Winboard Forum

Hello,

It's been some time that we were talking about PSWBTM starting with the same game of the opening game file for each pairing in a gauntlet (in fact now I am not sure that was the issue...). Has it been fixed? I'd gladly use WinBoard for my gauntlets among WB compatible engines but I'd like to see random opening choice.

Cheers,

Hmm, I usually want all pairings to start from the same position, e.g. to let every gauntlet opponent play a 20-game Nunn match.

It will be difficult to make PSWBTM randomly chose a game from the loadGameFile, as it does not know how many games are in it. It never eamines it by itslf, it just passes the file name and game number to winBoard. Of course the "rewind" input field of PSWBTM could be used by the user to tell how many games are in the file (or how many of the games you want it to use).

I guess we could ask Pradu for another tickbox, to ask for random selection of a game (or position) upto the maximum specified by the rewind parameter.

H.G.Muller wrote:Hmm, I usually want all pairings to start from the same position, e.g. to let every gauntlet opponent play a 20-game Nunn match.

It will be difficult to make PSWBTM randomly chose a game from the loadGameFile, as it does not know how many games are in it. It never eamines it by itslf, it just passes the file name and game number to winBoard. Of course the "rewind" input field of PSWBTM could be used by the user to tell how many games are in the file (or how many of the games you want it to use).

I guess we could ask Pradu for another tickbox, to ask for random selection of a game (or position) upto the maximum specified by the rewind parameter.

Thanks for the reply. However, I provide PSWBTM with a file containing enough games for the gauntlet, e.g. 320 games if I want to play a 10-opponent gauntlet with 32 games against each. This file already contains random games. All PSWBTM has to do is tell WinBoard to increment the game counter each time it starts a new game.
Note that if I want to make a really random selection I will have to provide a very large file else all openings will be chosen from a limited number of samples. E.g. if I provide a file of 1000 games there will be duplicates for sure in a 200-game gauntlet, while if I provide an already random file all openings will be unique.

Best regards,

Gábor Szots wrote:

H.G.Muller wrote:Hmm, I usually want all pairings to start from the same position, e.g. to let every gauntlet opponent play a 20-game Nunn match.

It will be difficult to make PSWBTM randomly chose a game from the loadGameFile, as it does not know how many games are in it. It never eamines it by itslf, it just passes the file name and game number to winBoard. Of course the "rewind" input field of PSWBTM could be used by the user to tell how many games are in the file (or how many of the games you want it to use).

I guess we could ask Pradu for another tickbox, to ask for random selection of a game (or position) upto the maximum specified by the rewind parameter.

Thanks for the reply. However, I provide PSWBTM with a file containing enough games for the gauntlet, e.g. 320 games if I want to play a 10-opponent gauntlet with 32 games against each. This file already contains random games. All PSWBTM has to do is tell WinBoard to increment the game counter each time it starts a new game.
Note that if I want to make a really random selection I will have to provide a very large file else all openings will be chosen from a limited number of samples. E.g. if I provide a file of 1000 games there will be duplicates for sure in a 200-game gauntlet, while if I provide an already random file all openings will be unique.

Best regards,

Hi Gabor,
I never had this problem,
maybe you have accidently activated "rewind" in the "new tournament " screen?
Best
Fonzy

F. Bluemers wrote:Hi Gabor,
I never had this problem,
maybe you have accidently activated "rewind" in the "new tournament " screen?
Best
Fonzy

Hi Fonzy,

Not impossible although I am generally quite scrupulous with these things. I'll check next time.

Best,

F. Bluemers wrote:Hi Gabor,
I never had this problem,
maybe you have accidently activated "rewind" in the "new tournament " screen?
Best
Fonzy

How can you never have had this problem? (which for me is not a problem ata all, but eactly what want.) Has PSWBTM been changed since I started using it? Initially I always played Nunn matchs from a file of 20 games, in gauntlets woth 20 games per pairing. And I never ran out of games, all Nunn positions were played by each engine.

But this was before the rewind parameter was added to PSWBTM (at my request). Perhaps since this parameter was added, PSWBTM never rewinds, not even between pairings, unless you reach the specified count? I guess thi would work, one can always acheive what I want by setting the rewind xactly to the number of games in a pairing.

One thing I now for sure: if you set the rewind to a value larger than the number of positions, as soon as you get to a position that does not exist in the file, WinBoard fails to start, and the tourney pauses.

H.G.Muller wrote:

F. Bluemers wrote:Hi Gabor,
I never had this problem,
maybe you have accidently activated "rewind" in the "new tournament " screen?
Best
Fonzy

How can you never have had this problem? (which for me is not a problem ata all, but eactly what want.) Has PSWBTM been changed since I started using it? Initially I always played Nunn matchs from a file of 20 games, in gauntlets woth 20 games per pairing. And I never ran out of games, all Nunn positions were played by each engine.

But this was before the rewind parameter was added to PSWBTM (at my request). Perhaps since this parameter was added, PSWBTM never rewinds, not even between pairings, unless you reach the specified count? I guess thi would work, one can always acheive what I want by setting the rewind xactly to the number of games in a pairing.

One thing I now for sure: if you set the rewind to a value larger than the number of positions, as soon as you get to a position that does not exist in the file, WinBoard fails to start, and the tourney pauses.

I used the pgn file/set from marc larcosse to play lots of gauntlets with zct.
I only had problems IF rewind was activated
otherwise it would cycle nicely through the set of positions/games.
("two games" was activated as well,same pos wirh white and black)
best
Fonzy

Gábor Szots wrote:Hello,

It's been some time that we were talking about PSWBTM starting with the same game of the opening game file for each pairing in a gauntlet (in fact now I am not sure that was the issue...). Has it been fixed? I'd gladly use WinBoard for my gauntlets among WB compatible engines but I'd like to see random opening choice.

