chessgames.com
Members · Prefs · Laboratory · Collections · Openings · Endgames · Sacrifices · History · Search Kibitzing · Kibitzer's Café · Chessforums · Tournament Index · Players · Kibitzing

Computer
  
Number of games in database: 40
Years covered: 1987 to 1999
Overall record: +6 -16 =18 (37.5%)*
   * Overall winning percentage = (wins+draws/2) / total games.

Repertoire Explorer
Most played openings
C47 Four Knights (2 games)

Search Sacrifice Explorer for Computer
Search Google for Computer


 page 1 of 2; games 1-25 of 40  PGN Download
Game  ResultMoves YearEvent/LocaleOpening
1. Kasparov vs Computer 1-0281987RotterdamD15 Queen's Gambit Declined Slav
2. Alburt vs Computer  ½-½561989Harvard (USA)A40 Queen's Pawn Game
3. Silman vs Computer ½-½431991It Chicago (USA)D27 Queen's Gambit Accepted, Classical
4. Computer vs W Schmidt 1-0511993Katowice mB84 Sicilian, Scheveningen
5. Anand vs Computer  1-0751993It (active)B19 Caro-Kann, Classical
6. Computer vs Zsofia Polgar  ½-½661993Oviedo rapidD47 Queen's Gambit Declined Semi-Slav
7. Joel Benjamin vs Computer  ½-½47199401, Boston Harvard CupD00 Queen's Pawn Game
8. Computer vs K Burger  0-1531995ICC 2 12 08/26/95 Internet Chess ClubD52 Queen's Gambit Declined
9. Joel Benjamin vs Computer 0-161199502, Cup Harvard New York USAA11 English, Caro-Kann Defensive System
10. K Burger vs Computer  1-0411995ICC 2 12 08/26/95 Internet Chess ClubD90 Grunfeld
11. Computer vs Bisguier  ½-½331995It Chicago (USA) (01)C67 Ruy Lopez
12. Zsofia Polgar vs Computer  ½-½611995The Hague AEGONC00 French Defense
13. L Christiansen vs Computer 1-0241995The Hague AEGOND02 Queen's Pawn Game
14. Computer vs Zsuzsa Polgar ½-½411995The Hague AEGONC12 French, McCutcheon
15. Computer vs Zsofia Polgar  0-1721995The Hague AEGOND05 Queen's Pawn Game
16. L Christiansen vs Computer 0-1941995The Hague AEGOND32 Queen's Gambit Declined, Tarrasch
17. Computer vs Zsofia Polgar  ½-½451995The Hague AEGONB48 Sicilian, Taimanov Variation
18. Computer vs Zsuzsa Polgar 1-0571995The Hague AEGONB22 Sicilian, Alapin
19. Zsofia Polgar vs Computer  ½-½231995The Hague AEGONC47 Four Knights
20. Zsuzsa Polgar vs Computer  1-0541995The Hague AEGONA46 Queen's Pawn Game
21. Computer vs Zsuzsa Polgar 0-1261995The Hague AEGONC44 King's Pawn Game
22. Computer vs Zsofia Polgar  ½-½541995The Hague AEGONB23 Sicilian, Closed
23. Zsofia Polgar vs Computer  1-0321995The Hague AEGONC26 Vienna
24. Computer vs Zsuzsa Polgar 1-0481995The Hague AEGONB32 Sicilian
25. K Burger vs Computer  1-0531996BR 3 12 06/04C30 King's Gambit Declined
 page 1 of 2; games 1-25 of 40  PGN Download
  REFINE SEARCH:   White wins (1-0) | Black wins (0-1) | Draws (1/2-1/2) | Computer wins | Computer loses  

Kibitzer's Corner
< Earlier Kibitzing  · PAGE 5 OF 6 ·  Later Kibitzing>
May-15-13  Capabal: <pbercker: @ <capabal>

I don't completely understand the answer, but the FAQ addresses practically the very question you ask!> http://www.truechess.com/web/champs...

I could see complexity related to the distance between what the computer evaluates as the best move, and what it evaluates as the 2nd, 3rd etc. best moves. So that the shorter the distance is between the “best move” and the others, the more complex the position may be said to be.

Now, finding ways to adjust for these differences in complexity when evaluating the “accuracy of play””or whatever it is we call it, is where the problem lies. The adjustment seems unnecessary in the sense that the extent of the deviation from the “best move” is already present in the raw evaluation. I'll try to explain.

Assume two positions, A and B.

Let’s say the best move in position A is evaluated as 200 centipawns better than the 2nd best move, 220 centipawns better than 3rd best etc.

And let’s say the best move in position B is only 20 centipawns better than the 2nd best, and 27 centipawns better than 3rd best and so on.

You could say that position A would tend to be considered simpler than position B.

Now, a player choosing anything other than the best move in position A (the simpler position) will see his score for “accuracy of play” take a hit of at least 180 centipawns. Whereas a player choosing for example the 2nd or 3rd best moves in position B (a more complex one) will only see a substraction of 20 or 27 centipawns.

