|
|
Post by michael on Feb 3, 2015 6:20:45 GMT
The link below is a shot video of the issue. The simple trebuchet I was attempting to make conserves energy that it shouldn't, and ends up with perpetual oscillation. I suspect the devs are aware of this issue, but if you guys would like any help, I am a physics major with significant computation physics background. I'd love to help with this project if I can.
www.youtube.com/watch?v=LC852FEE8yA&feature=youtu.beAlso, the way pistons are written, there seems to be a divide by near 0 exponential displacement issue making pistons behave quite creeply when large stacks of them cause individual ones to bend off of their main extension axis. I guess you could refer to it as their shearing motion. |
|
|
|
Post by Praesumo on Feb 3, 2015 6:35:13 GMT
Excellent video. Could you also post the .bsg (save file)?? It would help to know what the ballast blocks weight is set at, and would allow the devs ( and me) to quickly load it up and look at the issue.
|
|
|
|
Post by michael on Feb 3, 2015 6:53:26 GMT
I didn't save it so I had to remake it real quick. All of the ballasts are maxed at 2.00 mass units.I added struts to connect the pillars to make it a little more stable. The effect is still produced. See attachment.Scratch that... Adding the struts to make it more stable seems to at least usually diminish the problem. This means that the issue is in oscillations in the support legs and through the swivel joints. The wobbles are giving it energy. Delete the horizontal cross struts and it will do what it did before. Attachments:funkyphysicstreb.bsg (3.65 KB)
|
|
|
|
Post by michael on Feb 3, 2015 7:07:48 GMT
With more struts things are more consistent with real physics. This is an issue KSP has been polishing out for quite some time. It isn't really a simple solution given standard computation physics modeling methods lol. I misunderstood it to be a simpler issue at first. I'll just add more struts for now. I'm sure the devs will get to it. Content is more of a priority at this point. Attachments:workingphysicstreb.bsg (4.56 KB)
|
|
|
|
Post by Zeblote on Feb 3, 2015 13:13:48 GMT
I assume this is because they are using unity's default physics engine (physx), which isn't meant for wobbly stuff like the wooden beam. I'd rather be happy it works at all.
|
|
|
|
Post by michael on Feb 3, 2015 19:22:07 GMT
I assume this is because they are using unity's default physics engine (physx), which isn't meant for wobbly stuff like the wooden beam. I'd rather be happy it works at all. Ya. Unity's physics are really nice for basic needs, and to most of the player base it isn't an issue. I always code my own physics in unity because I love simulation games for learning. I love the art style and optimization straight from alpha. Unfortunately that optimization is due likely in large part to using unites own built ins, which means making some of these physics changes could be cumbersome. |
|