Cheers,

Hi Gabor. It was quite a while ago and I remember working on it but I don't remember if I fixed it or not. Here's a debugging compile of PSWBTM:

http://www.pradu.us/home/cc/PSWBTM/PSWBTM_debug.zip

In the details tab it will say what parameters it sends to Winboard. Please let me know if you find a problem with this PSWBTM.

H.G.Muller wrote:One thing I now for sure: if you set the rewind to a value larger than the number of positions, as soon as you get to a position that does not exist in the file, WinBoard fails to start, and the tourney pauses.

True, you must have enough games in the PGN or you must set the rewind parameter less than or equal to the number of positions in the PGN. This is bad by design (PSWBTM should check for the number of positions in the FEN/EPD/PGN file and make sure the user doesn't input a number greater than this) but I was lazy .

Pradu wrote:In the details tab it will say what parameters it sends to Winboard. Please let me know if you find a problem with this PSWBTM.

Hi,

I have just run a small test gauntlet which I stopped after 6 games because of a problem with one of the engines.
It seems to me that PSWBTM sends "-lgi 1" to WB which makes it start with game 1 for each pairing. The rewind box was unticked.

Here is what I got in the details window after I restarted PSWBTM:

Event: debug
BugChess2 1.6.3 32-bit Gauntlet
2 games per pairing
Time Control: 40 moves in 0 minutes 50 seconds
Pondering is off
Starting Moves: C:\Sakk\WinBoard_F\64.pgn
5 participants

Pairing 4
Game 1

Games Played: 6
Games Left: 4

Playing time: 14 minutes 34 seconds
Average time per game: 2 minutes 25 seconds
Estimated time to finish: 9 minutes 43 seconds

Winboard Parameters: -sgf "temp.pgn.pswbtm" -autosave -popupExitMessage false -thinking -xperiodic -mg 1 -xreuse -xreuse2 -saveSettingsOnExit false -cp -pgnEventHeader "debug" -xponder -tc 0:50 -inc=-1 -movesPerSession 40 -lgf "C:\Sakk\WinBoard_F\64.pgn" -lgi 1

PGN File: default.pgn


And here is the default.pgn file:

[Event "debug"]
[Site "SZOTS"]
[Date "2008.11.05"]
[Round "1.1"]
[White "Frenzee feb08 32-bit"]
[Black "BugChess2 1.6.3 32-bit"]
[Result "0-1"]
[TimeControl "40/50"]
[Annotator "11. +0.09 11... -0.06"]
[Number "1"]

1. e4 c5 2. Nf3 d6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 a6 6. Be3 e6 7. f3 b5 8.
g4 h6 9. Qd2 Nbd7 10. O-O-O Bb7 11. h4 {+0.09/9 2} b4 {-0.06/10 1} 12.
Nce2 {-0.23/10 1} d5 {+0.03/9 1} 13. Ng3 {-0.23/10 1} dxe4 {-0.03/8 0} 14.
Nxe4 {-0.57/10 2} Nxe4 {+0.61/9 0} 15. fxe4 {-0.57/11 0} Bxe4 {+0.47/9 1}
16. Rg1 {-0.51/10 1} Bd5 {+0.40/8 1} 17. Kb1 {-0.46/9 1} Ne5 {+0.61/9 1}
18. Qh2 {-0.36/9 1} Bd6 {+0.41/10 2} 19. Nf5 {-0.14/9 2} Bb8 {+0.61/9 2}
20. Qe2 {-0.14/10 1} g6 {+0.62/9 0} 21. Nd4 {-0.01/9 1} Qxh4 {+0.61/9 0}
22. Bg2 {-0.48/10 2} Qh2 {+0.38/9 0} 23. Rdf1 {-0.48/10 3} g5 {+1.05/9 0}
24. Nf3 {-0.37/8 1} Nxf3 {+0.58/10 1} 25. Rxf3 {-0.30/10 1}
Ba7 {+0.49/10 0} 26. Rh3 {-0.09/10 0} Qe5 {+0.50/10 1} 27. Rf1 {-0.40/10 2}
O-O {+0.33/10 1} 28. Bxd5 {-0.28/10 1} exd5 {+0.44/11 1} 29.
Qd3 {-0.41/10 1} Bxe3 {+0.59/11 1} 30. Rxe3 {-0.41/11 1} Qd6 {+0.73/11 1}
31. Ref3 {-0.83/9 1} Rae8 {+0.73/11 1} 32. Qd1 {-0.85/10 1} f6 {+0.94/10 1}
33. R3f2 {-0.85/10 1} Kg7 {+1.40/10 1} 34. Ka1 {-0.99/10 1}
Re5 {+1.57/10 1} 35. Kb1 {-1.65/10 2} Qe6 {+1.59/11 1} 36. Kc1 {-1.83/10 1}
a5 {+1.72/11 2} 37. Kb1 {-2.28/10 1} f5 {+1.83/11 1} 38. Qd4 {-2.18/11 1}
Kg8 {+1.81/10 0} 39. Qd1 {-2.56/11 1} f4 {+1.99/11 1} 40. Kc1 {-3.23/11 2}
Re3 {+2.19/11 1} 41. Rd2 {-4.06/10 2} d4 {+2.08/10 0} 42. b3 {-4.27/11 6}
Rg3 {+1.93/10 0} 43. Rh2 {-4.26/11 4} d3 {+2.39/9 0} 44. cxd3 {-3.69/10 2}
Rd8 {+2.05/9 0} 45. Re1 {-3.64/10 1} Qa6 {+2.27/9 1} 46. Rc2 {-3.38/9 1}
a4 {+2.46/9 0} 47. bxa4 {-3.70/8 1} Qxa4 {+2.39/9 0} 48. Kb1 {-3.95/9 1}
b3 {+2.05/8 0} 49. axb3 {-4.20/11 1} Qxb3+ {+2.46/10 0} 50.
Rb2 {-4.63/12 1} Qxd1+ {+2.54/10 0} 51. Rxd1 Rxg4 {+2.58/11 0} 52.
Rh1 {-4.63/10 1} Rd6 {+2.59/10 0} 53. Rb7 {-4.63/10 1} f3 {+2.77/11 1} 54.
Kc2 {-5.35/9 4} Rgd4 {+2.90/11 0} 55. Ra1 {-5.45/9 2} Rd7 {+2.95/11 0} 56.
Rxd7 {-5.53/9 0} Rxd7 {+3.03/12 0} 57. Kd2 {-5.53/10 0} Re7 {+3.46/12 0}
58. Ra8+ {-8.74/9 0} Kf7 {+3.61/13 0}
{White resigned} 0-1

