AddForce behavior

[xavram]

I'm trying to use AddForce for a vehicle. Not that I'm not using the Newton vehicle, its a mesh with a convex hull (sometimes has wheels, sometimes treads, sometimes "hovers", so I'm trying to do something generic that covers any of the scenarios).

My question is, it seems that you can add force to an object over and over again and it won't move for a while, based on the force applied and the mass of the object. Makes sense but when a certain threshold is hit, the object seems to start slowly moving...and then takes off big time.

I guess I'm not understanding how to control this better. I assumed (maybe incorrectly) that the simpliest way to make an object move in a realistic manner was to add force to it in the direction its supposed to be going, until the object is moving at the velocity that you want, then stop adding force...but my tests/experiments with this don't seem to bear this out.

Can someone point me in the right direction for how to do this?

[Julio Jerez]

If the objet is resting it will be under the influence of the static friction.

when this is teh case adding forces do not move the body until the force is larger the static friction force.
when the body start moving it goes into kinetic friction mode,
In kenetic friction the frictin force is smaller so if the same force is applied generate a net acceleration that will keep moving the body faster and faster.

this the basic realistic behavior of kenetic/static friction action on solids in real life.
you can check on the internet or on your text book so that you refresh your memory from highscholl basics physics.

[xavram]

So, that being the case, is AddForce not the best way to simulate vehicle movement? Would SetLinearVelocity be better?

I've tried that but I see issues with just setting the velocity in the direction that the vehicle is facing...if it goes up a jump and flies off the edge, it keeps going up and up, since it's direction is now pointed upward.

I know I'm missing something obvious here, I'm a total rookie on this whole physics thing. Got my feet wet with a relatively simple "shoot ball at wall of bricks" app but this vehicle movement stuff is an entirely different kettle of fish.

[Julio Jerez]

why don't you they the ray cast car?

[xavram]

I'm using Newton through TrueVision 3d 6.5, I don't believe that's available to me.

Plus, isn't that ray cast car using a vehicle with 4 wheels? For my project, the "car" is actually a tank and it could have treads or wheels...or even "jets". What I need is some kind of simulation that I can apply to all types, that's why I was using just a basic convex hull added to the mesh.

[Stucuk]

What version of Newton does TrueVision use? If they are using 1.53 (Which is old) then that could cause problems (As it has more bugs than the latest verison). If they are using 1.53 you should bug them to update there SDK to use 2.0 .

[xavram]

I'm fairly sure its currently using 2.0

[Julio Jerez]

you can add a series of wheels to teh ray castcar to simulate a track or a car.

[xavram]

So if the materials involve have the "standard" friction settings (.9 static, .5 kinetic), is there a formula I can use, knowing what the mass of the object is and the gravity setting to ssay, "adding this much will result in movement"?