Chess Genius 3 Beat Kasparov on a Pentium 90?

[Milos]

About Kramnik's tactical ability : 19.Nxf7??
[pgn]
[Event "Brains in Bahrain"]
[Site "Manama BAH"]
[Date "2002.10.15"]
[EventDate "2002.10.04"]
[Round "6"]
[Result "0-1"]
[White "Vladimir Kramnik"]
[Black "Deep Fritz (Computer)"]
[ECO "E15"]
[WhiteElo "2807"]
[BlackElo "?"]
[PlyCount "68"]

1. d4 Nf6 2. c4 e6 3. Nf3 b6 4. g3 Ba6 5. b3 Bb4+ 6. Bd2 Be7
7. Bg2 c6 8. Bc3 d5 9. Ne5 Nfd7 10. Nxd7 Nxd7 11. Nd2 O-O
12. O-O Rc8 13. a4 Bf6 14. e4 c5 15. exd5 cxd4 16. Bb4 Re8
17. Ne4 exd5 18. Nd6 dxc4 19. Nxf7 Kxf7 20. Bd5+ Kg6 21. Qg4+
Bg5 22. Be4+ Rxe4 23. Qxe4+ Kh6 24. h4 Bf6 25. Bd2+ g5
26. hxg5+ Bxg5 27. Qh4+ Kg6 28. Qe4+ Kg7 29. Bxg5 Qxg5
30. Rfe1 cxb3 31. Qxd4+ Nf6 32. a5 Qd5 33. Qxd5 Nxd5 34. axb6
axb6 0-1
[/pgn]

19.Nxf7 wouldn't be enough but than it came 26. hxg5+?? which is a big mistake and then 27. Qh4+??? which is even worse

[hammerklavier]

In some Sources mention Pentium 120Mhz... (Penitum 100Mhz released March 1994 ) but those chips were easily overcloked... for example an Pentium 100 Mhz run perfectly at 120Mhz in DOS mode .

The Streng of ChessGenius always was unequal. In some positions play like an Real Master, and in other play horrible positional mistakes.

For many year I use Chessgenius (DOS versions), in wrong mode, only with 512kb of Hash Tables. In Beginning of 2000s I discover the /x parameter... ( MG2 /x in MS-dos enable all ram in your system for HASH TABLES...with Himem.sys disable-F5 boot trick).

in DOSBOX emulator can easily enable hash tables , editing the dosbox-0.74.conf file with parameters===>
memsize=32
cycles=max
xms=false
ems=flase

and run with /x parameters. Example:

(for Mephisto Genius 2) ==> C:\<directory MG2>\mg2 /x
(Genius 3) ==> C:\<directory CG3>\cg3dos /x

[Uri Blass]

It's very interesting how computer chess progresses over time.

Chess Genius 3 playing in the 1994 Intel Grand Prix on a Pentium 90 defeated Gary Kasparov, and also beat GM Nikolic (a very strong GM) 2-0.

With this in mind, I was enormously proud when, as a teenager I drew a 30 0 game against Chess Genius 3.

Recently, I pulled it out and was shocked by how easily I crushed it.

Obviously I have become a stronger player, but for example, I have a lot more trouble against even Comet A90 (another old fav), so it's kind of weird how strong GMs lost to this computer.

Here's my game anyway, clearly 11...0-0?? is a terrible, losing move which proves Kramnik's old assertion that "computer's weakness is tactics".

These words perplexed me at the time, but this game (and others) prove his point.

I think in those days (1994) players assumed that "computers see everything" (they do NOW) and played too safely against them. I'm pretty sure the typical Kasparov attack would have roasted CG3 like a beginner.

[pgn][Event "?"]
[Site "Macbook Pro"]
[Date "?"]
[Round "?"]
[White "Norman"]
[Black "Chess Genius 3"]
[Result "1-0"]

1.e4 Nf6 2.e5 Nd5 3.Nc3 e6 4.d4 Bb4 5.Bd2 Nxc3 6.bxc3 Be7 7.Qg4 g6 8.Bd3 d6 9.h4 dxe5 10.dxe5 Nc6 11.Nf3 O-O 12.h5 Bc5 13.hxg6 Bxf2+ 14.Kxf2 fxg6 15.Kg1 Qe7 16.Re1 b6 17.Rh3 Qc5+ 18.Kh1 Ne7 19.Bg5 Nf5 20.Bf6 Qxc3 21.Rg1 Bd7 22.Ng5 h5 23.Qf4 Qc5 24.Ne4 Qd4 25.Qg5 Kf7 26.Rxh5 Rg8 27.Rh7+ 1-0[/pgn]

I think that chess genius used a different opening book against kasparov so kasparov could not play this game.

I do not believe that there was a chance for 1.e4 Nf6 in the book of Genius against kasparov.

I also know that kasparov believed he was playing genius2 and not genius3 during the game.

I think that getting conclusions based on a single game is wrong and I doubt if you can beat Genius3 with different openings when it does not choose 1....Nf6.

You're hilarious man.

You are so motivated to sh!t on me that you ignore logic.

Firstly, I wasn't bragging, so there's no need to "bring me down" anyway, but let's handle this properly...

"I think that chess genius used a different opening book against kasparov so kasparov could not play this game."

Beginning a biting statement with "I think" kills any form of credibility in the words that follow, but regardless of this...

I am probably 2180-2200 ELO and scored this easy win against Chess Genius 3.

