Fluid Dynamic and Physics simulation improvements

[Nechochwen]

I am certain most everyone here has built and launched a tall, skinny, rocket and experienced the "Noodle Effect" where as the rocket ascends it begins to bend around the point connections between parts.

 

How difficult would it be to pre-model the envelope and dry mass distribution of a vehicle for each stage and then only use that baked object for physics calculations?

[MilongaDelAngel]

I think, to a certain degree, this is intentional. It might be a bit extreme at the moment, but real rockets have to be built and flown carefully so they don't fold in half too. Tall skinny rockets are supposed to be more unstable. The unrealistic part is how flexible they are, in real life they would just snap and explode long before going banana shaped. I guess that was ruled out for being too unforgiving, but it would be way too easy to build rockets that don't look like rockets if they were processed as one rigid body. I'm sure they will find the right balance between rigid and noodle sooner or later.

[Nechochwen]

True. And I would be all for having crumple thresholds and other deformation parameters, the problem is that the current connection points are mathematical points and they have nonzero torque limits. Frankly I believe that this is the root cause of "The Kraken".