Re: RWBC new results (+ top 7 PGN)

[Anthony Cozzie]

Geschrieben von:/Posted by: Anthony Cozzie at 11 August 2004 15:11:53:
Als Antwort auf:/In reply to: Re: RWBC new results (+ top 7 PGN) geschrieben von:/posted by: Tord Romstad at 10 August 2004 22:44:40:

[Event "RWBC_Class_B_4th_Edition"]
[Site "ESPRESSO"]
[Date "2004.08.09"]
[Round "5"]
[White "Fruit_X0705"]
[Black "Terra_33b11"]
[Result "0-1"]
[ECO "B12"]
[PlyCount "53"]
[EventDate "2004.??.??"]
1. e4 c6 2. d4 g6 3. c3 d5 4. Nd2 Bg7 5. Ngf3 Nh6 6. h4 O-O 7. Bd3 Bg4 8. Qb3
b6 9. Nf1 dxe4 10. Bxe4 Qd6 11. g3 Nd7 12. Bf4 Bxf3 13. Bxd6 Bxe4 14. Bxe7 Rfe8
15. Rg1 Rxe7 16. Ne3 Bf3 17. Kd2 Ng4 18. Rae1 Bh6 19. Kc2 Be4+ 20. Kc1 Nxf2 21.
g4 Nd3+ 22. Kd2 Nxe1 23. Rxe1 Bf4 24. Qa3 Rae8 25. Rf1 Bxe3+ 26. Kxe3 Bg2+ 27.
Kf2 {White resigns} 0-1

This is a really cool game. Black sacrifices a queen and a pawn against three
minor pieces, but gets a promising attack in compensation. Well worth a look.
It would be interesting if some strong chess player could analyse this game and
tell us whether blacks sacrifice was really sound.
I am pleased to see that Terra, one of my personal computer chess heroes, is
doing remarkably well so far in the tournament. It has already faced Zappa and
Fruit, arguably the two strongest participants in the tournament, but still has
an impressive 4.5/5 score. I am less pleased to see that my own engine is
unfortunate and has to face the fearsome Terra in the next round.
Tord

You are on crack if you think Zappa is one of the strongest participants. Zappa will have to be _very_ lucky to promote. OTOH, the version of Zappa that is playing is about 4 months old With, as Gunther has happily informed me, completely broken time management.
I am going to install windows soon (once I fix some hardware difficulties) and re port; at that point I may have an update for Gunther (parallel & deeper search, better eval, UCI support, etc). Sadly the nps of my program keeps dropping. It is down to 650K or so on my opteron, which would be I guess 350K on Gunther's P4.
Too bad it will be too late for Class B . . .
anthony

[Tord Romstad]

Geschrieben von:/Posted by: Tord Romstad at 11 August 2004 17:30:15:
Als Antwort auf:/In reply to: Re: RWBC new results (+ top 7 PGN) geschrieben von:/posted by: Anthony Cozzie at 11 August 2004 15:11:53:

You are on crack if you think Zappa is one of the strongest participants. Zappa will have to be _very_ lucky to promote. OTOH, the version of Zappa that is playing is about 4 months old With, as Gunther has happily informed me, completely broken time management.
I am going to install windows soon (once I fix some hardware difficulties) and re port; at that point I may have an update for Gunther (parallel & deeper search, better eval, UCI support, etc). Sadly the nps of my program keeps dropping. It is down to 650K or so on my opteron, which would be I guess 350K on Gunther's P4.
Too bad it will be too late for Class B . . .

If you had said this before the start of the Open Class, there is a remote
possibility that a few people would have believed you. But now, when we have
all seen Zappa's impressive results in the Open Class, Class D, and Class C,
it is simply too late to fool us into believing that Zappa is anything less
than a fearsome opponent. Sorry!
I wish I knew how to write such a fast program myself.
Yeah, these tournaments always seem to start at an inconvenient moment.
Tord

[Günther Simon]

Geschrieben von:/Posted by: Günther Simon at 11 August 2004 18:56:05:
Als Antwort auf:/In reply to: Re: RWBC new results (+ top 7 PGN) geschrieben von:/posted by: Anthony Cozzie at 11 August 2004 15:11:53:

...

You are on crack if you think Zappa is one of the strongest participants. Zappa will have to be _very_ lucky to promote. OTOH, the version of Zappa that is playing is about 4 months old With, as Gunther has happily informed me, completely broken time management.
I am going to install windows soon (once I fix some hardware difficulties) and re port; at that point I may have an update for Gunther (parallel & deeper search, better eval, UCI support, etc). Sadly the nps of my program keeps dropping. It is down to 650K or so on my opteron, which would be I guess 350K on Gunther's P4.
Too bad it will be too late for Class B . . .
anthony

Depends on you According to my rules you have exactly 2 weeks from now
to send a bugfix for Zappa.
(Don't know though how much rounds would be left after that time...)
Best regards,
Günther
Quote from of my rules:
6. Only engines with a 'really ugly bug' may be replaced by a new bugfixed version, when
the bug occurs in the first 6 rounds! If such a bug is found by other testers and is
also fixed during the first 6 rounds, it will be used too!
The description of a 'serious' bug is more precise now.
(The bugfix should arrive at least two weeks after the bug was found here!)
It must cause one of this things:
- crashing during a game
- illegal move(s)
- time losses
- _serious_ time management problem
- throwing away _whole_ pieces (not pawns) without reason
- stop playing
- disabling/malfunction of a normal feature of the program, like hashsize or EGTBS
- repetition/stalemate in a won position
- being unable to win a _simple_ endgame mate position
- not detecting 50 moves rule/repetition/
- wrong mate/draw claims/resigning
Programs which have a previous stable version without this bug can be replaced by this one
along the whole tourney.