrite it). Then after the second popnet, use the parent_id attribute in the add sop to connect up those groups of points. Check the attached hip file To make it a tube, add a polywire sop at the end. pop_lines.hip
×