[BUG] Fragments flung off at high velocity

[Lord DC]

If I fire an asteroid at a moon orbiting something, or just something in general, the fragments are accelerated way beyond the speed of the impactor.

I don't know if this is truly a bug, however to be safe I'm posting it here.

[gabriel.dac]

I noticed that too. The fragments seem like missiles lifting off and crashing back on the planet

[C7]

If you fire a body at 10km/s it will typically accelerate as it falls into the gravity well of the planet you're aiming at. It's possible they're getting sped up a good bit. However, they shouldn't be ejecting faster then that, if you take a look at the speed just before impact.

If they are, please provide some reproduction steps. What body did you fire at which? Were you in a specific simulation or time speed?

We're still tuning this stuff up after such a large update.

Thanks for the feedback!

[Lord DC]

How to reproduce:

1. Load Earth and Moon sim

2. Fire a Ceres at the Moon.

3. Before impact, take the speed of the Ceres

4. check the speed of the fragments, it is usually a LOT faster than the Ceres (right before impact)
Will Record Video

[smjjames]

I'm also seeing the fragments go significantly faster.

What I'm also seeing is that the fragments have incredibly fast rotation periods, in the frame of seconds. Fast enough to rip themselves apart if it were real world. The rotations that I'm seeing in the current sim are from maybe 6 seconds (the lowest I saw was around 7 seconds) to 15 or so.

[Retsof]

I've noticed this happening too.  For instance, if using the random Jupiter system, after a collision of moons the new body will have gained a lot of velocity.  And if using the new total fragmentation, some of the fragments will launch off into infinity once they start colliding with each other.

EDIT:  I may have figured something out.  It seems that collisions change the "Rotation Axis" planet parameters, instead of the "Obliquity" parameter.  And if the "Rotation Axis" parameter is anywhere other than X0 Y1 Z0, the planet completely spazzes out.  This probably messes up the angular momentum calculations the next time it impacts something.

[C7]

Thanks for the reports, it does seem that there are some issues with the rotations. We've also seen some problems when a body becomes collide-able when it's inside another.

We're in the middle of some performance optimizations on the physics engine, and once that is complete, we can go back to stability and polish. I'll add this problem to the our Trello board, so we don't forget about it.