Very easy to understand. Random numbers do not guarantee that when you XOR a few a few times with eachother, that not the same hashes appear. This is basic Zobrist hashing knowledge. No serious Chess programmer uses "just random", they optimize the heck out of their random's using methods that Arthur and Erik already explained.That's misleading:) You claim that although PRNG may be very good at producing random numbers, the latter are no good for Zobrist hashing. I'm not convinced:(
I have never said that hand-picked 60-bit values outperform 64-bit random values.I've never seen any evidence or discussion of this. I certainly cannot imagine that specially chosen 60-bit numbers are better than randomly chosen 64-bit values. So I see no basis for your claim above.