shur, I''m working on your problem and I think these are only very distantly related. Sorry to dissapoint you d9930380, but I don''t want to touch this problem with a 39.5 foot pole.

George D. Filiotis
Are you in support of the ban of Dihydrogen Monoxide? You should be!

What about using bezier curves? The start and end of your bridge can be the start and end point of a bezier curve and then for each person on the bridge, add a bezier point. By default, there''d be two end points and one in the middle. As soon as someone stepped onto the bridge, move the middle bezier point to the location of the player. You''d still need some maths algo''s in there to calculate height of the point and its control point movements etc. You could even wave oscilate the control points to simulate rope tension being released. I''m no mathematician and I''ve no idea how well or if it''d work but just an alternate idea. Beziers are pretty fast and easy.
Jules.

If you check out Advanced Character Physics (here), you could use this to solve the system.

It is an approximation (aren''t they all ) but I''ve found it to give some excellent looking results.

Just model the ropes as a string of particles joined by rods (or very stiff springs) and then planks by rods/tetrahedra joining them. The nice thing is that even without the full simulation you can run the constraints to achieve the equilibrium.

What is it with you people and equilibrium? I think his first idea was good. Rope bridges sway and wiggle alot when you''re walking on them, and if you want it not to very much, just apply a dampening force proportional to the velocity.

I think in some of the posts above, especially Symphonics, he was using fancy words and he didn''t even know what he was talking about.

Also, the problem with having no elasicity is that it creates infinite acceleration when you try to solve it analytically.. think about it.

Anyways you''re on the right track there..

quote:

---

Also, the problem with having no elasicity is that it creates infinite acceleration when you try to solve it analytically.. think about it.

---

The great thing about the Verlet system is that you don''t solve it analytically so no problems with infinite acceleration. What it does require is many iterations.

Ability to achieve equilibrium is good, especially when dealing with rods or stiff springs because it means your simulation is not going to blow up (or less likely).

His initial idea was indeed a good model, however it all depends on what purpose the model is going to be put to (which was never mentioned).

Really a rope bridge only has three primary behaviors.

First it sags SOMEWHAT beneath a load at the point of contact.

Second it sways laterally when a load is applied.

Third it bobs up when a load is removed, which can result in a vertical oscillation as well.

So bottom line create a system that handles these in a manner that looks good and be done with it. None of these three behaviors should be difficult or should require complicated physics applications. After all a bridge sags in an arch shape which I''m sure you could find a simple enough function to model...how about the cosine function? Then treat the entire structure like a simple rope attached at both ends. Whichever section attaches to the plank that is stepped on will be pushed down...as you go out on either side push them down less according to a function (again just figure something out that looks nice). When it is removed make it pop back up, but maybe introduce a harmonic oscillation in there. As for the lateral sway, that''s just a lateral wave.

So the SECOND bottom line ( ) is this: Pretend like your bridge is a plain old rope. Then introduce vertical deformation and possible waves due to a load, and lateral waves due to a load. Make it look good and use physics if you want, but if you get into something much more complicated than sine and cosine functions you''re probably making it too complicated.

He should be able to exactly model the "sag" of the bridge with a caternary, but I think that trying to approximate the physics of a bridge with trigonometric function would probably be more trouble than just modelling the elastic connections.

------------------------------------------------
The wind shear alone from a pink golfball can take the head off a 90-pound midget from 300 yards.

-Six String Samurai