On Dec 6, 5:44 am, David Richerby
wrote:

Don't bother. Sanny knows nothing about chess and nothing about
ratings. The ratings on his site bear no relation to any other rating
system and Sanny consistently overestimates the strength of his
program.

It would be more accurate to say that Sanny has
consistently claimed high numbers for his program,
and that for a long time he was way, way off. But
not anymore; now the program is a heck of a lot
tougher, and his numbers more difficult to judge.

Let me give an example: I occasionally see
players leave a piece en prise in tournaments,
where a simple Queen check will pick it off; but
now Sanny's program only rarely hangs a piece
in that way, generally misjudging the value of
spite checks and pawns versus a piece. More
often, it seems to now hang onto its men as
well as -- or better than -- most humans.

With no time limit for the human opponent, it's
not easy to say what the rating strength really
might be; let's just say that Sanny's program is
more dangerous an opponent than most humans
I've played lately, even if it still has severe
weaknesses in some areas.


-- help bot