[ODE] Negative Penetrations ok?

[Pascal Rettig]

Hi,

   I'm trying to intergrate ODE with another collision system and
I had a couple of questions that someone might have the
answer to (I've looked through the faq and about 1/2 of the individual
  archives and haven't been able to find the answers):

  1. Is it ok to create contact joints with negative penetration distances?
     in other words for collisions that are close to happening but haven't
     happened yet. And if so, does it help at all, or does it just bog
     down the system with more contacts?

  2. All my initial tests have been with boxes and I've noticed that they
     sometimes stick together when moving. I copied most of the code from
     the boxstack example (I do delete the contactGroup after each
     worldStep). Could this just be a standard stability issue that needs fine
     tuning, or might it be a problem with what I'm feeding ODE for
     contact joints?

(BTW: I'm using dWorldStep, CFM=0.001, ERP=0.1, a timeStep of 0.032 and
            a gravity of -9. My boxes all have mass of 1.0 and are 2X2X2.
           I'm also using single precision as the engine i'm using -- 
torque -- uses
           single precision for everything)

    Thanks,
                 Pascal
-------------- next part --------------

---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.545 / Virus Database: 339 - Release Date: 11/27/2003