Introducing the *.EBF project

[Florentino]

I just had a look at the EBF probing code, which I downloaded from http://rebel13.nl/misc/ebf/index.html. If I understand it right the positions are stored with a 64 bit hash key, but without taking en-passant possibilities and castling rights into account. Isn't this a problem or do I miss something (a move starting a sacrificial attack in the centre might have a good score, because the engine analysed a position where there was no more castling possible, but if you are looking at the same position with castling rights the move might simply lose).
Of course the situations where you have this ambiguity are rare, but why not add castling rights and e.p. square to the hash key (like in polyglot)?

[Rebel]

I am aware of the glitch, need to fix it. Thanks for the reminder.