What I see is that the adjustment for complexity, in this sense, is already in-built in the raw evaluation system.

Sullivan seems to define (or at least calculate) complexity as the difference between the evaluation of the “best move” and the evaluation of <that same move> right before it became the best move (at whatever iteration this happens, it appears).

This already seems arbitrary enough. Then you have the business of translating whatever “basic complexity” you calculate by this method into an adjustment of the deviation between the best move and whatever the player chose.

So my problem with the complexity notion remains the same. Its definition is slippery. Its quantification is whimsical. So adjustments for it are intolerably arbitrary by necessity. And (in at least some sense) complexity adjustments are already inbuilt in the raw evaluation as I've explained above.

What these kinds of games lead to in one way or another is the old Von Neumann admonition about the dangers of free parameters. As quoted by Antonino Zichichi:

<Von Neumann was always warning his young collaborators about the use of these free parameters by saying: “If you allow me four free parameters I can build a mathematical model that describes exactly everything that an elephant can do. If you allow me a fifth free parameter, the model I build will forecast that the elephant will fly."> http://www.justpax.it/pcgp/dati/200... LIMATE.pdf

May-15-13
Premium Chessgames Member
  nimh: <Capabal>

As I have observed, the average error increases with the difference between the best and the second best move, which means it's the other way around than you insist.

<What I see is that the adjustment for complexity, in this sense, is already in-built in the raw evaluation system.>

I'm not sure what you mean by 'built in', but basically you are right, certain aspects of difficulty of positions can be derived from evaluations of moves. However, it doesn't exempt us from measuring them and their effect on the accuracy of play, and using the data to estimate the hypothetical level of accuracy in positions with average difficulty.

<This already seems arbitrary enough.>

If the method Sullivan used were abitrary, then there shouldn't be any correlation to the accuracy of play. Please pay attention to the graph:

http://www.truechess.com/web/champs...

<What these kinds of games lead to in one way or another is the old Von Neumann admonition about the dangers of free parameters. As quoted by Antonino Zichichi:>

This is funny; you accuse complexity measurements of being abitrary, but at the same time you make an abitrary comparison to free parameters. :)

May-15-13  Capabal: Because the magnitude of the error is measured by the magnitude of the distance between best move and other moves. So even if there are fewer errors in positions where the best move is far ahead of all others, when these errors occur they weigh more. So it’s not really surprising that the average error increases that way. It simply shows that the adjustment for the complexity of the position is already in-built in the error measurements.

It is comparable to a free parameter because:

1. The way the complexity is quantified is necessarily arbitrary (although this is also true of of the way the raw error is quantified).

But especially because:

2. The amount of adjustment applied once the complexity is quantified is even more arbitrary.

3. The adjustment seems unnecessary because the raw error measurement already takes into account the complexity, in the sense that in unclear positions where there is a small distance between “best move” and several other moves, the error that is measured will be much lighter than in those where the distance between best move and other moves is greater.

The more adjusting parameters you introduce, the better your chances of losing sight of what you are trying to measure.

May-15-13
Premium Chessgames Member
  nimh: What does 'in-built in' even mean in the context we're arguing in? And how does it follow that it's not necessary to take that factor into account when trying to determine one's accuracy of play in a hypothetical scenario where the positions were of average difficulty?
Jun-21-13
Premium Chessgames Member
  OhioChessFan: <An Apple 1 from 1976, one of the first Apple computers ever built and forerunner of today's MacBooks, IPads and IPhones, goes on the auction block at Christie's next week. The bidding starts at $300,000, with a pre-sale estimated value of up to $500,000.>

http://money.msn.com/business-news/...

Aug-17-14
Premium Chessgames Member
  alexmagnus: Playing a machine can be a risk for life now :D https://www.youtube.com/watch?v=8Jc...
Feb-08-16
Premium Chessgames Member
  OhioChessFan: <Last week, news broke that the holy grail of game-playing AI—the ancient and complex Chinese game Go—was cracked by AI system AlphaGo.

AlphaGo was created by Google’s DeepMind, a UK group led by David Silver and Demis Hassabis. Last October the group invited three-time European Go champion Fan Hui to their office in London. Behind closed doors, AlphaGo defeated Hui 5 games to 0—the first time a computer program has beaten a professional Go player.>

http://singularityhub.com/2016/02/0...

Feb-08-16  Shams: <OCF> Thanks, good article. Those AI guys sure seem cocky for being in a field whose historical successes are so rare.

Also cocky:

<Next up for AlphaGo is a match this March in Seoul against legendary Go world champion, Lee Sedol, who in the 2014 Wired article was quoted saying, “There is chess in the western world, but Go is incomparably more subtle and intellectual.”>

I doubt there is any reason to take Sedol's opinion of chess seriously.

