Very cool! Interesting how the choice of solver affects the solution. Euler doesn't handle misbehaved equations very well. You can see this in the Helix setup where the bodies just fly off.
Notes re: Multi-body problems on "Show HN: Browser-based interactive 3D Three-Body problem simulator" (2025)
https://news.ycombinator.com/item?id=45981336 which has Velocity Verlet integration method, 4th-order Runge-Kutta (RK4), and Kepler solvers
- i have seen a tonne of these
- would be nice if someone at this point actually made a tutorial on how to go about making one from scratch with threejs
Feedback : The ui covers more than half my screen on mobile browser
How many of these are there? https://trisolarchaos.com/
When I hit Random 3d to see a normal three-body they yeet out of existence far far away.
The tri-Solarians could have just used this!
They did :) That's why they said fuck that and moved to Earth.
OP's account posts LLM generated comments almost exclusively, this is likely not an exception
Very cool! Interesting how the choice of solver affects the solution. Euler doesn't handle misbehaved equations very well. You can see this in the Helix setup where the bodies just fly off.
Random3D seems to always have all the stars heading off to infinity.
Occasionally two will tight spiral off together.
(Experimentally 5/250, including 1 cool wide, almost flat spiral.)
Very cool! nice to be able to export json too
No Weber very sad.
needs a WebXR bookmarklet
Notes re: Multi-body problems on "Show HN: Browser-based interactive 3D Three-Body problem simulator" (2025) https://news.ycombinator.com/item?id=45981336 which has Velocity Verlet integration method, 4th-order Runge-Kutta (RK4), and Kepler solvers
Had a similar idea and built a prototype for it. :)
https://solgrbit.d9ng.co.kr/ Had a similar idea and built a prototype for it.
[dead]