Same for me from Germany.

I wanted to try the MSVC intrinsic implementation of popcount, since its speed looks promising, but the above code seems to present a problem.

return popCount( (bb & -bb) - 1 );

how can you take the negative of a unsigned integer?
What is this code suppose to do?
Does anyone have an ...

Thank you! This is fantastic!
Is the Vice code available somewhere?

Hmm, I would have expected 3668 instead of 1834 dictionary entries since hash(piece,a,b) != hash(piece,b,a) The paper indicates that for ZobristXOR hashing we do have equality of these hash() calls.
In chess there are 7,336 reversible move hashes or half that number if we equal (piece, a, b) to ...

I attach code that generates a 4096 entry collision-less table for the 1834 dictionary entries, in the manner indicated above. It seems to hang on Qh8 for 2048 entries, maybe the linear algebra over F2 demands collisions (other random seeds get as far as Re7 or so, but I've never reached completion ...

Wow, interesting reading and insights in Cuckoo hashing.

I once proposed the cheapo trick with direct move repetition applied two plies before the repetition might be occur, to prune "unmake" of a previous reversible move of the stronger side (alpha > 0), which would allow a refutation of the ...

Even though magic bitboards are very fast my engine still runs very slowly and only gets about 60k nps. I've done some logging and tests and I've concluded that it is my make / unmake methods that are horrible slow, as I anticipated. Currently, I have a Move_t struct type that represents a move, it ...

Another fast approach:
http://www.dcs.bbk.ac.uk/~mark/download ... _final.pdf

Nice reading and insight in congruent modulo arithmetic, but not particular fast approach for each line as the author claim.
https://chessprogramming.wikispaces.com ... +Bitboards

What I'm trying to say is, lets say that I have an attack table for a rook. Is there a certain order that these attacks have to be put in? If I took the array of all the possible attacks, and jumbled up the order, would there still be a magic number that works?
You don't put all attack sets in any ...

The issue with a structure of elements of the same type versus an array is to index the array by variables, i.e. in move generation by side to move and piece type, or one-dimensional with piece-index including both color and type, to utilize the array in a branchless way. See f.i. Beowulf with ...