[Event "debug"]
[Site "SZOTS"]
[Date "2008.11.05"]
[Round "1.1"]
[White "BugChess2 1.6.3 32-bit"]
[Black "Frenzee feb08 32-bit"]
[Result "1-0"]
[TimeControl "40/50"]
[Annotator "11. +0.09 11... -0.16"]
[Number "2"]

1. e4 e5 2. Nf3 Nc6 3. Bb5 a6 4. Ba4 Nf6 5. O-O Be7 6. Re1 b5 7. Bb3 d6 8.
c3 O-O 9. h3 Nb8 10. d4 Nbd7 11. a4 {+0.09/8 0} bxa4 {-0.16/9 1} 12.
Bxa4 {+0.36/10 1} c5 {-0.35/10 4} 13. dxe5 {+0.23/8 0} Nxe5 {-0.07/10 1}
14. Nxe5 {+0.28/9 0} dxe5 {-0.13/9 0} 15. Nd2 {+0.23/10 0} Qc7 {-0.21/11 1}
16. Bb3 {+0.25/10 0} Bb7 {+0.24/10 1} 17. Qe2 {+0.28/10 1}
Rfd8 {-0.18/10 1} 18. Qc4 {+0.26/10 0} Bd6 {+0.19/11 1} 19.
Rd1 {+0.24/10 0} h6 {+0.34/9 1} 20. f3 {+0.32/10 0} Rab8 {+0.04/10 1} 21.
Qa4 {+0.52/10 1} Bc6 {-0.16/10 2} 22. Qa2 {+0.57/11 1} Bf8 {-0.16/10 1} 23.
Nc4 {+0.53/11 1} Rxd1+ {+0.00/11 1} 24. Bxd1 {+0.55/11 0} Rd8 {+0.00/12 1}
25. Ne3 {+0.51/10 1} Qb8 {-0.53/9 2} 26. Qxa6 {+1.15/11 1} Bb5 {-0.67/10 1}
27. Qa2 {+1.16/12 2} Bc6 {-0.92/10 1} 28. Bb3 {+1.42/11 1} Qc7 {-1.23/10 2}
29. Qa7 {+1.45/11 1} Bb7 {-1.30/11 1} 30. Qa5 {+1.53/12 1} Qb8 {-1.30/11 1}
31. Nc4 {+1.68/11 1} Be7 {-1.30/10 1} 32. Qa2 {+1.65/11 0} Bf8 {-1.29/9 1}
33. Na5 {+1.80/12 1} Rd7 {-1.61/10 0} 34. Nxb7 {+1.83/11 0}
Qxb7 {-1.70/12 1} 35. Be3 {+1.82/12 1} Rd3 {-1.64/12 1} 36.
Kf2 {+1.89/12 3} Rd8 {-1.74/11 1} 37. Bc4 {+1.88/11 0} Ne8 {-1.92/11 1} 38.
Kg1 {+1.79/10 1} Qc7 {-1.89/12 1} 39. Qa7 {+1.95/10 0} Rd1+ {-1.74/12 1}
40. Kh2 {+1.99/12 0} Rxa1 41. Qxa1 {+1.99/12 0} Nd6 {-1.82/15 2} 42.
Be2 {+2.00/11 0} Kh7 {-2.19/13 8} 43. Qa8 {+2.12/11 0} Kg8 {-2.49/12 1} 44.
Ba6 {+2.14/12 0} g5 {-2.45/12 1} 45. h4 {+2.31/12 0} c4 {-2.48/11 1} 46.
hxg5 {+2.59/12 0} hxg5 {-2.80/12 1} 47. Bxg5 {+3.16/12 0} Kg7 {-2.80/13 1}
48. Qd8 {+3.45/12 0} Qxd8 {-2.46/14 1} 49. Bxd8 {+3.40/13 0}
f6 {-2.46/16 1} 50. Bc7 {+3.54/15 1} Kf7 {-2.51/14 1} 51. Bxd6 {+3.57/14 0}
Bxd6 {-2.51/15 0} 52. Bxc4+ {+3.63/14 0} Ke8 {-2.70/17 1} 53.
b4 {+3.75/17 1} Kd7 {-2.70/17 1} 54. Bd5 {+3.75/15 0} Be7 {-3.16/15 1} 55.
Kg3 {+3.83/16 0} Bf8 {-3.16/16 3} 56. Kg4 {+3.83/16 0} Bh6 {-4.05/16 1} 57.
Kf5 {+4.30/16 0} Ke7 {-4.05/16 0} 58. b5 {+4.27/18 0} Be3 {-4.41/17 0} 59.
g4 {+4.45/18 0} Bf2 {-4.26/15 0} 60. Bb3 {+4.50/16 0} Be3 {-4.26/14 0} 61.
Be6 {+4.85/18 0} Bb6 {-4.26/15 0} 62. g5 {+4.85/18 0} fxg5 {-4.26/16 0} 63.
Kxe5 {+4.88/18 0} Ba5 {-4.78/14 2} 64. c4 {+4.88/18 0} Kd8 {-6.83/12 0} 65.
Kd6 {+5.03/16 0} Bc7+ {-7.82/13 0} 66. Kc6 {+6.24/19 0}
{Black resigned} 1-0

