Author: Robert Hyatt
Date: 19:29:37 09/17/99
Go up one level in this thread
On September 17, 1999 at 17:56:56, Ralf Elvsén wrote: >On September 17, 1999 at 16:21:40, Robert Hyatt wrote: > >>On September 17, 1999 at 15:44:00, Ralf Elvsén wrote: >> >> >>I would make you the same offer I would make to anyone else... come over >>to ICC and play 'crafty'. You can pick the time control, and see how you >>do. You are playing crafty using a non-crafty book. Which is ok, but it >>isn't 'crafty', any more than you should play hiarcs after forcing it to >>play moves from a crafty book. >> >>However, there is an IM on ICC that _only_ plays to draw, generally, and he >>has very poor luck. Since your rating is so much lower than his, I really >>don't think you can pull this off there. But I am open to being proven >>wrong if you have the time and interest... >> >>Bob >> > >Relax Bob :) > >Ok, what I said was to the effect that "On MY hardware THESE versions >of THESE programs with THIS book APPEARS to play not much higher >than 2000 (if at all)." > >I did NOT say: "I can get a high percentage of draws vs >e.g. Crafty on its present hardware." > >I don't know, maybe there is a jump between my hardware and yours >that makes me "lose my grip" ? I tell you, I really have to >concentrate hard and always have the feeling I might "fall over >the edge". Maybe the edge is much closer vs your hardware? >(Hmm, maybe Scrappy would be fair?) scrappy is only about 1/2 the speed, which is not a huge difference, although it is a bit weaker.. but it can be a real handful... > >Also, as you point out, perhaps Craftys book would suite me worse. >In the second draw vs Crafty "the book" traded queens early, and maybe Crafty >said "who traded my queen?". I also "know" Hiarcs book, I know almost nothing >about Craftys preferences. I am too lazy creating a book for the Winboard >versions... > >I don't have an ICC-account, I would have to pay $$ to play many hours >*because* if I decided to play I want to play several games and at the same >time control (40/120 + 20/60 + game/60). (Sorry, I forgot to include >the exact time control.) I mean, if I play two games at say >game/60, which I would feel very uncomfortable with, and lost both >that would certainly prove nothing to me but a lot to other people. >If you care to look at the pgn:s I sometimes think over 20 minutes >for a move, which I sometime have to do. At game/25 I very rarely draw. you can log on as a guest and play... the games have to be unrated, of course, but I play many people like this, including a couple of GMs that want to remain unknown when they fiddle with some odd opening they want to work on... > >Btw, how could you tell I wasn't cheating on ICC? I am a pretty experienced >chess software user. I bet you would have a hard time telling whether I >was running e.g. Crafty in analysis mode over here. Since I know I can >play decent anti-computer chess, with a blunder checker I could be hard to beat >for any computer. If this is a credibility issue (is it?), that wouldn't solve >much... > I can tell. :) If you make it thru enough games, there are many 'signatures' that using a computer exhibits... >I don't think I will accept your offer (although I am not categorical >on this point), simply because I have nothing to prove. >I asked about opinions on the quality of the programs chess, that is all. >Feel free to do so, if you can find the time. > >Ok, ok, I admit I was provocative, I could have excluded the >rating thing, sorry ... :) > >Once again, I didn't say that Crafty (or Hiarcs) sucks. I just showed >some games and gave my opinion. You, if anyone, should know >that claims like these aren't taken seriously by people with >a minimum of discrimination. Should that stop me from >showing these games? Maybe :) I didn't take it that way either. I am simply _always_ interested in someone that claims to be able to work around something I have tried hard to prevent from happening. My code to prevent locked positions is not perfect. In particular, while choosing book lines, it doesn't know what it is doing until it runs out of the book (something a GM friend of mine is on my case about regularly). > >Let me rephrase what I (should have) meant: If you (I mean anyone) believe >these games are genuine, maybe you should reconsider your opinion on the >programs performance vs low-rated players *as inplemented in the >ChessBase interface, with this book, on this hardware, on this time >control, etc AND if the players have a good knowledge about the programs* . >Implications for other circumstances are drawn at your own risk! > >If people don't believe in the games, I couldn't care less... >I'll just be happy if they don't bore me with their snide remarks... > >Regards, Ralf > >PS. I consider myself to be a "Crafty-friend". >Thanks for an (in most cases) excellent program :) Hope you didn't think I was responding in an obnoxious way. I am always interested in real examples of where something isn't done very well. I spend enough time working on this sort of problem that I want to make sure it is working as well as I can make it work...
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.