Inertial dampeners

[Lurkily]

I'm proposing a solution to rigidity that requires no rebuild of how connectivity works, doesn't challenge physics with interconnectivity, and with energy consumption, has a metric you can adjust to balance it with. 

The Inertial dampener would have an adjustable radius, energy consumption scaled to radius and other objects would dampen their response to any physics impulse away from a part's configured position. This would be more forgiving when you have intentionally moving parts, and avoids weird physics calculations with frames of reference and such that might make a mess. 

The main challenge I see is with moving parts - establishing that a part's configured position should be relative to the hinge, spring, or whatever, and that decoupled parts should be relative to the ultimate parent.

Edited October 31, 2018 by Lurkily
Autoco-wreck

[ManTheMister]

So kind of like a shield, except instead of blocking damage it greatly decreases physics? What happens when the drone crashes into terrain when the dampener is on?

[Lurkily]

 The parts take increased damage from impact- with less displacement, there'd be less impact mitigation, so I expect that would just happen without planning it.  They'd still be mobile to seem degree, but they'd resist it more. 

[ManTheMister]

I don’t feel that some way to simply strengthen all connections is the right way to balance things. Drone stability should require some thought, not just an extra block and a few more batteries.

[Lurkily]

There are a lot of solutions to coherence through creative construction, such as 'zippering' or 'lacing' children so that the bottom pulls on the top and vice versa, and magnets can already be used this way to help coherence.  

But with rigidity solutions resurging, I thought I'd test something else. 

Interconnected drones have been tried, and apparently physics became taxing very quickly. 

[ManTheMister]

1 minute ago, Lurkily said:

Interconnected drones have been tried, and apparently physics became taxing very quickly. 

I know that. I don’t think that stability should come at the cost of CPU power. I think that I have a solution that might help with 2 problems in one. Right now, basic blocks aren’t that useful (except for the “L” one, purely because of it’s shape). What if the connections between basic blocks were rigid? That way they would be very useful in stabilizing drones. They would work almost like a “Skeleton”.

[Lurkily]

I seem to recall the devs having spoken about experiments with rigidity, and calling it boring; one reason I stepped away from ideas about rigid chassis or skeleton blocks/assemblies - it doesn't take many fully stable blocks to make a fully stable drone.  

I'm not against the idea, but I'm trying to find solutions for coherence rather than rigidity, since rigid doesn't seem to be preferred.  I think perhaps they want to encourage the creative solutions to coherence, rather than obsolete them. 

[Micha]

The thing I tried is to modify the mass of all parts in a post process step to make parts at the root of the hierarchy heavier than the ones at the end. This shifted the center of mass towards the drone core and completely removed the wobblyness. 

A sideeffect was that each drone felt very similar when flying it and the building got boring fast as you could almost not fail anymore.

Also I needed to adjust the thrusters dynamically because of the mass change. 

I also tried adding struts at one point. But then I just started to connect each part to all the others for extra stability and this resulted in an unmanageable mess. Also the performance got worse because of all the additional joints. If we go this route it needs some limitation

Having a stable core skeleton is something I have to try sometime.

 

[Lurkily]

My chassis idea was awkward.  I'll think on something better, and try to have a post tonight. 

[unmog]

6 hours ago, Lurkily said:

There are a lot of solutions to coherence through creative construction, such as 'zippering' or 'lacing' children so that the bottom pulls on the top and vice versa

Uhm, noob question coming. What is zippering and how do I do it?

6 hours ago, Micha said:

Having a stable core skeleton is something I have to try sometime.

 

'nother noob question coming. Have you tried just making the part connections a little stronger? >.>

 

[ManTheMister]

44 minutes ago, unmog said:

'nother noob question coming. Have you tried just making the part connections a little stronger? >.>

 

I believe that the devs have, and if I remember correctly they decided that it made drones to boring.

[Lurkily]

1 hour ago, unmog said:

Uhm, noob question coming. What is zippering and how do I do it?

1

1

Here is a subdrone I use to courier ore I don't want to carry back myself.  The ore containers were quite unstable when I tacked them on - because they're attached to my ship while I mine, they're subject to the full capabilities of my drone's maneuvering, not just the thrusters that are directly attached.

EDIT: Ignore the logic.  It's a bit messy, still.  I had to rebuild it, and haven't yet cleaned it up.

  So what I did here is 'lace them up' like a pair of shoelaces.  The left parts can't pull away without pulling on the center mass, and on the bottom, pulling on the other chain of parts.  They can't 'peel away' without encountering resistance from the whole subdrone's mass.