Changing physics engine

Earlier this week, me and Georges had a meeting with representatives of Havok to discuss licensing their physics engine for Dragontwin. Georges has worked with Havok in the past during his time at Ubisoft since it was used in Assassin's Creed, so he still had a contact there and got us a fast track to evaluation. We're getting access to it next week and have 3 months to test it out before deciding if it's worth it or not.

But wait, don't we already have a physics engine for all the physics stuff we've done so far? Yes, it's called Chaos and comes with UE5 (the game engine we're using). In the good old days of UE4 there was a different physics engine called PhysX. I remember the switch from UE4 to UE5 all too well because suddenly all my physics projects stopped behaving. Chaos had issues. Over time it has gotten better and it is now totally fine for normal physics usage, but we're going beyond normal with Dragontwin. We want next-gen, and Chaos ain't it. Meanwhile, Havok (which lets you do this ) has just made a full integration for UE5 and they just announced a new pricing plan that is meant to be more accessible to smaller (non-AAA) developers. How accessible? Well, 50K USD licensing fee per project, so you can't be too small. It's not completely beyond our reach though, but the question is - is it worth it?

The main benefit we'd get from Havok is that we can just do more, because it is much faster than Chaos. I'll be careful with specifics just in case because we signed an NDA, but let's just say we'd get several times as many simulated bodies. So that means we can simulate destruction on a much bigger scale without everyone's PCs catching fire. We should also be able to get more accurate simulations for the dragon bodies, we're not sure exactly what that will look like so we'll have to see when we get to test it, but theoretically it will look even better!