Author: Bas Hamstra
Date: 05:27:11 09/15/99
Go up one level in this thread
On September 15, 1999 at 07:49:45, Alessandro Damiani wrote: >On September 15, 1999 at 05:36:27, Bas Hamstra wrote: > >>On September 15, 1999 at 03:59:18, stefan wrote: >> >>>What is the mystery of fast programs like Fritz? >> >>My guess: it generates captures 1 by 1 (so not de whole batch) and does very >>limited sorting. Very carefully optimized assemblercode. Waste no ticks. Hardly >>any endnode eval (only pawns/kingsafety). >> >>I hear Junior is even faster than Fritz, nowadays, nps wise. >> >>Does Fritz throw out SEE wise losing captures? I don't know. Probably not. If >>so, and you would factor in the nodes it does not SEE prune it's NPS would be >>lower. But still pretty good. >> >> >> >>Regards, >>Bas Hamstra. > >If you generate captures one by one then you are not able to sort them by a SEE. Correct. But that's not what I said :) You can generate captures one by one, NOT sort SEE, but still throw out SEE wise losers. (BTW you can do SEE faster than normal if its only for pruning losers) Regards, Bas Hamstra. 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.