Works great Ferdinand !!
found src.pgn !!
Game: 67750
reading pgn time: 42.68s
A. Drawn (1/2-1/2) games:
sub_total = 39501
B.1. White wins by mate (1-0):
sub_total = 3040
B.2. White wins by non-mate (1-0):
sub_total = 14306
C.1. Black wins by mate (0-1):
sub_total = 2033
C.2. Black wins by non-mate (0-1):
sub_total = 8870
grand_total = 67750 game(s)
summarizing time: 27.25s
total time: 69.93s
done!!
Press enter key to exit
I need a new tool ... maybe for Ferdinand / Norm!?
Moderator: Ras
-
jpqy
- Posts: 556
- Joined: Thu Apr 24, 2008 9:31 am
- Location: Belgium
-
Norm Pollock
- Posts: 1087
- Joined: Thu Mar 09, 2006 4:15 pm
- Location: Long Island, NY, USA
Re: I need a new tool ... maybe for Ferdinand / Norm!?
Hi Frank,
What's the point of getting games ending with mate because so many games end 1 or 2 moves before an inevitable mate? And especially computer games that end after a few repetitions of a +9.00 evaluation (or equivalent).
The new tools seem to work great.
With regard to my tools, I would first separate the file by 1-0, 1/2-1/2, 0-1 using "resultSplit".
Then I would use "joined" by Andreas Stable, available at the bottom of my page. Actually I would use the batch file "join.bat" as follows:
join filename.pgn
output goes to the monitor and to a text file name "j1".
"joined" is a very advanced data producer. It even gives you the number of stalemates, # moves by each piece, etc. Here is a sample output:
What's the point of getting games ending with mate because so many games end 1 or 2 moves before an inevitable mate? And especially computer games that end after a few repetitions of a +9.00 evaluation (or equivalent).
The new tools seem to work great.
With regard to my tools, I would first separate the file by 1-0, 1/2-1/2, 0-1 using "resultSplit".
Then I would use "joined" by Andreas Stable, available at the bottom of my page. Actually I would use the batch file "join.bat" as follows:
join filename.pgn
output goes to the monitor and to a text file name "j1".
"joined" is a very advanced data producer. It even gives you the number of stalemates, # moves by each piece, etc. Here is a sample output:
Code: Select all
Scanning file gm1830.pgn containing 11371707 bytes !
Read 15363 games with 1379074 moves, 293372 lines, 11371707 bytes !
Got max. 327 moves in game 10164 before line 193792 !
Got 42337908 legal moves in 1394437 positons !
Got max. 70 legal moves after move 44 in game 4781 line 91631 !
Total move statistics:
Pawn moves = 348189, Knight moves = 235253, Bishop moves = 212861
Rook moves = 269521, Queen moves = 169077, King moves = 171574
Check moves = 76803, Mate moves = 281, Stalemate moves = 16
Hit moves = 279343, En passant moves = 866, Pawn two moves = 92505
Short castlings = 24670, Long castlings = 2731
Promotion to queen = 1196, Promotion to rook = 8
Promotion to bishop = 2, Promotion to knight = 22
Total game statistics:
White won = 5965, Drawn = 4433, Black won = 4965
Unknown result = 0, Illegal result = 0, Conflicting results = 0
White mated = 107, White stalemated = 7
Black mated = 174, Black stalemated = 9
Number of games with different lengths:
51: 180, 52: 162, 53: 192, 54: 174, 55: 178, 56: 185
57: 205, 58: 218, 59: 300, 60: 380, 61: 375, 62: 255
63: 290, 64: 240, 65: 247, 66: 231, 67: 254, 68: 215
69: 222, 70: 232, 71: 228, 72: 214, 73: 237, 74: 213
75: 230, 76: 211, 77: 219, 78: 238, 79: 224, 80: 228
81: 228, 82: 207, 83: 209, 84: 192, 85: 178, 86: 200
87: 197, 88: 156, 89: 188, 90: 241, 91: 200, 92: 191
93: 203, 94: 164, 95: 175, 96: 153, 97: 156, 98: 177
99: 158, 100: 155, 101: 136, 102: 130, 103: 159, 104: 149