Aug-12-16
Premium Chessgames Member
  AylerKupp: <Giraffe vs. Ziggurat, Match 1 (30 min / game)> (part 1 of 3)

There have been many posts on this site denigrating computers' and chess engines' approach to chess because it is not "human like". Fair enough although I think it's a misguided comment. After my recent reply to <Sally Simpson> in this post, Sinquefield Cup (2016) (kibitz #457) (please read first to put this post in context) I decided to run a test. Giraffe had seemed very slow to me, certainly compared with the top engines, so out of curiosity I ran a 4-game 30-min per game match between it and Ziggurat 0.22, the lowest ranked (#305) engine in the same CCRL list and rated at 1750.

The result was a wipeout. Giraffe won all 4 games. So it's clear that the claim that it is a mid-level ranked engine is credible, although since computer engine ratings and human ratings are not comparable, the claim that it plays at an IM level is questionable, since it clearly did not fulfill any IM norms. Still, it shows that a human-oriented chess engine can hold its own against maybe half of the computationally-oriented engines.

For those interested here are the scores of the 4 games, with the numbers in square brackets indicating the engines' evaluation, search depth, and selective search depth respectively. As you can see, Giraffe was able to reach almost twice the search depth as Ziggurat and I'm sure that had a lot to do with its results. And it shows that neural network-based engines can be more efficient than at least some computationally-oriented engines.

[Event "Match 1"]
[Date "2016.08.12"]
[Round "1"]
[White "Giraffe_w64"]
[Black "Ziggurat_x64"]
[Result "1-0"]
[WhiteElo "2440"]
[BlackElo "1750"]
[Variation "2...Qxd5 3.Nf3"]
[Opening "Scandinavian"]
[ECO "B01"]
[TimeControl "1800"]

1.e4 [+0.12/15 26] d5 [0.00/9 24] 2.exd5 [+0.42/13 22] Qxd5 [-0.02/9 20] 3.Nf3 [+0.52/15 42] Bf5 [+0.03/8 15] 4.Nc3 [+0.81/14 25] Qe6+ [-0.01/9 15] 5.Be2 [+1.08/15 31] Nc6 [+0.04/9 15] 6.d4 [+0.97/15 25] Nb4 [+0.71/9 20] 7.d5 [+0.93/15 39] Qg6 [+0.73/8 17] 8.O-O [+1.81/12 29] Nxc2 [+0.72/8 18] 9.Rb1 [+2.23/14 30] O-O-O [+0.71/7 16] 10.Nh4 [+6.74/16 35] Qf6 [-3.39/9 15] 11.Nxf5 [+6.71/16 29] Na3 [-3.56/8 15] 12.bxa3 [+7.76/15 33] Qxc3 [-3.49/7 15] 13.Bf4 [+7.60/14 22] Qxa3 [-3.73/8 15] 14.Qc2 [+8.14/15 24] e5 [-5.65/8 15] 15.dxe6 [+8.50/15 33] Qc5 [-6.95/8 15] 16.Qa4 [+8.44/14 33] g5 [-7.61/7 15 Black resigns] 1-0

Aug-12-16
Premium Chessgames Member
  AylerKupp: <Giraffe vs. Ziggurat, Match 1 (30 min / game)> (part 2 of 3)

[Event "Match 1"]
[Date "2016.08.12"]
[Round "2"]
[White "Ziggurat_x64"]
[Black "Giraffe_w64"]
[Result "0-1"]
[WhiteElo "1750"]
[BlackElo "2440"]
[Opening "Reti Opening"]
[Variation "1...Nf6 2.e3"]
[ECO "A05"]
[TimeControl "1800"]

1.Nf3 [0.00/10 17] Nf6 [+0.02/15 30] 2.e3 [0.00/10 15] d5 [+0.17/15 44] 3.Bb5+ [+0.06/8 15] c6 [+0.38/15 28] 4.Be2 [+0.08/9 15] Bf5 [+0.42/14 42] 5.O-O [+0.11/9 16] e6 [+0.29/14 41] 6.Nh4 [+0.14/9 16] Bd6 [+0.54/14 24] 7.Nxf5 [+0.11/10 15] exf5 [+0.33/16 38] 8.c4 [+0.08/9 15] h5 [+0.22/15 23] 9.cxd5 [+0.28/9 16] cxd5 [+0.43/15 20] 10.Nc3 [+0.29/7 15] Nc6 [+0.56/14 19] 11.Qb3 [+0.09/8 15] Qd7 [+0.61/14 37] 12.f4 [0.00/8 15] d4 [+0.96/15 30] 13.Nb5 [+0.19/8 20] Bc5 [+1.32/15 26] 14.Qc4 [-0.01/8 16] Bb6 [+1.92/14 19] 15.b3 [0.00/7 15] a6 [+2.20/15 19] 16.Na3 [-0.13/8 15] h4 [+2.29/14 17] 17.Nc2 [0.00/8 15] h3 [+2.56/15 25] 18.gxh3 [+0.06/8 15] Rxh3 [+2.84/14 19] 19.Bb2 [+0.18/7 15] Ne4 [+2.75/14 21] 20.Rad1 [+0.16/8 15] Rc8 [+4.09/14 21] 21.Kg2 [+0.04/7 15] Rh6 [+3.82/13 15] 22.Na1 [-0.50/7 15] Qd6 [+6.19/14 21] 23.Bf3 [-2.96/7 15] Qg6+ [+29.98/17 19] 24.Kh1 [-M4/2 0] Ng3+ [+29.98/18 20] 25.Kg2 [-M6/2 0] Ne2+ [+29.99/18 15] 26.Bg4 [-M6/2 0 White resigns] 0-1

