Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: CCT5 & NoonianChess

Author: Tony Werten

Date: 03:56:58 01/25/03

Go up one level in this thread


On January 25, 2003 at 06:50:19, Uri Blass wrote:

>On January 25, 2003 at 03:36:54, Tony Werten wrote:
>
>>On January 25, 2003 at 01:14:27, Charles Roberson wrote:
>>
>>>
>>>   The tournament was great fun. Tiring, but fun. I prefer the format at
>>>   WCCC 2002, but it would be too much for all tournaments to be like that.
>>>
>>>   The competition was great. I have lots of analysis data. The talk on ch 64
>>>   "compfortably numb" (pun intended). I enjoyed the IM commentary and Volker
>>>   performed excellently as TD. I hope to see more like these several times a
>>>   year. You can't get better competition without an expensive trip to Europe.
>>>
>>>   Version 3.4 was used. It is 2x faster than 3.3 and more knowledgable:
>>>       better king safety, mobility, ...
>>>   It averaged a ply deeper than the Maastricht version even at half the time
>>>   control.
>>>
>>>   I'll comment on some of the games as the theme seems repeatable.
>>>
>>>   Game 1: Searcher - NoonianChess 1-0
>>>       I tried our a virtually untested new book. Ran into a bug on move 6.
>>>      This caused the drop of a bishop for 2 pawns. Amazingly, Noonian played
>>>      quite well after this -- it used its center pawn advantage to control
>>>      the center and the game. Then a mistake on move 34; KxB should have been
>>>      the move (a free bishop and now Noonian would be up a full two pawns).
>>>      But seems there is a bug that caused the mistake. Analysis from CM8000
>>>      revealed that after 34 ... KxB, Noonian can force the win of the other
>>>      bishop but alas Searcher could for a draw by perpetual check or 3frp.
>>>
>>>      So, I was happy with the game and use my old tournament book for the
>>>      rest of the tournament.
>>>
>>>   Game 3: NoonianChess - PostModernist  0 - 1
>>>      Noonian gained a draw from PM in WCCC 2002. It was a great experience
>>>      competing against Andrew so, I was quite happy to do it again. Noonian
>>>      stayed in book for 10 or so moves. Noonian makes questionable bishop
>>>      moves on moves 17 and 18. (remember game 1 -- mistake in not taking a
>>>      free bishop). Noonian does not make the same mistakes when given the
>>>      positions but it does if the game is played to that point (a bug a bug
>>>      -- or is it something to dig into). After this, PostModernist performs
>>>      excellently in controlling the game.
>>>
>>>    Game 4: Matacz - NoonianChess 0 -1
>>>      Noonian won a pawn and created a passer on the A file by move 21. The
>>>      rest of the game centered on this one pawn. Noonian pushed the pawn and
>>>      Matacz spent most of the rest of the game keeping the a pawn from
>>>      promoting. There were times when I thought Noonian had stronger moves but
>>>      I'm not sure (I haven't analyzed it deeply but the stronger moves
>>>      involved a bishop -- hmm bishop problem sounds like a theme to me).
>>>      Finally, Matacz is able to capture the pawn on a2. However, Noonian has
>>>      too many threats on the king which causes the win of material and the
>>>      eventual mate of Matacz. A long game. After the first two, I kept
>>>      wondering when is the bug going to happen again and lose this one.
>>>      The great part of live games!!!!
>>>
>>>    Game 5: NoonianChess - Aristarch 1 - 0
>>>      The opening had me on the edge of my seat. Noonian was agressive from the
>>>      start with an attack on Aristarch's king side. I had some of
>>>      that "Australian speed skater luck": Aristarch lost its connection
>>>      for more than 15 minutes and forfeited the game.
>>>
>>>     Game 6: Pepito - NoonianChess  1 - 0
>>>      I was happy that Noonian held its own to move 35. Pepito sacs a rook for
>>>      the bishop and things go down hill from there. A rook for a bishop --
>>>      hmm is there an issue with bishops?
>>>
>>>     Game 7: NoonianChess - Amyan 0 - 1
>>>      NoonianChess plays well from opening to midgame. I was happy to move
>>>      25. Noonian has a nice position. I need to analyze this deeply to see
>>>      what really happened. At move 35, Noonian has a passer and seems there is
>>>      a lot of potential for a win. However, at move 41 Noonian gives up a rook
>>>      for a pawn and knight. I suspect it liked 3pawns and a knight vs a rook
>>>      and a pawn -- especially considering 2 of Noonians pawns are passers.
>>>      Amyan forces the trade of all this and the ending is a forced draw.
>>>      However, I didn't have egtb's and move 65 is a mistake. I loaded this
>>>      position into Noonian latter and it doesn't make the mistake. Another
>>>      bug to fix. Also, this the second game in live tournament competition
>>>      lost due to lack of egtb's -- the first is Goliath Lite - NoonianChess
>>>      Maastrict WCCC 2002.
>>>
>>>    Game 8: Czolgista  - NoonianChess 0 - 1
>>>      I expected to win this one as the programs met in competition on ics this
>>>      week. However, Czolgista froze up on move two and lost on time having
>>>      made only one move. Wow, that Austrailian speed skater luck was with me.
>>>
>>>   Game 9: NoonianChess - XiniX  1 - 0
>>>      This is my favorite game. I have not analyzed it yet but I can't wait.
>>>      This game had classic horizon effect issues. Both sides had chances in
>>>      the middle game. So, it was quite exciting. Forget the edge of my seat --
>>>      I was pacing during this one. At move 30, Noonian considers itself up
>>>      2.8 pawns. The kibitzing made this far more interesting than without it.
>>>      For a series of moves Noonian and Xinix agree with the scores that
>>>      Noonian is up 3 pawns. Then Xinix searchers a little deeper and claims
>>>      the score is even. A move later (move 45), Noonian thinks it is down
>>>      4/10 of a pawn. After several moves of 3 pawns up, now both programs
>>>      call it even. (so where is this bug?). Oh no -- on move 46 both programs
>>
>>This game showed a new bug in XiniX :( At move 45 XiniX thinks it's winning (it
>>was) but then somehow at 46 got a different position on the board. At move 49
>>where I (manually) resigned, XiniX is actually trying to give a checkmate move,
>>but winboard keeps saying "illegal move" :(
>
>Is not it the opposite?
>
>I do not see how 49.Rxf1 in the game is legal without 46.Ra1 so I guess that
>Xinix said illegal move about Rxf1 of Noonian.
>
>What was the problem?

I haven't looked again, but IIRC, a bishop at b7 had shown up (!) making Rf1 a
doublecheck ( and a checkmate ? )

Tony

>
>Did Xinix ponder on the move that it expected(Maybe Ne6) instead of the move
>that was played?
>
>Uri



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.