https://chessprogramming.wikispaces.com ... tion+TableWhile early implementations and programs relying on root pre-processing to guide search and evaluation were obligated to clear the hash table between root positions, most todays programs do not, to profit from entries of previous searches. Nevertheless, to avoid persistence of old entries which may no longer occur from the current root, aging is used to likely replace those entries by new ones, even if their draft and flags would otherwise protect them.
This is helpful for describing a theoretical purpose of transposition age, but is vague on implementation. My current empirical tests (without using any transposition age information) show that erasing the hash table between iterations will slightly weaken overall play. Another observation is that erasing everything but the hash best move between iterations produces the same results as not erasing anything. That is, erasing the hash scores between iteration makes no difference. This indicates the most important aspect of a hash table is to keep a best move to search first, and not to hold too strict to a value bound.
How does aging differ from hash depth?
How many bits should be allocated for a time stamp in the hash save?
What happens if the transposition age is not saved, and the hash table is not erased between root moves?
How is transposition age affected by threading, or is threading duplication the only reason for transposition age?
Will lazy SMP make a new adjustment to aging?