[Event "Match 1"]
[Date "2016.08.12"]
[Round "3"]
[White "Giraffe_w64"]
[Black "Ziggurat_x64"]
[Result "1-0"]
[WhiteElo "2440"]
[BlackElo "1750"]
[Opening "Scandinavian"]
[Variation "2...Qxd5 3.Nf3"]
[ECO "B01"]
[TimeControl "1800"]

1.e4 [+0.12/15 23] d5 [0.00/9 17] 2.exd5 [+0.54/16 38] Qxd5 [-0.03/8 15] 3.Nf3 [+0.47/15 43] Bf5 [+0.03/8 15] 4.Nc3 [+0.72/15 37] Qe6+ [-0.01/9 15] 5.Be2 [+1.00/15 36] Nc6 [+0.04/9 15] 6.d4 [+0.97/15 23] Nb4 [+0.71/9 15] 7.d5 [+0.93/15 39] Qg6 [+0.74/9 15] 8.O-O [+1.99/15 37] Nxc2 [+0.72/8 15] 9.Bb5+ [+2.08/14 38] Kd8 [+0.92/9 15] 10.Rb1 [+1.26/16 35] Na3 [+0.91/9 15] 11.bxa3 [+2.25/16 26] Bxb1 [+1.89/9 15] 12.Ne5 [+2.87/16 29] Qf6 [-0.15/10 22] 13.Qe1 [+2.84/16 21] h6 [-1.41/8 15] 14.Nxb1 [+5.38/15 21] a6 [-1.45/9 16] 15.Bc4 [+5.76/15 20] b5 [-1.74/8 15] 16.Bb3 [+6.73/15 22] a5 [-2.30/8 15] 17.d6 [+7.59/15 17] e6 [-4.35/8 15] 18.dxc7+ [+7.94/15 16] Kc8 [-5.52/7 15] 19.Nxf7 [+8.04/15 17] Rh7 [-6.49/8 15] 20.Bxe6+ [+8.70/15 26] Kxc7 [-9.19/8 16 Black resigns] 1-0

Aug-12-16
Premium Chessgames Member
  AylerKupp: <Giraffe vs. Ziggurat, Match 1 (30 min / game)> (part 3 of 3)

[Event "Computer chess game"]
[Date "2016.08.12"]
[Round "4"]
[White "Ziggurat_x64"]
[Black "Giraffe_w64"]
[WhiteElo "1750"]
[BlackElo "2440"]
[Opening "Reti Opening"]
[Variation "1...d5 2.e3"]
[ECO "A06"]
[TimeControl "1800"]

1.Nf3 [0.00/10 18] d5 [+0.02/13 32] 2.e3 [+0.09/9 16] c5 [+0.14/14 26] 3.Be2 [+0.11/8 16] Nf6 [+0.09/14 27] 4.b3 [+0.11/9 15] Nc6 [+0.61/14 32] 5.O-O [+0.06/8 16] e5 [+0.99/15 27] 6.Bb2 [+0.07/8 15] e4 [+1.15/16 29] 7.Ne5 [+0.17/9 15] Nxe5 [+1.05/16 33] 8.Bxe5 [+0.11/10 15] Bd6 [+1.06/15 39] 9.Bxf6 [+0.04/9 15] Qxf6 [+1.62/15 38] 10.Nc3 [0.00/8 16] Qe5 [+1.76/15 22] 11.g3 [-0.05/9 15] d4 [+2.15/15 22] 12.Nb5 [+0.04/9 15] Be7 [+2.16/15 36] 13.Rb1 [-0.12/8 20] a6 [+2.74/13 21] 14.Na3 [-0.23/10 21] b5 [+3.07/13 26] 15.exd4 [-0.34/8 26] cxd4 [+3.77/11 23] 16.Qc1 [-0.49/8 28] Bh3 [+4.37/11 47] 17.Re1 [-0.56/8 71] O-O [+4.38/10 17] 18.b4 [-0.36/8 46] Rac8 [+4.34/10 21] 19.Bd3 [-0.06/6 25] f5 [+4.50/8 38] 20.f3 [+0.39/3 77] Rc6 [+4.52/7 41] 21.Qb2 [+0.42/8 28] Rg6 [+5.33/6 28] 22.Qb3+ [+0.67/6 75] Kh8 [+6.32/8 28] 23.Kh1 [-0.05/8 63] Rxg3 [+5.95/9 29] 24.hxg3 [+4.06/5 66] Qxg3 [+5.95/9 19] 25.Bf1 [+1.44/8 52] exf3 [+7.02/8 25] 26.Re2 [-12.68/9 68] Bxf1 [+9.69/12 18] 27.Rxf1 [-11.74/9 30] fxe2 [+9.75/14 22] 28.Rc1 [-11.51/9 26 White resigns]

I will next run a classical time control match between the two engines to see if the results will be different but, frankly, I doubt it.

