Further develoment of Jolt physics ? Scientific resources ?

[luigicaval]

Hello everyone, I would like to know where to find all the sources of physical modeling of Jolt routines? Is it Mass-Spring type methods or PBD, XPBD (sources Matthias Müeller) or SPH or others? By the way, is there an implementation of “height field” or wavelet fluids?

In short, I would like to discover the link between the existing functions of Jolt Physics and the theoretical resources on which they rely, in order to be able to extend certain models. Thank you all in advance!

Presumably it is XPBD, but I ask to be sure, as there are a lot of algorithms. What do you think of VBD (2024) already mentionned in another post ?

[jrouwe]

Hello,

The soft body functions make use of XPBD but the rigid body dynamics uses the traditional rigid body formulas. There are a lot of references to articles in the source code, e.g.:

• Soft Constraints: Reinventing The Spring - Erin Catto - GDC 2011
• Solving Constraints' by Erin Catto presented at GDC 2007
• GDC 2013: "Physics for Game Programmers; Continuous Collision" - Erin Catto
• "Constraints Derivation for Rigid Body Simulation in 3D" - Daniel Chappuis
• Implementing Quickhull - Dirk Gregorius.
• Rotation Joint Limits in Quaternion Space by Gino van den Bergen
• Proximity Queries and Penetration Depth Computation on 3D Game Objects - Gino van den Bergen
• A Fast and Robust GJK Implementation for Collision Detection of Convex Objects - Gino van den Bergen
• Ray Casting against General Convex Objects with Application to Continuous Collision Detection - Gino van den Bergen
• Gino van den Bergen - Rotational Joint Limits in Quaternion Space - GDC 2016
• Stop my Constraints from Blowing Up! - Oliver Strunk (Havok)
• Contact generation for meshes - Pierre Terdiman (https://www.codercorner.com/MeshContacts.pdf)

Especially the presentations by Erin Catto were very helpful in getting Jolt up and running.

By the way, is there an implementation of “height field” or wavelet fluids?

If you mean by height field a static collision volume then yes that's implemented in the HeightFieldShape. Fluids are not currently supported.

What do you think of VBD (2024) already mentionned in another post ?

I haven't spent much time looking at VBD so I can't really say much about it.

[luigicaval]

Thank you very much for your detailed answer.

Concerning the "heightfield" I referred to this presentation (there are articles too): "Fast Water Simulation for Games Using Height Fields" from Matthias Müeller-Fischer.

I wonder if you think you can include this kind of model in your engine, or give us some hints to think about where we can do something in the code? Because your engine is superb, the result of a very long work, and congratulations for that, including the update for Godot 4.4.

2010 Fast Water Simulation for Games Using Height Fields.pdf

2010 Real-time Simulation of Large Bodies of Water with Small Scale Details.pdf

Ant also the SPH solution, with an emphasize on unified solution. If you think it is interesting for your engine.

2014 Unified Particle Physics for Real-Time Applications.pdf

I am curious to have your point of view. Thanks again.