Author: Manuel Leone
Date: 09:38:49 12/08/99
Go up one level in this thread
On December 07, 1999 at 19:50:16, Thorsten Czub wrote: >On December 07, 1999 at 19:18:12, Timothy J. Frohlick wrote: > >>It is an insult to Marty Hirsch to state that MChessPro v.8 was a half-hearted >>attempt to program a computer chess adversary. If I know Hirsch, he is probably >>planning some chess program that has over 10 Gigabytes of chess knowledge in it. >> You will of course need to be running Win2000 or NT and have a 30 Gigabyte hard >>drive on a 1500 Megahertz computer. > >I don't like the way the people talk about martys program here, too. > >Mchess has a lot of knowledge in it. >it is a professional program. >you cannot say: oh - mchess is not anymore state of the art. >it has something unique still in it that was from the very first >moment in it, and was tuned and tuned and brought to its best. > >and - although other programs have (maybe) overtaken mchess8 due to >new (later found) search algorithms or new inventions in search algorithms, >you should better not underestimate mchess. > >we should better always count with marty hirsch. he does not seem to be the >guy you should underestimate. > > >i have come to the conclusion (by testing his program and talking with him and >watching him talk and discuss) >that he has a very strong will and mind and goes his own way. >you should not underestimate a genius mind going its own way. >never. > > >>Those of you who have not played the Hirsch program should consider it. It is a >>DOS program but it kicks butt. Marty put his heart and soul into this program. > >exactly. and that can be felt in almost any mchess version right from the >beginning. > > >>The reason that it plays a bit weaker against computers is the fact that it is a >>slow searcher. Just be brave enough to play it head to head with your own >>noggin and it will scramble your gliae. > >if you give mchess 40/120 time, the advantage of the fast-searchers is not that >big ... than mchess has enough time. > >>Mchess Pro v 8 has received excellent reviews by Komputer Korner and others at >>this site and it really is a good program to have in your collection. > >exactly. >marty hirsch is not dead. many other programmers have shown before that they >are always good for a showdown with the best NEW programmers i the scene. >remember how kittinger after a long time only doing dedicated chess >computers/machines did w-chess and made an amazing tough monster again. > >marty is alive. it may be he is working on other things too in the moment. >but i am sure that he is also from time to time working on his chess program. >maybe he needs time. we should give him the time. >and in the moment he has something he thinks it is a progress, i will buy it. >because i do trust in his judgement, because i have bought his products >before and will do it again. >mchess was really the first intelligent chess programs for ibm-compatible-pc's. >mchess was a new quality in a.i. in computerchess. >hardware progress has made it easy for the fast searchers to compete against >intelligent programs with their dump programs. > >but as last championship (paderborn) has shown (and i know from interviews >stefan meyer kahlen has given) : >it seems that >we have gone over a turning-point. >the intelligent programs have enough hardware-power >to stand the dump fast searchers. >the dump searchers cannot anymore outsearch the intelligent programs. >they have to change their strategy when they want to make wins. >a smart mchess can suddenly be on top of any list. >its just a matter of tuning. >i am sure if marty has help of a team of people, mchess can be the leader >again. just a question of tuning. > Some of you may recall that a few years ago it was suspected that Mchess 5.0 included a secret opening book which was designed to beat Genius 3.0 and move to the top of the SSDF list. If you play these two programs against each other, you will notice that many of Mchess's moves are instantaneous even when has completed all of its moves from the regular opening book. This suggests to me that Mchess got many of its rating points by fraudulent means--trying to increase sales by beating the top program and getting on top of the rating list, instead of actually developing a better program.
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.