cs_

4 ms·

1. 1.▲
   by cs_ 8y ago

   It would be interesting to view a skyplot of the GPS satellite trajectories, or at the least output the Dilution of Precision values comparing GPS with GPS+Galileo. This would help to illustrate the effect of additional satellites on measur
2. 2.▲
   by cs_ 8y ago

   A quick test with C++'s <random> library agrees. For n = 100,000,000 points, the rejection method beat out the improved spherical by 7.8%. Using the normal distribution was even slower. XYZ: 16.418057 seconds Gauss: 23.4155
3. 3.▲
   by cs_ 9y ago

   I think that tends to produce a distribution clustered at the poles. At high latitudes, the impact of the longitude is much smaller than it would be around the equator. Mathworld has a nice write up that visualizes the effect and details mo