Author: Mark Young
Date: 22:25:26 01/11/00
Go up one level in this thread
On January 12, 2000 at 00:21:05, Robert Hyatt wrote: >On January 11, 2000 at 20:32:17, Mark Young wrote: > >>On January 11, 2000 at 04:35:56, blass uri wrote: >> >>>On January 10, 2000 at 18:34:41, Mark Young wrote: >>> >>>>On January 10, 2000 at 11:44:03, James Robertson wrote: >>>> >>>>>On January 10, 2000 at 00:56:19, Mark Young wrote: >>>>> >>>>>>On January 10, 2000 at 00:20:31, James Robertson wrote: >>>>>> >>>>>>>On January 10, 2000 at 00:15:09, Mark Young wrote: >>>>>>> >>>>>>>>On January 10, 2000 at 00:01:34, Havergal Brian wrote: >>>>>>>> >>>>>>>>>On January 09, 2000 at 23:40:42, Mark Young wrote: >>>>>>>>> >>>>>>>>>>On January 09, 2000 at 23:00:44, Robert Hyatt wrote: >>>>>>>>>> >>>>>>>>>>>On January 09, 2000 at 22:13:03, Marc Plum wrote: >>>>>>>>>>> >>>>>>>>>>>>On January 09, 2000 at 17:43:32, Robert Hyatt wrote: >>>>>>>>>>>> >>>>>>>>>>>>(snips) >>>>>>>>>>>> >>>>>>>>>>>>>Kasparov is in ChessBase's pocket. However, from this point forward, since >>>>>>>>>>>>>we are going to continue to see this, I believe that I will simply choose to >>>>>>>>>>>>>say that "no more crafty versions will be available for ChessBase products". >>>>>>>>>>>>> >>>>>>>>>>>>>If it is so bad, they really don't need it anyway. If I see any future versions >>>>>>>>>>>>>on their web site, I will let my attorney do the talking. >>>>>>>>>>>> >>>>>>>>>>>>Obviously, you are entitled to do whatever you want with your program. I >>>>>>>>>>>>appreciate your making this excellent program available for free. It's nicer, >>>>>>>>>>>>IMHO, to use it in the ChessBase playing interfaces than within Winboard, but >>>>>>>>>>>>thanks for making it available as long as you have. >>>>>>>>>>>> >>>>>>>>>>>>That being said, I'm not sure why you seem to be blaming ChessBase for what this >>>>>>>>>>>>Dutch amateur said on Kasparov's web page. Is there more information you could >>>>>>>>>>>>share that makes this clearer? >>>>>>>>>>>> >>>>>>>>>>>>Marc Plum >>>>>>>>>>> >>>>>>>>>>> >>>>>>>>>>>No... And I am much less concerned now. I had been sent a copy of excerpts >>>>>>>>>>>that seemed to imply that it was a chessbase or kasparov-like article. Now I >>>>>>>>>>>see the actual author, and don't really care what his opinion is. If you read >>>>>>>>>>>the article slowly and carefully, it looks idiotic anyway. I'd have to have >>>>>>>>>>>my name on the 'byline'... >>>>>>>>>> >>>>>>>>>>Glad to hear it, I was really confused how you were coming to some of your >>>>>>>>>>conclusions in your first post. I did not find the article anti-crafty, but you >>>>>>>>>>are correct it is a poorly written article with little value pro or con about >>>>>>>>>>any chess program. >>>>>>>>>> >>>>>>>>>>I would have thought you would have read the original article before jumping to >>>>>>>>>>conclusions...but mistakes happen. >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>>I am amazed that you did not find the article to be "anti-crafty". How about >>>>>>>>>reading the article again and concentrate on the top few paragraphs. >>>>>>>> >>>>>>>>I read the article, it has a paragraph giving the author's observation of crafty >>>>>>>>16.6. His observations are accurate in regards to Crafty 16.6 on a single >>>>>>>>processor. To make out the article as anti-crafty or venomous or the use of >>>>>>>>other such words is nonsense. >>>>>>> >>>>>>>He is obviously biased. He picks a tournament that supports his preconcieved >>>>>>>ideas on Crafty, and derives he strength estimate from that. But why not use the >>>>>>>tournament later in his article where Crafty comes ahead of other commercial >>>>>>>programs? >>>>>>> >>>>>>>This is the anit-Crafty writing we are talking about. >>>>>> >>>>>>I disagree, there is nothing anti-crafty about this since he put the good result >>>>>>in his article! It seems some are just pissed off that is one good result shown >>>>>>did not change is overall opinion of Crafty based on his overall impression of >>>>>>crafty's play. I also seen the one good result in the article, but I still >>>>>>concur with his opinion, because I own Crafty 16.6 also. >>>>>> >>>>>> >>>>>>> >>>>>>>James >>>>> >>>>>Ok, since you dismiss that, why not try another tack. Why does he even take time >>>>>to MENTION a supposedly weak FREE program IN AN ARTICLE ABOUT COMMERCIAL >>>>>PROGRAMS?? >>>>> >>>>>It is the same as