Aug-12-16
Premium Chessgames Member
  WannaBe: Computer, Tea, Earl Grey, Hot.
Aug-12-16
Premium Chessgames Member
  alexmagnus: While NN games are hopeless when it comes to identifying the NNs, I wonder how these "Computer" games came about. I mean, these are all human vs. computer games - not exactly a common type of a game - and nobody knows which computer was involved?!
Aug-13-16
Premium Chessgames Member
  AylerKupp: <Giraffe vs. Ziggurat, Match 2 (40 moves / 2 hrs> (part 1 of 2)

These are the results of a classical time control match between Giraffe and Ziggurat. Another wipeout, Giraffe won 4-0. Again, even at the much longer time controls, Ziggurat could seldom get to search depths > 9 ply while Giraffe routinely got to search depths near or above 20 ply. Here are the scores of the games, with the numbers in square brackets following the moves are each engine's evaluation, search depth, and selective search depth:

[Event "Giraffe vs. Ziggurat, Match 2"]
[Round "1"]
[Date "2016.08.12"]
[White "Giraffe_w64"]
[Black "Ziggurat_x64"]
[Result "1-0"]
[WhiteElo "2440"]
[BlackElo "1550"]
[Opening "Scandinavian"]
[Variation "2...Qxd5 3.Nf3"]
[ECO "B01"]
[TimeControl "40/7200:40/7200:40/7200"]

1.e4 [+0.03/18 192] d5 [0.00/9 17] 2.exd5 [+0.51/19 248] Qxd5 [-0.03/8 15] 3.Nf3 [+0.49/18 246] Bf5 [+0.03/8 15] 4.Be2 [+0.79/17 236] Nc6 [+0.14/9 15] 5.d4 [+0.61/17 190] O-O-O [+0.14/8 15] 6.Nc3 [+0.57/17 177] Qa5 [+0.20/8 15] 7.Bd2 [+0.67/17 169] Nb4 [+0.17/9 15] 8.Rc1 [+1.15/17 286] Nxa2 [+0.01/8 15] 9.Nb5 [+2.69/19 254] Qb6 [+0.78/8 15] 10.Ra1 [+3.27/19 148] c6 [+0.76/8 15] 11.Na3 [+3.26/20 242] Nb4 [+0.66/8 15] 12.Nc4 [+3.37/19 234] Nxc2+ [+0.04/10 15] 13.Kf1 [+3.60/19 217] Qc7 [+0.16/8 15] 14.Rxa7 [+3.71/18 217] Kb8 [-0.32/9 15] 15.Ba5 [+3.41/18 206] Qf4 [-0.33/9 15] 16.Rxb7+ [+3.62/17 196] Kxb7 [-0.56/8 15] 17.Bxd8 [+3.70/17 186] Be4 [-0.26/8 16] 18.Nfe5 [+3.94/16 133] Nxd4 [-0.11/8 15] 19.Na5+ [+7.76/18 121] Ka8 [+M23/2 0] 20.Qa4 [+7.58/19 127] Qc1+ [-4.27/9 15] 21.Bd1 [+7.62/19 141] Qc5 [-4.72/9 15] 22.Nac4+ [+7.59/18 151] Qa5 [-M10/2 0] 23.Qxa5+ [+29.99/21 143] Kb8 [-M6/2 0 Black resigns] 1-0

[Event "Giraffe vs. Ziggurat, Match 2"]
[Round "2"]
[Date "2016.08.12"]
[White "Ziggurat_x64"]
[Black "Giraffe_w64"]
[Result "0-1"]
[WhiteElo "1550"]
[BlackElo "2440"]
[Opening "Reti Opening"]
[Variation "1...Nf6 2.e3"]
[ECO "A05"]
[TimeControl "40/7200:40/7200:40/7200"]

1.Nf3 [0.00/10 17] Nf6 [+0.14/19 255] 2.e3 [0.00/10 15] d5 [+0.06/19 291] 3.Bb5+ [+0.06/8 15] c6 [+0.28/18 233] 4.Be2 [+0.08/9 15] Bg4 [+0.34/18 292] 5.O-O [+0.05/9 15] e6 [+0.36/19 271] 6.Nc3 [+0.10/9 15] Nbd7 [+0.63/18 191] 7.h3 [+0.05/9 15] Bxf3 [+1.21/20 226] 8.Bxf3 [-0.03/9 15] h5 [+0.96/19 271] 9.b3 [+0.03/8 15] g5 [+2.35/18 136] 10.Bb2 [+0.11/8 15] Qc7 [+2.86/19 127] 11.Re1 [+0.09/8 15] g4 [+3.56/19 151] 12.hxg4 [+0.11/10 15] hxg4 [+4.90/19 189] 13.Qb1 [-M31/4 1] Qh2+ [+9.12/21 162] 14.Kf1 [-M4/2 0] gxf3 [+9.17/21 134] 15.gxf3 [-311.81/9 15 White resigns] 0-1

