Jump to content


  • Content count

  • Donations

    0.00 CAD 
  • Joined

  • Last visited

  • Days Won


Everything posted by michaelb-01

  1. Can I set the IPR background image via python?

    I was trying to do this recently too but couldn't figure it out and assumed it wasn't possible through python. Did you manage to figure it out?
  2. Mplay and python

    @trzanko Have you managed to run a python script from an Mplay menu? I've tried importing it from: $HOUDINI_PATH/scripts $HOUDINI_PATH/scripts/python $HOUDINI_PATH/python2.7libs In Houdini all of those paths are fine but Mplay doesn't see them.. And I can't seem to do a relative import either..
  3. I'm exporting some particles from Houdini via alembic (importing via exocortex crate), but how do I control the size of the particles when rendered (rendering in Arnold)? I tried setting pscale and radiusPP (on points and verts) in Houdini but nothing seems to work. And thinking about it I guess I would need to get opacity per particle exported too.. Or I could re-map these based on age in Maya.. Has anyone done this before?
  4. I'm using Arnold and Aces (1.0.3). Our backplates are written out as Aces 2065-1. However there are no options to set the colour space when loading a background image in the render view. It seems to use the default output of the Arnold Rop which is AcesCG so the colours are slightly off. I thought maybe I could load it through COPs but again there are no real controls for setting the colour space (when using ACES). I can just project an image onto a grid as a backplate but does anyone have a way to colour correct the background image in the render view?
  5. Transform packed fragments at render time

    I never did unfortunately.. Let me know if you do!
  6. I am working on a big destruction job and I’m trying to optimise our workflow as much as possible. my usual workflow: - cache a single frame of my fractured geo as packed fragments - import rbd sim and create point per rbd piece (dop import) then lighting would import these two caches and use transform pieces. This works well since we only cache a single frame of the fractured geo, and the cache for the rbd points is very small. however the geo is still written into the ifd file every frame, which can be quite large. So, is it possible to specify the geo needed for the whole sequence and then have mantra transform the geo at render time (the same as transform pieces). This would make the ifd’s tiny an alternative method is to write each fractured piece out as a separate file then copy empty packed disk primitves to the rbd points then use the unexpandfilename intrinsic to expand it at render time. This makes tiny and fast ifd files which is great but seems quite slow to render - probably because it has to pull so many files from disk (1000s of pieces).. is it possible to do the render time transform pieces approach or does anybody have a better method? (The two I’ve mentioned are fine I’m just trying to optimise it!)
  7. Houdini's hairgen node displays the hair polygon lines with thickness defined by the width point attribute. How can I visualise this in the viewport on normal curves/polygon lines?
  8. If I have parameters on digital assets that are expressions I sometimes find it useful to have another parameter that shows the result of the expression. I can obviously middle mouse on a parameter to show the evaluated result but is there a way to show the evaluated result by default? I feel like the answer is no so my solution would be to use callbacks for any parameter that affects the expression then update the 'result' parameter as a raw string, unless anyone has a better suggestion?
  9. I'm exporting some simulated geometry to Maya from Houdini and I keep getting the following error: UVs aren't per-vertex or per-polygon per-vertex, skipping And it won't load any UVs... All of the geometry has UVs in houdini and there are no "null" uvs (one point has 0,0,0 but that is correct). I've deleted all groups except the inside and outside primitive groups of the fracture and all attributes except: point: P vertex: N, uv Alembic format is HDF5. I don't know what else to try!
  10. Get full path to paneTab

    Sorry I thought pane.pwd() was just returning a string but I think its actually the hou.node so I can just do hou.pwd().path()
  11. Get full path to paneTab

    I'm trying to get the full path to the parent node in the current paneTab under the users mouse. I'm currently doing the following to return the paneTab: pane = hou.ui.paneTabUnderCursor() I can get the current parent node with: pane.pwd() So if i'm in /obj/geo1/subnet1 this will return subnet1 but how do I get the full path? (/obj/geo1/subnet1)
  12. I'm trying to add pop forces to ragdolls so they flail their limbs a bit however my pop force seems to have no effect (though gravity does..). I've attached a simple scene falling_mlb_v001.hiplc
  13. Speed up Mantra IPR

    Unified noise seems to be much worse than a turbulent noise. When rendering a unified noise the scene takes 3 or 4 seconds to generate, however with turbulent noise its milliseconds. Obviously unified noise is a much more complex noise but i'm surprised its that slow..
  14. Speed up Mantra IPR

    I'm having the same issue. Usually I render in Arnold and i'm used to very fast start and refresh times. I'm developing my material at the top level (in /mat) then i'm going to collapse it into a material later. This way I can drag shading nodes directly onto geo in the ipr and it will make me a temporary material and render just that shading node - this doesn't seem to work if you are inside a material builder.. So i've isolate a small patch of my geometry, i'm rendering a very small region of the IPR and i'm rendering a unified noise just to tweak the parameters, but every time I make a change on the noise it re-generates the scene which takes a few seconds. It makes look dev very painful... Anyway to prevent mantra from re-generating the scene (only refresh the materials..?)
  15. Ocean - no displacement with Deadline

