Abstract
This note completes the comparison of the performances of seven replacement schemes. The performances are presented as functions of the transposition-table size. Some 200 chess middle-game and endgame positions have been studied. It turns out that the number of nodes of a subtree is a better estimate for potential savings than the depth of a subtree. A two-level table, using the number of nodes in the subtree searched as the deciding criterion, performs best and is recommended. Previous results based on fewer experiments are confirmed.
Get full access to this article
View all access options for this article.