Hi Justin and Artur: Nice to know that I am not the only one doing OT at the end of the year :)
What I had given out in my previous e-mail was just a "hint" to Justin to see if he has any kind of (theta,phi) parameterization in his overall logic to bring about a spherical effect. Just thought I will add a few comments... To Justin: I DID mention that the values of theta and phi can be precomputed...this means that the sin/cos values and their products can also be precomputed and stored. I am not exactly sure what kind of explosive effect you're simulating - probably might be very apparent to particle system experts - but if your particles explode radially something along the lines of what I suggested might bring in a spherical effect; like a Borg sphere getting torpedoed in the outer space of Delta Quadrant. As an aside, if you are interested in generating random and uniform distribution of points on the sphere there is one due to Joseph O'Rourke, IIRC, available in comp.graphics.algorithms FAQ. Each of these locations can be described as r(theta,phi) with theta and phi being constant for each particle throughout their lifetime or be used simply to determine their starting locations. To Artur: I DIDN'T mention anything about the distributions of theta and phi in my e-mail. What I gave out are simple coordinate transformation relations from spherical to Cartesian. There is nothing intrinsic in them which will cause clustering near the poles. It is only when, for example, you choose UNIFORM distributions in theta and phi you will get into the pole problems that you mentioned. Whatever be the distribution, the orientation can be described by (theta, phi) - it is simply the position vector of a point on a unit sphere. Anyways, if you have the stuff working Justin, will check it out sometime. Regards and Happy Holidays Vaidya =========================================================================== To unsubscribe, send email to [EMAIL PROTECTED] and include in the body of the message "signoff JAVA3D-INTEREST". For general help, send email to [EMAIL PROTECTED] and include in the body of the message "help".
