In this series of lessons, we will learn how to create dynamic and high quality fracturing simulations. Dynamic fracturing of rigid bodies is a powerful and production ready tool in Houdini. The shelf tools are a great place to start but there's a lot more under the hood that can be leveraged to get high quality and highly tunable results in less time. This tutorial covers dynamic fracturing starting in the context of dynamic simulations, then using dynamic fracturing building blocks to visualize fracturing without the need to simulate anything, and other tip and tricks to attain complex effects and improve robustness of the workflow. The material is based on the assumption that the viewer has a working knowledge of other aspects of Houdini and might have some experience with other DOP simulations. Dynamic fracturing was introduced in Houdini 9.5 but this tutorial was created using Houdini 11.0 so there may be some differences with earlier versions of Houdini. Software required: Houdini 11.0.
Luke Olson is a visual effects artist with Reel FX Creative Studios in Dallas.
Section Introduction Transcripts
Section Introduction Transcripts
Introduction and Project Overview [Autogenerated] My name is Luke Wilson on the visual effects artist, and I've had the great pleasure of working on some amusement park rides, animated features, feature length films and some commercial projects. Houdini is my primary tool in this tutorial. Siri's I'd like to show you the dynamic fracturing on Houdini. We're gonna start with a fairly basic scene with just a ball in a piece of glass to get an idea at an overview of how those dynamic fracturing networks work. And then we'll move on to some more advanced visual ization of the fracturing of applying that to live simulation for the last project. For this one, let's start with the ball breaking glass hip file. So this has just the geometry that we need, and then we'll set up the dynamic networks. Right now, ones, the wall object is just a box. The ball is just the platonic solid set to a soccer ball, the ground just a grid. And let's take a quick look at the glass network. It has a few sheets of glass here, which are just just boxes transformed into place up against the wall, and there's a few more if you'd like to unlock. Just uncheck this bypass toggle. But for the sake of the interactive, sometimes I'm gonna leave it with just a single piece of glass. All those sheets get merged into a single object, and I use connective ity soft and a partition salt to break those into different groups. So whenever we import those into the top network, we can important with just a single single import and old. Create a top object for each group where we have shed 01 and two there. But again, I'll just stick with just the one. For now, waken create the dynamic network, either by using any of these rigid body shelf tools. Or I could just go down here to the little same menu and create a new simulation. Trains that to Clarence Breaking Cem and now under where it says current simulation. When I create any of those shelf tools, they'll go to whatever network is selected here. We'll set up the ground first as a terrain object. The wall as a static, rigid aesthetic object the glass as a dynamic object and the ball as a dinner mint dynamic object as well. So as we can see we've got ground the wall glass and a ball. Thanks for all these work. Give the ball a initial state, moving it towards the wall. It looks like the positive Z access and slightly negative X axis is where we want to send the soccer ball. Okay, they can see it collides with object. So in order to make this glass breakable, now we can select it. And then up in this rigid body shelf, there's a make breakable tool, and this will add a fracture parameter node after whatever object you have. And okay, now we have this fracture, Parma's node, and see what that does just with the defaults. Thanks. So right off the bat, it doesn't fracture on the same frame is the ball actually hits, so the fracture takes at least one step to apply. So if we don't want that delay, we can add atleast to sub steps to the simulation. Just two. See how that goes, depending on the speed of the object. That may take more than that. But in this case, to you is probably enough. Okay, so now we see the fracture occurs at the same time is the impact Let's go over a few basic troubleshooting tips. So few. If the object is not fracturing, I'm gonna make sure that this fracture par mes is set up under the object that you want to fracture and make sure that fracturing is enabled. And also that the item that you want to fracture is included in the impact group. So say the if we want the ball to trigger the fracture. Since the start includes all objects that I work or we could set it specifically just to the bowl, you can see the ball still causes the fracture. That said, we wanted only the ground to cause a fracture. I'm sorry into the ground to cause a fracture when the ball hits, It shouldn't fracture. We can leave that with all objects. Also, the rigid body silver needs to be set to Houdini's own rigid body. Silver O. D. E is a very efficient, dynamic silver, but it does not work with the dynamic fracturing and a couple other features in Houdini. So make sure that this is set to R B D. And also the force of the impact sometimes may not be enough to cause a fracture with the default parameters. So this minimum and maximum impact force, you can find out how forceful an impact is. If the details view in the context of adopt, we can see it's going forward here. Impacts and this impulse call him. Here is how forceful of the impacts are. So right now, since the ball hasn't hit yet, there's not a whole lot going on. But when the ball hits, there will be impacts greater than 50. Since that minimum is 50 when the fracture occurs, I'm gonna pause the sim. I'm sorry. Positive recording and let the SIM go See how it looks with the complete frame send. The simulation is finished and it feels a little odd. Maybe too much friction. S o. I think I will reduce the friction some and add a few more sub steps before I run that again. Let's take a look at the notes that have been created at the soft level for each of those objects that were simulated specifically the glass inside of the glass network. This partition note was previously the last node, so this rest node will create a rest position at the time when the object was fractured. So that way, whatever you shade the object later, the rest position will follow the object even after a fracture. Because if you can't really rely on things like you've ease with dynamically fracturing objects, because once the object fractures, how do you map this based on the inside, the U. V s wouldn't exist there yet. This stop import brings back the objects that were originally imported into the dock network, and the second op is for bringing in the fractured pieces and then emerge gives you both of those. I make those changes to the network trying maybe five substance on and produced the friction on the glass a little bit. Pause this recording again and see how that seven turns out the SIM has completed and reducing the friction. Help Thio have the glass come down a little easier instead of sticking together so much. If you'd like to cash this simulation in the commercial editions of Houdini, you can use a filed up to ride out the same data. But in the non commercial additions, that note is not functional. I was going to create a another geometry node called cash and create some object merge nodes and bring in that spring in both the glass and the bowl. Like to label these, keep things organized, create a group for each will merge those and write those out with a rob output driver. We can read those back in with a final note. One side of the switch will have the live Sam on the other will have the cash to sin topping. That way, you can check that out later and be able to quickly scrub through it instead of having to simulate again. And I had a pas this again while the cash runs, the cash is completed running. So when we set this cash switch to our final note and we can scrub through what we've got here, I think now's a good time to save, and we can proceed on to the next section of the more advanced control of the fracture parameters. Save this and we'll move on