Welcome to od|forum

Register now to gain access to all of our features. Once registered and logged in, you will be able to contribute to this site by submitting your own content or replying to existing content. You'll be able to customize your profile, receive reputation points as a reward for submitting content, while also communicating with other members via your own private inbox, plus much more! This message will be removed once you have signed in.

Martin Schan

Members
  • Content count

    12
  • Joined

  • Last visited

Community Reputation

1 Neutral

About Martin Schan

  • Rank
    Peon

Personal Information

  • Name
    Martin
  • Location
    Hamburg, Germany

Recent Profile Visitors

721 profile views
  1. two flip object

    Hi diless, I took a look at your file. The problem is you can't simply merge multiple Fluid Objects and have interaction between them. I don't know if it is possible at all. Would be nice to have that verified by a more experienced FLIP-user. But I solved your problem by merging your two drops before feeding them into one single FLIP Object. Drops_Collide_MS.hiplc
  2. Sorry to hear that. Maybe you have gone through all of the following, but here is what I would do, to find the source of the problems. I guess it is either your hardware or your houdini installation. 1. I would try a file from a different houdini user. 2. I would let others try out my problematic file and see if they encounter the same problem. 3. And install the latest version of Houdini. This might give u some hints for further troubleshooting. I hope it helps a little. If all that doesn't help you should probably report a bug to SideFx.
  3. Hi Marco, this might be very obvious, but you should check your Frame Increment Value on your Mantra Node. For a default behaviour you would want this to be on 1. @Atom Is this what you meant as well? Or is there another option?
  4. Thanks for the hint, Juraj. Definitely an interesting thread. For the moment I could only took a brief look into your hip, but I'll catch up on it later. Again, thanks.
  5. Hi Sebastian, thanks for your help. Your Setup really reads the goalPos. Yeehaa, great! So the solution is not to use a VOP Force but a POP VOP. How the heck could I have known that? The Bind node works and even the Import Point Attribute node. To your proposal, to reverse the effect and start with the end position. I assumed it might look wrong or at least odd, especially for the first few pieces. But yeah definitely worth a try. Thanks for the visualize node as well, pretty cool for debugging and just what I've been looking for. I even found a Print node for VOPs. Both work great together. It makes me wonder, why you never see those (freaking important) helper nodes in a professionell tutorial?!
  6. Thanks Atom for your reply. That is a useful expression. I tested it on my Destination Pos Parameter and it reads out the goalPos on my packed Geometry. But the real problem isn't the access (and probably never was...my bad). I tested the expression with the hardcoded goalPos of piece 10, which works. They move to that small tube. (see screenshot) If I understand this correctly, according to the screenshot, the $OBJID or $OBJNAME can only give back 1 for a packed geometry. And are thus not useful for any cycling expression, right? So what variable or method can I use to feed in a different value for every processed piece in that VOP Force Node? Any hints or advice would be appreciated. A great weekend to everyone. Morph with VOPs in DOPS 03b.hiplc
  7. Hi guys, this is what I'm after: I wan't to build up a fractured ring around a static cylinder object. At the end, when the fractured pieces come closer to their end position I morph them into their final position. To get interesting collision I tried to build that in DOPs. But I struggle at getting the desired goal position into VOP Force, to calulate the proper velocity for every point to reach it's final position. The initial Geometry in DOPs has already a goalPos attribute, but I can't figure out how to access it. Import Point Atribute doesn't work at all in VOP Force, right?! So I tried to get the position from the TARGET Null with point() expression, which works on RBD Point Objects but not on Packed Primitives or RBD Objects. Some sort of referencing or match up problem I guess. I'm sure there is a way to realize this with ease. Hopefully you guys can help me out. Thanks in advance. Cheers Martin P.S. DOPs are really frustrating for a beginner like me. I have experience in coding and such things and what I'm missing are real debugging options. I have a hard time to trace and fully understand what happens with the data at what point. Do you guys have any additional tips besides Geometry Spreadsheet? How do you debug your far more complex DOP Networks? Morph with VOPs in DOPS 03.hiplc
  8. I have been fiddling around with generating procedural caps for the past two days. The Polycap and Divide SOP work well for a simple cap of one outline. But i want to cap more complex shapes from several illustrator curves. So basically I have a bigger outline and several inner curves within that outline. The only solution I came up with for now is to polyknit the pieces on one end, to generate a large connected geometry, hand-pick the loop and finally polycap it with that loop group. But that is all useless when I wan't to be flexible with the resolution of the curves. So I'm looking for a more procedural way for this problem. I'm grateful for any ideas. I made up an example with simplified setup. caps_for_arbitrary_shapes.hipnc
  9. Uh nice. I'll check that out. Thanks Konstantin.
  10. Hi Konstantin, thank your for sharing your idea. To delete the unwanted cap by area is a nice trick. And the profile approach is also very useful. Might be more flexible to control the shape of the edge that way. Maybe with two ramps, one for the X-Position and one for Y. I'll definitely try that. Meanwhile I managed to get my cap loops with boundary edges and unshared edges. But your way is simpler and probably even less likely to break. Greetings Martin
  11. I don't know why I didn't try this or saw this before, but the option to convert the points to edges under the Edit-tab works well. The only thing left is that the proximity-color-approach doesn't really create a clean caps loop. The polycap seems to be forgiving about that though. But it also does not work well with smaller segments from the resample node. Any suggestions for a cleaner cap loop? I will try the Boundary Groups that Unshared Edges gives. edge group by proximity 2.hipnc
  12. Hi I’m Martin and I’m new to this forum. I have been trying to get my head around Houdini for a year now, whenever I find the time. Fortunately with increasing success lately. Not least because of this great forum and it’s smart heads. So my thanks to you already. Now to my question. I'd like to have a nice beveled edge geometry, with several polybevels that are feed automatically with the right selection of edges. And all that based on only one curve input. Sure, I can realise it with a fixed edge group selection, but I wan’t this to be flexible. So my idea was to create an edge group by proximity to the curve that defines that edge, with Attribute Transfer. That works fine with the downside, that I only get a point group and that doesn’t result in a nice beveled edge of course. The file probably makes my problem clearer. I did a little research and found a thread from 2006 with a similar problem. https://www.sidefx.com/index.php?option=com_forum&Itemid=172&page=viewtopic&p=24138 The suggestion was to use a python node to generate an edge group. I pluged that python node into my scene and it did nothing. I also took a look at the code and yeah…ähm didn’t quite understand it. I did a little VEX already, but that python stuff exceeds my knowledge for now. I also found a suggestion to format a string of points to a string that represents the edges. But I have no idea how to access the pattern field in a Group SOP and feed the formated string back into a new pattern field. Do you guys have any suggestions or directions you can point me to? That would be great. edge group by proximity.hipnc