Rudinie Posted October 20, 2008 Share Posted October 20, 2008 Hi everyone, when i use the points of a sphere as the source to copy a couple of boxes to, the boxes are nicelly rotated according to the point-normals. Whenever i use the same sphere as input for a Popnet and use the Popnet as the source instead of the sphere in the copyNode, the rotation of my boxes is all set to align to the z-axis. So i think my particles don't use the normals from the input, allthough i did use "inherit attributes" in the source node in popnet. How can i get the point normals from my sphere as normals for my particles? Or is there something else i'm doing wrong? I attached the file so you can have a look at what i mean. I hope someone could give me a little help and point me (and my boxes) in the right direction. Thanks in advance. Rudi ball.hipnc Quote Link to comment Share on other sites More sharing options...
Rudinie Posted October 20, 2008 Author Share Posted October 20, 2008 Ah, i found out myself, adding a point sop before sending the points to POP did the trick. I had to set the inherit velocity to a very low value to avoid the particles flying away straight away. Now i need to find how to control the orientation in POPS. Rudi Quote Link to comment Share on other sites More sharing options...
Rudinie Posted October 21, 2008 Author Share Posted October 21, 2008 Now i need to find how to control the orientation in POPS. This is actually much harder than i thought. How do i use the Normals (that get in from the points i use as input to POPS) to rotate all the particles so they face the same direction. A "lookat" doesn't work because the normals are all off. How do i correct for the normals? Can i calculate how to rotate the particles so i correct for the normals? (Does this make sense?) Craig Zerouni writes in his book (on page79) that "it is also possible to add rotation attributes afterwards, but that is more difficult, because they are in quaternions" Quaternions, does that mean the fact that you have an angle and three axis? So how do i transfer the normals (3 axis) to a rotation? I could use some help on this, so if anyone would care to explain that would be great! Rudi (confused) Quote Link to comment Share on other sites More sharing options...
Owl Posted October 21, 2008 Share Posted October 21, 2008 is this what you need? ball_2.hipnc Quote Link to comment Share on other sites More sharing options...
Rudinie Posted October 21, 2008 Author Share Posted October 21, 2008 (edited) is this what you need?ball_2.hipnc Thanks for your reply. I get an error however when i open your file, related to "Owl_redoguides1", using incomplete asset definition (full definition not found.) Edited October 21, 2008 by Rudinie Quote Link to comment Share on other sites More sharing options...
Rudinie Posted October 23, 2008 Author Share Posted October 23, 2008 I've found that using a voppop (never used that until now) will do the trick. I had to use a keyable parameter outside voppop (keyed from 0 to 1) and get that in to slowly set the normals from the first input geometry of the pops network to the second. Is there a way to do this without keyframes? Quote Link to comment Share on other sites More sharing options...
Owl Posted October 23, 2008 Share Posted October 23, 2008 (edited) Thanks for your reply. I get an error however when i open your file, related to "Owl_redoguides1", using incomplete asset definition (full definition not found.) oh darn my bad, that otl doesn't belong in there can you show that file? Edited October 23, 2008 by Owl Quote Link to comment Share on other sites More sharing options...
Miles Posted November 7, 2008 Share Posted November 7, 2008 I just added an point sop, and turned on normals. ball_sol.hipnc Quote Link to comment Share on other sites More sharing options...
Rudinie Posted November 11, 2008 Author Share Posted November 11, 2008 can you show that file? Sorry for the late reply, totally slipped my mind. Here is a file where i use a grid as the second input and in a voppop i retrieve the normals from that grid to make the squares all turn to look in the z-direction. Rudi normals_control.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.