Using joints for "special" stuff

[PJani]

I am working currently on "realistic" petrol engine, but i came up with few other ideas and before i continue with development of petrol engine i need to redesign some stuff which extends custom joints.
I want to know if its good idea to use Custom Joint interface or doing update in outside callback.
SpringJoint(creating spring between two bodies),
ThrusterJoint(for "pushing" one attached body with some force),
AirLiftJoint(body acting as wing, with some coefs, direction vectors and aabb of body)

Thankyou.

[JernejL]

Whatever you use, the updates / calls always need to be within the forcetorque or customjoint callback intended for it.

[PJani]

Would this be the right code for thrusters, because i am getting some weird effects. Rocket which is tilt for 45 degrees will not change direction to the ground.

the m_locDir is local direction of thruster
the m_locPos is local position of thruster
the m_thrustForce is thrust force.

Code: Select all

   dMatrix mtrx;
   NewtonBodyGetMatrix(GetBody0(),&mtrx[0][0]);

   dVector com;
   NewtonBodyGetCentreOfMass(GetBody0(),&com[0]);

   dVector thrustMuzzleForce (mtrx.RotateVector(m_locDir).Scale (m_thrustForce));
   dVector thrustMuzzleTorque (mtrx.RotateVector(m_locPos - com) * thrustMuzzleForce);
   NewtonBodyAddForce (GetBody0(), &thrustMuzzleForce[0]);
   NewtonBodyAddTorque (GetBody0(), &thrustMuzzleTorque[0]);


And NewtonBodyAddForce is applied at center of mass or at center of object(local 0,0,0)?

[Dave Gravel]

Maybe NewtonBodyAddImpulse is good for this, No ?
Do you have try ?

[PJani]

The NewtonBodyAddImpulse is too "explosive". I want to solve this by forces. Forces are soft vs impulse.

[PJani]

One thing the NewtonBodyAddForce is applied at center of object or at center of mass?

[JernejL]

One thing the NewtonBodyAddForce is applied at center of object or at center of mass?

Code: Select all

// adds force to body at a specified point in global space
procedure AddPointGlobalForce(const body: PNewtonBody; PrevBodyMatrix: pointer; Force, Point: Vector);
var
  GlobalForce: vector;
  bodymatrix:  Tmatrix4F;
  Torque:      vector;
begin

  if PrevBodyMatrix = nil then
    NewtonBodyGetMatrix(body, @bodymatrix)
  else
    move(PrevBodyMatrix^, bodymatrix, sizeof(Tmatrix4F));

  GlobalForce.x := Point.x - bodymatrix[3][0];
  GlobalForce.y := Point.y - bodymatrix[3][1];
  GlobalForce.z := Point.z - bodymatrix[3][2];

  Torque := CrossProduct(GlobalForce, Force);

  NewtonBodyAddForce(body, @Force.x);
  NewtonBodyAddTorque(body, @Torque.x);
end;

// adds force to body at a specified point in local space
procedure AddPointLocalForce(const body: PNewtonBody; Force, Point: Vector);
var
  GlobalForce, GlobalPoint: Vector;
  bodymatrix: Tmatrix4F;
begin
  NewtonBodyGetMatrix(body, @bodymatrix);

  GlobalForce := MatrixRotateVector(Force, bodymatrix);
  GlobalPoint := MatrixTransformVector(Point, bodymatrix);

  AddPointGlobalForce(body, @bodymatrix, GlobalForce, GlobalPoint);
end;

does that make sense? scroll down, there's 2 functions there

I wish those two functions would eventually become part of newton itself some day.. along with GetPointGlobalVelocity and GetPointLocalVelocity.. Julio: would that be possible? i'm sure it would help a lot of people, as adding forces off-body centre is a very intiutive way to perform physics tasks and not everyone will immediately understand how to do it properly (the asociation with addtorque)..

[PJani]

I will rewrite this two functions for c++.

Yep these two function could be great and useful feature in newton, even function call overhead could be reduced this way. So the calculations for global torque and global force can be optimized.

[PJani]

Is mtrx.TransformVector(point); right in AddLocalForce? Is this code right?

Code: Select all

      static void NewtonBodyAddGlobalForce(const NewtonBody* bdy, const dVector& force, const dVector& point)
      {
         dMatrix mtrx;
         NewtonBodyGetMatrix(bdy,&mtrx[0][0]);

         dVector glbForce =  point - mtrx.m_posit;
         dVector glbTorque = glbForce * force;

         NewtonBodyAddForce(bdy, &force[0]);
         NewtonBodyAddTorque(bdy, &glbTorque[0]);
      }
      static void NewtonBodyAddLocalForce(const NewtonBody* bdy, const dVector& force, const dVector& point)
      {
         dMatrix mtrx;
         NewtonBodyGetMatrix(bdy,&mtrx[0][0]);

         dVector glbForce = mtrx.RotateVector(force);
         dVector glbPoint = mtrx.TransformVector(point);

         //NewtonBodyAddGlobalForce(bdy, glbForce, glbPoint);

         dVector newForce =  glbPoint - mtrx.m_posit;
         dVector newTorque = newForce * glbForce;

         NewtonBodyAddForce(bdy, &glbForce[0]);
         NewtonBodyAddTorque(bdy, &newTorque[0]);
      
      }


[JernejL]

I have no idea how you implemented it, but here's my funcs to do that:

Code: Select all


function MatrixRotateVector(vec: Vector; m: Tmatrix4f): Vector; // tested, works.
begin
  result.x:= vec.x * m[0, 0] + vec.y * m[1, 0] + vec.z * m[2, 0];
  result.y:= vec.x * m[0, 1] + vec.y * m[1, 1] + vec.z * m[2, 1];
  result.z:= vec.x * m[0, 2] + vec.y * m[1, 2] + vec.z * m[2, 2];
end;

function MatrixTransformVector(vec: Vector; m: Tmatrix4f): Vector;
begin

   Result.x := vec.x * m[0][0] + vec.y * m[1][0] + vec.z * m[2][0] + m[3][0];
   Result.y := vec.x * m[0][1] + vec.y * m[1][1] + vec.z * m[2][1] + m[3][1];
   Result.z := vec.x * m[0][2] + vec.y * m[1][2] + vec.z * m[2][2] + m[3][2];
end;



[PJani]

I am using Julios dMath lib, because i am adding new joints in jointlibrary.

[PJani]

buahahaha, its Alive, its Alive! Joke, its all most nothing

[Julio Jerez]

Hey that was cool.