Simulate empty bullet casings, try two!

Become a member of the CGSociety

Connect, Share, and Learn with our Large Growing CG Art Community. It's Free!

REPLY TO THREAD
 
Thread Tools Search this Thread Display Modes
Old 02 February 2013   #1
Simulate empty bullet casings, try two!

Seeing as my last thread on the same subject was kind of based on a misfire idea, i.e. using nparticles to simulate something that has an irregular shape, I'll start over:

What's the best way to make a controllable animation of empty bullet casings (not a predefined amount) being ejected from a turret at a time of the animators' (me) choosing? Since they need to act somewhat physically correct, for instance using a lowres cylinder to simulate and bind a highres mesh to it, it seems that rigid bodies is the way to go for me. The biggest concern is just how to spawn the actual rigid bodies at a constant rate, based on incoming controller input, and how to spawn them with the collision mesh already attached to each casing. Making a number of bullet casings prior to animation is not really an option.
 
Old 02 February 2013   #2
I would like to know this as well. Currently I cant think of anything outside of the box to achieve this but hopefully somebody well, nParticles would not be appropriate as you know.

Ive used a RB solver built from bullet with Maya integration that can spawn RB via an emitter obviously not available to the public. But since it is achievable I assume its present in the bullet API and could be harnessed by a programmer to build something similar.

This would be the best solution I believe.

Having said that though Bullet is still an incomplete solver but could still serve your needs.
__________________
Vimeo
 
Old 02 February 2013   #3
Originally Posted by hellspawned: Seeing as my last thread on the same subject was kind of based on a misfire idea, i.e. using nparticles to simulate something that has an irregular shape, I'll start over:

What's the best way to make a controllable animation of empty bullet casings (not a predefined amount) being ejected from a turret at a time of the animators' (me) choosing? Since they need to act somewhat physically correct, for instance using a lowres cylinder to simulate and bind a highres mesh to it, it seems that rigid bodies is the way to go for me. The biggest concern is just how to spawn the actual rigid bodies at a constant rate, based on incoming controller input, and how to spawn them with the collision mesh already attached to each casing. Making a number of bullet casings prior to animation is not really an option.


in maya? no way really. This is one of the big reasons that pushed me into XSI ICE and houdini. They have a vastly superior Bullet/Particle integration. Hopefully maya 2014 changes this. Even C4D is better for this. - if that is not vomitous, I dont know what is.
__________________
From Russia, with love @ stooch.tv
 
Old 02 February 2013   #4
well, this is a tough one in maya. as mentioned before Houdini would be a better solution because bullet is better integrated. I won't rely on Maya 2014 to come up with something usable like this. Autodesk just seems to be totally in love with their viewpoint 2.0 so I guess a lot of dev goes into this.

what I would try, not really fancy but might do the trick. create a bunch of shells before hand and attach the bullet solver to it. I guess you can even duplicate a already attached geo. put them in a group and move them out of the way. use a particel emitter to emit particles then grab their position and velocity with an expression and move an existing bullet mesh to this position and transfer the velocity to initial velocity and turn on the solver on that object. As I said, not really fancy but could do the trick.
 
Old 02 February 2013   #5
Come on guys, lets face it.

Maya is primarily used by Studios.. ones that can afford to write their pipelines and solvers or tweak Bullet or Maya to do what they want... so why would AD spend time to make it easier for the lil guy?

Blender, LW, C4D, etc are all tailored towards the lil guy... and of course why they have to have better implementations...

Sad, I know.. but what can we do?

Beg Duncan? LOL!?
__________________

.

Reel - 3DInk.com
Filmography -IMDB
R&D - VIMEO

*please read- cyanbsl.blogspot.com/*
 
Old 02 February 2013   #6
Originally Posted by HowardM: Come on guys, lets face it.

Maya is primarily used by Studios.. ones that can afford to write their pipelines and solvers or tweak Bullet or Maya to do what they want... so why would AD spend time to make it easier for the lil guy?

Blender, LW, C4D, etc are all tailored towards the lil guy... and of course why they have to have better implementations...

Sad, I know.. but what can we do?

Beg Duncan? LOL!?


I started a thread earlier to shed light on Bullets shortcomings hopefully drawing some attention to Duncan et al. It only needs a few implementations made to it and it would be a fairly decent RB solve, RB emission is one, RB sleep threshold is another.

In a perfect world of course it would react to nParticles and fluids but then we dont live in one.
__________________
Vimeo
 
Old 02 February 2013   #7
Yeah Yang the guy who wrote Miarmy has implemented Rigid Body (PhysX) emitter inside Maya...currently its restricted to primitive shapes I guess but feel free to contact him for more info.

