Author: Robert Hyatt
Date: 07:54:41 10/05/00
Go up one level in this thread
On October 05, 2000 at 00:46:08, Chessfun wrote: >On October 04, 2000 at 16:27:26, Robert Hyatt wrote: > >>On October 04, 2000 at 10:24:13, Chessfun wrote: >> >>>On October 04, 2000 at 10:14:49, Robert Hyatt wrote: >>> >>>>On October 04, 2000 at 00:35:34, Timothy J. Frohlick wrote: >>>> >>>>>On October 03, 2000 at 20:46:43, Robert Hyatt wrote: >>>>> >>>>>>Err.. exactly _what_ program won't kick a GM at blitz? I have seen GNU do it. >>>>>>I have seen everybody else do it too... >>>>>> >>>>>>This is hardly a distinguishing event. >>>>> >>>>>Dr. Hyatt, >>>>> >>>>>You are correct. One game does not mean much. Gambit Tiger made it look so >>>>>easy though and Mecking made some decent moves. We will see how good this >>>>>program is within the next six months. >>>>> >>>>>The audience that will be buying these type of programs will be most amused by >>>>>the style of play of Gambit Tiger. I am certain that you would enjoy it too. >>>>> >>>>> >>>>>Tim Frohlick >>>> >>>>My impression after watching hundreds of games between the new beta versions >>>>and Crafty is that it is very solid, has filled a couple of horrible endgame >>>>holes in the previous version, but I have not noticed any tendency to wildly >>>>attack at all. At least against Crafty on ICC and chess.net. >>> >>> >>>The Gambit program has only been out for 2 days. >>>Most you probably have so far were against 12.92 Beta. >>>As far as I'm aware there are two or three accounts at >>>ICC now running Gambit 095. And only one at Chessnet probably >>>cchess. >>> >>>Sarah. >> >> >>I have seen several 'cchess' games this week. Nothing caught my eye. > > >You had originally said after watching hundreds of games. >Now you have seen several games....but you still don't know >if the operator was running Gambit 095. I said I had played hundreds of games vs tiger. That is true. And it can easily be verified. Someone said that since gambit tiger had been released, cchess on chess.net was running it. I said I had seen several games this week vs cchess, which I can certainly publish if need be. I don't _know_ anything about what he is running. Someone else said he was running the new gambit tiger. I assume that when someone makes a direct statement like the one posted here, then the statement is true. > >I myself honestly doubt that Crafty has played hundreds of games >against the Tiger beta's, that includes the three main servers. Feel free to doubt whatever you want. That does _not_ make anything a fact. Ask any of the tiger operators on ICC. >Probably 1/2 the computer accounts are noplayed anyway, then you >have a formula that don't exactly allow computers to easily play. This shows your ignorance. Ask one of the ICC admins how large the noplay list is. Then I'll be happy to show you Crafty's version, of which over 90% are crafty clones added by xboard's ZIPPYNOPLAYCRAFTY option or the equivalent in my interface. Please don't make statements where you know exactly _nothing_ about the topic you are talking about. > >"((computer & rating>myrating-200 & inc>2 & !bullet) | (!computer & >rating>myrating-400)) & (wild=0 | wild=2 | wild=3 | wild=4 | wild=5 | wild=8) & >!freeweek & autocolor & established & rated & (bullet | (blitz & time<=5 & >inc<4) | (standard & time<61 & inc<31))" How does that prevent computers? It plays computers so long as the inc is > 2, and their rating is within 200 points of crafty's. That is restrictive? The formula is simple: standard and time < 60 minutes and inc < 31 seconds. blitz and time <= 5 minutes and inc < 4 seconds. any bullet time control. Only restriction on computers is inc of at least 3 seconds and the rating restriction. That is "restrictive"??? > >You would think using a; >Intel SC450NX quad xeon/550mhz, 512mb RAM, 5X9 10K LVDS disks >that Crafty should easily "smash" these others running on PIII 500's, >so why the rating tolerance of 200. I know more time to "smash" Old Gm's. Study some statistics for a bit. Crafty is typically rated 3000+. I don't expect to win every game even with a program on a P5/200 machine. The statistics don't support that. My formula simply handles the case of where a program has been running on a P5/200 and has a rating of 2400, and then he goes off and buys a new PIII/900. That causes rating distortion and I choose to not participate. If he gets his rating into a reasonable range, I will play. > >Yet there are those constant seeks of 5 3 Crafty sends out to play these >GM's. And what is interesting about smashing Shirov over and over on ICC? >Yea, yea I know heard it before. > >Sarah. Crafty seeks 5 3 because _most_ GMs want to play that time control. All you have to do is ask them. You might have heard lots of things before. It would help if you remembered them..
This page took 0.01 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.