Aug-13-16
Premium Chessgames Member
  AylerKupp: <Giraffe vs. Ziggurat, Match 2 (40 moves / 2 hrs> (part 2 of 2)

[Event "Giraffe vs. Ziggurat, Match 2"]
[Date "2016.08.12"]
[Round "3"]
[White "Giraffe_w64"]
[Black "Ziggurat_x64"]
[Result "1-0"]
[WhiteElo "2440"]
[BlackElo "1550"]
[Opening "Scandinavian"]
[Variation "2...Qxd5 3.Nf3"]
[ECO "B01"]
[TimeControl "40/7200:40/7200:40/7200"]

1.e4 [+0.03/18 181] d5 [0.00/9 17] 2.exd5 [+0.51/19 236] Qxd5 [-0.03/8 15] 3.Nf3 [+0.49/18 226] Bf5 [-0.03/9 15] 4.Be2 [+0.79/17 211] Nc6 [+0.14/9 15] 5.d4 [+0.61/17 168] O-O-O [+0.14/8 15] 6.Nc3 [+0.57/17 156] Qa5 [+0.20/8 15] 7.Bd2 [+0.67/17 151] Nb4 [+0.17/9 15] 8.Rc1 [+1.15/17 293] Nxa2 [+0.01/8 15] 9.Nb5 [+2.59/19 256] Qb6 [+0.78/8 15] 10.Ra1 [+2.99/19 265] c6 [+0.76/8 15] 11.Na3 [+3.32/20 226] Nb4 [+0.66/8 15] 12.Nc4 [+3.44/19 197] Nxc2+ [+0.04/10 15] 13.Kf1 [+3.21/19 203] Qc7 [+0.16/8 15] 14.Rxa7 [+3.67/19 215] Kb8 [-0.32/9 15] 15.Ba5 [+3.75/18 137] Qf4 [-0.33/9 15] 16.Rxb7+ [+3.85/18 169] Kxb7 [-0.56/8 15] 17.Bxd8 [+3.72/18 185] Be4 [-0.26/8 15] 18.Nfe5 [+3.99/16 132] Nxd4 [-0.11/8 15] 19.g3 [+8.18/17 98] Qh6 [-3.33/9 15] 20.Qxd4 [+9.07/18 153] Qc1+ [-6.46/8 15] 21.Bd1 [+9.22/17 166] Ka8 [-12.46/9 15] 22.Qxe4 [+9.72/19 144] Qxd1+ [-18.31/10 15 Black resigns] 1-0

[Event "Giraffe vs. Ziggurat, Match 2"]
[Date "2016.08.12"]
[Round "4"]
[White "Ziggurat_x64"]
[Black "Giraffe_w64"]
[Result "0-1"]
[WhiteElo "1550"]
[BlackElo "2440"]
[Opening "Reti Opening"]
[Variation "1...Nf6 2.e3"]
[ECO "A05"]
[TimeControl "40/7200:40/7200:40/7200"]

1.Nf3 [0.00/10 17] Nf6 [+0.14/19 254] 2.e3 [0.00/10 15] d5 [+0.06/19 284] 3.Bb5+ [+0.06/8 15] c6 [+0.28/18 233] 4.Be2 [+0.08/9 15] Bg4 [+0.34/18 291] 5.O-O [+0.05/9 15] e6 [+0.36/19 271] 6.Nc3 [+0.10/9 15] Nbd7 [+0.63/18 191] 7.h3 [+0.05/9 15] Bxf3 [+1.21/20 227] 8.Bxf3 [-0.03/9 15] h5 [+0.96/19 272] 9.b3 [+0.03/8 15] g5 [+2.44/19 206] 10.Bb2 [+0.11/8 15] Bd6 [+3.08/19 147] 11.Qe2 [+0.09/8 15] g4 [+3.72/20 145] 12.hxg4 [-0.74/11 15] hxg4 [+4.13/19 233] 13.Bxg4 [-0.74/9 15] Qe7 [+4.01/18 222] 14.Bh3 [+1.04/9 15] O-O-O [+3.92/17 210] 15.Rad1 [+1.01/8 15] Rdg8 [+4.39/17 200] 16.Kh1 [+1.13/9 15] Rg4 [+4.25/16 190] 17.Na4 [+1.20/8 15] b5 [+4.52/17 130] 18.Nc3 [+1.17/8 15] Kb7 [+4.26/16 136] 19.a4 [+1.19/8 15] b4 [+4.48/17 91] 20.Na2 [+1.13/8 15] e5 [+4.41/16 108] 21.f3 [+1.12/8 15] Nh5 [+5.61/17 157] 22.Kg1 [+1.06/9 15] Ng3 [+5.63/17 82] 23.Qe1 [+1.09/9 15] Rgg8 [+5.67/17 80] 24.Rf2 [+1.09/8 15] Ne4 [+6.06/18 99] 25.fxe4 [+0.24/9 15] Rxh3 [+6.66/17 76] 26.Re2 [-0.14/8 15] Rgg3 [+8.50/18 71] 27.exd5 [-4.08/9 15] Qh4 [+8.68/18 93] 28.Qxg3 [-4.12/10 15] Qxg3 [+8.95/17 96] 29.dxc6+ [-4.60/9 15] Kxc6 [+9.09/17 71] 30.c3 [-4.52/8 15] e4 [+29.99/21 113] 31.Nxb4+ [-M4/7 0 White resigns] 0-1

