aesthete

I didn't know the example, thanks! So to make the shape what I want, you'd suggest firstly bend the lines and mesh them with the main kernel using VDB. Right? I also think that it's difficult to make the shape by displacing… may be challenging to me. Using vel field would be a nice method than I tried above, thanks! I'll give it a try!

Hi guys. I've stuck in making a displacing geometry which forms the shape of popcorn these days. My ultimate goal is to make a popcorn generator, which can give different shapes of the popcorn procedurally. https://www.youtube.com/watch?v=jUHzcXrX2_U As you can see, kernel's peels are bending toward bottom and I think this is the most important thing to make the generator. To aim that, firstly I started a concept of curling line, but it wasn't proper when it applied to surface or geometry. There were a lot of things to take into consideration for me. So now, I'm trying to use displacement instead. In my thoughts, the way of making it is roughly like below.. - set the part of bottom, where displacement won't be occurred, or slightly. - set the part of middle which is going to be displaced and finally form the curly shape through a certain process. - with solver or loop, displace the area with adjusted normal. - smooth or remesh to reset the displaced area properly. - reset the displaced area to make it bend curly... But in this way, might be wrong and nasty..., there's the problems adjusting normal and reseting or regroupig the displaced area to make displaced curly. I couldn't let the area bend inwards with normal adjusted by simple vector, so tried to rotate the vector and it may work well, but to reset the displaced area remains big problem... attached a hip file, and any suggestion and idea for making the shape of popcorn would be appreciated. Thanks. displaceCurly.hip

It's because using a Solver SOP instead of a Trail SOP gives me the result what I want. Yeah, as you said, there might be no difference of those things to do in the hip file, but what I wanted is a dense object which is filled with wires to capacity from inner space, so I emitted particles on inner space. Trail SOP could show me the almost same result when increasing Trail Length(when this value is lower, lines seem to move on their direction so the inner space does not look dense), but it's slightly slower than using Solver SOP in my test. However I'm sure that there is better and smarter way to get the result using Trail SOP or other nodes. Appreciate for your reply Atom, Really helpful. Replacing with @width can make a faster render, but in that case, It looks slightly different from the polywires I rendered former scenes(I mean, not attached images.) 'cause the curves are flat as you mentioned, so I use N vector such as Hair Normal VOP to shader so that the curves look as tubes. However, since I rendered polywires with bump and normal map on the primitives for increasing wire's quality, I'm facing a new issue about applying those maps. Thanks again.

Hi guys, I'm now working on a work kind of growth animation.. and I've faced a problem that it takes tooooo long time when render or even import, since what I want to make is the extremely dense object wrapped with polywire like second image or even more than. In my setup, I made particle trails and applied them to polywire to make mesh. As you know, like the attached image, the denser(the higher frame like a second image), the more expensive for render. Actually, I rendered the half of my scenes barely, but after them, the primitive will be dense in earnest. To solve, I reduced polywire's Divisions and Segments but too many primitive causes expensive render time after all. Then what I was thinking after is to use the concept of Fur such as fur density using texture map. ? , but I couldn't get the clear idea from that, It might be too challenging for me. Or remove polywire out of camera view or at the back of it from that? Or using hair shader? but in this case, it's too late for me... How can I solve this matter with this setup using polywire to reduce render time? Any suggestions would be really appreciated. Due to attached error, add google drive link with images and a simple hip file. https://drive.google.com/drive/folders/0B2PqEaLsEVgWZlBZVGJoc29GYms Thanks, Lee

Thanks a lot for your help again and Atom's image! Really helpful! I'll give it a try.

Thank you so much for your replying and the hip file. I did not even know that using particle's velocity can make them lied on what I determined! If you don't mind, I have another question. After trying out the way you suggested, I also wanted to make the particles move around on each particle's surface where they lie on. It's like P + ((normalized gradient * negate) * (signed distance)) gives the result that particles are snapped onto the surface and slightly move around the surface. How can I do this?

Hi guys. I've been trying to make particles snap onto surface using vdb sdf.(volume sample and gradient) Specifically, I am trying to emit the particles from sdf volume and let them lie on some range of a surface. I know how to snap particles on a surface but what I really want to make is exactly not that all particles stick on the surface. I want some particles to lie down a short distance away from the surface outside or inside.(randomly) It may be softball question but I can't get easily figured out… Any suggestions would be appreciated! lee prac_sdf_distance.hip

Thanks a lot both of you! I'll give it a try.

Hi all. I've wrapped up my head for the for loop in SOP network. I think i've tried to make quite a complicated setup, using the for loop in SOP. First of all, what i want to make is the looping with a bounding box which has a certain size enough to contain specific number of points. When it contains that specific number of points then the loop will stop, and will start again without the points which were contained previously. So for example, like attached hip file, given that the bounding box can contain "27" points, if it finds 27 points when it is located at the 91st point, the loop will stop because it found 27 points. And it will start again from the 92nd point, since the looping was executed from the 0th point to 91st point, searching whether it found the 27 points. To accomplish that, i used a bounding box with a specific size, which can contain 27 points to loop the group. Then i gave the metadata to the location of the bounding box so that the looping can be executed starting from the first point number to the last as iterations, of the source geometry (a cloud of points). But it seems like "Stop Condition" at the Block End node is evaluated before it performed the iteration. What i really wanted was "Stop Condition" to be evaluated afterwards. Sorry for long winded explanation. To summarize, - How can i stop the condition after it performed the iteration, not before? - When the loop stops, is there way that i can exclude the points which were contained before, and then start again? I don't want to use "Delete" because i want to keep the original point number. I heard that for this kind of setup, i need to set 2 for loops (a loop in a loop) but it is very complicated. How can i make this setup using 2 for loops? Any help and tips would help a lot. Thank you for your help. forloop.hipnc