[Event "debug"]
[Site "SZOTS"]
[Date "2008.11.05"]
[Round "1.2"]
[White "BugChess2 1.6.3 32-bit"]
[Black "Matacz 1.4"]
[Result "1-0"]
[TimeControl "40/50"]
[Annotator "11. +0.27 11... -0.30"]
[Number "3"]

1. e4 c5 2. Nf3 d6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 a6 6. Be3 e6 7. f3 b5 8.
g4 h6 9. Qd2 Nbd7 10. O-O-O Bb7 11. Be2 {+0.27/8 0} d5 {-0.30/10 1} 12.
exd5 {+0.19/9 0} Nxd5 {-0.33/11 1} 13. Nxd5 {+0.19/9 0} Bxd5 {-0.26/10 1}
14. Kb1 {+0.25/10 2} h5 {-0.31/10 1} 15. gxh5 {+0.72/9 1} Rxh5 {+0.05/10 1}
16. f4 {+0.59/9 0} Rh7 {-0.18/10 1} 17. Bf3 {+0.59/9 0} Nf6 {-0.29/10 1}
18. Qg2 {+0.91/9 2} Qd7 {-0.19/9 1} 19. f5 {+0.89/9 1} O-O-O {-0.42/9 1}
20. fxe6 {+0.78/8 0} fxe6 {-0.38/10 1} 21. Rd3 {+0.75/9 0} e5 {-0.01/9 1}
22. Bxd5 {+1.10/9 1} Qxd5 {-0.53/10 1} 23. Nf3 {+1.33/10 0}
Qe4 {-0.83/10 1} 24. Rxd8+ {+1.61/9 0} Kxd8 {-1.20/12 1} 25.
Rd1+ {+1.71/10 1} Ke8 {-0.91/11 1} 26. Qg3 {+1.49/10 0} Qf5 {-0.95/10 1}
27. Nxe5 {+2.00/10 1} Ne4 {-1.98/10 1} 28. Qg2 {+1.88/10 1}
Bd6 {-1.94/10 1} 29. Ng4 {+1.88/10 0} Be7 {-2.82/10 1} 30. Rd4 {+2.09/9 1}
Nf6 {-2.52/11 1} 31. Nxf6+ {+2.19/9 1} Bxf6 {-2.53/12 1} 32.
Rf4 {+2.23/10 1} Qh5 {-2.77/11 1} 33. Qc6+ {+2.39/9 1} Kf7 {-2.72/12 1} 34.
Qd7+ {+2.03/10 1} Kg8 {-2.54/13 1} 35. Qc8+ {+1.85/10 2} Kf7 {-2.32/11 1}
36. Qd7+ {+2.03/10 0} Kg8 {-2.70/13 1} 37. Rf1 {+1.91/9 0} Rh8 {-2.30/11 1}
38. Rf5 {+2.07/10 0} Qf7 {-2.27/13 1} 39. Qc8+ {+1.97/10 0}
Qf8 {-2.08/13 1} 40. Qxa6 {+1.95/11 0} Qe8 {-2.02/12 1} 41. a3 {+1.93/10 0}
Qxe3 {-1.74/11 1} 42. Qc8+ {+1.96/11 0} Kf7 {-1.92/14 1} 43.
Rxf6+ {+1.98/11 0} gxf6 {-1.93/14 1} 44. Qxh8 {+1.96/11 0} f5 {-1.93/13 1}
45. h4 {+2.06/8 0} Qg1+ {-1.93/11 1} 46. Ka2 Qc5 {-1.98/10 1} 47.
Qh5+ {+2.00/10 0} Kf8 {-2.02/11 1} 48. Qd1 {+2.01/11 1} Qc4+ {-2.02/9 1}
49. Kb1 {+2.16/11 1} Qxh4 {-2.11/10 1} 50. Qf1 {+2.24/11 0}
Qc4 {-2.16/10 1} 51. Qxf5+ {+2.29/10 0} Kg7 {-2.26/11 1} 52.
c3 {+2.47/10 0} Qe2 {-2.14/11 1} 53. Qd5 {+2.46/11 1} Kf6 {-2.15/12 1} 54.
Ka2 {+2.57/11 0} Ke7 {-2.45/11 1} 55. Kb3 {+2.57/11 0} Kf6 {-2.60/12 1} 56.
c4 {+2.65/10 1} Qe3+ {-2.75/12 1} 57. Ka2 {+2.71/12 0} bxc4 {-2.80/12 1}
58. Qxc4 {+2.78/12 1} Ke7 {-2.85/11 1} 59. a4 {+2.73/11 0} Qe5 {-2.85/9 1}
60. b4 {+3.04/10 0} Qh2+ {-3.26/9 1} 61. Kb3 {+3.01/11 0} Qh1 {-3.36/9 1}
62. Qd3 {+3.17/10 0} Qh5 {-3.92/10 1} 63. a5 {+3.41/11 0} Qf7+ {-4.06/11 1}
64. Ka3 {+3.44/11 0} Qf6 {-4.09/10 1} 65. a6 {+3.52/10 0} Qf2 {-4.12/10 1}
66. b5 {+3.79/10 0} Qb6 {-4.42/10 1} 67. Kb4 {+3.85/10 1} Qf2 {-5.00/9 1}
68. a7 {+9.78/10 0} Qf4+ {-12.50/9 2} 69. Kb3 {+12.00/11 1}
Qf7+ {-12.51/9 1} 70. Ka3 {+12.10/11 0} Qf6 {-50.37/12 1}
{Xboard adjudication} 1-0

