In supplement to being an gathering tongue programmer, I am additionally an electronics engineer/technician (with likely "similar flat of experience", or equal more), and therefore, alongside cognition of the two areas, I can think of hundreds of distinct ways of being capable to build item that generates random numbers, and of elevated norm (for example, according to pale noise generators, which are the thermal noise of semiconductors, try to foretell this alongside supercomputers). Therefore, I don't think this is by far the most difficult part to do on an x86-64 processor, for the Engineering elite, current in the Intel and AMD teams, for example.
Furthermore, I must say that I have not yet received any additional feedback from AMD concerning the case. But they have already escalated my case internally alongside them, so I am motionless current waiting for a exact update from them, to portion it here. And, since 9 days before posting here, until the current moment, not a sole 0 came out of the 2 AMDs that I had admission to!
With the new update I made to my application to test this, it is now additionally imaginable to benchmark the random figure generator, and it is additionally now imaginable to run the test on processors that do not have the 'rdseed' instruction. As for the benchmark, it is amazing how distinct the results are between distinct processors (e.g.: a Core i5 from 2012 generates 12.6 M numbers per second, during a Core i7 7700 from 2017 generates "only" 750 k numbers per second, and my AMD Ryzen 7 about 2.6 M numbers per second, and the AMD generator being entirely inert to clock variation, and the Core i5 from 2012 100% reliant on the clock, as it seemed to me in the tests I did). Which leads me to acknowledge that the random figure generator circuit must have ample differences, and likely entirely distinct versions/revisions, equal for models from the identical manufacturer.
So far, I additionally have not established decent data on how their RNG component works.
