Author: Uri Blass
Date: 13:57:13 05/30/03
Go up one level in this thread
On May 30, 2003 at 16:30:18, Robert Hyatt wrote: >On May 30, 2003 at 07:55:30, Jeremiah Penery wrote: > >>On May 30, 2003 at 03:45:13, Uri Blass wrote: >> >>>On May 29, 2003 at 22:47:57, Jeremiah Penery wrote: >>> >>>>Little Goliath and Genius (IIRC) would show 0.00 score also, but it doesn't mean >>>>they knew what they were talking about, any more than Fritz. >>> >>>Genius? >>> >>>I remember that Genius3 at the time of the tournament could not find Qe3. >>> >>>I do not know if Little Goliath shows 0.00 there but I know that it shows 0.00 >>>when there is some checks with no draw. >> >>I think Genius 5 showed some strange drawing line or something. I could be >>wrong. > > >I was up most of the night flipping between ICC and FICS where these >discussions were going on. Not a single computer showed a draw for Qe3 at >any point during the night. I haven't seen a _correct_ draw score for that >move yet. Little Goliath might well produce 0.0 but it is highly unreliable >because it considers N checks in a row to be a draw, which is often wrong. Yace showed a draw score in this discussion and yace does not have the goliath problem. I can understand changing the evaluation to consider N checks in a row as more close to draw but considering it as a draw is simply bad decision and it seems that this kind of decision is not done in order to play better but in order to try to impress other people. Uri
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.