Author: Robert Hyatt
Date: 08:31:58 11/06/00
Go up one level in this thread
On November 06, 2000 at 02:34:04, Thorsten Czub wrote: >On November 06, 2000 at 01:00:53, Christophe Theron wrote: >>>Minor eval changes (commands any user can type directly into crafty) will >>>yield this: >>> 5 0.35 -- 1. Rdg1 >>> 5 0.40 4.17 1. Rdg1 Bf3 2. Bg5 Qc7 3. Bxh7+ Kxh7 >>> 5 0.64 ++ 1. Rhg1!! >>> 5 0.74 4.72 1. Rhg1 Bf3 2. Rdf1 Qc7 3. Bxh7+ Kxh7 >>> 5-> 0.88 4.72 1. Rhg1 Bf3 2. Rdf1 Qc7 3. Bxh7+ Kxh7 >>> 6 1.15 4.61 1. Rhg1 Nf6 2. Bh6 Ra7 3. Bxh7+ Kxh7 >>> 4. Bxg7 Rxg7 >>> 6-> 1.40 4.61 1. Rhg1 Nf6 2. Bh6 Ra7 3. Bxh7+ Kxh7 >>> 4. Bxg7 Rxg7 >>> 7 1.85 ++ 1. Rhg1!! >>> 7-> 4.66 5.00 1. Rhg1 Nf6 2. Bh6 Ra7 3. Bxh7+ Kxh7 >>> 4. Bxg7 Rxg7 >>> 8 7.20 5.35 1. Rhg1 Qc8 2. Bh6 e4 3. Bxe4 Re8 4. >>> Rxg4 Qxg4 >>> 8-> 13.21 5.35 1. Rhg1 Qc8 2. Bh6 e4 3. Bxe4 Re8 4. >>> Rxg4 Qxg4 >>> 9 25.25 5.23 1. Rhg1 Qc8 2. Bh6 Rf4 3. Bxg7 Nxg7 >>> 4. Nxf4 Bxd1 5. Ne6 Nc6 >>> 9-> 31.06 5.23 1. Rhg1 Qc8 2. Bh6 Rf4 3. Bxg7 Nxg7 >>> 4. Nxf4 Bxd1 5. Ne6 Nc6 > >nice nice bob. > >>>Which shows what your position proves. Namely nothing. The first issue is >>>to _play_ the right move. Whether your eval is overly optimistic or overly >>>pessimistic doesn't really matter, in this position... >>>There is no "new" paradigm... >>>That is just a buzz-word... > >aha. and how do YOU know ? without knowing the program ? Perhaps because I _listen_ to the author of that program when he explains what he did? Vs trying to dream of what I _hope_ he might have done? >without seeing its evaluation live in many many games ? I _have_ seen the evaluation live in "many games". Some of the beta testers have asked to see my kibitzes (scores, etc.) and they do the same. That is how I first noticed that its eval was often 2-3 pawns higher than mine with absolutely no justification, based on the board position. It because pretty obvious that it has lots of big scores for its own pieces, but it doesn't pay any attention at all to what kind of defensive resources are availab.e Not _every_ open file around the king is fatal. Particularly if your opponent gets two rooks doubled on that file himself. >i say it is different, and i came with a position >difficult to understand for some computerchess programs that >are on the market. You are simply wrong. _every_ program anyone tried found the right move. But that isn't good enough for you. You have some exaggerated idea about what the _score_ for that move should be. I can adjust crafty's king safety quite easily to make the score for Rhg1 -5, 0, or +5. Yet it plays that move _every_ time. The number assigned to the score is not nearly so important as the move that is chosen. >IMO you talk about "hear-said" with showing your prejudices. that is "hear-say" and I did "hear" what Christophe "said". >you don't want that a new paradigm exists. >that it even COULD exist. >you are like newton talking with einstein, >or the catholic church talking with giordano bruno. >or the catholic church talking with galileo galilei. >you have your point of view (it does not exist!) >and you claim your point of view. nothing against this. >its a free world. >you don't see the thing because you don't want to see it. because >it does not fit in your ideas how to make a chess program. >its not possible for you. >but it IS possible. >i do not say that any program that "solves" the position is a new >paradigm. i say that christophes program is very strong AND >works also in this position. if cstal or gandalf would be complete >losers, and would behave good in this position, it would not be >a big thing, or ? but they are not complete losers. and even more >for gambit-tiger. it won 2 tournaments, dutch and french championships, >and IMO the reason it did so is that is is not working the "crafty/bob hyatt" >way/paradigm. Then I assume you believe it will rise to #1 on the SSDF when it