Part of the Khronos Group
OpenGL.org

The Industry's Foundation for High Performance Graphics

from games to virtual reality, mobile phones to supercomputers

Results 1 to 8 of 8

Thread: OT: ODE and other RigidBodyDynamics API's

  1. #1
    Junior Member Regular Contributor
    Join Date
    Oct 2000
    Location
    England
    Posts
    103

    OT: ODE and other RigidBodyDynamics API's

    Really off-topic, but sort of relevant to simulations...

    Has anyone any experience in using ODE (Open Dynamics Engine)?
    I'm asking out of curiousity, as I'm currently using it in a test bed, but am hesitant about integrating it into my scenegraph for a number of reasons:-
    1) Is ODE's API typical of other API's, such as Havok and MathEngine? I need to know this before I implement a generic physics class, so I could change between them easily. Does, say Havok, work with body ID's etc. in the same way as ODE?
    2) Is ODE stable enough and flexible enough to be used (in a basic way) in a commercial simulation? (I've noticed its box stacking is a little unstable, but maybe I've not put enough relaxation into them).

    I'd appreciate some responses from people who have experience in more than one 3rd party physics API...

    Cheers.
    --kieran

  2. #2
    Junior Member Regular Contributor
    Join Date
    Nov 2001
    Posts
    102

    Re: OT: ODE and other RigidBodyDynamics API's

    Im not sure how ODE stacks up to other simulators. I know from personal experience that it does work well commercially, but you will need to add a good amount of your own code to massage it to that level.( im sure this would be the same with any simulator, the only exception is how much you are paying for support to do the massage for you.) No kidding off topic for an OpenGL forum.

  3. #3
    Junior Member Regular Contributor
    Join Date
    Oct 2000
    Location
    England
    Posts
    103

    Re: OT: ODE and other RigidBodyDynamics API's

    Ah, thanks for answering.
    Off-topic yes - but I assumed lots of people here have felt the need to actually move their opengl meshes at one time or another, so valuable experiences may be here somewhere...
    So you're saying ODE is sufficiently stable - which is good news.
    But you can't assure me that it's API concepts are similar to other physics engines?
    --kieran

  4. #4
    Junior Member Regular Contributor Julien Cayzac's Avatar
    Join Date
    Aug 2001
    Location
    Yokohama, Japan
    Posts
    230

    Re: OT: ODE and other RigidBodyDynamics API's

    You should consider subscribing to ODE's mailing list, which is quite reactive, and where your questions would be *ON* topic...

    Julien.

  5. #5
    Junior Member Regular Contributor
    Join Date
    Oct 2000
    Location
    England
    Posts
    103

    Re: OT: ODE and other RigidBodyDynamics API's

    O...K....sorry to have bothered you, Julien.
    Next time you want help with an *on-topic* OpenGL question, I'll be sure to show you the same empathy.
    --kieran

  6. #6
    Advanced Member Frequent Contributor
    Join Date
    Feb 2000
    Location
    San Diego, CA, USA
    Posts
    604

    Re: OT: ODE and other RigidBodyDynamics API's

    Man you guys are harsh. It's not like Kieran is some n00b who just signed up and asked a google-answerable question....he's been here a long time and made a lot of valuable contributions.

    Give him a little leeway.

    Unfortunately, I don't have any first hand experience with the physics API's, but GameDeveloper magazine did a side-by-side comparison of the major commercial ones. I can't remember the exact issues (it was a two-parter) but i can look later tonight if you want. You may have some luck on their web site.

    -- Zeno

  7. #7
    Junior Member Regular Contributor
    Join Date
    Nov 2001
    Posts
    102

    Re: OT: ODE and other RigidBodyDynamics API's

    Yea, the ODE mailing list has alot of great information in it definitally worth looking into. Yes thats exactly what I'm saying (in reference to its similarity with other API's). And its definitally very stable as far as an average commercial simulation will go, however im not sure the results are always perfectly correct, especially for complicated simulations that need to be solved(stacking, funnels, ect).

    The only problem with asking this question on an ODE mailing list is your likely to get answers that are only from people that enjoy using ODE.

  8. #8
    Junior Member Regular Contributor
    Join Date
    Oct 2000
    Location
    England
    Posts
    103

    Re: OT: ODE and other RigidBodyDynamics API's

    Yep, cheers. I've read through a fair amount of their archives.
    I only ask here because there are a fair few names here that I know I can trust, because I've followed their conversations on stuff that I do know about (opengl) and therefore know they aren't bullshi*t merchants. I'm not familiar with anyone on the ode mailing list, except russell smith. But here I've got the shared experiences of the likes of zeno, jellofish, nutty, dorbie, davepermen, sirknight etc. All familiar household names!
    --kieran

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •