About Populate3D, what does Seed mean?

Hi All
I am confused with the “Seed” of Populate2D/3D, I don’t know how it work.
For example, I randomly put 5 points in a cube, so the “Count” = 5.
If I change the “Seed”, I will get some different distributions.

And I’d like to discuss one thing, If I want to get evenly radom points, what do you have a good idea?
Any reply is welcome.

Thanks in advance.

seed is basically exactly that, it allows you to set a different starting point for the randomisation.
evenly distributed points are often achieved with some sort of packing e.g circle packing which are a bunch around on the forum if you use the search function. but there are also other methods to choose your distribution. i am not fully sure but i think it is some sort of noise distribution working in populate.

The pseudo-random algorithm that we normally use has this Seed input which is like the initial value that determines the random result, so if you re-compute that algorithm with the same Seed, you get the same random values, if changed, you get another random values.

So to create random points, you first create a random object using the Seed. Then you use this randomizer to get random values for each point coordinate. This gives you a very iregular distribution, different from what Populate returns. So what it does is at each new random point, it evaluates if it is too close to the rest of the points already included, and if it is, it discards it and tries other random coordinates, looking for a uniform density.

It’s that basic from the outside, and there are no parameters to vary its distribution in GH1. However, you could use a similar algorithm to iteratively add a random point to the collection and choose its distribution, or you could do it in other ways such as creating extra points and discarding the ones you don’t want.