Author: Anthony Cozzie
Date: 07:54:16 01/15/04
Go up one level in this thread
On January 14, 2004 at 17:20:46, Filip Tvrzsky wrote: >On January 14, 2004 at 16:56:20, Anthony Cozzie wrote: > >>On January 14, 2004 at 16:26:42, Ed Trice wrote: >> >>> >>>> >>>>Have you considered trying MTD(f) instead of PVS? I am not sure it is any more >>>>efficient in practice, but it is easier to code, and has the additional benefit >>>>of making you feel different, original, interesting, intelligent, handsome and >>>>attractive. >>>> >>> >>>Well Aske Plaat would love to hear that :) >>> >>>But doesn't MTD(f) trigger a great deal of researches? I remember trying that >>>once and it bloated the tree. >> >>---- opinion mode on ---- >> >>MTD(f) has two big problems. >> >>1, you ponder the wrong move occasionally because your PVs are less accurate. >>If you are pondering the wrong move 20% of the time that is equivalent to a 10% >>time loss. >> >>2, MTD(f) is at its worst when the score is dropping. A fail high in MTD(F) is >>much faster than a fail low (1 child node vs all child nodes). Unfortunately, >>this is when you need your search the most: you are in trouble, and you need to >>make exact moves to win/draw (you might already be lost, but thats just the way >>it goes). I remember some Zappa-Gothmog games where Gothmog had been searching >>8 ply, got in a tight spot, made a 6 ply search, played a huge blunder, and went >>from -1 to -5 the next move. >> >>---- opinion mode off ---- >> >>Most people that try MTD(f) will give up very fast because it requires a >>two-limit hash table rather than a bound hash like most people implement. > >I have noticed this in several postings before but still do not follow: why is >with MTD(f) necessary to store two limits in hash table instead of only one? >Filip > >>Its a >>difference of style, but in my opinion worst case performance is key when for >>search. There is some interesting room for work IMHO with MTD(f)/PVS hybrids. >> >>anthony There is no reason you _have_ to, but the fact is you are much more likely to come across the case. In PVS it is very unlikely you will ever have two bounds: your node types are ALL, CUT, and PV: so you have one of [lowerbound, upperbound, exact]. This is what Hyatt does in Crafty. In MTD(f) you do multiple searches with windows, so you know that a node is "better than -15 but less than -2". Intuitively it seems that throwing out this would hurt you as you enter an oscillating pattern. Of course, my intuition has been wrong before . . . anthony
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.