Scalar Tardis
SL Scientist/Engineer
Join date: 5 Nov 2005
Posts: 249
|
12-23-2005 03:32
I'm trying to see if the physics engine understands angular momentum and centripetal forces, by making a "governor", as seen spinning around on old steam-powered devices.
This is a gear with a vertical Hinge joint, and then another horizontal Hinge joint attached to the gear with a weighted swing arm out to the side. When the gear spins slowly the weight will hang very low, and as the gear speed increases the weight SHOULD lift up into the air and hang out to the side due to rotational forces.
But, I cannot find any combination of joints that will allow me to add a hinge onto any other hinge joint, whether as a parent or child or with linked objects or not.
This hinge-design limitation is also breaking my attempts at making sliding control arms that transfer motion from one hinge point to another hinge point.
Is this jointing functionality intentionally disabled/broken for some reason?
|
Introvert Petunia
over 2 billion posts
Join date: 11 Sep 2004
Posts: 2,065
|
12-23-2005 04:06
Joints were declared "essentially unusable" by the Linden who implemented them a year or so back, so if you can get them to work at all you've done well. Chaining them is thought to be especially troublesome.
As to why they've not been removed is anyone's guess. They aren't expected to be worked upon or fixed until the Havok 2 physics engine is ported which was oringinally scheduled for 2003-2004; I wouldn't hold my breath.
The physics engine is probably not accurate or reliable enough to make a flyball governor; you'd do better to visually simulate it.
|
Scalar Tardis
SL Scientist/Engineer
Join date: 5 Nov 2005
Posts: 249
|
12-23-2005 04:24
Well, I'm trying to figure out real physical hole-and-pin joints that don't use any special constraints, but these don't work for some reason. A simple horizontal pin lying through the center hole of two rings will not just lay there quietly at the bottom of the trough... no, the pin instead gathers momentum and jiggles around wildly, until it works itself out of the rings and either shoots out or falls to the ground. If I put caps on the rings to contain the pin inside the holes, it will just jiggle around in there endlessly. I've seen a mention that the engine doesn't like physical concave surfaces, which may be part of it, but how else do you fix the location of a loose pin through a hole? Holes are by defintion, concave surfaces. 
|
Introvert Petunia
over 2 billion posts
Join date: 11 Sep 2004
Posts: 2,065
|
12-23-2005 04:29
I found the primer from the source: /111/58/1577/1.htmlGood luck. Oh, for what little I played with them, there seems to be damping on all physics so that balls (and others) don't bounce endlessly. Putting the ball inside the pendulum was clever; any persistent kinetic system needs to have energy added periodically. It is not clear that you can count on the captured ball to not damp itself. llApplyImpulse() is more deterministic (if you can get the parameters right).
|
Jesrad Seraph
Nonsense
Join date: 11 Dec 2004
Posts: 1,463
|
12-23-2005 04:32
Joints are still useful for adding 31 prims to your vehicle 
_____________________
Either Man can enjoy universal freedom, or Man cannot. If it is possible then everyone can act freely if they don't stop anyone else from doing same. If it is not possible, then conflict will arise anyway so punch those that try to stop you. In conclusion the only strategy that wins in all cases is that of doing what you want against all adversity, as long as you respect that right in others.
|
Eloise Pasteur
Curious Individual
Join date: 14 Jul 2004
Posts: 1,952
|
12-23-2005 04:34
I'm not sure if he was serious or not, but between being snowballed Andrew Linden did say that he's been advised by the Havok people to move to Havok 3 - Havok 2 won't be supported for much longer!
Definitely don't hold your breath!
|