Jump to content

  • Log In with Google      Sign In   
  • Create Account

Banner advertising on our site currently available from just $5!


1. Learn about the promo. 2. Sign up for GDNet+. 3. Set up your advert!


Irlan Robson

Member Since 08 May 2012
Online Last Active Today, 05:12 AM

Topics I've Started

Bounce v1.0 Release

24 July 2015 - 12:51 PM

Good afternoon everyone.

I'm happy to distribute Bounce, a lightweight and portable 3D physics engine for games. 

Bounc3 is available at GitHub. 

https://github.com/irlanrobson/Bounce

Some of the physics engine features can be seen at the basic (.pdf) documentation, living inside the repository. Additionaly, a demo application lives inside the repository as well.

 

Cheers,

Irlan R.


Adding restitution to the MLCP

24 May 2015 - 08:22 AM

The compl. cond. for the contact and impact in a vel. based formulation is:

 

Ju(t + dt) + eJu(t) >= 0 _|_ lambda >= 0.

 

where u = system velocity, and J is the normal jacobian, e is the coefficient of restitution.

 

My question is:

 

Using an error correction term b, the condition becomes:

 

Ju(t + dt) + eJu(t) + b >= 0 _|_ lambda >= 0

 

?

 

So, defining low = 0, high = infinite, passing thourgh all simplification, assuming acceleration = ( u(t + dt) - u(t) ) / dt, we have

 

A = J*M^-1*J^T,

x = dt*Lambda

b = J * ( u(t)+dt*M^-1*f) + eJ*u(t) + b, (f == external forces),

 

w = Ax - b _|_ low <= w <= high,

 

?


3D SAT Problem

10 April 2015 - 07:55 PM

Hello, newbie to SAT algorithm.
 
What am I missing in the SAT implementation below?

http://pastebin.com/CqNbB7ph

I think the problem is between the edge-edge case. Code doing all computations in world-space.

Graphs tool for a thesis (needing an suggestion)

13 March 2015 - 10:20 AM

I need a graphs tool to write my graduation project. I'm writing a rigid body simulation so I need to graph the Euler and RKs methods. I would like to know if there is one which I can do this fast. A free software if possible.


Friction Impulses (boxes are sliding on the ground)

09 February 2015 - 08:07 AM

Hi. 

 

The problem I'm facing with is with correct impulses but wrong behaviour. The physics engine uses split-impulses (bias-velocities), but when objects fall over a plane it starts to slide like if there was an joint above them. Here's a video:

 

[This video was deleted because was messing up with my channel, but it is what the description says]

 

The loop (at 60HZ = 0.016ms):

 

Update contact positions (contact caching not implemented yet, so skip it);

Check for collisions;

Apply forces (integrate velocities);

Prepare constraint data (normal inverse mass, two tangents inverse mass, and bias values);

Warm Start last frame impulses (contact caching not implemented yet, so skip it)

Apply Contact Constraint Impulses (10 iterations, zero accumulated bias impulse, etc.);

Apply velocities (integrate positions, zero bias velocities);

 

I'm skiping the warm start, but that doesn't mean that I should get the behaviour above.

 

Has anyone faced with that problem?

 

Here's the overview of each task:

 

http://pastebin.com/K1iF06nH


PARTNERS