[Event "debug"]
[Site "SZOTS"]
[Date "2008.11.05"]
[Round "1.2"]
[White "Matacz 1.4"]
[Black "BugChess2 1.6.3 32-bit"]
[Result "0-1"]
[TimeControl "40/50"]
[Number "4"]

1. e4 e5 2. Nf3 Nc6 3. Bb5 a6 4. Ba4 Nf6 5. O-O Be7 6. Re1 b5 7. Bb3 d6 8.
c3 O-O 9. h3 Nb8 10. d4 Nbd7
{Error: first chess program ("Matacz_14.exe" ) exited unexpectedly} 0-1

[Event "debug"]
[Site "SZOTS"]
[Date "2008.11.05"]
[Round "1.3"]
[White "RomiChess P3k 32-bit"]
[Black "BugChess2 1.6.3 32-bit"]
[Result "0-1"]
[TimeControl "40/50"]
[Annotator "11. +0.41 11... -0.06"]
[Number "5"]

1. e4 c5 2. Nf3 d6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 a6 6. Be3 e6 7. f3 b5 8.
g4 h6 9. Qd2 Nbd7 10. O-O-O Bb7 11. a3 {+0.41/10 1} d5 {-0.06/9 1} 12.
g5 {+0.24/11 1} hxg5 {+0.13/9 0} 13. exd5 {-0.04/12 2} Nxd5 {+0.15/10 1}
14. Bxg5 {+0.26/12 2} Qc7 {+0.12/9 1} 15. Ne4 {+0.28/12 2} b4 {+0.59/9 1}
16. axb4 {+0.17/11 1} e5 {+0.38/9 1} 17. Qe2 {+0.48/11 1} Nxb4 {+0.27/9 1}
18. Nf5 {+0.38/11 2} Bd5 {+0.43/9 1} 19. c4 {+0.64/10 1} Rb8 {+1.15/8 0}
20. Kb1 {+0.42/11 1} Bxe4+ {+1.22/8 1} 21. Qxe4 {-0.56/12 1}
Nc5 {+1.67/10 1} 22. Qe2 {-0.29/12 2} Na4 {+2.09/9 2} 23. Qg2 {-0.23/11 2}
Nc6 {+2.10/10 5} 24. Rd2 {-1.45/11 1} Nc3+ {+3.95/9 1} 25. Kc2 {-3.04/12 1}
Qb6 {+1.85/8 0} 26. Kd3 {-3.61/12 1} e4+ {+119.98/9 0} 27.
fxe4 {-327.56/11 1}
{White resigns} 0-1

[Event "debug"]
[Site "SZOTS"]
[Date "2008.11.05"]
[Round "1.3"]
[White "BugChess2 1.6.3 32-bit"]
[Black "RomiChess P3k 32-bit"]
[Result "1/2-1/2"]
[TimeControl "40/50"]
[Annotator "11. +0.09 11... +0.41"]
[Number "6"]