Kasparov was 2800+ and Nikolic around 2680...

Are you saying that this "magical" opening book gave Chess Genius 500-600 ELO - even back in 1994? Wow! Impressive.

I also know that kasparov believed he was playing genius2 and not genius3 during the game.

Oh...so a guy who is arguably the strongest player in history and had never lost to a computer is going to play more cautiously against an older program?

Besides this, Kasparov played 1.c4 not 1.e4 in the game he lost, so how was he even to TEST how the program is going to respond to the king's pawn?

I think that getting conclusions based on a single game is wrong and I doubt if you can beat Genius3 with different openings when it does not choose 1....Nf6.

Who cares what you doubt?

Who was "getting an impression" anyway?

I made the comment that I have more trouble than I did in that game against Comet A90, but never did I say that I'm a stronger player than Chess Genius. I myself was referring to a single game.

I'm gonna just assume that you're another one of the negative-minded, "bring em down if they do something good" types on TC...

So many of you.

My point is that you said that it is weird that genius could beat strong GM's when I do not find it weird and this is the reason that I responded.

I did not say that opening book gave Genius 500-600 elo in 1994 against humans
and my point was only that your conditions is not the same as the conditions that kasparov played.

I also do not know if you used the same time control that kasparov played against Genius(25 minutes per game based on my memory and not tournament time control when kasparov could easily win 2-0 even against deep thought that was stronger by tactics)

I believe that Genius is not stable and in some openings it can play like 2800 and in other openings it can play significantly weaker
and getting conclusion from a single game that it is weird that genius could beat strong GM's is wrong.

I believe kasparov had Genius2 at home and he probably found that playing safe works against it when playing for tactics may work but may fail depending on the opening choice of Genius and he did not want to take risks.

I know that there was a rematch between kasparov and genius3 later when kasparov won 1.5-0.5 and I am sure Kasparov prepared at home
for the second match because it was important for him.

He also played with white 1.c4 in the game he won so I guess that he found 1.e4 is not a way for a sure win based on experience.

[hammerklavier]

Other common razon was in 80s & 90s the engines plays weak in direct King attacks... for example this game. Nothing spectacular, In a well known position Queen Gambit , Genius not see Danger on King side and play 19...Bb4 20....Rad8.

[pgn][Event "Blitz 10"]
[Site "?"]
[Date "2007.07.12"]
[Round "?"]
[White "Mallada, Gustavo"]
[Black "Comp ChessGenius 2.01, Palm Zire 21 126Mhz"]
[Result "1-0"]
[ECO "D55"]
[BlackElo "2200"]
[Annotator "Mallada,G"]
[PlyCount "55"]

1. d4 d5 2. c4 e6 3. Nc3 Nf6 4. Bg5 Be7 5. e3 O-O 6. Nf3 Nbd7 7. Rc1 b6 8. cxd5
exd5 9. Bd3 Bb7 10. Ne5 Nxe5 11. dxe5 Nd7 12. Bf4 Nc5 13. Bc2 c6 14. h4 Re8 15.
Qh5 g6 16. Qg4 Ne6 17. Bh6 Qc7 18. f4 Bc5 19. Qg3 Bb4 20. h5 Rad8 21. hxg6 hxg6
22. Qh3 Qxe5 23. fxe5 Rd7 24. Bg5 Bxc3+ 25. bxc3 f5 26. Qh8+ Kf7 27. Qf6+ Kg8
28. Rh8# 1-0[/pgn]

This game was played in 2006 on Palm zire 21 126mhz... ChessGenius 2017 android on 1.2 GHz Cortex-A53 at 260.000/nps plays the same errors

[Vinvin]

...

The themes I marked in my set "hard-CCC-2009-beta1" are :

Code: Select all

Attack the castled king
crippled position
Deep combination
deep mate threats
Deflection sacrifice
Endgame
Fortress
King's attack
King's attack threats
Long mate threats
Manoeuvres
Mate
Mate threats
Pawns endgame
perpetual threats
Pertpetual check
Pin
Positional move
Positional squeeze
promotion
promotion threat
queens endgame
Rising the pressure
sac for 2 linked passed pawns
sac winning tempo
Sacrifice
Sacrifice for the initiative
Simplification
Stalemate
Trapped piece
underpromotion
Zugzwang

Awesome list...How can I check them out?

The 2009 is quite old and a lot of positions are solved quickly now : http://home.scarlet.be/vincentlejeune/hardsets.html
The 2016 version (beta 6 currently) is here :
EPD : https://www.dropbox.com/s/d3pqj6amz3mrb ... P.epd?dl=0
PGN : https://www.dropbox.com/s/s4qeumzcxz4yu ... 5.pgn?dl=0

[Paloma]

For 2009 >> Het spijt ons, de pagina die u zoekt, bestaat niet.

[Vinvin]

For 2009 >> Het spijt ons, de pagina die u zoekt, bestaat niet.

Fixed now, you can retry.

[Paloma]

By clicking on http://home.scarlet.be/vincentlejeune/h ... -beta1.pgn

comes this up:


edit: click on image

[Vinvin]

By clicking on http://home.scarlet.be/vincentlejeune/h ... -beta1.pgn

comes this up:


edit: click on image

Reload the page ! the correct link is now : http://home.scarlet.be/vincentlejeune/h ... -beta1.pgn

[Paloma]

Now it works!