Search found 25 matches

by h4tt3n
Tue Oct 23, 2018 4:11 pm
Forum: Links, Papers, Libraries, Demos, Movies, Comparisons
Topic: Tutorial: How to make rigid, stable damped springs
Replies: 0
Views: 3404

Tutorial: How to make rigid, stable damped springs

Hello folks, Here's a link on my tutorial on how to improve stability and rigidity the infamous old Hooke's law damped spring. Practically speaking, it explains how to express a damped spring as a rigid constraint between two point masses. The tutorial contains a detailed walkthrough on how to deriv...
by h4tt3n
Tue Oct 23, 2018 3:26 pm
Forum: Research and development discussion about Collision Detection and Physics Simulation
Topic: Constrain point on rigid body to line or line segment
Replies: 0
Views: 3415

Constrain point on rigid body to line or line segment

Hello folks, I am implementing a number of rigid body constraints for a small proof-of-concept 2D game demo, and now I'm working on a constraint that limits a point on one rigid body to only move along a line or line segment mapped onto another rigid body. I have a working prototype, but I'd like to...
by h4tt3n
Sat Sep 10, 2011 8:05 am
Forum: Research and development discussion about Collision Detection and Physics Simulation
Topic: Newbie physics simulation question (collision detection)
Replies: 1
Views: 2979

Re: Newbie physics simulation question (collision detection)

Hello there, There are several issues with your current setup, the most notable beeing the use of sqrtf(). Square root functions always take up a relatively large portion of cpu-power, especially when nested inside a double loop. Instead, you could test the circle radiii squared against the distance...
by h4tt3n
Thu Mar 31, 2011 9:21 am
Forum: Research and development discussion about Collision Detection and Physics Simulation
Topic: Angular Velocity 2 questions
Replies: 2
Views: 5412

Re: Angular Velocity 2 questions

In the real world, a unconstrained body not influenced by any forces can only rotate along one axis. Look at the Earth as an example. If Bullet allowed you to make an object rotate about more than one axis at a time, it wouldn't be physically correct.

Cheers,
Mike
by h4tt3n
Sun Jun 20, 2010 5:56 pm
Forum: Research and development discussion about Collision Detection and Physics Simulation
Topic: 2D angular velocity
Replies: 1
Views: 2369

Re: 2D angular velocity

Well, let me google it for you :-) http://lmgtfy.com/?q=2d+cross+product the 2d cross product between vectors a and b is defined as: ax*by - ay*bx where x and y are the vector components. Notice that the result is a scalar, not a vector. If you normalize the vectors first, it gives you sine to the a...
by h4tt3n
Sun May 30, 2010 4:40 pm
Forum: Research and development discussion about Collision Detection and Physics Simulation
Topic: Mass-spring and dynamic relaxation
Replies: 9
Views: 9287

Re: Mass-spring and dynamic relaxation

horyd wrote:As far as I can tell they appear to be quite similar, however all relaxation is based from F = Kx + Cv + Ma while mass spring systems are generally confined to F=Kx?
Where exactly did you read this?
by h4tt3n
Wed Nov 11, 2009 8:15 pm
Forum: Research and development discussion about Collision Detection and Physics Simulation
Topic: Stiff Deformable Materials in Realtime
Replies: 3
Views: 2760

Re: Stiff Deformable Materials in Realtime

Below is a link to a thread, where I recently described my approach, which I call an "iterative" velocity verlet algorithm. It'll handle spring stiffnes values of several millions (for particle mass 1 and timestep 0.01) with a reasonable number of iterations. It will definitely do what you...
by h4tt3n
Wed Nov 11, 2009 3:58 pm
Forum: Research and development discussion about Collision Detection and Physics Simulation
Topic: Stiff Deformable Materials in Realtime
Replies: 3
Views: 2760

Re: Stiff Deformable Materials in Realtime

Why exactly can't you do this mass-spring mesh in realtime using explicit integration? Any relatively new machine should be able to handle 10.000's of springs without sweating. It seems to me like you're overdoing the integration part a bit. Did you try the relatively simple velocity verlet algorithm?
by h4tt3n
Wed Oct 21, 2009 9:03 am
Forum: Research and development discussion about Collision Detection and Physics Simulation
Topic: Finite Elements Method: are there proble,s with stability?
Replies: 57
Views: 37662

Re: Finite Elements Method: are there proble,s with stability?