1. e4 e5 2. Nf3 Nc6 3. Bb5 a6 4. Ba4 Nf6 5. O-O Be7 6. Re1 b5 7. Bb3 d6 8.
c3 O-O 9. h3 Nb8 10. d4 Nbd7 11. a4 {+0.09/8 0} Bb7 {+0.41/11 1} 12.
Nbd2 {+0.00/8 0} h6 {+0.40/11 2} 13. Qe2 {+0.33/9 1} Rb8 {+0.57/11 2} 14.
axb5 {+0.42/9 1} axb5 {+0.03/12 2} 15. Rd1 {+0.33/9 1} Re8 {+0.44/11 1} 16.
Bc2 {+0.49/7 0} Bf8 {+0.12/10 2} 17. dxe5 {+0.47/8 1} Nxe5 {-0.26/12 1} 18.
Nxe5 {+0.26/9 0} Rxe5 {-0.07/11 1} 19. f3 {+0.32/10 1} Ra8 {-0.45/11 2} 20.
Rxa8 {+0.44/9 0} Qxa8 {-0.48/13 1} 21. Nb3 {+0.43/10 0} d5 {-0.63/12 2} 22.
Qxb5 {+1.02/9 1} Qa7+ {-0.76/11 1} 23. Nd4 {+1.13/10 2} c5 {-0.46/13 1} 24.
Nc6 {+0.79/10 0} Bxc6 {-0.35/14 1} 25. Qxc6 {+0.97/10 0} dxe4 {-0.62/13 1}
26. fxe4 {+0.64/10 0} Re6 {-0.91/13 1} 27. Qa4 {+0.65/10 1}
Qxa4 {-0.25/14 1} 28. Bxa4 {+0.63/11 0} Nxe4 {-0.26/14 1} 29.
Bb3 {+0.74/12 1} Rb6 {-0.28/14 1} 30. Bc2 {+0.72/11 0} Re6 {-0.35/14 1} 31.
Bb3 {+0.67/12 0} Re7 {-0.52/13 1} 32. Bc4 {+0.83/11 1} Re5 {-0.73/13 1} 33.
Re1 {+0.79/11 0} Bd6 {-0.60/13 1} 34. Bd3 {+0.76/13 1} Ng5 {-0.40/13 1} 35.
Rxe5 {+0.81/13 0} Bxe5 {-0.72/15 1} 36. Kf2 {+0.83/14 0} Ne6 {-0.74/14 0}
37. g3 {+0.76/14 1} Kf8 {-0.74/12 0} 38. Be3 {+0.79/14 0} Ke7 {-0.32/14 0}
39. h4 {+0.78/14 2} Kd7 {-0.44/12 0} 40. Kf3 {+0.79/14 0} Kc6 {-0.37/13 0}
41. Bc4 {+0.79/14 0} h5 {-0.81/15 1} 42. Bg1 {+0.80/13 0} Bd6 {-0.53/15 1}
43. Bf2 {+0.79/13 1} g6 {-0.56/14 1} 44. Ke4 {+0.80/12 0} f5+ {-0.53/15 1}
45. Ke3 {+0.78/11 0} Nf8 {-0.36/15 1} 46. Bf7 {+0.66/13 0} Kd7 {-0.66/15 1}
47. Be1 {+0.70/13 0} Ke7 {-0.85/15 1} 48. Bd5 {+0.71/13 0} Nd7 {-0.57/15 1}
49. b4 {+0.59/12 0} Be5 {-0.52/15 1} 50. b5 {+0.68/11 0} Nb6 {-0.45/15 1}
51. Ba2 {+0.45/14 1} c4 {-0.85/15 1} 52. Kf3 {+0.57/15 0} Kd6 {-0.92/16 1}
53. Bf2 {+0.46/15 0} Nd7 {-1.18/17 1} 54. Bxc4 {+0.72/13 0}
Bxc3 {-1.18/16 1} 55. Ba2 {+0.68/13 0} Bf6 {-1.18/14 1} 56.
Ke2 {+0.76/13 0} Bc3 {-1.18/15 1} 57. Ba7 {+0.76/13 0} Bb2 {-1.20/14 1} 58.
Bf7 {+0.76/12 0} Nf8 {-1.48/15 1} 59. Bf2 {+0.85/10 0} Bf6 {-1.48/14 1} 60.
Kd2 {+0.88/13 1} Bd8 {-1.22/13 1} 61. b6 {+0.93/12 0} Kc6 {-0.84/16 1} 62.
Be8+ {+0.94/13 0} Kb7 {-1.36/18 1} 63. Bc5 {+0.94/13 0} Ne6 {-1.38/18 1}
64. Bd6 {+0.78/13 0} Bxb6 {-1.00/15 1} 65. Bxg6 {+0.80/12 0}
Kc6 {-1.50/14 1} 66. Be5 {+0.79/13 0} Kd5 {-1.08/14 1} 67. Bb8 {+0.80/13 0}
Ng7 {-1.10/15 1} 68. Bf7+ {+0.81/14 0} Ke4 {-1.04/17 1} 69.
Bf4 {+0.80/15 1} Bd4 {-0.91/17 1} 70. Ke2 {+0.66/14 0} Bf6 {-1.10/17 1} 71.
Bc4 {+0.75/14 2} Be5 {-1.00/14 1} 72. Bd3+ {+0.75/14 0} Kd4 {-0.71/15 1}
73. Be3+ {+0.76/14 1} Kc3 {-0.78/16 0} 74. Bf2 {+0.84/14 1}
Bc7 {-0.80/16 1} 75. Be1+ {+0.83/14 0} Kd4 {-0.86/14 0} 76.
Bc2 {+0.79/14 1} f4 {-0.94/14 0} 77. gxf4 {+0.71/14 2} Bxf4 {-0.57/14 0}
78. Kf3 {+0.68/14 0} Bc1 {-0.52/14 0} 79. Bg6 {+0.72/14 1} Ba3 {-0.41/13 0}
80. Be4 {+0.71/14 1} Bd6 {-0.30/13 0} 81. Bh7 {+0.68/14 0} Ne8 {-0.61/14 1}
82. Bg6 {+0.67/14 0} Ng7 {-0.39/15 1} 83. Bh7 {+0.71/13 1} Ne8 {-0.17/17 1}
84. Bg6 {+0.69/13 0} Ng7 {-0.16/18 1} 85. Bd2 {+0.76/12 0} Be7 {-0.34/15 1}
86. Kg3 {+0.49/13 0} Bd6+ {-1.36/15 1} 87. Kg2 {+0.57/14 0}
Be5 {-1.36/15 1} 88. Kf3 {+0.72/12 0} Bf6 {-0.46/15 1} 89. Be1 {+0.54/15 0}
Bd8 {-0.36/15 1} 90. Bc2 {+0.59/13 1} Ne6 {-0.42/14 1} 91. Bg6 {+0.70/14 0}
Ng7 {-0.09/17 1} 92. Be4 {+0.60/13 1} Be7 {-0.34/15 1} 93. Bc2 {+0.69/14 1}
Ba3 {-0.44/14 1} 94. Bb1 {+0.70/13 1} Ne6 {-0.40/14 1} 95. Bg6 {+0.83/12 0}
Ng7 {-0.05/18 1} 96. Bd2 {+0.70/12 0} Be7 {-0.04/18 1} 97. Kg3 {+0.39/14 1}
Bd6+ {-0.03/17 1} 98. Kg2 {+0.57/14 1} Be5 {-0.02/17 1} 99.
Bc2 {+0.53/13 1} Ne6 {-0.14/14 1} 100. Bf5 {+0.54/13 0} Kd5 {-0.16/15 1}
101. Bb1 {+0.53/12 0} Bb2 {-0.45/13 1} 102. Ba2+ {+0.55/12 0}
Kd6 {-0.14/14 1} 103. Kf3 {+0.55/12 1} Ke5 {-0.16/15 1} 104.
Kg3 {+0.51/12 0} Kd6 {-0.26/14 1} 105. Bb4+ {+0.55/13 0} Ke5 {-0.09/14 0}
106. Bc4 {+0.54/12 0} Ng7 {-0.14/14 1} 107. Kh3 {+0.45/13 0}
Ke4 {-0.13/15 1} 108. Bb3 {+0.41/13 0} Ke5 {-0.12/14 1} 109.
Ba2 {+0.49/11 0} Bd4 {-0.13/15 1} 110. Bf7 {+0.01/14 1} Kf6 {-0.11/15 1}
111. Bd5 {+0.02/14 0} Ke5 {-0.13/15 1} 112. Bf7 {+0.01/15 1}
Kf6 {+0.12/18 1} 113. Bd5 {+0.02/15 0} Ke5 {+0.13/17 1} 114.
Bc4 {+0.01/14 0} Ne6 {-0.15/14 0} 115. Be1 {+0.19/13 1} Ng7 {-0.17/13 0}
116. Be2 {+0.15/14 2} Ke4 {-0.14/13 0} 117. Bd1 {+0.12/14 0}
Be5 {-0.17/13 0} 118. Bb3 {+0.01/15 2} Kf5 {-0.15/13 0} 119.
Kg2 {+0.13/13 1} Ne6 {-0.13/12 0} 120. Ba4 {+0.08/13 1} Ke4 {+0.14/11 0}
121. Bc6+ {+0.05/12 0} Ke3 {+0.05/15 1} 122. Bf2+ {+0.03/10 0}
Kf4 {-0.13/16 1} 123. Bd5 {+0.00/13 1} Ng7 {+0.10/15 1} 124.
Kh3 {+0.00/15 0} Nf5 {+0.27/16 1} 125. Bb6 {+0.00/17 0} Ne7 {+0.27/17 1}
126. Bf7 {+0.00/19 0} Nc8 {+0.27/18 1} 127. Bc5 {+0.00/22 0}
Nb6 {+0.27/60 0}
{Romi claims, draw by 50 move rule} 1/2-1/2

