# What kind of Values are we dealing with? (2d Rigid Body Collisions)

This topic is 2484 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

## Recommended Posts

Hi there,

For a long time now I have had a a simple physics engine set up that handles collisions between 2d rigid bodies, with thanks to numerous articles online. However, I am noticing errors occurring on occasions where the collisions appear broken. What is reassuring is the fix to these errors doesn't lie in the equations, but instead it lies in the values I have assigned to my Rigid Bodies.

Admittedly, I am both new and self taught in this area of physics and mathematics, and as a result in building my physics engine I have plucked together a wide array of equations from as various different articles with misunderstanding of the very basics. Therefore I come with a fairly ignorant question;

In the following article : http://chrishecker.com/images/e/e7/Gdmphys3.pdf

What values would you deem as common for say, the vectors rBP, rAP, the inertia etc, and should they be scaled respective to the velocity and angular velocity?

I'm self taught, so unfortunately I haven't seen the equations I've learned demo'd with actual values before.

cheers,
Murray

##### Share on other sites
There are no common values - it all depends on the scale of your objects and world, both in size and mass.

A common problem with values when simulating rigid body physics is having two objects collide with masses greater than one order of magnitude apart. This will generally cause objects to sink into each other.

Your inertia should be calculated based on the geometry of the object and its mass.

##### Share on other sites
Hi wildbunny.

I noticed the sinking you described in an attempt to fix these errors.

I have decided the cause for the errors I mentuioned is I'm calculating both the inertia and the mass of objects incorrectly. The scale of my geometry is uniform, but my method of calculating value of inertia and mass are corrupted causing problems in my collision response. I'm hoping that by fixing these values the problems will solve themselves once and for all.

1. 1
2. 2
3. 3
4. 4
Rutin
13
5. 5

• 13
• 10
• 9
• 9
• 11
• ### Forum Statistics

• Total Topics
633692
• Total Posts
3013355
×