New Lagged Fib. PRNG gen and random2.d

[Joseph Rushton Wakeling]

On 26/08/13 11:35, monarch_dodra wrote:
> On Monday, 26 August 2013 at 09:24:57 UTC, monarch_dodra wrote:
>> That's my biased review/comment on the numbers. I'll post them here, so you
>> can interpret yourself.
>
> Units are ms. 1_000_000_000 iterations

Cool -- so, you're getting faster performance than Mersenne Twister for 
higher-precision/wider range numbers.

That's not the same as statistical quality of course.  There are some general 
concerns about Lagged Fibonacci -- see:
https://en.wikipedia.org/wiki/Lagged_Fibonacci_generator

I really must implement Mt19937_64 for comparison, and get on with creating 
something that tests D random numbers against Dieharder.