Gábor Szots wrote:I have just run a small test gauntlet which I stopped after 6 games because of a problem with one of the engines.
It seems to me that PSWBTM sends "-lgi 1" to WB which makes it start with game 1 for each pairing. The rewind box was unticked.

This is also how it behaves with me, and how I always thought it was supposed to behave. (Remember that originally, the "rewind" parameter was not implemented.)

But I agree that now te "rewind" option exists, it would be much more versatile to simply continuing incrementng the loadGameIndex across pairings, and only reset it when the "rewind" value gets exceeded. This still would allow you to simulate the original mode, by putting the rewind equal to the number of games per pairing (or half that if you play each position twice).

Question to Pradu:

Why are you specifying the -xreuse ? If you play a match of only one game, it should not have any effect if you reuse the engine process or terminate it, as there is never a second use.

Second request:

Could you replace the "-mg 1" by "-mm"? This has the same effect, (putting WinBoard in matchMode with 1 game per match), but would allow the user to overrule the number of matchGames in the .ini, and is thus more versatile.

H.G.Muller wrote:

Gábor Szots wrote:I have just run a small test gauntlet which I stopped after 6 games because of a problem with one of the engines.
It seems to me that PSWBTM sends "-lgi 1" to WB which makes it start with game 1 for each pairing. The rewind box was unticked.

This is also how it behaves with me, and how I always thought it was supposed to behave. (Remember that originally, the "rewind" parameter was not implemented.)

But I agree that now te "rewind" option exists, it would be much more versatile to simply continuing incrementng the loadGameIndex across pairings, and only reset it when the "rewind" value gets exceeded. This still would allow you to simulate the original mode, by putting the rewind equal to the number of games per pairing (or half that if you play each position twice).

Question to Pradu:

Why are you specifying the -xreuse ? If you play a match of only one game, it should not have any effect if you reuse the engine process or terminate it, as there is never a second use.

Second request:

Could you replace the "-mg 1" by "-mm"? This has the same effect, (putting WinBoard in matchMode with 1 game per match), but would allow the user to overrule the number of matchGames in the .ini, and is thus more versatile.

There is also a "-inc=-1" parameter the meaning of which I could not find.
I remember you have created a documentation for WinBoard F, where can I find it?

Regards,

The "-inc -1" s the time increment (spcifying it is not an incemental TC), which has nothing to do with what you remember.

What I added in the alpha version for 4.3.15 was "-lgi=-1" to specify auto-incrementing of the game index, so that WinBoard in matchMode would play all games starting with a different opening from the loadGameFile. (while "-lgi=-2" would cause each opening to be played twice.) I even added a "-rewindFile N" option for this, which would reset the gameIndex (or positionIndex) if it exceeded N.

The docs for this does not yet exist, as 4.3.15 is yet an unreleased alpha version, and docs are usualy what I make last...

I am not sure this WB feature is useful in combination with PSWBTM, as PSWBTM is perfectly capable of stepping the games / positions by itself. They were meant for stand-alone matchMode. OTOH, peple that want to play ultra-fast games might not want the overhead of PSWBTM starting up a new WinBoard (and thus necessarily new engine processes). As a kludge they could replace every single game that PSWBTM orders by a match of more games (using the option "/sameColorGames N" in the winboard.ini), reusing the engine pocesses, with a very small matchPause). In that case the optons might be useful. You woud need a kludge to overrule the value for -lgi that PSWBTM passed, though (to make it -1). So perhaps it is better if I would actually not specify the autoIncrement by negative value of the -lgi option, but in stead use a separate -autoInc option, which woud make WinBoard ignore the value of -lgi that was actually given. Then the overrule is automatic, if you use tht option in the .ini, as PSWBTM woud not set it. (Normally, what PSWBTM gives on the WB command line would overrule what is in the .ini.)

