Fill a jar with little spheres
#1
Hi,
I try to fill a jar with little spheres with particle physics ( with physX shapes it would be tooooo long)
but instead of accumulate on each other and fill the jar, they intersect each other and just cover the bottom.
Collision shape is ok ( draw collision shapes show something correct ).
Any idea ? Thanks
  Reply
#2
(04-30-2019, 08:50 AM)HHbomb Wrote: Hi,
I try to fill a jar with little spheres with particle physics ( with physX shapes it would be tooooo long)
but instead of accumulate on each other and fill the jar, they intersect each other and just cover the bottom.
Collision shape is ok ( draw collision shapes show something correct ).
Any idea ? Thanks

It sounds like something is wrong with your particle physics. Weird of you toggle the draw collision shape and it is correct. Can you share a screenshot or your file?
  Reply
#3
(04-30-2019, 10:29 AM)mazi710 Wrote:
(04-30-2019, 08:50 AM)HHbomb Wrote: Hi,
I try to fill a jar with little spheres with particle physics ( with physX shapes it would be tooooo long)
but instead of accumulate on each other and fill the jar, they intersect each other and just cover the bottom.
Collision shape is ok ( draw collision shapes show something correct ).
Any idea ? Thanks

It sounds like something is wrong with your particle physics. Weird of you toggle the draw collision shape and it is correct. Can you share a screenshot or your file?

Here is the file !   Big Grin
  Reply
#4
(04-30-2019, 12:04 PM)HHbomb Wrote:
(04-30-2019, 10:29 AM)mazi710 Wrote:
(04-30-2019, 08:50 AM)HHbomb Wrote: Hi,
I try to fill a jar with little spheres with particle physics ( with physX shapes it would be tooooo long)
but instead of accumulate on each other and fill the jar, they intersect each other and just cover the bottom.
Collision shape is ok ( draw collision shapes show something correct ).
Any idea ? Thanks

It sounds like something is wrong with your particle physics. Weird of you toggle the draw collision shape and it is correct. Can you share a screenshot or your file?

Here is the file !   Big Grin

You have very high gravity and only 1 Timestep per frame, so the particles go so fast and with too few time step so one calculation it's above another particle, next calculation it's below. Meaning it won't collide with other particles. I turned the gravity down by 500% and time steps up to 1/16.


Attached Files
.max   pb01 2017.max (Size: 668 KB / Downloads: 322)
  Reply
#5
(05-01-2019, 11:57 AM)mazi710 Wrote:
(04-30-2019, 12:04 PM)HHbomb Wrote:
(04-30-2019, 10:29 AM)mazi710 Wrote:
(04-30-2019, 08:50 AM)HHbomb Wrote: Hi,
I try to fill a jar with little spheres with particle physics ( with physX shapes it would be tooooo long)
but instead of accumulate on each other and fill the jar, they intersect each other and just cover the bottom.
Collision shape is ok ( draw collision shapes show something correct ).
Any idea ? Thanks

It sounds like something is wrong with your particle physics. Weird of you toggle the draw collision shape and it is correct. Can you share a screenshot or your file?

Here is the file !   Big Grin

You have very high gravity and only 1 Timestep per frame, so the particles go so fast and with too few time step so one calculation it's above another particle, next calculation it's below. Meaning it won't collide with other particles. I turned the gravity down by 500% and time steps up to 1/16.
Hi, sorry I didn't see your answer !!!  Blush 
Thank you very much, I know now for time step  Big Grin  but I wasn't aware gravity matter so much... (sorry for my poor english)
  Reply
#6
(05-03-2019, 08:08 AM)HHbomb Wrote: Thank you very much, I know now for time step  Big Grin  but I wasn't aware gravity matter so much... (sorry for my poor english)

Gravity doesn't matter technically be the amount the particles move each frame matters. So lets say your gravity is -0,1 you need 1/2 Substep for example, then with -0,2 gravity you need 1/4 Substep to get the same calculations. So the more the particles move each frame the more substeps you need.
  Reply
#7
(05-03-2019, 11:35 AM)mazi710 Wrote:
(05-03-2019, 08:08 AM)HHbomb Wrote: Thank you very much, I know now for time step  Big Grin  but I wasn't aware gravity matter so much... (sorry for my poor english)

Gravity doesn't matter technically be the amount the particles move each frame matters. So lets say your gravity is -0,1 you need 1/2 Substep for example, then with -0,2 gravity you need 1/4 Substep to get the same calculations. So the more the particles move each frame the more substeps you need.

Thanks, that is clear for me now. But as objects accelerate with gravity, more they travel more they go fast (if they don't collide) and not linearly. So more they travel more I need to decrease substeps... That need some practice !
  Reply


Forum Jump: