itriix Posted October 2, 2008 Share Posted October 2, 2008 Hello everyone, I have a question in reference to ordering points. Here is the basis of what i'm trying to accomplish: If you can imagine a deforming peice of geometry, moving along the Z-axis and clipping a slice of that geometry so that you can essentially get a SLICE of the geometry. Then fusing the points together to get a silhouette of points... however, when you do this, the point numbers are all in some random order. I've tried to think of some different ways that I could reorder them but none of the normal SORT methods or other methods i'm used to have worked. The reason for doing this is I want to achieve the effect of a blurred, trail left behind from something moving very fast... if you can imagine a person running, and leaving kind of a trail of their motion behind them, by taking interesections *slices* of the person at moments in time, you can build up many of these slices over time and SKIN them all together, then use an interesting shader... So, as of now, if anyone could help me i'd really appreciate this Thanks, Jonathan Quote Link to comment Share on other sites More sharing options...
aracid Posted October 3, 2008 Share Posted October 3, 2008 Hey Jonathan try looking at the trail sop, then divide sop * remove shared edges Then skin sop, it might be a solution for what ur after take a look at the file and scrub through the time line, its a really basic example but it should give u a better idea of what to do. hope this helps trail.hip Quote Link to comment Share on other sites More sharing options...
itriix Posted October 3, 2008 Author Share Posted October 3, 2008 (edited) thanks aracid, here is my .hip file... the problem still occurs for my situation -- so what i am doing in Houdini is: finding the points that make up the point sillhouette of the geometry *my method being clipping both sides of the geometry, to leave a 1 pt wide slice*. this makes a point sillhouette. if i make many of these slices as the geometry translates through space, it will leave a trail of them, which i can then skin, and have a long whispy, surface. so the problem is, when i get the sillhouette of points, all of the points are out of order, and no ordering method that I can find will put them in order... so that means i can't skin it correctly... the hip file i included shows the main process. I was outputting the individual silhouettes to bgeo by a rop node and then i was bringing in those sillhouettes with a file node, and then merging them... you will not be able to see those, unless you render out new bgeos... however, if you just turn on POINTS you can see the points of the silhouette in the main network *not the one filled with a bunch of file nodes and a merge* i hope i explained alright thanks for your help Jonathan points_problem.hip Edited October 3, 2008 by itriix Quote Link to comment Share on other sites More sharing options...
anim Posted October 3, 2008 Share Posted October 3, 2008 (edited) what about using cookie with XY plane? look at the file points_problem_cookie.hip Edited October 3, 2008 by anim Quote Link to comment Share on other sites More sharing options...
petz Posted October 3, 2008 Share Posted October 3, 2008 you can use a divide-sop after the clip-sop and turn on "remove shared edges". if the incomming geometry is closed you should get just the section. petz Quote Link to comment Share on other sites More sharing options...
itriix Posted October 3, 2008 Author Share Posted October 3, 2008 thank you all, anim: i checked out the .hip file, and i think using the cookie is the answer! thank you so much! i need to try and apply it to my *complicated* example now, and see if this still works. i've been using a *simple sphere deforming* but in actuality, i am going to be attempting to do this to a running human model... do you forsee any issues with this on something such as a 3d model of a human? i'm a little concerned about how the points are going to react, when the arms are moving and the legs are moving... *not sure how it will trace the silhouette*, and if this method will even still work for that... hope so, i'll let you know what i find!! thanks J Quote Link to comment Share on other sites More sharing options...
itriix Posted October 3, 2008 Author Share Posted October 3, 2008 thanks petz! i'll check this method also thanks Quote Link to comment Share on other sites More sharing options...
petz Posted October 3, 2008 Share Posted October 3, 2008 "try looking at the trail sop, then divide sop * remove shared edges" sorry aracid, repeated unnecessarily what you have mentioned in your post before. seems that i have to read the postings more carefully. petz. Quote Link to comment Share on other sites More sharing options...
anim Posted October 3, 2008 Share Posted October 3, 2008 ...do you forsee any issues with this on something such as a 3d model of a human? ... actually you can get into some issues with the cookie itself especially with some complicated geometry the clip and divide seems to be better solution another issue is the number of primitives(closed sections) in silhouette if this number differs(from frame to frame) it will be hard to use skin sop Quote Link to comment Share on other sites More sharing options...
itriix Posted October 4, 2008 Author Share Posted October 4, 2008 another issue is the number of primitives(closed sections) in silhouetteif this number differs(from frame to frame) it will be hard to use skin sop hmm yeah i was worried about that 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.