Author: Uri Blass
Date: 17:27:48 01/18/05
Go up one level in this thread
On January 18, 2005 at 20:18:32, Dann Corbit wrote: >On January 18, 2005 at 20:01:21, Uri Blass wrote: > >>On January 18, 2005 at 19:45:36, Anthony Cozzie wrote: >> >>>Chandler, your statement "chess cannot be solved by computers" is patently >>>WRONG. Zappa on my opteron can solve chess, it would just take a VERY long time >>>to do so. TSCP on a PDA can solve chess, it would just take even longer. >> >>I know nothing about zappa but I am sure that tscp cannot solve chess. >> >>It has limited depth of 32 plies based on my memory so it will never search >>lines that are longer than 32 plies. >> >>I believe that zappa also cannot solve chess. >> >>possible reasons except the reason of maximal depth are: >>1)zappa use null move pruning that is not correct so it may miss some zugzwang. >>2)zappa has some bug that will cause it to crash after 12345678910111213 nodes >>It never search that number of nodes so you never found that bug. >>3)There is no hardware that live forever and you will get an hardware crash >>after 1234 years of search regardless of the hardware that you use. >>4)God decided that this world has only 123456789 years to live and zappa needs >>more time to solve chess so it simply not fast enough. > >What if there is a forced checkmate and it is 40 plies away from the root? > >You cannot know that the universe will end before Zappa finds it, even >pragmatically. I only said :"I believe that zappa cannot solve chess" > >Turning off null move will halve the search depth, but going to proof search >instead of evaluation may double it. So even the loss of null move may not >matter. Besides, even if it is still halved, you just double the time. > >It is a mistake to say that something cannot happen, no matter how absurd. >IMO-YMMV. I did not say that it cannot happen but only that I believe that it cannot happen. > >I believe that many smart people at one time believed that a computer would >never beat a GM. And probably after that, many believed that a computer could >never beat the strongest player in the world. > >We do not know how hard it is to solve chess, except for the worst case. It is >possible to stumble onto a solution tomorrow. Everything is possible but I believe that it will not be solved in the near future and that it will not solved by a program that exist today. Uri
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.