Quote from: StatsMan on August 24, 2015, 08:02:57 PM

In regards to the .5 BTC - I did state you refunded that amount, I was eluding to it happening 2 more times last night and the second time it happened it was on a .7 BTC wager.

About the MT comment - instead of writing it all out for you I will quote TrevorXavier from reddit.

"When the second round of shuffling comes into play with Mersenne Twister ("MT"), I am genuinely concerned about the poor distribution quality due to frequent reseeding. MT is known to suffer when poorly initiated (see the original paper), and randomizing a deck first with a stronger random number generator before MT does not imply it will be immune to a poor reshuffle, as MT may not exhibit uniform distribution across multiple reseeding operations. Many statistical tests will be needed to confirm this from a fair-game standpoint, so at this point I only consider it a "concern" and not a proof.
So, you make a wonderful point. It may actually be better to use a primitive for the "cutting" of the deck before dealing, rather than an algorithm. As libertaad stated in another comment, it looks like he intends to push bitZino away from MT and onto something different.
Thank you again for your excellent comment. Very good points made."

FYI - libertaad was the guy that came up with the system used by CoinRoyale and was the owner of BitZino. Liberated also stated that the MT was going to be changed because of the limitations t has.

The points above are exactly what we stated prior to finding the write up.

We know who libertaad is

We understand the limitation of MT as well that's why our dealer shuffle doesn't just use MT and Fisher-Yates, it also uses the Java RNG and therefore the process as a whole has enough randomness.