Author: Robert Hyatt
Date: 20:06:51 09/10/02
Go up one level in this thread
On September 10, 2002 at 18:29:02, Jesper Antonsson wrote: >On September 10, 2002 at 13:40:54, Robert Hyatt wrote: >>On September 10, 2002 at 09:26:14, Eli Liang wrote: >>>>(7) Dennis Breuker has a paper comparing 5 different transposition table >>>replacement schemes. Has any else been able to validate his results that >>>replacement should be based on the number of nodes searched and not search >>>depth? >> >>Hard to say. The issue will become hash entry size. You can store "depth" in >>a byte. You will need at least 4 (and really more) bytes for nodes searched. >>Your table will therefore be smaller, and _that_ effect was not measured. > >Just a simple thought: You can store the logarithm of nodes searched, i.e. >1024-2047 nodes make you store 10, 4096 => 12 and so on. Perhaps good enough, >and 5 or 6 bits should suffice. > >/Jesper The only downside is quantization errors. IE if you store log2(nodes) then you collapse any value from 2^N to 2^(N+1)-1 into one value...
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.