    Probably a stupid question but i'm not a mantra user... I'm currently rendering an ocean in Mantra on our farm with Deadline. I've written out the ocean spectra, foam etc and everything renders correctly on my box. I'm then submitting an IFD generation job and a mantra job (frame dependent on the IFDs) however it renders without any displacement on the ocean...
  16. Show curve thickness in viewport

    Perfect, thanks all!
  17. Flip Splash - Grid Scale Optimisation

    I'm doing a flip sim of a splash. My particle separation is 0.006 which produces 14mil particles so pretty low. However because its quite a big splash the bounding box of the sim gets quite large so by the end of the sim I have 600mil voxels in the surface volume and 115mil voxels in the vel volume. So this is crashing my machine (which has 128GB ram)... My grid scale is the default of 2 but the vast majority of those voxels will be wasted since there are no particles in most of the bounding box. Is there any way to optimise this? The obvious way would be to use sparse volumes in dynamics instead of Houdini's native volumes but I imagine that would be creating a flip solver from scratch.. Or can I increase the grid scale without it negatively affecting the sim too much (i'm going to test this now..) I've attached a flip book of the sim https://drive.google.com/open?id=1k3Kexf-DJ5_ULPLt6dpsfx5f9LtW2OVz
  18. Default autosave

    I'm trying to set this up for the whole studio, I have it working in my 456.py with: hou.hscript('autosave on') But autosave defaults to save every 1 minute, how can I change this (through the 456.py or some other method) to every 10 minutes for example? I can't find anything relevant in hscript or hou..
  19. FLIP Viscosity - Collision object grid stepping artifacts

    Hey Dave, I know this is quite old now but I've had problems with collisions when using viscosity in the past too. I've found I need to increase the exterior band voxels on my collision vdb to greater or equal than the surface extrapolation on the flip solver (default of 0.5). I wonder if this could fix your problem too?
  20. Grains - Disable Constraint Averaging

    I posted this on the SideFX forum but haven't had any response so I thought i'd try my luck here: I'm getting some unbalanced forces when using grains, its the same problem as described in the grains masterclass around 32:50. As Jeff says the solution is to disable constraint averaging and then insure all the internal forces add up to 1. I've done this but I still have an imbalance.. I wanted to check which forces are meant to add up to 1, in the video it implies that Internal Collisions + Clumping + Explicit Constraints = 1. However in the help document it mentions that Friction with Particles should be included as well. So should I set Friction with Particles, Internal Collisions, Clumping and Explicit Constraints to 0.25? (
  21. Grains cloth - bend resistance

    Using grains for cloth is extremely fast but since it’s not a fully fledged cloth solver it lacks control for creating realistic cloth sims. There is only really self collision and a basic stretch resistance through explicit constraints (through to avoid stretch you have to put constraint iterations very high..), but it doesn’t seem like there’s any concept of bend resistance in the explicit constraints. Does anyone know of a way of creating bend resistance or faking it somehow?
  22. Grains cloth - bend resistance

    Thanks for your reply Jamie. Yes thats similar to what I had in mind but would be a fair amount of work.. Obviously cloth does all this but its much much slower and I always revert to nCloth instead.. But when you have very high res cloth geometry and collision meshes (especially animated) Houdini's cloth is too slow and nCloth can't handle that amount of geo so i've been looking for workarounds. I had thought of adding another layer of grains but not ideal as it will slow it down a lot and there is not much control..
  23. Grains cloth - bend resistance

    Hi Chris, That makes sense, I suppose thats the best approach for getting bend resistance with grains, thanks for looking into it!
  24. Grains cloth - bend resistance

    Hey Alvaro, I think (though I may be wrong) increasing the simulation substeps (and/or the constraint iterations) will effectively increase the stretch resistance (so make it stiffer in that sense) but it won't have any effect on how much the grains (cloth) can bend
  25. Granular Sheet - Pin Constraint

    If you just want to freeze some particles you can use the ‘stopped’ integer attribute. If you want to animate the pinned particles you will need to update targetP in pops (as has been mentioned above). I’ve been doing a lot of ‘cloth tearing’ recently with grains and it requires a bit more of a setup but it’s doable. Essentially you need to: - prefracture the mesh - store a class attribute on the points (that will get picked up by the grain source) - for your constraints, in a primitive wrangle, check the two primitive points (using primpoint function) and compare their class values, if they are the same the constraint exists ‘within’ a fractured piece so we set its strength attribute really high (1x10^8 for example), if the class value is different you know the constraint connects two different fractured pieces so we want to make it tearable, so set the strength much lower (maybe 0.5) - enable break constraints on the grain solver and tweak the break threshold until you get some nice tearing. Post sim: - use a point deform node to drive the mesh by the points, but we want each fractured piece to be driven only by its corresponding grains points otherwise the mesh will stretch. You could try doing this in a for loop but that is very slow so I think the following is better: - time shift the mesh to frame 1, offset the point position by the class value (@P += i@class) and feed this into the first input of a point deform - time shift your grain sim to frame 1 and offset it by class too (it should line up in space with the mesh), and feed that into the rest input of the point deform - feed the animated grain sim points into the template points input of the point deform - if done correctly the mesh should tear cleanly with the points. To get a nicer tear there it’s actually a bit more work to be done, I’ll try and upload an example hip file if I have time. Maybe someone knows of another method?