@ 3.26
https://www.youtube.com/watch?v=bjo...8990466&index=4

Last edited by viki164 : 02 February 2013 at 09:05 PM.
 
Old 02 February 2013   #8
Oh man, the avalanche toppling people is awesome wonder how fast it sims!
__________________

.

Reel - 3DInk.com
Filmography -IMDB
R&D - VIMEO

*please read- cyanbsl.blogspot.com/*
 
Old 02 February 2013   #9
Originally Posted by HowardM: Come on guys, lets face it.

Maya is primarily used by Studios.. ones that can afford to write their pipelines and solvers or tweak Bullet or Maya to do what they want... so why would AD spend time to make it easier for the lil guy?

Blender, LW, C4D, etc are all tailored towards the lil guy... and of course why they have to have better implementations...

Sad, I know.. but what can we do?

Beg Duncan? LOL!?


well.. Houdini is definitely aimed at the "big boys" and they seem to think that a good bullet integration is useful. i tend to agree. Anyway, im more than happy using houdini until maya gets there, i think bullet integration isnt really the main issue here, more like an overall integrated and non redundant dynamics solution that allows third party solvers. but im sure they are on it. just might take a while.
__________________
From Russia, with love @ stooch.tv
 
Old 02 February 2013   #10
Houdini had a pretty lame implementation of Bullet initially too, so there is hope!

Last edited by Cheesestraws : 02 February 2013 at 11:06 PM.
 
Old 02 February 2013   #11
Originally Posted by viki164: Yeah Yang the guy who wrote Miarmy has implemented Rigid Body (PhysX) emitter inside Maya...currently its restricted to primitive shapes I guess but feel free to contact him for more info.

@ 3.26
https://www.youtube.com/watch?v=bjo...8990466&index=4


Miarmy is pretty cool, though I think it can only import one nCloth nCache node with offsets for dynamic cloth. Not really ideal but still a nice tool, looks like tonnes of fun to play with.
__________________
Vimeo
 
Old 02 February 2013   #12
Here is a relatively fast way to do shell casings( see attached scene ). You could pretty much have armageddon without it being a performance issue.

One uses 2 nParticle systems, both with exactly the same emission rate. The emitters for each system are separated by the length of the casing creating a particle pair. A constraint using component order links the particle pairs together. The casing is instanced to the second emitter, which also has an expression setting an added per particle attribute named "lookAtRotate" to the position of the first particle system. It is important that the expression be on the second system... if the first system has an expression that references the second then the second system will not properly evaluate( it messes up the dg attribute dirtying). Then on the second system the instancer aim position is set to "lookAtRotate" (the position of the first particle system).

This is pretty basic, and handles the orientation, but the twist might cause the casings to twirl rapidly when the aim is near the up vector direction. For casings this spinning might not be a concern.

I added an additional expression to set the second emitter rate from the first, so one can just play with the rate on emitter1.

As well on the constraint I set restLength method to constant and set restLength to 0.6( the desired separation of the particle pairs ). The emission speed randomizations along with collisions caused too much variation in the spring lengths, so explicitly setting them in this manner helps.

The casings in the scene rotate at the start frame... you might set the lookAtRotate in the creation expression, as well.
Attached Files
File Type: zip shellCasings.zip (11.0 KB, 233 views)

Last edited by Duncan : 02 February 2013 at 07:19 PM.
 
Old 02 February 2013   #13
Genius! That is such a great idea!
__________________

.

Reel - 3DInk.com
Filmography -IMDB
R&D - VIMEO

*please read- cyanbsl.blogspot.com/*
 
Old 02 February 2013   #14
Duncan to the Rescue Again !!
 
Old 02 February 2013   #15
Duncan, that is absolutely amazing. Blown away that you could do that lol. Which is kind of a problem in itself. You have demonstrated a very cool technique, which is based on your intimate knowledge of these things you created. But wouldn't it be cool if we can do what you just did but with a clear nodal workflow ? is that even technically possible? Also, can you describe more complex shapes with particle "clumps" etc?
__________________
From Russia, with love @ stooch.tv
 
reply share thread



Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is On
HTML code is Off
CGSociety
Society of Digital Artists
www.cgsociety.org

Powered by vBulletin
Copyright 2000 - 2006,
Jelsoft Enterprises Ltd.
Minimize Ads
Forum Jump
Miscellaneous

All times are GMT. The time now is 10:59 PM.


Powered by vBulletin
Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.