Concept Question: Combined Fluid / Object Movement Simulations

Support Particles and Physics Simulations
1

Greetings. It’s getting close to 6am here in IL so I’m going to make an attempt at brevity while also getting this off my chest.

First, a quick introduction to me, my level of Blender experience, and why I am here. My name is Ethan and I live in IL. I’ve recently graduated from high school and part of my “summer job” this summer is feverishly working on scholarships. I am here for some assistance regarding a short video for a scholarship contest about teens and driving safety. I downloaded Blender several days ago and have watched a bunch of videos to get my bearings. I recently finished an image, which you can check out here: www.goput.it/2ct.png (I will do a more formal forum-style link after I sleep). It’s definitely not unique nor did I branch much from the tutorial, but it gave me the basis for what I want to do.

So, anyways, my idea consists of these parts:

*Modeled car interior
*Glass filled with liquid beer (i.e. Blender-generated)
*Flying glass/beer
*Shattering windshield

In essence, I want the beer glass to be launched during a collision and hit the windshield, shattering both the windshield and the glass and flinging bits of “glass-glass,” auto glass, and beer all over the place. Naturally it will be in slow-motion for a nice, dramatic effect. I’ll run over it with Vegas after the animation is complete to add music, sound, and text effects.

My question is basically, “Is this doable and if so, how can I do it?” I don’t need extreme specifics and already-existing tutorials are just fine - I just wanted some expert opinions on this.

Oh, yes. The catch is that this is all due by 11:59PM today, so I’m really riding the wire on this one. Honestly, though, I don’t feel I stand much of a chance at this point (this is something that could end up on TV, for example), so I am really more concerned about learning something new, doing this all from scratch, and possibly having the opportunity to enter a truly refined finished product in to other competitions down the line.

Please let me know what you guys think. I’m going to hit the sack now. Also, my attempt at brevity was a failure but I think I’ve provided you all with enough information to (hopefully) get the ball rolling.

Thanks in advance! :slight_smile:

2

The beer would not shatter the windshield, would it?
Anyway, considering you want to fake it, there are some tutorials out there about shattering objects. I have never done them. I guess you would need to simulate the shatter of the windshield with a rigid body, then the glass against the baked shield animation as a rigid body, then considering you want to use the grid fluid simulator, use the animated glass as an obstacle to it.

As for the interactions, as I see it, they would be one way only. You could have one affecting the other (the baked glass affecting the fluid, for instance) but not both. Obviously, this would not be very reallistic.

It is an interesting shot, but very complicated.

3

Yeah, I’m not sure if I’ll be able to pull off something realistic. But yes, a standard drinking glass would likely not shatter a windshield. Auto glass is amazingly resilient and doesn’t shatter anyways. Perhaps I could simply simulate the cracking instead? That might make the animation simpler, actually. The glass could shatter and the windshield merely crack.

I’ll poke around and learn about shattering and then do some tests to see if I can create something along the lines of what I want. Dunno how it’s going to go.

4

Yes, that would be simpler. You could probably solve the whole windshield problem with textures that way. Though the cracks should have depth, and that should show up. I guess you could do it with a very high detail displacement.

But the simulation should be with just a simple plane.

As for the shattering of the glass… as far as I’ve seen blender doesn’t have a very reallistic way of doing it, like realflow. But it should be enough.

Just keep in mind that all the distance the fluid travel will have to be inside the domain. So, if it travels a lot, you will have a very big domain and you won’t be able to put a lot of resolution in it.

But you could fake it with various shots and various simulations.

Like I said, a very interesting project. I wish to do something like that. But something that might take months to look good.

And you wanted to do this in one day? If you spent years doing simulations like that and had some massive render and simulation farms… I would still doubt it.