Cylinders gain momentum

Report any bugs here and we'll post fixes

Moderators: Sascha Willems, Thomas

Re: Cylinders gain momentum

Postby Julio Jerez » Thu Sep 18, 2014 3:04 pm

yes they will no slow down only there is an angular damping, the old method was no correct.
the old method was calculation an intersection plane at the lower point of the cylinder.

this produced 4 contact points a the base separated for a tolerance.
this made cylinder and round object act as of the where no perfect round and act as a dry friction.

how ever that become a problem for small objects, because a small objet, say 0.02 (in radius)
a tolerance of 0.005 is a huge fraction of the radius and the object act weird.

now the contacts are feature based. if the cylinder hit edge on over a polygon the feature is the line that define the cylinder and the contact are the tow lower point for the end of line of the cylinder.
the problem Is that the cylinder may roll for a long time, but shi is better.
adding a small angular drag should take car of that.
Julio Jerez
Moderator
Moderator
 
Posts: 11039
Joined: Sun Sep 14, 2003 2:18 pm
Location: Los Angeles

Re: Cylinders gain momentum

Postby pHySiQuE » Thu Sep 18, 2014 3:27 pm

I have not changed the angular damping value from the default. What default angular damping value do you suggest?
pHySiQuE
 
Posts: 608
Joined: Fri Sep 02, 2011 9:54 pm

Re: Cylinders gain momentum

Postby Julio Jerez » Thu Sep 18, 2014 3:41 pm

just try a few values, there is always a first time for everything.

The behavior now is the correct one, by definition a perfect cylinder rolling on a flat plane will roll for ever if there is not friction.
The important part f the angular velocity is preserved.
So if the cylinder land with some velocity it will preserver it that unless they is some grad of friction force draining the momentum.
Julio Jerez
Moderator
Moderator
 
Posts: 11039
Joined: Sun Sep 14, 2003 2:18 pm
Location: Los Angeles

Re: Cylinders gain momentum

Postby Julio Jerez » Fri Sep 19, 2014 9:14 pm

a better way to deal with the problem is by setting a CustomDryRollingFriction to the cylinder you you want the momentum to dissipate linearly.

The behaviors now is the correct behavior, if a round object hit a friction less surface with some angular velocity, it should roll indefinitely.

The old behavior was nice but was wrong. for example if you want the opposite, the cylinder role for ever, you will not be able to do in any way, now that the natural behavior.

now if you want to apply some friction, you have two ways,
1-apply a angular drag which will give viscous friction (may no what you want)
2-set a dry rolling friction which let you control precisely the how fast or slow the angular momentum will dissipate.

you do not have to do it fro all cylinder, just the one you need the functionality. (an option)
Julio Jerez
Moderator
Moderator
 
Posts: 11039
Joined: Sun Sep 14, 2003 2:18 pm
Location: Los Angeles

Previous

Return to Bugs and Fixes

Who is online

Users browsing this forum: No registered users and 1 guest