When making a bloomfilter for the x points of scalars 1...2**30 , what is the best hashing algorithm to use to get the lowest false positives candidates possible ?
The X coordinate itself is a good source of randomness, you don't need additional hash functions.
h1 = hash & 2**32 - 1
h2 = hash >> 32 & 2**32 - 1
h3 = hash >> 64 & 2**32 - 1
You will get 3 32-bit numbers.
Adjust the percentage of false positives by the size of the filter and the number of bits per record (hash functions).