Timeline for Unity falling body pendulum behaviour
Current License: CC BY-SA 4.0
16 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| S Dec 26, 2018 at 20:07 | history | suggested | Ruzihm | CC BY-SA 4.0 | uses imgur mirror of hotlinked image |
| Dec 26, 2018 at 17:00 | review | Suggested edits | |||
| S Dec 26, 2018 at 20:07 | |||||
| Mar 9, 2017 at 17:30 | history | edited | CommunityBot | replaced http://upload.wikimedia.org/ with https://upload.wikimedia.org/ | |
| May 20, 2016 at 15:38 | comment | added | teodron | @MLM sorry for resurrecting this, but knowing this problem to be a good benchmark for illustrating the properties of numerical integration methods, I noticed your remark where you stated you did not want to cheat by using the analytical (theta-based) solution. As far as I know, the numerical integration (in your case Symplectic Euler) cannot completely solve the issue of not being able to keep a reasonably constant swing radius. Then I saw this line: return this.GetPointOnLine(pivot_p, currentStatePosition + movementDelta, distance <= this.ropeLength ? distance : this.ropeLength);. | |
| May 29, 2014 at 2:32 | comment | added | MLM | @user3447980 I updated the code with a fix for drift in side to side motion when there shouldn't be any movement. I noticed this some setups. I also tried a double precision version of the Pendulum script but there is no difference. There is still a problem where the pendulum slowly doesn't make it as high as when it started which I will try to figure out. | |
| May 29, 2014 at 2:29 | history | edited | MLM | CC BY-SA 3.0 | fixed current position issue in code |
| May 29, 2014 at 0:17 | history | edited | MLM | CC BY-SA 3.0 | added 238 characters in body |
| May 28, 2014 at 19:15 | history | edited | MLM | CC BY-SA 3.0 | deleted 10 characters in body |
| May 28, 2014 at 19:12 | comment | added | MLM | @user3447980 I get this slightly "chaotic" effect as well. Part of it has to do with floating point imprecision in the decimal range. Increasing dt may help as deltaTime will be smaller making it interpolate less but also means more imprecision in the decimal range. I think the real issue is that since the rope can't stretch, I limit the distance away from the pivot(see the bottom of PendulumUpdate). Also keep in mind that this is a rope and not a rigid rod. So if you put the bob in the middle it will free fall, before hitting the end of the rope and starting to swing. | |
| May 28, 2014 at 18:59 | vote | accept | user3447980 | ||
| May 28, 2014 at 18:57 | comment | added | user3447980 | This is utterly phenomenal. Really incredible. Been playing around with this and you can really see just how realistic this when you push it into a turbulent state and see it come apart. My one query would be that it appears to be mildly chaotic in increasing and decreasing amplitude, is there a way to explain what is happening there? Still trying pick this apart and digest it all... | |
| May 28, 2014 at 16:59 | history | edited | MLM | CC BY-SA 3.0 | Explained update() loop |
| May 28, 2014 at 16:02 | history | edited | MLM | CC BY-SA 3.0 | added 98 characters in body |
| May 28, 2014 at 15:15 | history | edited | MLM | CC BY-SA 3.0 | deleted 110 characters in body |
| May 28, 2014 at 15:09 | history | edited | MLM | CC BY-SA 3.0 | added 218 characters in body |
| May 28, 2014 at 14:58 | history | answered | MLM | CC BY-SA 3.0 |