I'm sorry.... WHAT?

[Mushroom Kerman]

...

...

I was, at first, confused as to why my ships were flying around like rocket powered ramen noodles... but NOW THAT I KNOW WHY I'm even more confused as to why these numbers are... barely helping?

10,000x more rigid and the wobble only gets.... a lil bit better? Can someone who is good with computers explain this to my tiny mushroom brain?

[selfish_meme]

I don't know why the fix does not work for you, I don't even know why you have it, I built an SLS alike Mun mission and had no wobble, I can only assume you guys must be trying to put some super heavy upper stages into orbit, come back to that sort of play once the devs have released some gameplay patches, make some smaller stuff.

I know why rocket wobble happens if you want the explanation for that?

[sumeee]

From my very limited testing it seems that the joint rigidity is based on size of the connected parts, XL fuel tanks have higher rigidity than MD fuel tanks.  A rocket with XL fuel tank will behave pretty good but a rocket of the same height build with MD fuel tanks will wobble pretty hard, combine that with the over tuned SAS and you will have a bad time.

[dr.phees]

Isn't it weird that Kerbals have very cool animations now, but the actual physics engine is still in its beginnings?

If multiplayer worked and we had a way to play Kerbals from first person view, that would actually be cool and I could have lots of unique fun. But if one is rebuilding basically the same game, the foundation should come first. I hope they can recover from this and not just patch their physics engine with workarounds.

Edited February 27, 2023 by dr.phees

[tstein]

3 hours ago, selfish_meme said:

I don't know why the fix does not work for you, I don't even know why you have it, I built an SLS alike Mun mission and had no wobble, I can only assume you guys must be trying to put some super heavy upper stages into orbit, come back to that sort of play once the devs have released some gameplay patches, make some smaller stuff.

I know why rocket wobble happens if you want the explanation for that?

6 parts rockets wobble, then you reload and they do not wobble, then you reload again and they wobble again.

[kerbiloid]

Just reload an even number of times.

[Gotmachine]

5 hours ago, Mushroom Kerman said:

10,000x more rigid and the wobble only gets.... a lil bit better? Can someone who is good with computers explain this to my tiny mushroom brain?

The short answer is that PhysX, the physics engine KSP 2 (and 1) uses is not capable of simulating infinitely rigid joints. That's not what it's designed for.
They will always flex to some extent, and additionally, it's dependent on the mass ratio between connected parts.
Said otherwise, a joint connecting a very heavy and a very lightweight part will flex (a lot) more than a equally strong joint connecting two parts of equal mass.
This is why you get a lot of flex in KSP 1/2 when you connect a several tons fuel tank to a 100 kg decoupler.

And to be clear for all the people thinking having wobbly joints is a design decision and a physically accurate challenge, it's not.
The behavior is unphysical and largely unpredictable, it exists because PhysX is the default physics engine available in Unity and requires the least effort to implement.

The only workaround is to spam more internal joint connections between each part, which is what the KSP 1 autostruts does (badly) and the KerbalJointReinforcement KSP 1 mod does (a lot better).

Edited February 27, 2023 by Gotmachine

[The Aziz]

1 hour ago, dr.phees said:

Isn't it weird that Kerbals have very cool animations now, but the actual physics engine is still in its beginnings?

You're saying this like if the same people worked on both, or making them were equal in difficulty. Dude working on Kerbal animation won't help the physics engineer working on part joints.

[tstein]

3 minutes ago, Gotmachine said:

The short answer is that PhysX, the physics engine KSP 2 (and 1) uses is not capable of simulating infinitely rigid joints. That's not what it's designed for.
They will always flex to some extent, and additionally, it's dependent on the mass ratio between connected parts.
Said otherwise, a joint connecting a very heavy and a very lightweight part will flex (a lot) more than a equally strong joint connecting two parts of equal mass.
This is why you get a lot of flex in KSP 1/2 when you connect a several tons fuel tank to a 100 kg decoupler.

And to be clear for all the people thinking having wobbly joints is a design decision and a physically accurate challenge, it's not.
The behavior is unphysical and largely unpredictable, it exists because PhysX is the default physics engine available in Unity and requires the least effort to implement.

The only workaround is to spam more internal joint connections between each part, which is what the KSP 1 autostruts does (badly) and the KerbalJointReinforcement KSP 1 mod does (a lot better).

And that is the reason I have been parroting "Give us procedural tanks" for the last year!

Just now, The Aziz said:

You're saying this like if the same people worked on both, or making them were equal in difficulty. Dude working on Kerbal animation won't help the physics engineer working on part joints.

But the money used to prioritize hiring one could have been  prioritizing other things.. that is the point.

[The Aziz]

Were they supposed to fire the animator or what? Or should the game arrive with Kerbals T-posing?

[tstein]

2 minutes ago, The Aziz said:

Were they supposed to fire the animator or what? Or should the game arrive with Kerbals T-posing?

They were supposed to FOCUS better when they hired people for the project.  When a project ends up with so many problems  it generally is because of bad planning and management. Also yes,  if there is a team member that cannot help on what the product needs, it should be fired as it happens in every company worldwide, or retrained to do something useful if he/she is willing, or  leased to another company (to get funds for what you need). That is MANAGEMENT of a company.

