Author: Ricardo Gibert
Date: 11:02:04 01/20/00
Go up one level in this thread
On January 19, 2000 at 21:01:47, Robert Hyatt wrote: >On January 19, 2000 at 19:24:37, Ricardo Gibert wrote: > >>[snip] >>> >>> >>>OK... change that to "middlegame position". Same result. I don't reach >>>endgame positions very often from middlegame positions. Usually not until >>>around move 40 or so in a real game. That leaves 40 moves to search with >>>no regard to null-move failures at all. And if the program is smart enough to >>>switch null-move off when it is not appropriate, rather than just turning it off >>>at the root, this is a total non-issue... for _any_ position you care to name... >>> >>>So my original statement remains accurate... A bigger tree is _not_ more >>>prone to null-move failures than a smaller one... >> >>Keep changing it. Ignoring what Dan wrote, here is what _you_ wrote: >> >> "what makes you think that in a 10 ply search, where there are N zug positions, >>that in a search space 10 times bigger there are more than 10*N zug positions?" >> >>The assumption _you_ make is there are a non-zero number of "zug positions" >>within a 10 ply search. Obviously, _you_ made the tacit assumption that the >>position in question is one very much like one encountered at around move 40 or >>so. This is _your_ premise. This is why _your_ argument fails. Your argument >>clearly does not consider just any random middlegame position. Most middle game >>positions will not yield a zugzwang position within a 10 ply search. Your >>argument addresses those middlegame positions where zugzwangs _do_ arise within >>a 10 ply search, so the game _is_ necessarily pretty far along in the vast >>majority of such cases. >> >>You keep trying to re-invent the premises to make your argument work. Wouldn't >>it be simpler to admit you made a mistake? Then you could amend your argument so >>it works. Instead you want to create the pretense your argument was fine all >>along. Why do you want to do this? > > >I didn't make a mistake. I ran the following easy to reproduce test. I created >a new variable in the eval, "average material left". I ran some 6 ply searches >on 10 positions chosen from the problem set I have, none of them starting out >with only pawns left. I computed the 'average material' at each endpoint by >just summing the total material on the board into a 64 bit int, and adding one >to a counter. I then did the same, but searched each position to 12 plies. > >The numbers didn't vary significantly. In about 1/2 the positions, the >average material dropped somewhat more with 12 plies than without. In others >it didn't change at all. > >I consider that ample evidence that zug doesn't become a problem. If you would >like to run the test yourself, the crafty source is available. The modification >was obviously trivial to make. And you can answer the question once and for all >rather than continually trying to explain why my (and Bruce's) answers were >wrong. > >I can't do 10 and 20 ply tests. But I can easily (and did) do 6 and 12. >If the average material doesn't go down significantly between the two tests, >then the zug potential does not go up either. > >Feel free to refute this... but please do so with data rather than opinion. >It is easy to produce... You're joking, right? 10 positions? There are about about 10e48 positions in chess. You're pulling my leg right? It would be fascinating to see which 10 you chose. Surely there is more convincing argument to support your conclusion. Why cling to one that is unconvincing?
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.