What metric weight does dMass represent?

[colm...@gmail.com]

Does a value of 1.0 represent 1 gram, 1 kilogram or is there any
relationship at all? I think it's relevant when using dBodyAddForce.
It seems to me that whenever I give an object on a simple hinge joint
a weight more than 1.0, or as low as 2.0, the body begins to sag
deeply and become wayward. Is it 1 ton or is 1.0 just the max and only
values betweeen 0 and 1 should be entered?

Thanks for any replies.

[Daniel K. O.]

2008/3/23, colm...@gmail.com <colm...@gmail.com>:

> Does a value of 1.0 represent 1 gram, 1 kilogram or is there any
> relationship at all?

The interpretation is up to you. Sane people prefer to use SI, so they
measure mass in Kg, force in Newtons, and so on. If you use another
interpretation for mass values, remember to adjust any dependent unit
(like, say, forces).

> I think it's relevant when using dBodyAddForce.
> It seems to me that whenever I give an object on a simple hinge joint
> a weight more than 1.0, or as low as 2.0, the body begins to sag
> deeply and become wayward. Is it 1 ton or is 1.0 just the max and only
> values betweeen 0 and 1 should be entered?

As a general consequence of floating point arithmetic, operating
values with very different magnitudes might introduce a lot of errors.
Thus it's recommended to use values as close to 1.0 as possible, for
numeric stability. This is probably the only sane reason to use non-SI
units in your simulation.

Also, time steps should be as small as possible (but making it too
small might again introduce errors). Your problem might be related to
this. Try dumping the simulation state (dWorldExportDIF); if you still
can't figure out the problem, post the DIF output, with more details.

--
Daniel K. O.
"The only way to succeed is to build success yourself"

[colm...@gmail.com]

Brilliantly detailed answer, Daniel. Thanks. It's exactly what I was
looking for :)