blows out all those other "crafty/bob hyatt" paradigm programs? :) > >that was my point. and to give you examples i presented this position. >of course what happens now is that the one with a different opinion >make easy jokes, the ones who like to bean count come with results >of other programs to show: there are MORE programs solving it. Yes there are. _all_ programs solve it. At least all the programs that ran the test position. That was my _point_. > >its not about solving. this is just an example. >there is nothing to solve. >you have to make your own plan to COME in such a position. >you have to create the key-move positions you want to win with >a big punch. therefore you have to create ideas. > >the new paradigm is not about finding the best move in a position. >its about finding the best move for the idea! Can you spell hyperbole? vaporware? nonsense? The "new paradigm" is a modified evaluation function, as directly stated by the author. It is not a new search paradigm. New search extensions unused before. Just some speculative additions to a pretty decent evaluation. That doesn't qualify as a "new paradigm" by any definition of the word I can find in my dictionary. Of course, don't let that stop you... > >and only if it works, i will be able to convince you, >since people of the old paradigm are this kind oif blind that they >need RESULTS to see something that is there. >no problem: gambit-tiger shows you RESULTS. > I'm watching these "results". At the momemt, I am much more concerned with the results of the new Fritz program. It looks very serious. > >>If it's so simple, Bob, why don't you play with this tweaked Crafty in a serious >>tournament? > >exactly. >they don't understand. Somebody doesn't understand. I don't think it is me. > >>Remember that Gambit Tiger 1.0 has WON the two tournaments he entered recently: >>the french computer championship and the dutch computer championship. It has not >>only made a nice show, but it has also WON the tournaments. > >right. the thing is: it works. its not a "buzz-word" as bob, >the "boss of the old paradigm", believes. > >Bob completely controls and oversees anything in the old-paradigm. >he knows everything about it. >but it seems he knows or does not want to know or learn or understand >about the new thing. Thorsten, I can't tell you how bad you sound writing that kind of stuff. Before you rattle on and on about "the new thing" why don't you ask Christophe about what he did differently? Or look up the post where he explained this in the CCC archives. You talk about it reverently, as though it is some sort of sensational break-through. And you sound silly in doing so, when everybody here read Christophe's explanation and knows what part of his program he modified to produce "gambit-tiger". >thats a pity. i thought professors would be interested in new ideas >and have the job to promote development. The first issue is to be able to _recognize_ "new ideas". I'm pretty good at that. This doesn't qualify. Several of us have tried lots of speculative things to produce exciting attacks. It isn't new. All the way back to the super-constellation. Seems to me a good test would have been to jump into the CCT2 tournament, if you are so sure it would have won. > >>How would Crafty have performed in these two strong tournaments? How would the >>tweaked Crafty perform? > >good idea. let it play against gambit-tiger. > I do play against it regularly. Ask the beta testers... >>The one who proves nothing here is YOU. Gambit Tiger has proved what it can do. >>Your tweaked Crafty is just a joke. > >it works in THIS position. >still the old paradigm people try to solve chess like solving cross-word-puzzle. >they want to solve positions ! instead of teaching their programs >to play a whole game of chess. > >hm. >oh man. I agree. "oh man..."
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.