Jump to content

Hartman-

Members
  • Content count

    14
  • Joined

  • Last visited

Community Reputation

4 Neutral

About Hartman-

  • Rank
    Peon

Contact Methods

  • Website URL
    ianmhartman.com

Personal Information

  • Name
    Ian

Recent Profile Visitors

266 profile views
  1. store point number in array

    It's what made sense to me at the time, it works but I should probably revisit my logic. Was a bit tired when I made the example!
  2. store point number in array

    Those are much simpler than mine, now I'm learning!
  3. Hey! Yeah, look at the hip file I've attached. I'm using a sop solver with a single wrangle that gets piped into the rbd solver, post or pre solve, doesn't matter. If you are calculating mass in the rbd object, then you should have a good range of values to use in that wrangle. I'm just doing an animated float channel to determine whether or not a piece is active. Hope this helps! RBD_Active_By_Mass.hiplc
  4. store point number in array

    What you want to do here is to run a wrangle as a detail (so it will only run once) and then loop through the points. This way, you can build upon the array and then set the array to each point. See attached file! Create_Point_Array.hiplc
  5. Manual cook on open/save

    As a quick update... it looks like the best solution for the frame range to start at $FSTART would be to do something like this: firstFrame = hou.playbar.timelineRange()[0] hou.setFrame(firstFrame) timelineRange() returns the first and last frame of the animation settings ($FSTART, $FEND), not necessarily what the timeline is set to ($RFSTART, $RFEND)
  6. Manual cook on open/save

    For the frame range... You can try to execute hscript via python using hou.hscript() http://www.sidefx.com/docs/houdini/hom/hou/hscript Similar to what this guy was trying to do: https://www.sidefx.com/forum/topic/43552/ I'm not at my box to confirm but these seem like they are what you want. Also look into the hou.playbar module, lots of good stuff in there. I'm not sure about saving in manual mode... I'll have to do a bit more research myself, I've always just defaulted to opening in manual mode, no matter what.
  7. wind tunnle smoke

    Hey, Have you watched this Illume Webinar about pyro fx? (https://www.sidefx.com/tutorials/houdini-illume-pyro-fx-tips-and-tricks/) They talk about driving pyro using particles as source and the like. I've jumped around, so I can't speak to the content directly, but what I caught they were using pop's that were converted into fog volumes. If you want a simple fog volume, you could use the vdbfromparticles node and then add your points attribs to the fog with that node as well (such as velocity for motion blur). Hope this helps!
  8. Manual cook on open/save

    Is the 456.py file in the correct location? It should be located in $HOME/houdiniXX.X/scripts. On windows this will most likely be C:/Users/USERNAME/Documents/houdini16.0/scripts. I'm not on a mac box right now and I'm blanking on the location, you should be able to get a idea for it through some quick commands. http://www.sidefx.com/docs/houdini/basics/config_env#browsing-environment-variables ... you could even go into the save dialog and type in $HOME to path to the home directory. The first 10-15 minutes of this video talks the 123.py and 456.py files and how they are used. https://www.sidefx.com/tutorials/illume-webinar-houdini-in-the-pipeline/ My best guess is that the file isn't being found. I was able to setup a quick 123.py file that set the update to manual and the frame to 30 without trouble. Hope this helps!
  9. hou.ui.selectFile() not selecting folders in h16

    Juraj, Do you have any parameters set when calling the hou.ui.selectFile() method? I'm doing some quick tests in the Python shell and it looks like its working fine for me. I'm selecting a folder and its printing out the correct path to the shell, even when I specify a file type.
  10. grow solver problem + attribute copy

    So, the logic that you've setup means that every frame you are merging the previous frames scattered points and the current frame. In the first example this works fine as the chances that two points will be on top of each other is fairly slim but not impossible. But for switch 2 and 3 the points never change their position so your point cloud function in the solver will always find points that are nearby (aka they are on top of each other) and the hexes will never grow. The simple, but super costly at scale, solution is to put a fuse sop after your merge... But after a couple seconds the simulation chugs... so not recommended. Also, what kind of look are you going for with the second/third examples? Do you just want them to grow after being activated or do a wave where they grow and then shrink as the wave passes? I'll try and whip up a file to show. Hope this helps!
  11. I believe it has to do with the unique points along seams option you have on... if its the surfaces that are offset from the fractured bits (circled in attached image). booleanShatter_withVDB_spheres_fix.hiplc
  12. [SOLVED]Scatter By Inverse Volume Density?

    I whipped up a quick file to demonstrate this. invert_density_volume.hiplc
  13. Hey all! I'm pretty stuck in a rut trying to think my way through what I want to accomplish with a brick tool I've been working on. Currently I can feed in a box, tube, or an L-shaped box that I've modeled and I can get well stacked bricks around the perimeter. What I'm stuck on is then taking this and adding in holes for windows/doors/etc. Similar to what is seen here: https://letshoudini.wordpress.com/2016/06/19/031-procedural-modeling-for-destruction/ What I want to be able to do is copy a pre-fractured brick to the points, like the simple box I have now in the attached hip file, with rotations to help vary the layout a bit. Thus, should I boolean out holes from the mesh after generating bricks? Maybe taking into account where windows are (will be) and then not placing brick points there? (then I would have to have half bricks to copy). Not sure the booleans would behave as I want to be able to instance high detail geo to the pieces after simulation. The solution in the link looks to be a boolean as the simulation then only runs with simple bricks clustered to add some interest to their breakup (https://letshoudini.wordpress.com/2016/09/18/037-earthquake-building-collapse-rd-01/). I don't think I'm seeing this problem clearly right now and would really appreciate some nudges in a direction to help grease the logic wheels! Hartman- BuildingForDestruction.hiplc
  14. I'm trying to debug why upres doesn't work when importing a pyrocluster. It looks like the upres node is only recognizing the first set of pyro cluster data being read in, i.e. the data from the pyro_0 group which would correspond to the first cluster. I can manually tell the upres node to use the different groups of data that would correspond to the other clusters (pyro_1, pyro_2, etc), but not multiple groups... only one. I have no clue how to go about reading in a pyro cluster cache so that I can upres it... I feel like this should possible in Houdini without having to setup upres nodes for each cluster in the simulation. I've attached the most basic scene (copy sop with spheres that are the cluster points for a pyro cluster sim) that I'm using to try and get the upres to work before integrating into any other scenes. pyrocluster_forupres.hipnc
×