[MarcAbaddon]

Procedural tanks are not a solution to wobbly rockets, they'd be only a bit of balm by reducing the number of joints. There are still times when you are going to stack other parts. 

[tstein]

20 minutes ago, MarcAbaddon said:

Procedural tanks are not a solution to wobbly rockets, they'd be only a bit of balm by reducing the number of joints. There are still times when you are going to stack other parts. 

Yes but they reduce the stack by half. That is engineering solution. If you cannot remove the problem you remove the frequency of the condition. The point is.. it is super SIMPLE and it could have already delivered improvements.

[MarcAbaddon]

1 minute ago, tstein said:

Yes but they reduce the stack by half. That is engineering solution. If you cannot remove the problem you remove the frequency of the condition. The point is.. it is super SIMPLE and it could have already delivered improvements.

True enough, they would make life simpler at this stage. I don't really think we need fully procedural tanks, but just being able to adjust length and fuel type would be enough. Just saying they will need to give way to solve wobbling anyway. 

[Gotmachine]

1 hour ago, tstein said:

Yes but they reduce the stack by half. That is engineering solution. If you cannot remove the problem you remove the frequency of the condition

Due to the nature of the problem, that doesn't really solve anything, it would make it even worse.
As I mentioned, the main issue is that PhysX joints become less rigid when the mass ratio between connected parts is higher.
So for example having a single huge procedural fuel tank part connected to a decoupler is actually worse than several smaller fuel tank parts connected together and then to the same decoupler.

[tstein]

11 minutes ago, Gotmachine said:

Due to the nature of the problem, that doesn't really solve anything, it would make it even worse.
As I mentioned, the main issue is that PhysX joints become less rigid when the mass ratio between connected parts is higher.
So for example having a single huge procedural fuel tank part connected to a decoupler is actually worse than several smaller fuel tank parts connected together and then to the same decoupler.

Not really because you should only have  a joint in the  decouplers. And usually the parts in front and behind a decoupler are roughly in the same mass (but ineed you do need limits on the side to avoid extremes).  

 

[MarcAbaddon]

What counts as the part in front of the decoupler? The engine? Then you would have a pretty bad mass ratio between it and the preceding fuel tank. I haven't really looked into the implementation details yet.

[tstein]

5 minutes ago, MarcAbaddon said:

What counts as the part in front of the decoupler? The engine? Then you would have a pretty bad mass ratio between it and the preceding fuel tank. I haven't really looked into the implementation details yet.

Engines  should not  even  be computed They are the force originators and the force always  is exerted upwards of most of the engine mass.   Effective simplification  do not even USE Them in the rigid  body mechanics.  No it woudl nto reduce realism because   real life rockets are NOT even remotely as in gae so when you realism is already zero, reducing does nto change much.

[kerbiloid]

16 hours ago, The Aziz said:

Were they supposed to fire the animator or what?

Why? The Kerbal animation looks the most perfectly developed part of the game.

Thus, this specialist(s) probably could be used more effectively.

***

Had they even looked the KJR mod before release?
 

[RocketRockington]

17 hours ago, Gotmachine said:

The short answer is that PhysX, the physics engine KSP 2 (and 1) uses is not capable of simulating infinitely rigid joints. That's not what it's designed for.
They will always flex to some extent, and additionally, it's dependent on the mass ratio between connected parts.
Said otherwise, a joint connecting a very heavy and a very lightweight part will flex (a lot) more than a equally strong joint connecting two parts of equal mass.
This is why you get a lot of flex in KSP 1/2 when you connect a several tons fuel tank to a 100 kg decoupler.

And to be clear for all the people thinking having wobbly joints is a design decision and a physically accurate challenge, it's not.
The behavior is unphysical and largely unpredictable, it exists because PhysX is the default physics engine available in Unity and requires the least effort to implement.

The only workaround is to spam more internal joint connections between each part, which is what the KSP 1 autostruts does (badly) and the KerbalJointReinforcement KSP 1 mod does (a lot better).

Came here to say this.  PhysX uses an iterative relaxation solver.  Depending on the iteration count, there's a limit on how precisely connected bodies can be solved for.

I'm kinda disappointed that KSP2 continued to use this as their solution at all.  I didn't expect them to go do something 100% awesome like using BeamNG.drive's  soft body physics for really good destruction effects (I wish) but I would have hoped they'd use something other than PhysX, which is (imho) one of the worst physics engines for simulating rigid body physics for mechanical systems with linkages.  PhysX is more for game physics like ragdolls and single-rigid-body collisions to be simulated fast.

[cfds]

23 hours ago, dr.phees said:

Isn't it weird that Kerbals have very cool animations now, but the actual physics engine is still in its beginnings?

It is not weird if you consider two points:

* People who can create an engine that acts as a solid foundation for future development are more expensive than 3D modelers and animators

* Engines that act as a solid foundation for future development do make far less flashy promo materials than 3D models or animations

So if you want to create a good game, you start with the solid foundation and add the flashy parts later. Otherwise, you create a flashy game, sell it to the hyped masses,  and hope that the problems of a wobbly foundations magically vanish (or are excused by the aforementioned masses high on copium).

Edited February 28, 2023 by cfds
Typo