So this once again shows that basing an engine on human-like playing reasonable both from the viewpoint of capability and computational efficiency. Too bad that its developer, Matthew Lai, decided to stop his work on it because of possible conflicts with his job. But the Giraffe code is open source so maybe someone else will be willing to continue its development.

Aug-13-16
Premium Chessgames Member
  AylerKupp: <alexmagnus> I picked the Zsofia Polgar vs Computer, 1995 game at random and saw that it was played in Tournament (computers), Hague (Netherlands) (1995). A Google search indicated that Z. Polgar's opponent was Dappet and led me to the game on this site (Zsuzsa Polgar vs Dappet, 1995). A search on Dappet let me to this site: https://chessprogramming.wikispaces.... So I suspect that the information is out there and it is likely that if one was sufficiently motivated(I'm certainly not!) one could find out the chess engines involved and something about them. But nobody really cares much in either case anymore. The most notable exception, of course, were the Kasparov vs. Deep Blue where description of both the computer hardware and software has been described in detail.

I think that the reason that these human vs. computer games are rare is that prior to 1990 the result of a human top player vs. computer game was usually pretty much a foregone conclusion and after 2000 the result was also pretty much a foregone conclusion – but the other way! So I think that only during the 1990s, when the outcome was possibly somewhat in doubt were these type of games played and publicized.

Aug-16-16
Premium Chessgames Member
  AylerKupp: <Who was the greatest chess player of all time?> (part 1 of 2)

My concern with the Truechess.com paper, http://www.truechess.com/web/champs..., and the others are two:

1. The search depth used to determine the reference best move is laughably inadequate to provide much confidence in the accuracy of the computer's evaluation, particularly in close positions. Bratko et. al. in their original papers limited the search depth to fixed 12 ply or less (not even search extensions were used), and the Truechess.com study refers to an average search of 17.4 "iterations" which I assume to be ply. And that average search was conducted for an initial "full six minutes", also laughably inadequate to provide much confidence in the accuracy of the computer's evaluation. Later studies used early versions of Houdini at higher search depths, but the articles did not address the fact that Houdini needs a greater search depth than Rybka (which the authors used once they abandoned Crafty) in order for one to have comparable confidence in its evaluation. So, while the greater search depth used with Houdini might seem on the surface to be a significant improvement over the search depth used with Rybka, it is effectively not much more accurate in determining the "best" move in a given position.

I'm not trying to disparage the authors. The affordable technology at the time that they made their studies was primitive compared to what's available today. And they probably had a limited amount of time to devote to their projects. So they were caught between a rock and a hard place; they needed to analyze a sufficiently large number of games and positions in order for their results to be statistically significant but, given that the time to achieve greater search depths increases exponentially, they just didn't have the time available to devote to their projects.

The limited search depth was defended by claiming that the computer move rankings did not change much as the search depth was increased, so the results obtained by low search depths was adequate. Well, I know from personal experience that this is false; depending on the position an engine's top ranked move can change several times as the search depth increases.

So I think that the authors have compiled a flawed database of positions and evaluations which would take much time to correct, and have used this flawed database to arrive at questionable conclusions. I therefore do no have much faith in the accuracy of their results.

Aug-16-16
Premium Chessgames Member
  AylerKupp: <Who was the greatest chess player of all time?> (part 2 of 2)

2. The authors used only one engine at any one time. I again know from personal experience that different engines produce different move evaluations and rankings, so who’s to say that one engine's top ranked move is better than another engine's top ranked move? No only that but, because of the non-determinism of chess engines, particularly multi-core engines, the same engine running on the same computer, when used to analyze the same position and run to the same search depth, will give different evaluations and rankings. Not MAY, WILL. Guaranteed. Again, this aspect of computer position analysis has never been addressed to my knowledge by any of the authors, and might not even have been appreciated at the time that they did their research.

So determining a reference move to be used to grade the player's choice of move is not easy and would be very time consuming. Imagine having to run an analysis of <each> position multiple times by multiple engines, each analysis at a much greater depth. Such an analysis would require much greater computing resources and much more time. And in the end there is no good known (at least to me) to determine the confidence level in one or several engines' evaluation of a position, so we don't really know how good the agreed upon reference move is.

<Sally Simpson> If you have gotten this far you might remember the discussion we recently have about ways to assess the complexity of a position with regards to including within an engine ways to determine that a position might be difficult for a human to analyze. The http://www.truechess.com/web/champs... article provides one possibility to attempt to do this.