Why are you specifying the -xreuse ? If you play a match of only one game, it should not have any effect if you reuse the engine process or terminate it, as there is never a second use.

Indeed it should not have any effect. I just put early in the design in case I changed around PSWBTM to have it play multiple games per Winboard launch but this didn't happen. I could remove it.

Could you replace the "-mg 1" by "-mm"? This has the same effect, (putting WinBoard in matchMode with 1 game per match), but would allow the user to overrule the number of matchGames in the .ini, and is thus more versatile.

I thought I had done this already but I guess I didn't. I'll change it to use -mm.

Gábor Szots wrote:It seems to me that PSWBTM sends "-lgi 1" to WB which makes it start with game 1 for each pairing.

This is the intended behavior. I believe that if you use a PGN/FEN/EPD for starting positions, you are trying to create a test that is a bit more deterministic. If so, then for all pairings you should use the same starting positions.

Pradu wrote:

Gábor Szots wrote:It seems to me that PSWBTM sends "-lgi 1" to WB which makes it start with game 1 for each pairing.

This is the intended behavior. I believe that if you use a PGN/FEN/EPD for starting positions, you are trying to create a test that is a bit more deterministic. If so, then for all pairings you should use the same starting positions.

Whenever I start a new gauntlet I provide a new, different PGN file to be used as opening book. So what I try to do is just the opposite: I try to make opening choice as random as possible.
What I would like to ask for is a option that I can play (for example) positions 1 to 10 against the first opponent (with or without alternating the colours during the match), then positions 11 to 20 against the second one, and so on. For a 5-opponent gauntlet with 10 games against each opponent I would provide a 50-game PGN file.

Note that I would like to keep the current behaviour as well.

Thanks for your efforts,

Code: Select all

[b]Opponent Games

Match 1 Hiarcs                      Positions 1-10

Match 2 Chess Tiger              Positions 11-20

Match 3 Deep Shredder          Positions 21-30

Match 4 Deep Junior UCI        Positions 31-40

Match 5 Fruit                        Positions 41-50[/b]


Hello Pradu, H.G.,

Above is a tournament which I wish to play.

Can it be done exactly this way using Winboard with/without PSWBTM?

Any answer would be appreciated.

Later.

Roger Brown wrote:

Code: Select all

[b]Opponent Games

Match 1 Hiarcs                      Positions 1-10

Match 2 Chess Tiger              Positions 11-20

Match 3 Deep Shredder          Positions 21-30

Match 4 Deep Junior UCI        Positions 31-40

Match 5 Fruit                        Positions 41-50[/b]


Hello Pradu, H.G.,

Above is a tournament which I wish to play.

Can it be done exactly this way using Winboard with/without PSWBTM?

Any answer would be appreciated.

Later.

Hi Roger,

Yes this can be done via a batch file that would play all those matches
with the desired positions consecutively. You must wait though until I do
some research on my HD, because my knowledge about it is a bit rusty meanwhile.
A hint for now, Uri once showed a batch either here or in
CCC which contained the command for looping through a certain pgn
respectively position file which I had just modified and extended
for several matches in a row, but I have no idea how long ago that was.

Regards,
Guenther

Well an extensive example with unbelievable long PGN file/debug names
I only modified the loop through the pgn file out of lazyness, but I guess
you could deduce the idea how it should work from here?
Note that each FOR should be a single line in the batch but for some reason it is displayed different here. You have to modify for your
time control and your engine and paths and number of games(mg)
per position. I even included the variable in the debug name I see.
One note if you have an engines exe with a command line needed it must be set between '...' e.g. scp='x.exe xboard'.
I hope that helps.

Regards,
Guenther

Code: Select all

FOR %%P IN (1,2,3,4,5,6,7,8,9,10) DO start /w winboard.exe /mg=2 /tc=5 /mps=40 /debug /nameOfDebugFile M%%PKtulu_75-Rybka10b-32_GS34a.deb  /autoCallFlag true /ponderNextMove false /cp /lgf GS34.pgn /lgi %%P /fcp=Ktulu_75.exe /fd=C:\WB\General\Ktulu_75 /scp=Rybka_10b-32.exe /sd=C:\WB\General\Rybka_10b-32 /sgf=.\Test\Ktulu_75-Rybka_10b-32_GS34a.pgn
FOR %%P IN (11,12,13,14,15,16,17,18,19,20) DO start /w winboard.exe /mg=2 /tc=5 /mps=40 /debug /nameOfDebugFile M%%PKtulu_75-Rybka10b-32_GS34b.deb  /autoCallFlag true /ponderNextMove false /cp /lgf GS34.pgn /lgi %%P /fcp=Ktulu_75.exe /fd=C:\WB\General\Ktulu_75 /scp=Rybka_10b-32.exe /sd=C:\WB\General\Rybka_10b-32 /sgf=.\Test\Ktulu_75-Rybka_10b-32_GS34b.pgn

Pradu wrote:

Gábor Szots wrote:It seems to me that PSWBTM sends "-lgi 1" to WB which makes it start with game 1 for each pairing.

This is the intended behavior. I believe that if you use a PGN/FEN/EPD for starting positions, you are trying to create a test that is a bit more deterministic. If so, then for all pairings you should use the same starting positions.

Pradu, I know that this originally was intended behavior. But now that the rewind option exists, wouldn't it be more versatile to just let the game index run across pairings, like the gauntlet Roger proposes? The other behavior can always be reproduced by setting rewind = nr of games per pairing. But what Roger wanted in his example, can not be achieved with the current control logic.