Author: Bas Hamstra
Date: 16:49:19 10/08/99
Tired of all the experiments, but quite a few facts wiser I wanted to play again
and started to work on the old version. And it starts to perform! Since I have a
Celeron 466, bugfixed some eval code, improved sorting and hashing, *no* human
has won from it, at FICS. Not even a chance. In fact it even defeated CM6000 on
slower hardware. Well, that was a time loss :)
And it plays without opening book, doing that quite well! Of course all that
opening eval slows it down. It does 50 k on average.
It deserves a name. Tao 1.1 :/
- Still the incremental attacks, with 32 bits for every possible attacker
- Incremental movegenerator
- No SEE sorting, but SEE losers pruning
- No piece square tables yet, but that will change, because a lot of what I do
can be stuffed in it
- 400k hash entries
- Null R=2
- No opening book, but reasonable opening eval. Centre control, loss of
castling rights, tempi, it knows its way with it. It even stands the King's
gambit. Occasionally it tries to defend the wrong gambit pawns :)
- Check, One reply, Recap and Mate extensions
Statistics for SuperBooker(C) On for: 2 hrs, 1 min, 25 secs Idle: 0 secs
(playing game 7: MrZetaOfMaine vs. SuperBooker)
rating RD win loss draw total best
Blitz 2030 58.5 403 346 125 874 2431 (29-Apr-98)
Standard 1961 129.1 16 16 2 34
Lightning 2281 110.2 30 15 0 45
Wild 1631 318.5 0 1 0 1
Bughouse ---- 350.0 0 0 0 0
Crazyhouse ---- 350.0 0 0 0 0
Suicide ---- 350.0 0 0 0 0
Address : bas.hamstra@wxs.nl
No revolution, I know, but I feel it is approaching master level now and it is
climbing fast, defeating some 2200 programs. And this is FICS, no ICC, don't
forget.
Regards,
Bas Hamstra.
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.