Author: James B. Shearer
Date: 21:57:29 11/20/98
Go up one level in this thread
Ok, my times were a bit off because of some windows 95 stupidity
involving touching the hash table for the first time. If I analyze another
position first to make sure the hash tables are in memory then I get the
following times. Ply 7 completes at 4.56. qxb2 fails low in ply 8 at 7.77.
But the fail low is not resolved until 73 seconds. Your times are 2.46, 4.95
and 7.95. So 16.1 resolves the fail low much faster (perhaps because you use
48m hash vrs 12m for me). Your time might still be too slow however because
this was a 0 4 bullet game. As noted the other programs I tried reject qxb2 in
under one second. I guess this just shows null move has its drawbacks
particularly for bullet games. It must not hurt too much because the clone was
rated over 2900.
Btw c5 is also bad (losing after bh6). qd6 or qe6 seems to be
required.
James B. Shearer
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.