Physics warp forces

[NERVAfan]

Why does physics warp amplify forces? Is that supposed to happen?

[*Aqua*]

It's a technical limitation. The faster time passes the less exact physics calculation will be. When your computer can't keep it up (indicated by a yellow or red time display) it will skip some stuff so it can keep up. This can result in phantom forces and more instability as the intervalls between calculations are getting larger.

[JonoRig]

ok. Well there was an above better answer that didn't show up until I posted, so I'm deleting this ha

[Sirrobert]

I believe it's because the CPU has to take bigger steps in the calculation.

So if a force would normally last only 1 second, the bigger steps cause it to be calculated for a full 2 or even 3 seconds

(offcourse it's not actually seconds, but more in the order of miliseconds.)

[NERVAfan]

OK, thank you.

So if you had a sufficiently good computer it would not happen? Or is it a limitation of KSP itself rather than the computer?

[federicoaa]

I guess is both. On one hand your processor may not be able to run all the calculations on time. On the other hand, KSP is single threaded, so having a multicore CPU does not help

[Claw]

I also think part of the proble with physics warp is that some things (like thrust) are multiplied while other things (like joint stiffness) don't seem to be. So ships that arent overdesigned tend to experience unplanned disassembly.

[Red Iron Crown]

I assumed that all the calculations in physics warp worked exactly the same as normal time, just that 1 sec in real life corresponded to >1 sec in game time. Does it really change the minimum time step like on-rails warp does?

[Claw]

I don't know what math it is doing because I haven't dug into it that much. However, it doesn't seem like it's just skipping physics frames. It seems more like it multiplies some of the values by the warp speed. So you end up with things like broken couplers or parachutes that rip off, when at 1x they don't do that.