# Re: Time problem

```I know the problem I will expose is classical, but I´m a newbie in these
things. I´m coding a computer graphics work wich have a OpenGL doll made
of spirals. This doll must do many movements. My problem is making this
doll to be animated independent of machine clock. I mean, the animation
must play at same speed in any machine.
```
```The best approach is to measure the time at the return from 'swapbuffers'
using gettimeofday() - it's easiest and quite accurate under Linux.

Subtracting the time at the start of the previous frame from the time
now gives you the amount of time the previous frame took to calculate - which
is a REASONABLE estimate for how long this one will take.

Now, you need to know the speed (and perhaps accelleration) for each moving
object in each translational and rotational axis (X,Y,Z,H,P,R).  Now, multiply
the speed (in - say - meters per second or degrees per second) by the time
we expect this frame to take to compute - and add that on to the current
position to give the new position that your models will be in when you've
finished computing and displaying.

The simple Newtonian equations of motion normally suffice:

s = u * t + 0.5 * a * t * t
v = u + a * t

Where:

t -- the time step
s -- the distance moved in time 't'
u -- the velocity at start of frame
v -- the velocity at the start of the NEXT frame
a -- the accelleration throughout the frame

This can be computed separately for each of X, Y and Z - and also works
for heading, pitch and roll angles so long as you are careful to account
for the wrap-around at 360 degrees and you know about the 'gymbal lock'
problem assocoated with animating rotations.

---------------------------- Steve Baker -------------------------