saying "here is my article about strong commercial programs. >>>>>I'm leaving out some commercial entrys because they are not strong enough (IE >>>>>LChess). But I definately have time to talk about how weak the free Crafty is." >>>>> >>>>>What kind of writing is that? >>>>> >>>>>James >>>> >>>>Let it go....a few words about crafty in a full page article does not make the >>>>article anti-crafty. He gave his opinion on a program you get for free when you >>>>BUY ALMOST ANY CHESS BASE PROGRAM. So I don't have a problem with him talking >>>>about crafty. And I will never understand your logic when the article shows a >>>>favoralbe result for crafty(that he did not have to put in) then you use this a >>>>proof that article is anti-crafty. Sorry it does not wash. He gave his opinion >>>>on Crafty 16.6 play running in a chessbase interface, and he is more then less >>>>accurate in his observations. If you think he is wrong about crafty 16.6 in >>>>chessbase post your own results that show otherwise. Last Night I started My own >>>>match to confirm his observation, playing at the same time control of game 30. I >>>>have 20 games played against Fritz 6 vs Crafty 16.6. The results are Fritz 6 15 >>>>points Crafty 16.6 5 points. This is not an uncommon result for crafty playing >>>>ANY of the top commercial programs. >>> >>>I am interested to know if there is a difference between crafty as a chessbase >>>engine and crafty as the original program with crafty's book. >>> >>>I remember that crafty16.6 not as an engine for fritz lost 16:14 against Junior5 >>>and 16:12 against Fritz5.32 in James walker games(I think they were 1 hour/game) >>> >>>Crafty16.6 was weaker than the commercials of the same time if it has the same >>>hardware but there was no big difference like 17:5 >> >>I strongly disagree.... >> >>This question has been asked before, so I dug through RGCC and recoverd my old >>post. The answer to you question is there is little difference between them in >>terms of strength. All the below results are from full blown Craftys with the >>best books, and huge tablebases running on multi cpu computers. The games were >>played against computers that were or still on ICC, RGCC, CHESSNET >> >>************* >>Here are Hiarcs7 results vs the Crafty's I have played. >> >>Drunken +2 -0 =0 2 cpu's = over 1000mhz. >>Crafty +6 -2 =1 Quad 400 >>BarfusII +2 -0 =0 PII 450 >>Cubebox +3 -0 =1 Dual PII 350 >>Quart +2 -0 =0 Celeron 450 >>FlobII +7 -3 =2 PII 300 >>Elminster +6 -0 =3 PII 233 >> >>(T) +28 -5 =7 Avg. mhz. for crafty per game. 667.425 MHz. >> >>Hiarcs7(PII450) has scored 79% vs the Crafty's, running with a avg. computer >>speed of 667.425 Mhz. As of now Hiarcs7 is showing a + rating of 226 points over >>the Crafty's. All games were played at 10 20 or faster. >>*************** >> >> >> >> >> >>> >>>Uri > > >If you look at the dates when that was done, that was in the 'broken' section >as I mentioned several times. For better data, here are some results from the >past month vs various commercial programs on ICC: I agree crafty has closed the gap in the last year... but Uri was asking about that ventage of Crafty. The Crafty versions I played against was not 16.6 as it was not out yet. They lastest version it could have been was 16.4 but most games were with the older 16.x versions and some 15.xx versions. If memory serves you said you used an experimental version of crafty which had same new king safety code in it for our match with Hiarcs 7, but you found it to be broken during the match games with Hiarcs 7 and it was making Crafty think it had more then one king to protect which caused Crafty major problems. > >Crafty is on the usual quad xeon/400 machine. > >opponent/hardware wins draws losses >shredder 3/PIII-550 9 5 4 >genius/shredder/PIII-600 36 23 9 >Tiger 12/PII-525 2 0 1 >Nimzo 7.32/xeon-612 23 21 17 >Hiarcs 7.32/PIII-600 12 6 5 > >There are more, but that is enough. All games were played in December or >January of this year. As you can see, that gives a bit different view of >things. Granted that I have better hardware. But this is far different from >the results you posted against older versions. > >This version is fairly stable as I am trying to slowly tune/tweak for the >upcoming ICC chess tournament. Versions tested during major modifications >are often much weaker...
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.