Somebody can tell my what is the right way to do a bomb force (from point to all directions)

I try to do with scaling sphere, but is too slow and is´t constant.

Assuming you are talking about DOPs, the way to get a radial force is to use the Magnet Force Dop. Check out the attached.

I've also found that using a few points scattered on a sphere with radiating normals of varying magnitudes which are fed into a Field Force DOP work quite well- and you can get easily cause asymmetrical spikes which look quite good.

Another thing you can try out is overriding the initial divergence field with a field of your own.

In this example, I set a “peanut” of divergence in the center of my density cloud and pull it in into DOPs. Try change the position of the divergence field to see a sense of how it causes a radial ejection.

jason_iversen
I've also found that using a few points scattered on a sphere with radiating normals of varying magnitudes which are fed into a Field Force DOP work quite well- and you can get easily cause asymmetrical spikes which look quite good.

Hi Jason,

I brought some points loaded with force attribute (which are copied from their point normals) into DOPs with the Sop Geo DOP.
The force field is fed into a Field Force DOP.

I'm trying to break up an RBD Glue object (pre-shattered) with this force field–like a bomb–but all it's doing is moving the glue object, not breaking it apart.

Attached is the hip file. Can you take a look at what I'm not doing right?

Your setup seems correct - and what I notice is that the pieces will only fly apart correctly if you allow the glue object to collide with the ground first, even if you set all the Internal Glue to 0. If you replace the Glue Object with a Fractured Object it will work as expected. I don't know why this is the case - perhaps even a glue strength of zero needs a bona fide impact first in order to break the (ultra weak) glue bond.

Another thing to be aware of: the Field Force can be pretty darn slow when used with RBD objects. How it works is it applies the field force to every cell of the Collision Volume for every object. There is no way to tell it to just apply at the centre-of-mass instead - which might be the desired behaviour sometimes. SESI is aware of this slowness and this limitation.

I hope this helps,
Jason

jason_iversen
Another thing to be aware of: the Field Force can be pretty darn slow when used with RBD objects.

Yep, I'm running the sim with a Fractured object as I type.
So RBD objects aren't really recommended if I don't have a lot of time–so fluid objects? The 2D smoke example you posted ran pretty fast. Any other dop objects where it'd be good for?

-Dave

A slight side note. But it is my understanding that if you import something from SOPs (metaball magnent, points in a field force) you'll lose multithreading.
Is this still the case in 9.5?

I've done this before by simply scattering a few points over a sphere (which you animate scaling up, and then copying some more spheres onto these points.

Make this a static object in dops that deforms, and you should be fine.

It may not be useful depending on what you're doing, but it can be useful for breaking objects apart from the inside out.

IndyZoneCo

jason_iversen
Another thing to be aware of: the Field Force can be pretty darn slow when used with RBD objects.

Yep, I'm running the sim with a Fractured object as I type.
So RBD objects aren't really recommended if I don't have a lot of time–so fluid objects? The 2D smoke example you posted ran pretty fast. Any other dop objects where it'd be good for?

Yeah, a fluid sim is, say, one 50x50x50 grid - or 50x50x1 for a 2D sim. If you're using RBDs where you might have 50 objects with Collision Volumes each at 30x30x30, that's a lot more cells to apply forces to. I do hope that they allow the force to applied at the object centres only - as I can hardly imagine how well a Torque field force would work on many adjacent cells like that.

For Cloth simulations and SPH fluids, I think the force is applied at every vertex, so it should perform ok for those solvers.