sfdex opened this issue on Jan 18, 2008 · 10 posts
sfdex posted Fri, 18 January 2008 at 6:53 PM
I searched the forum and didn't find an answer to this, so here goes. Is there a way to set particle collisions to happen at the geometry level, as opposed to at the hotpoint level? The still above shows what I mean. These balls are particles, and they bounce off the geometry of the stairway at their hotpoint, not when the outer geometry collides with the stair. So, as you can see, it looks like the balls are sinking half way into the stairs.
Am I missing a setting?
BTW: This is a work project, so I'm on C5P.
Thanks!
Dex
GKDantas posted Fri, 18 January 2008 at 7:13 PM
bwtr posted Fri, 18 January 2008 at 8:59 PM
I think it's one of those little annoying things where you have to change the hot point to place it at the botom of the objects "box".
You need to go into 4 screen view and, using "Caps Lock", move the hotpoint down to the bottom of the object.
Watch to unclick caps lock when you are finished,
Hope that helps.
bwtr
GKDantas posted Fri, 18 January 2008 at 9:13 PM
MarkBremmer posted Fri, 18 January 2008 at 9:42 PM
Particles are hot point based and since there may be rotation information in the particle settings, moving the hot point won't solve the problem. In cases like this, I've created a duplicate object of the object the particles are supposed bounce on and offset it the distance of the particles. Then I just hide the object with an alpha setting of black. Not ideal, but it works.
GKDantas posted Fri, 18 January 2008 at 11:06 PM
Great idea... but need to know if theres a bug track about this issue, not a bug but if they add realvolumeto the articles we can simulate fluids...
Follow me at euQfiz Digital
MarkBremmer posted Fri, 18 January 2008 at 11:52 PM
Actually, it's not a bug. Particles are instances and are not calculated as real geometry. Therefore they are only really just a hot point. This was actually brought up during C4 and we were given the reason then. So, it's legacy code from C3. While the code can be totally rewritten to accomplish this, I don't think that it will move anywhere near the top of the wish list for awhile. Fluids and calculated fluid/particle dynamics actually use a completely different scheme. I'll be keeping this on my wish list. ;-)
GKDantas posted Sat, 19 January 2008 at 5:59 AM
Yes I know, but adding this simple feature for the particles we can do many things more with it... Ihope someday they work on some fluid too. This isnt in my wish list top... paint in 3D is in my wish list top and a fix for the physical system.
Follow me at euQfiz Digital
sfdex posted Sat, 19 January 2008 at 12:26 PM
Thanks, all, and Mark, your solution is brilliant, as always! I'd thought about moving the hotpoint to the edge, but as Mark says, the balls roll while they're moving, so hotpoint to the "bottom" edge wouldnt' work, as the "bottom" doesn't stay on bottom for long!
I discovered that particles will bounce off objects that do not have "visible" checked, so whether the alpha trick or making the objects invisible, that's a great solution.
GKDantas, I suspect that this isn't a "simple" feature to add; as Mark mentioned, the code would have to be rewritten. But I agree, a good fluid dynamics system would be most welcome, as would the 3D paint.
Those of us who have been here since RayDream and Infini-D may remember that RayDream used to have a 3D paint system. What happened to that? When MetaCreations was retooling RayDream and incorporating Infini-D archetecture, why did they drop the 3D paint? It was slow on machines of the day, but I'd bet it would scream on a modern system.... Anybody know where it went?
:D
GKDantas posted Sat, 19 January 2008 at 12:46 PM
I never used Raydream, but i had a painter called MeshPaint 3D that wasgreat too, but donthave more support for years...
Follow me at euQfiz Digital