@ExperienS Thanks a ton for the link. The solver described is very similar to mine, except they seem to use the leap-frog method where I use velocity verlet. I found chapter 4.2 about stability and the size of the time step very interesting. For quite a while I've been looking for a method to theore...
by h4tt3n
Tue Oct 20, 2009 5:19 pm
Forum: Research and development discussion about Collision Detection and Physics Simulation
Topic: Finite Elements Method: are there proble,s with stability?
Replies: 57
Views: 37662

Re: Finite Elements Method: are there proble,s with stability?

This reminded me of this paper: http://leri.univ-reims.fr/~nocent/papers/kacic03.pdf Sorry, the link is broken, and google doesn't know of any "kacic03.pdf". Have you got another link? Also I've spoken with the programmer who made Gish/Bridge Builder/etc. and they use a similar approach: ...
by h4tt3n
Sun Oct 18, 2009 12:35 pm
Forum: Research and development discussion about Collision Detection and Physics Simulation
Topic: Finite Elements Method: are there proble,s with stability?
Replies: 57
Views: 37662

Re: Finite Elements Method: are there proble,s with stability?

h4tt3n Could you explain the method in more details or show us some links on what you're calling iterative velocity verlet scheme? Certainly, it's almost ridiculously simple. The idea is as follows. Let's say we want to model a chain of particles held together by stiff constraints. Normally you wou...
by h4tt3n
Thu Oct 15, 2009 5:19 pm
Forum: Research and development discussion about Collision Detection and Physics Simulation
Topic: Finite Elements Method: are there proble,s with stability?
Replies: 57
Views: 37662

Re: Finite Elements Method: are there proble,s with stability?

I'd like to hear more about how h4tt3n avoids the unexpected displacements - do the ridiculously high K (spring rate) values still behave stably if you have a bunch of them in series, say a pendulum with a bunch of links and a heavy mass at the end? Yes they would. The stability of a force-based si...
by h4tt3n
Thu Oct 15, 2009 1:04 pm
Forum: Research and development discussion about Collision Detection and Physics Simulation
Topic: Finite Elements Method: are there proble,s with stability?
Replies: 57
Views: 37662

Re: Finite Elements Method: are there proble,s with stability?

Just out of curiosity, does anything prevent you from implementing a force-based Hooke's law spring and solve for velocity and position with an integration algorithm? I'm using an iterative velocity verlet method which allows me to implement undamped springs with ridiculously high K values without b...
by h4tt3n
Mon Sep 07, 2009 3:31 pm
Forum: Research and development discussion about Collision Detection and Physics Simulation
Topic: Newton Was Wrong ?
Replies: 14
Views: 6835

Re: Newton Was Wrong ?

You're welcome. Try this: float halfdt = 0.5f * dt; float invmass = 1.0f / mass; ... do ... r += v * dt + a * halfdt * dt; v += a * halfdt; a = pState(this, r, v, a) * invmass; v += a * halfdt; ... loop This should be even faster (but since I'm not sure how your pState func works the syntax might no...
by h4tt3n
Sun Sep 06, 2009 10:17 am
Forum: Research and development discussion about Collision Detection and Physics Simulation
Topic: Newton Was Wrong ?
Replies: 14
Views: 6835

Re: Newton Was Wrong ?

Hi Google, Indeed f(n) and f(n+1) are two different forces: f(n) is the force you calculated in the previous simulation loop and f(n+1) is the force you calculate during the current loop. Still, all you need is one force update per loop! This also shows from the wikipedia article, where the new acce...