Aug-16-16  Sally Simpson: Hi Alerkupp,

I followed the link...

Glad to hear you are out on parole.

I doubt if anybody (or thing) can answer the question 'who was/is the greatest chess player'.

But no harm done in setting a top of the range computer onto it. Just don't expect everyone to agree.

If God almighty himself appeared and stated who the greatest player there would always be someone (especially on here) to argue with him and in some cases make a valid point.

(The rows would get that bitter some would end up on God's ignore list)

Regarding a computer knowing what is and what is not a difficult position for a human is nigh impossible. It all depends on who they are playing.

How can you tell a machine that does not even know it is playing a game that it's opponent is a GM or a beginner.

When to start setting cheapo's tricks and traps in a lost position is beyond the top computers. They are too powerful, they cannot anticipate a bad move.

Yes what is the point, they can beat humans so why the need to cheapo humans, which I have no doubt they could do.

I bet in some famous lost games they could have muddied the waters if only we could let them - it must be frustrating for them. It can see a shot that turns the game into turmoil. No problem for it to thread it way through but impossible for a human. And yet it cannot play it. It has to follow the 'play the best move' routine.

See:

Van der Wiel vs Short, 1990 (kibitz #2)

Sep-06-16
Premium Chessgames Member
  OhioChessFan: <Bill Gates wants you to send him your resume if you can finish this insanely difficult book>

http://www.aol.com/article/2016/04/...

<In the world of Silicon Valley, there are few books held in higher esteem than "The Art of Computer Programming," a multi-volume set by Stanford professor emiritus Donald Knuth.

"If you think you're a really good programmer... read (Knuth's) Art of Computer Programming... You should definitely send me a résumé if you can read the whole thing," read a quote from Bill Gates on the cover of the third edition of the first volume.>

Sep-06-16
Premium Chessgames Member
  diceman: (<OhioChessFan: <Bill Gates wants you to send him your resume if you can finish this insanely difficult book>

http://www.aol.com/article/2016/04/...

<In the world of Silicon Valley, there are few books held in higher esteem than "The Art of Computer Programming," a multi-volume set by Stanford professor emiritus Donald Knuth.

"If you think you're a really good programmer... read (Knuth's) Art of Computer Programming... You should definitely send me a résumé if you can read the whole thing," read a quote from Bill Gates on the cover of the third edition of the first volume.>>

If you were that good, you would have created something better than "Windows" and Bill would be sending you his resume.

Sep-06-16
Premium Chessgames Member
  WannaBe: Tried that (volumn I) e-book, think I got to page 8 before I decide to switch to Angry Bird.
Sep-07-16
Premium Chessgames Member
  zanzibar: I wonder if Ken Thompson or Dennis Ritchie read that book through, cover to cover?
Sep-07-16
Premium Chessgames Member
  zanzibar: <• Everyone spoke highly of Knuth’s The Art of Computer Programming, and plenty of people owned copies that they dipped into from time to time, but only one person had actually read it all the way through. (I don’t remember who, sorry).

• Everyone agrees that Knuth’s “literate programming” seemed like a pretty neat idea, but no-one had ever actually done it. (Er, except Knuth.)>

https://reprog.wordpress.com/2010/0...

Sep-07-16
Premium Chessgames Member
  john barleycorn: <zanzibar> that is the fate of outstanding books:

Everybody agrees they are a must read,
many pretend to have read them and very few actually did.

(I am not alluding to the pseudo-intellectuals on Rogoff)

Jump to page #    (enter # from 1 to 6)
search thread:   
< Earlier Kibitzing  · PAGE 5 OF 6 ·  Later Kibitzing>
NOTE: You need to pick a username and password to post a reply. Getting your account takes less than a minute, totally anonymous, and 100% free--plus, it entitles you to features otherwise unavailable. Pick your username now and join the chessgames community!
If you already have an account, you should login now.
Please observe our posting guidelines:
  1. No obscene, racist, sexist, or profane language.
  2. No spamming, advertising, or duplicating posts.
  3. No personal attacks against other members.
  4. Nothing in violation of United States law.
  5. No posting personal information of members.
Blow the Whistle See something that violates our rules? Blow the whistle and inform an administrator.


NOTE: Keep all discussion on the topic of this page. This forum is for this specific player and nothing else. If you want to discuss chess in general, or this site, you might try the Kibitzer's Café.
Messages posted by Chessgames members do not necessarily represent the views of Chessgames.com, its employees, or sponsors.
Spot an error? Please suggest your correction and help us eliminate database mistakes!


home | about | login | logout | F.A.Q. | your profile | preferences | Premium Membership | Kibitzer's Café | Biographer's Bistro | new kibitzing | chessforums | Tournament Index | Player Directory | Notable Games | World Chess Championships | Opening Explorer | Guess the Move | Game Collections | ChessBookie Game | Chessgames Challenge | Store | privacy notice | contact us
Copyright 2001-2017, Chessgames Services LLC