Jump to content


  • Content count

  • Donations

    0.00 CAD 
  • Joined

  • Last visited

Community Reputation

9 Neutral


About violalyu

  • Rank
  • Birthday 10/15/1994

Contact Methods

  • Website URL

Personal Information

  • Name
    Wanqi Lyu
  • Location
    Los Angeles, CA

Recent Profile Visitors

745 profile views
  1. Hi Mark, Thank you so much for the reply! The first part for setting env variables via hou.putenv(), I thought about that at first, then I realized that this would not work for multiple MPlay sessions parented under the same Houdini session( for example, user flipbooked from one angle, and then want to decide if they want to save/publish later, and they flipbooked from another angle/viewport, then we have 2 MPlay parented to this Houdini process,), unless we have a way in Mplay context to know which session it is in, we cannot distinguish which information is associated to which even if we serialized a dictionary and save it to Houdini env, or maybe you know how that could be achieved? Because as far as I know, the MPlay python exec seems to be running as a hython process? and doesn't have any context of the houdini session (nodes etc.) not even hou.ui is availble at that level... AND, I think you are absolutely right about the lock file part, it did created a lock file when it launched, before any flipbook call. I guess that is the mark of the socket connection finished? But thank you soooooooooooo much Mark, I think this is enough to confirm that the mplay session exists! This is really really helpful for me! Best, Viola Lyu
  2. Hi Mark, Thank you so much for the reply! I didn't have the .mplay_lock files existing, .flipbook_lock files do exist. I think the situation here is a little bit different, where the actual scenario here is, I was trying to launch an MPlay session with a given env dict where I stored some information I need in the environment variable via a subprocess.Popen in a dialog I created, and then I was calling the sceneviewer.flipbook() but with the desired settings passes in, so the session label matches the one I brought up, so that I can do things in the MPlay menu item etc. What I've noticed is that, sometimes the MPlay session I brought up may not been connected to the main Houdini process when the flipbook() is executed, so the flipbook() created its own MPlay session and that got the priority so the original one (the one I created before hand) got deleted, and I don't have the desired environment variable anymore. I was trying to see if there's a way that we can avoid this racing situation, other than have a magic number of sleep before the flipbook() runs. Is the .mplay_lock file created after it starts to listen to a flipbook session? Thank you sooooooo much again for the reply!
  3. Hi Houdini friends, Does anyone tried or know, how to get a list of active MPlay sessions from the main Houdini process? Basically like a list of port - Mplay session pairs or the like, or even session labels. Thank you so much in advance! Best, Viola Lyu
  4. Hi Houdini friends, Does anyone have any insights/experience on how you could get information within an MPlay session, for example the session label? The reason for this is that we want to have some support with the flipbook generation where for example we can take the rendered image sequence (filename is set and known) on disk and manipulate them. But when we are doing multiple MPlay session within one scene or, across multiple scenes, we kind of need some information from the MPlay session to distinguish which path to use. Or is there other way that would achieve similar things? Thank you so much in advance! Best, Viola Lyu
  5. Hi Houdini friends, I'm not sure if anyone had this before, I'm having this really weird issue where I specify a digital asset to show some info with its "Descriptive Parm" field, and I specified a parameter for it to read, say 'display_parm'. The result of evalParm('display_parm') shows the correct result, but the actual displayed info is different from this result. In this case it is a string containing 3 parts, the actual displayed part is missing the middle part. It is as if the actual displayed info is using another parm or something. And this is only happening to some of the test cases not all of them. Has anyone had similar issues before? Thank you so much! Best, Viola Lyu
  6. Pyro collision with packed objects

    Hi Dragan, If you would post an example file maybe it is easier to spot on what's going on:)
  7. Pyro collision with packed objects

    Hi Dragan, Maybe try use volume sample in collision and link the VDB of your object as proxy volume?
  8. Emit particles from falling RBD object

    Hi Art3mis, I think maybe one thing is that since it is a packed rbd object in your scene, when you try to source it it was only a point, I did a quick test and it was emitting particles at the center point; it worked as in the video when the torus is a RBD object instead of packed (as shown in attached example). Hope it helps! Best, Viola Lyu particleDopTest_vl.hipnc
  9. growing dynamic curve

    Hi Jason, I've seen this video that is similar where Nema was using pop grain to achieve similar stuff, and he shared his scene file as well so if you want to check it out and have a starting point: Best, Viola Lyu
  10. Simple Circle Packing

    Hi Vusta, Not sure if you still have question about creating the button to reset the sim, but it you do you could try this: You can create a button in your UI and in callback script you can use hou.node('PATHTOSOPSOLVER').parm('NAMEOFBUTTON').pressButton() Best, Viola
  11. target position for particles

    Hi Evan, Maybe you could try popSteerSeek, it was for flocking/crowd but it allows for a certain level of control over the force and behavior. Hope it helps! Best, Viola
  12. Thank you so much! Somehow I cannot hit thank you on your entry...lol
  13. Falling Dust - Characteristic

    Still testing on this lol
  14. Falling Dust - Characteristic

    Hi Houdini friends! I've been trying to do a sim for falling dust, the kind of look I want to get is similar to this: The shape of the leading edge of the falling dust is "spiky" and has the grainy details. I kind of "cheated" by giving it a high disturbance and warp the velocity field so that all vel.y will be going down and scaled down vel.x and z, but when there's interaction with collision geo this will not work since the velocity field is manipulated, and it gives this "sliding" behavior along the collision geo. Attached is an example file that shows this issue. Has anyone done similar sim before or have any advice on how to get this kind of characteristic in the sim? Thank you soooo much in advance! Much appreciated! fallingDust_test.hipnc
  15. Pyro smoke localised drag

    Hi Agelos, you could try gas damp, you can check out the parameters and play with them, should be able to get that effect:)