Ggradest Posted May 2, 2020 Share Posted May 2, 2020 (edited) Hi. I try mixing hair and rbd destruction, but vellum "match animation" makes bad jittering. Here one point of each hair segment set to "Pin points". Disable "match animation" solve this jittering so this moving is caused by that. Every stiffness value and substep can't remove this. "Attach to geometry" makes more bad result. Can anyone solve this problem? Cheers! hair.hiplc Edited May 2, 2020 by Ggradest Quote Link to comment Share on other sites More sharing options...
Ggradest Posted May 2, 2020 Author Share Posted May 2, 2020 Now I found this jittering doesn't appear when hair density is low(1~10). But hair width has no effect... Quote Link to comment Share on other sites More sharing options...
Ggradest Posted May 2, 2020 Author Share Posted May 2, 2020 (edited) Oh I found low Damping rate (~0.001) solve this. But sometimes simulation breaks down. It's a little absurd that the simulation is so unstable against the simplicity of setup, so I may overlook something. Edited May 2, 2020 by Ggradest Quote Link to comment Share on other sites More sharing options...
Ggradest Posted May 2, 2020 Author Share Posted May 2, 2020 (edited) Here my final concept. I want to destruct these grasses, so use Convet line SOP. Any forces don't exist, so strange jittering appear on impact moment. https://imgur.com/lSXz03M.gif destruction.hiplc Edited May 2, 2020 by Ggradest Quote Link to comment Share on other sites More sharing options...
Noobini Posted May 2, 2020 Share Posted May 2, 2020 here, far less to wade thru, I don't like overcomplications, there's still a bit of jittering but do whatever you want with it. vu_hair.hiplc Quote Link to comment Share on other sites More sharing options...
Ggradest Posted May 2, 2020 Author Share Posted May 2, 2020 (edited) Thanks Noobni! The instability of this vellum seems to have many causes. The culprits I found were number of Hair segments, length of the Hair, Groom Bend, and original RBD situation. Bend in particular creates a mysterious bounce... It may not be compatible with vellum to express stability state. The trick at the moment is not to do too many segments (around 8, not 16 or so), make the hair somewhat longer (0.2~), and avoid Bend... Although, "match animation" seems to create a mysterious force in the simulation. If you visualize Stretch Rate or some stress values, you will see that it is displayed in red for a moment ! I've now added some Groom to Noobni's file. There is a mysterious bounce when Bend is turned on. This doesn't appear when the "match animation" is turned off, so it seems to be incompatible with each other. Here Match animation : strange bounce https://imgur.com/MzD3qVT.gif Here off it : nothing strange https://imgur.com/U1dkalE.gif vu_hair.hiplc Edited May 2, 2020 by Ggradest Quote Link to comment Share on other sites More sharing options...
anim Posted May 2, 2020 Share Posted May 2, 2020 the "strange bounce" or "mysterious force" it's not that mysterious, it is caused by orient attribute not being stable as it's being computed live on already animated geo at Hair Constraint stage, and since Match Animation for hair uses orient it's forcing it to twist as orient is changing randomly you better precompute orient on static hair, I just added Oriantation Along Curve SOP (green node) vu_hair_fix.hiplc Quote Link to comment Share on other sites More sharing options...
Ggradest Posted May 2, 2020 Author Share Posted May 2, 2020 Wow! Thaaaaaanks anim!! It's just I needed...!! I just realized "Compute Missing Orientation" parameter on vellum constraint property SOP... I'll keep in mind that some kind of direction information like Orient is important for point-based calculations. Thank you so much! 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.