Exploding Cube.

[Symbolic]

Hi,

I am still looking at the techniques to create a healthy fractured geometry. There are different ways. The MAX9 macro, the Cookie SOP or the Fracture SOP...

At the moment I am trying to make a cube that stays still in the air... and then explodes with an internal force...

I have tried constraining the corners of the cube... and then use a force to destroy it from inside... But it did not work.

Any ideas?

I am preparing a sample file... I have seen somewhere on the SideFx forums... that one needs a particle system to blow something from inside...

[Symbolic]

Ok. Hi every body.

This is the first test for an internal explosion.

Any idea about how that might be improved?

cube_explode_v2r5.hipnc

My starting point was this post on SideFx Forums:

http://www.sidefx.com/index.php?option=com...ghlight=explode

Although I did not understand the dual particle system mentioned in that post. It seems like one needs a primary strong particle system... and a secondary more weak one to create a more controlled internal force.

Here is a video of the *.hip file.

explode_test_1.mov

Edited December 30, 2007 by Symbolic

[Symbolic]

Happy new year to everybody!

I guess my server was down for a while... so if you were not able to access the sample movie... try it again some time...

[Symbolic]

Hi,

After making lots of tests... and exploring the fracturing techniques...

Here are some results:

I have a fractured object... I use RBD State DOP to make the glue strenght 0 at the 25th frame.

Then the particles kick in and cause an internal explosion.

But I could not figure out a way to control the explosion, it looks bad... and strange...

Here is the DOPs network:

I am using an example from the help file for the RBD State thing:

Here is the particle system. Nothing special. I have heard that the Speed Limit POP will add more control to the explosion. But it did not work.

Here is the video:

explode_test2.mov

Any ideas about how to improve the explosion? Thanks!

[johner]

Hi,

After making lots of tests... and exploring the fracturing techniques...

Here are some results:

I have a fractured object... I use RBD State DOP to make the glue strenght 0 at the 25th frame.

Then the particles kick in and cause an internal explosion.

But I could not figure out a way to control the explosion, it looks bad... and strange...

Any ideas about how to improve the explosion? Thanks!

You don't necessarily need to use particles if you can get the look you want with animated DOPs forces, which can be a little easier to work with. The DOPs FAQ available at:

http://www.sidefx.com/exchange/info.php?fileid=327

has a small example on explosions using animated forces. I've attached it fixed up a little for h9. Also I added a Point Force DOP to the Magnet Force example, since the Magnet won't impart any torque to the objects which looks strange to my eye.

Also, you can probably just use an RBDFractured object rather than animate the Glue strength, unless you're going for a specific, gradual effect, which can be cool. But if you're just turning all the glue off at once, you might just use the Fractured object, then adjust the timing using offsets and such at the DOPNet level.

7_explosion_h9.hipnc

[Symbolic]

Hi johner,

I have been playing around with your file for almost 3 hours... The DOP Forces approach seems to be a very good one. Thanks for the help!

...for now I am trying to get the correct relationship between the applied force, the density etc... just to make the explosion look good.

At this moment... there are some very fast chunks that go too far too fast... and the large chunks almost stay untouched...

I will post a video as soon as I get an acceptable result.

[johner]

Hi johner,

I have been playing around with your file for almost 3 hours... The DOP Forces approach seems to be a very good one. Thanks for the help!

...for now I am trying to get the correct relationship between the applied force, the density etc... just to make the explosion look good.

At this moment... there are some very fast chunks that go too far too fast... and the large chunks almost stay untouched...

I will post a video as soon as I get an acceptable result.

Yes, I've had that happen when you've got some very small pieces like it looks like you've got in that movie. What you can do is create a DOP group based on the computed mass of the object, and apply drag or even apply a separate magnet force.

To do this you can use the GroupDOP set up like in the screenshot below, with the mass cutoff obviously set by experimentation. Then use the groups to apply different forces or Drag to the light group or whatever. The only caveat here is that it seems like the mass is only computed after the first frame, so you might need to ramp up your forces so they're not too high on the very first frame.

Good luck.

Edit: I should add, it's also possible that you should just cull out those tiny shard pieces and not simulate them with DOPs, since they might end up being unstable no matter what you do, as they are close to degenerate geometry. You could use particles, geo instanced to particles (like BlastCode), or even a second-pass DOP sim, to get those small debris effects without introducing instability to your primary DOPS sim. But you might not have to go that far.

One more edit: And oh yeah, Details view is your friend, in case you're not intimately familiar with it already. In 9.1 (and I think 9.0), you can see the list of groups for a DOP object just by clicking on the name. Very handy debugging an expression like this.

Edited January 4, 2008 by johner

[Symbolic]

Hi johner,

Thanks for the information. I will try going deeper in this issue and experiment more in the light of your post... I will post any results here as soon as possible.