This week's theme, proposed by me, is — o yes — the final frontier.
The images were generated based on data from Nasa Exoplanet Archive. I only used the position of the planets in the sky in galactic coordinates plus one parameter. The former were plotted in a cartesian map:
I couldn't find a reason for the cluster on the bottom left of the image, but I suspect it is only an especially studied part of the universe and not an abnormally dense one.
The extra parameter would give me a radius for the "fuzziness" around the planets. The image at the top uses the Orbital Period (pl_orbper).
int pl_orbper = constrain(row.getInt("pl_orbper"), 0, 300);
float radius = log(pl_orbper) * 100;Others I tried were the Planet Mass (pl_masse),
float pl_masse = row.getFloat("pl_masse");
float radius = pl_masse/5;and the Planet Radius (pl_rade), each with slightly different results.
float pl_rade = row.getFloat("pl_rade");
float radius = pl_rade * 30;The arithmetic juggling is just a trick to bring the numbers to a usable range.
To generate the "fuzziness" I took one of the aforementioned parameters and used it to create a certain amount of random points around the position of the exoplanet. The larger the parameter (for instance, the larger the planet mass) the more points it would have around it.
positions = new PVector[int(diameter)];
for(int i = 0; i < int(diameter); i++){
float randomR = random(diameter);
float randomA = random(360);
float thisX = cos(radians(i * randomA)) * randomR + x;
float thisY = sin(radians(i * randomA)) * randomR + y;
positions[i] = new PVector(thisX, thisY);
}No points are actually drawn to the screen, but actually curves between the points. The color is generated randomly by cluster, but kept well within boundaries.