OSDev.org

My T key doesn't do anything if I press it on the bottom left side.. There must be something stuck under it.

Anyway.. I had my last day of college until February next year.

I had a 3D maths exam. It was broken into vectors, matrices, quaternions, physics, and space partitioning.

One of the questions asked us to make a 35* Y rotation quaternion, and I forgot how to, but I remembered how to make a rotation matrix and how to convert that in a quaternion so I did it the hard way

And the physics was just to do with movement (and to think I studied collision ). I screwed up one of the questions, since it gave info like the current position, the last position, the speed of acceleration (in per seconds), the number of updates per second, and we had to calculate where he object was after 2 updates. There's a quick and simple Verlet intergration formula to work it out, but at the time it didn't come into my head so I made up my own formula. It might not be how we were taught, or as optimised as verlet, but I hope my answer is still right (I took all the variables into account).

And I had a haircut today. They cut about half the length of my hair off. It always feels weird after having a haircut because you have to get use to less hair.

Collision physics annoys me. I have made a small physics simulator but it gets stuck when simulating newton's balls - when >2 collisions occur simulaneously it starts to die a painful death .

I use euler integration myself, but that's only because I haven't found a formula to apply linear verlet integration to rotational physics.

How do you do it?

JamesM wrote:it gets stuck when simulating newton's balls

[triestostiflelaugh]

Honestly, how childish! While writing that I made a mental note to check if anyone chortled about it. You are teh w!n!!!

JamesM wrote:Collision physics annoys me. I have made a small physics simulator but it gets stuck when simulating newton's balls - when >2 collisions occur simulaneously it starts to die a painful death .

I use euler integration myself, but that's only because I haven't found a formula to apply linear verlet integration to rotational physics.

How do you do it?

Have you tried using quaternions to represent rotations?

Then try:

Angular Velocity * delta time = ∆Θ(delta time) * axis of rotation

momentum of rotation = ((Axis of Rotation) cross product with(collision normal)) * Angular Velocity * radius it is spinning around * mass

Hmm, interesting! I must admit the only thing I have ever used quaternions for is SLERP (spherical linear interpolation) and then only under constant scrutiny of several tutorials! I may give them a go.

My head hurts...

On that note im starting a math and a english class in the new year, so hope fully my head wont hurt as much when i start at the prestart curse for DTU "danish technologic universite" (i really need that english class) after the summer vecation.

JamesM wrote:Honestly, how childish!

I've finally got all of trimester 3 grades back: That's 2 high distinctions, 1 distinction, and 1 pass.

A pass?!? That lecturer So I'm a great programmer and crappy level designer?