Author: Robert Hyatt
Date: 07:53:36 12/13/99
Go up one level in this thread
On December 13, 1999 at 10:28:54, Robert Hyatt wrote: >On December 13, 1999 at 00:44:53, Eelco de Groot wrote: > >> >>On December 12, 1999 at 10:19:26, Robert Hyatt wrote: >> >>>On December 12, 1999 at 02:26:26, Ed Schröder wrote: >>> >>>>>Posted by Robert Hyatt on December 11, 1999 at 21:08:12: >>>>> >>>>>Comp vs Comp will say nothing about how comp vs human goes. IE for an example, >>>>>Tiger 12 looks _very_ strong vs computers, but so-so against humans. I have >>>>>not yet studied its games very carefully, although I now have a couple of dozen >>>>>games vs Crafty on ICC and FICS. It seems to be perfectly tuned to beat >>>>>computers... it seems very materialistic and ready to accept any gambit offered, >>>>>and they try to make the opponent justify it accurately. >>>> >>>>>How it is going to do once it is out 'en masse' will be very interesting to >>>>>watch. >>>> >>>>Same here. >>>> >>>> >>>>>But it clearly isn't doing _nearly_ as well vs humans (even with anti-human on) >>>>>as it is doing against other programs... >>>> >>>>Do you have some game examples that supports your strong judgement? >>>> >>>>Ed >>>> >>>> >>>>>Which is completely not surprising. I said several years ago that to attempt to >>>>>write a program to blast to the top of the SSDF is a _totally_ different thing >>>>>from trying to write a program to blast to the top of the FIDE rating list. >>>>> >>>>>The games are too different... >>> >>> >>>Somebody else already posted a really bad result vs a humaon on FICS (winning >>>1/3, losing 2/3, against a player that isn't a "master" of anything but anti- >>>computer chess. I have watched "other" players (not often as I don't watch >>>very often, except when crafty/scrappy is idle) also cause problems... This >>>is the most striking example of comp-vs-comp strength being _far_ different than >>>comp-vs-human strength that I recall in recent years... >>> >>>But as I mentioned before, remember that "I am 10 years behind the commercial >>>programs". I don't see any reason to point out the weaknesses of someone that >>>is 10 years ahead of me, wouldn't you agree? But, in fact, the problems are >>>very obvious, so my analysis isn't needed anyway... >>> >>>Fixing the problems is going to adversely affect its currently great >>>anti-computer style of play, however... >> >>If Chess Tiger displays a great anti-computer style of play I think that is >>completely by accident, Robert. Christophe has stated on more than one occasion >>that he doesn't use games against computers at all for all the important parts >>of his testing. Maybe there are still holes in its positional play as you say >>but what program is without them? As Tiger is only tested on the severs now to >>see if it can run for a prolonged period of time unattended, - I understand >>sometimes the same machine is used for debugging too while still logged in? -, I >>don't think we can form a clear picture of its play against strong human >>opposition yet. >> >>Regards, Eelco > > > >I'm not psychic so I have no idea about how it was designed. I only know that >it is playing on ICC and FICS (not by the same person that started this thread, >I assume, since he said he isn't running on ICC. Yet both use the >'chesspartner' handle and claim to be tiger 12e or something similar). You can >certainly watch it play. > >Yes, _all_ programs have holes. Some are more tactical. Others appear to be >more strategic. It just appears (to me) that it is doing much better against >computers than against humans. And I am not paying attention to the games it is >losing on time. There are certain "warning signs" that you become familiar with >after living on ICC a few years. A program can do something and get away with >it, but you _know_ that against the strong IM/GM players there, doing that >"something" is going to ask for trouble. > >Best thing is to simply wait. If we wait for a year, and then look at how >many tigers, vs how many rebels, vs how many fritzs, vs etc (ignoring the >special case of crafty/gnuchess with automatic interfaces) there are on the >servers, you can probably conclude which program plays best against the strong >humans... there will simply be more of that particular program registered on >the server. IE log on, do =computer, then finger them one by one to see which >program they run. You might be surprised. :) (and again, ignore crafty users >since the auto-interface attracts many of them). That was an error. Right as I hit <submit> I realized that tiger 12 is 'chesspartner' on FICS. It is running under another handle on ICC, but for the life of me I can't recall it since I haven't watched it much in either place, except when it played Crafty or Shutka or one other player. It plays lots of games against 1200-1700 players and I don't watch those at all.
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.