karanjaura Posted May 18, 2018 Share Posted May 18, 2018 Hi guys I made this pyro cluster setup based on the technique showed by Atom in one of his tutorials. The problem I am facing is that the Vel Volumes are not getting transferred in the DOP. Any idea where the issue is ? Thanks a lot for your help.Cheers pyroClusterRnD_v01_ODFORCE.hip pyroClusterRnD_v01_ODFORCE.hip pyroClusterRnD_v01_ODFORCE.hip Quote Link to comment Share on other sites More sharing options...
bunker Posted May 18, 2018 Share Posted May 18, 2018 DOP's source volume only takes vel vector fileds for velocity, your fields are named vel_0, vel_1, ... if you're trying to source multiple vel volumes at the same time, you'll need some custom sourcing. Quote Link to comment Share on other sites More sharing options...
karanjaura Posted May 18, 2018 Author Share Posted May 18, 2018 1 hour ago, bunker said: DOP's source volume only takes vel vector fileds for velocity, your fields are named vel_0, vel_1, ... if you're trying to source multiple vel volumes at the same time, you'll need some custom sourcing. So you mean we can't pump velocities like that for cluster sims ?Then what would be the right approach to handle a situation like this ? Would really appreciate if anyone can guide me on that. Quote Link to comment Share on other sites More sharing options...
bunker Posted May 18, 2018 Share Posted May 18, 2018 if you convert your vel volumes to VDB (convertvdb+vdbvectormerge) you can access the volume data by name directly in a gasfieldwrangle inside DOPs, and add them all within a loop: // add all VDB velocity volumes to vel int clusters = 10; for(int i=0;i<clusters;i++){ v@vel += volumesamplev(0,"vel_"+itoa(i),@P); } 2 Quote Link to comment Share on other sites More sharing options...
karanjaura Posted May 18, 2018 Author Share Posted May 18, 2018 14 minutes ago, bunker said: if you convert your vel volumes to VDB (convertvdb+vdbvectormerge) you can access the volume data by name directly in a gasfieldwrangle inside DOPs, and add them all within a loop: // add all VDB velocity volumes to vel int clusters = 10; for(int i=0;i<clusters;i++){ v@vel += volumesamplev(0,"vel_"+itoa(i),@P); } Thanks a lot Julien for your help. I will try it out now.Cheers!! Quote Link to comment Share on other sites More sharing options...
karanjaura Posted May 20, 2018 Author Share Posted May 20, 2018 On 5/18/2018 at 3:06 PM, karanjaura said: Thanks a lot Julien for your help. I will try it out now.Cheers!! Hi Julien I tried the way you suggested but still the velocities are not getting pumped,not sure what I am doing wrong Would really appreciate if you could point out my mistake (pyroCluster_RnD_v02) on the other hand a friend of mine suggested me another way to make the same setup and the velocities are getting pumped perfectly fine in that (pyroClusterRnD_Optn02) Cheers pyroClusterRnD_v02_ODFORCE.hipnc pyroClusterRnD_Optn02_ODFORCE.hipnc Quote Link to comment Share on other sites More sharing options...
bunker Posted May 20, 2018 Share Posted May 20, 2018 (edited) The group in the convertVDB is incorrect. it should be @name=vel_*, same for the VDB vector merge (you can leave this one as default). In the gasfield wrangle, you need to specify an input 0 since it's used by the volumesamplev(0,...) pyroClusterRnD_v03_ODFORCE.hipnc Also, the whole point of clustering is to save simulation time or RAM by running each cluster on separate machines or one after the other on one machine. If you run all the clusters on one machine at the same time, you're just wasting time and RAM. Edited May 20, 2018 by bunker Quote Link to comment Share on other sites More sharing options...
karanjaura Posted May 20, 2018 Author Share Posted May 20, 2018 26 minutes ago, bunker said: The group in the convertVDB is incorrect. it should be @name=vel_*, same for the VDB vector merge (you can leave this one as default). In the gasfield wrangle, you need to specify an input 0 since it's used by the volumesamplev(0,...) pyroClusterRnD_v03_ODFORCE.hipnc Also, the whole point of clustering is to save simulation time or RAM by running each cluster on separate machines or one after the other on one machine. If you run all the clusters on one machine at the same time, you're just wasting time and RAM. Hey Julien Thanks a lot for your support :),I still need to learn many things from people like you, as I am still a newbie in the world of Houdini. cheers Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.