titor Posted May 31, 2011 Share Posted May 31, 2011 Hello everybody. I'm having some trouble to find a way of getting a group of mesh that are from a fractured mesh. To explain what I'm trying to achieve : -I have a box that i want to fracture but with high details on a precise place. -I then connect a sphere to an isoOffset and place it where I want to scatter points that the Voronoi fracture node will use. -Then i connect the Voronoi node. Here is what i get in the first attached file. Then i would like only the small pieces to be passed to a DOPNetwork, and i was thinking of using the isoOffset to achieve that but can't get it working. The Voronoi node cuts the mesh in groups, is there a way to get the groups that are in the isoOffset ? Here's the hip file if any of you have an idea. Thanks for your help RBD_Tentacles.hip Quote Link to comment Share on other sites More sharing options...
lukeiamyourfather Posted May 31, 2011 Share Posted May 31, 2011 Set a cluster attribute on the points handed to the Voronoi fracture node. Use zero for the pieces to be not connected then use another value other than zero for the other clusters. Or do all of this with the Voronoi fracture points node which has the clustering parameters integrated. Quote Link to comment Share on other sites More sharing options...
lukeiamyourfather Posted May 31, 2011 Share Posted May 31, 2011 Examples attached for each method if it helps. fractureClusterExample.hip Quote Link to comment Share on other sites More sharing options...
titor Posted May 31, 2011 Author Share Posted May 31, 2011 Hey, thank you for your answer Luke. But I don't quite get the way you transfer the cluster to the DOPnetwork. Is the digital asset ExplodeView yours ? If i get it right, it's just here in order to translate the pieces to show me, how the cluster handles that. Quote Link to comment Share on other sites More sharing options...
lukeiamyourfather Posted May 31, 2011 Share Posted May 31, 2011 Hey, thank you for your answer Luke. But I don't quite get the way you transfer the cluster to the DOPnetwork. Is the digital asset ExplodeView yours ? If i get it right, it's just here in order to translate the pieces to show me, how the cluster handles that. Before the exploded view node (just after the Voronoi fracture) is where the geometry should be sent to the DOP. The exploded view node is just to visualize the fracture. The RBD glue object tool on the rigid bodies tab of the toolbar will allow you to import multiple pieces of geometry based on group name and glue them together. By default it pulls groups that start with piece like "piece*" but that can be changed to something else if needed. Quote Link to comment Share on other sites More sharing options...
titor Posted May 31, 2011 Author Share Posted May 31, 2011 Ok, so here's the set-up i did to break a box with some sickness. (see the attached file) I used the user point, you used in your hip file. But when I try to just translate one of the groups called piece0, It just translate the inner part or the outside part but not what was in the cluster. I haven't tried to put it in DOP yet. Thanks again for your precious help Quote Link to comment Share on other sites More sharing options...
titor Posted May 31, 2011 Author Share Posted May 31, 2011 Here's the HIP file i forgot it RBD_Tentacles.hip Quote Link to comment Share on other sites More sharing options...
Macha Posted May 31, 2011 Share Posted May 31, 2011 You want only pieces of a certain size to be send to the dopnetwork? Quote Link to comment Share on other sites More sharing options...
titor Posted June 1, 2011 Author Share Posted June 1, 2011 I want only the pieces that were fractured in my isoOffset to be sent to de DOPnetwork, because I want to put the impact point on this pieces. For example I'll have a sphere, which is a static object with an animatioon, which is going to impact on those pieces. And i want them to blow away, whilst the others are staying put. If you need me to detail more tell me. And again thanks for your help guys Quote Link to comment Share on other sites More sharing options...
Macha Posted June 1, 2011 Share Posted June 1, 2011 (edited) The attached file attribute-transfers an attribute from the original seeds. Then, on a per-piece basis, it accumulates the attributes, and with a couple of attribute-promote nodes I get a single value for each piece. You can group those pieces according to what value they have and only pipe those into the simulation. Something like that perhaps? You get uneven parts though. I suggest you evenly divide the pieces to get a better result. Maybe just evenly divide them for the purpose of getting the initial selection and then use the original geometry (maybe with connectivity class attribute because they will remain the same, no matter how many times you subdivide) clusterme.hip Edited June 1, 2011 by Macha Quote Link to comment Share on other sites More sharing options...
Macha Posted June 1, 2011 Share Posted June 1, 2011 (edited) Here Titor, this is a bit better now; more balanced. Now you only need to use the other bits as static rbd. clustermeb.hip Edited June 1, 2011 by Macha Quote Link to comment Share on other sites More sharing options...
titor Posted June 1, 2011 Author Share Posted June 1, 2011 This seems to fit perfectly my needs Macha. I would just need some explanation on your foreach loop, cause this node is quite new to me. I just tried to launch quickly the simulation and set the glue strength to 0 an internal glue strength to 0 too, to see the pieces breaking apart, but they seem to stick together. I haven't looked deeper cause I can't right now but, I'll be back to you this evening with a lot of questions I'm sure Thanks to both of you guys you really helped me. Quote Link to comment Share on other sites More sharing options...
titor Posted June 6, 2011 Author Share Posted June 6, 2011 Hello Macha. I'm having some troubles using your last setup when it's about dealing with DOP. I succeeded in fracturing and grouping my pieces, but when I put them in the DOPNetwork and then launch the simulation, my pieces fall but stick together. I looked to the glue parameter and put it to 0 re-ran the simulation but nothing changed. Here is my file, could it come from the fact that once the pieces are fused in the foreach it only create one mesh instead of re-creating the pieces from the Voronoi ? ('m not sure if I'm still clear ^^) Thanks for your help clustermeb.hip Quote Link to comment Share on other sites More sharing options...
Macha Posted June 6, 2011 Share Posted June 6, 2011 Victor, you didn't create any pieces except for a single big one clustermec.hipnc Quote Link to comment Share on other sites More sharing options...
titor Posted June 6, 2011 Author Share Posted June 6, 2011 (edited) Oh thank you Macha once again. Now it works perfectly. So if I got it well, you're grouping the pieces together thanks to this expression @cluster>0.5 in the Blast node ? Is the @ sign some special character that can be used to do operations based on conditions ? Thanks again for your patience and help Edited June 6, 2011 by titor Quote Link to comment Share on other sites More sharing options...
Macha Posted June 6, 2011 Share Posted June 6, 2011 Well almost. I kept the original pieces but blasted away those for which the attribute cluster was>.5 The group that you created was just that; one group. I suppose the partition node could have been used as well. Quote Link to comment Share on other sites More sharing options...
titor Posted June 6, 2011 Author Share Posted June 6, 2011 Ok thanks for your explanations. 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.