rdg Posted August 27, 2007 Share Posted August 27, 2007 Hi, how could I get this effect using POPs only? In the attached file a circle is copied to a particle. The circle emits new particles - like shockwaves. I guess it would be possible to do this with just one popnetwork - resulting in a setup easier to control. A cross product might help, but I neither know how to get a plane from a crossproduct - nor if this plane will help ... I would highly appreciate, if you find the time to explain this! perpendicular__split__ring_3.hipnc Georg Quote Link to comment Share on other sites More sharing options...
MIguel P Posted August 27, 2007 Share Posted August 27, 2007 (edited) Maybe this helps perpendicular_split.hipnc Edited August 27, 2007 by MIguel P Quote Link to comment Share on other sites More sharing options...
rdg Posted August 27, 2007 Author Share Posted August 27, 2007 [quote name='MIguel P Quote Link to comment Share on other sites More sharing options...
rdg Posted August 27, 2007 Author Share Posted August 27, 2007 yes it helped. 1: you build a vector N perpendicular to T and V 2: you normalize V 3: you split the particles 4: ??? you randomize a angle 0-360 and build a rotation matrix from this angle and the velocity ... and multiply this by N ??? and feed this as new velocity. I see it, I understand it partly. thanks so far. Georg Quote Link to comment Share on other sites More sharing options...
MIguel P Posted August 27, 2007 Share Posted August 27, 2007 (edited) I create 2 vector attributes in the initial particle. One perpendicular to the velocity vector and I normalize it. The other is the normalized velocity. I called it "vel". So with this 2 vectors, I go to VOPs and use the Rotate VOP. This vops creates a rotation matrix using the axis (a, b, c) and x angles. If I multiply this by any vector, It will be rotated x degrees in the axis (a, b, c). So if I use as an axis my "vel" attribute and "rand($ID)*360" degrees it will give me a random rotated vector for each particle and ever perpendicular to the velocity of the initial particle. I set it as the velocity and the most important, I set the group to the group created in the split and not preserved, so the veocity is only applied the frame it borns. Anyway, I don't know if this is the best method. It just occurred to me. Edited August 27, 2007 by MIguel P Quote Link to comment Share on other sites More sharing options...
rdg Posted August 27, 2007 Author Share Posted August 27, 2007 [quote name='MIguel P Quote Link to comment Share on other sites More sharing options...
MIguel P Posted August 27, 2007 Share Posted August 27, 2007 good.One note: rotate seems to expect radians not degrees. but for random stuff like this it doesn't matter. Georg Ouch! Well, you said it, it doesen't matter. Quote Link to comment Share on other sites More sharing options...
rdg Posted August 27, 2007 Author Share Posted August 27, 2007 [quote name='MIguel P pp_split_ordered_1.hipnc Quote Link to comment Share on other sites More sharing options...
MIguel P Posted August 27, 2007 Share Posted August 27, 2007 In the angle, connect the result of 1/(2*$PI)*$ID Quote Link to comment Share on other sites More sharing options...
rdg Posted August 28, 2007 Author Share Posted August 28, 2007 [quote name='MIguel P pp_split_ordered_5_4.hipnc Quote Link to comment Share on other sites More sharing options...
MIguel P Posted August 28, 2007 Share Posted August 28, 2007 I couldn't find a way to query to number of particles in the birthgroup. Mmm... yes, that's the question... Quote Link to comment Share on other sites More sharing options...
rdg Posted August 28, 2007 Author Share Posted August 28, 2007 do you happen to know what to multiply with what to create an additional circular motion to the emitted circles? they should keep their expanding motion but rotate about their splitcenter. I can add a attribute $TX,$TY,$TZ to the emitter that the split particles inherit. But any orbit/torque/force after the voppop shows no effect. I think we need an addition vector that is perpendicular to the linear motion and add/multiply/??? this to the velocity. Quote Link to comment Share on other sites More sharing options...
rdg Posted August 28, 2007 Author Share Posted August 28, 2007 It will be something like this: $V + a perpendicular vector spiral_movement1.hipnc Georg Quote Link to comment Share on other sites More sharing options...
MIguel P Posted August 28, 2007 Share Posted August 28, 2007 Do you mean a vector like this? (showed in the normals). perpendicular_split.hipnc Quote Link to comment Share on other sites More sharing options...
rdg Posted August 28, 2007 Author Share Posted August 28, 2007 [quote name='MIguel P spiral_movement_v2_2.hipnc Quote Link to comment Share on other sites More sharing options...
MIguel P Posted August 28, 2007 Share Posted August 28, 2007 (edited) Yes, my vector could do that. (cannot work with h9 - as every voppop is filled with your 1st version ) Can't you see my scene? To get the vector do de cross product between "vel" and the normalized velocity of the new particles created in the split. Edited August 28, 2007 by MIguel P Quote Link to comment Share on other sites More sharing options...
Visual Cortex Lab Posted August 28, 2007 Share Posted August 28, 2007 hey you guys.. have you finished to talk like in a bar all alone!?!? ... there's other people here working.... LOL .. sorry .. slly joke.. interesting reading though Miguel cool as always. cheers! Quote Link to comment Share on other sites More sharing options...
rdg Posted August 28, 2007 Author Share Posted August 28, 2007 Miguel, yes I can see your file and fixed my 'bug' (I can edit vopops again ...). The way we set things up it looks like its impossible to add generic forces: http://www.sidefx.com/index.php?option=com...opic&t=9381 After a few test with my h8 scene I think that the new vector must not be perpendicular to the velocity. The angle needs to be modifyable to have controll over the angular motion. In the meantime I played with my original scene and added some colors. perpendicular__petal_v3_6.hipnc Need more time to understand this whole particle and matrix thing. thanks for your time! Georg Quote Link to comment Share on other sites More sharing options...
MIguel P Posted August 28, 2007 Share Posted August 28, 2007 (edited) Hehe, cool scene. I think that is the vector to use. I attach a scene (houdini 8) showing it. I generate the tangents from the circle using the cross betwen P (a vector pointing outwards the origin) and a vector pointing to y positive direction (the circle is in the xz plane). The idea is the same for the other scene, but instead of using a vector (0, 1, 0), using the direction vector of the first particle. Thanks yo you for the challenge. I really enjoy this aspect of houdini. circle_tangents.hip Edited August 28, 2007 by MIguel P Quote Link to comment Share on other sites More sharing options...
rdg Posted August 28, 2007 Author Share Posted August 28, 2007 [quote name='MIguel P perpendicular_split_mp_new_vector_2.hipnc 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.