105: 130, 106: 139, 107: 141, 108: 123, 109: 127, 110: 132
111: 123, 112: 114, 113: 108, 114: 123, 115: 104, 116: 83
117: 88, 118: 110, 119: 98, 120: 114, 121: 111, 122: 94
123: 105, 124: 102, 125: 69, 126: 74, 127: 72, 128: 66
129: 70, 130: 53, 131: 76, 132: 66, 133: 51, 134: 56
135: 38, 136: 46, 137: 38, 138: 42, 139: 43, 140: 38
141: 44, 142: 35, 143: 41, 144: 29, 145: 37, 146: 34
147: 28, 148: 27, 149: 28, 150: 33, 151: 21, 152: 24
153: 24, 154: 24, 155: 20, 156: 32, 157: 18, 158: 11
159: 18, 160: 13, 161: 17, 162: 17, 163: 20, 164: 18
165: 11, 166: 13, 167: 7, 168: 6, 169: 8, 170: 15
171: 11, 172: 9, 173: 12, 174: 11, 175: 6, 176: 8
177: 8, 178: 8, 179: 8, 180: 9, 181: 5, 182: 2
183: 5, 184: 7, 185: 3, 186: 6, 187: 4, 188: 3
189: 4, 190: 4, 191: 6, 192: 8, 193: 3, 194: 1
195: 4, 196: 4, 197: 6, 198: 4, 199: 2, 200: 8
201: 3, 202: 2, 204: 3, 205: 4, 206: 1, 207: 3
208: 1, 209: 3, 210: 2, 211: 2, 212: 3, 213: 3
214: 1, 215: 1, 216: 1, 217: 3, 218: 1, 219: 1
221: 2, 222: 1, 225: 4, 226: 1, 227: 1, 229: 1
230: 1, 238: 1, 240: 1, 241: 1, 242: 1, 245: 1
259: 1, 267: 1, 270: 1, 277: 2, 284: 1, 287: 1
292: 1, 318: 1, 322: 1, 327: 1
-
Ferdy
- Posts: 4856
- Joined: Sun Aug 10, 2008 3:15 pm
- Location: Philippines
Re: I need a new tool ... maybe for Ferdinand / Norm!?
It is good to know it is working, it is just a python script converted to exe file. One thing too is that it has a limit of only counting up to 700 moves in a game, I am not sure if Frank's games could reach that. But I will make an improvement in this area later.jpqy wrote:Works great Ferdinand !!
found src.pgn !!
Game: 67750
reading pgn time: 42.68s
A. Drawn (1/2-1/2) games:
sub_total = 39501
B.1. White wins by mate (1-0):
sub_total = 3040
B.2. White wins by non-mate (1-0):
sub_total = 14306
C.1. Black wins by mate (0-1):
sub_total = 2033
C.2. Black wins by non-mate (0-1):
sub_total = 8870
grand_total = 67750 game(s)
summarizing time: 27.25s
total time: 69.93s
done!!
Press enter key to exit
-
Frank Quisinsky
- Posts: 7520
- Joined: Wed Nov 18, 2009 7:16 pm
- Location: Gutweiler, Germany
- Full name: Frank Quisinsky
Re: I need a new tool ... maybe for Ferdinand / Norm!?
Hi Ferdinand,
wow!
I will try out it in the next minutes ...
Later
Frank
wow!
I will try out it in the next minutes ...
Later
Frank
-
Frank Quisinsky
- Posts: 7520
- Joined: Wed Nov 18, 2009 7:16 pm
- Location: Gutweiler, Germany
- Full name: Frank Quisinsky
Re: I need a new tool ... maybe for Ferdinand / Norm!?
Hi Norm,
with games ended with mate I can create better stats to the Topic: Tactic.
For myself:
A computer chess game, ended with resign-mode -5 or -9 ... is a half game only. Often engine have advantage with +6 and can't set mate (KNN-K) for an example, other examples ... pawn on h-a with wrong Bishop and different other examples.
Furthermore, a lot of engines have Problems to find the right mate way. Engines Need 30 moves for mate in 5 or game ended with draw after 50-moves rules (engine will not give mate). Engine lost on time 2 moves before mate ... I will see that. I will see the correct mate.
Main Problem today ...
Not checked by users, not checked by programmers because all using resing-mode for get more games!
I remember me on join, longer time not used.
I will try out it with your tools and join.
Thanks!
Best
Frank
with games ended with mate I can create better stats to the Topic: Tactic.
For myself:
A computer chess game, ended with resign-mode -5 or -9 ... is a half game only. Often engine have advantage with +6 and can't set mate (KNN-K) for an example, other examples ... pawn on h-a with wrong Bishop and different other examples.
Furthermore, a lot of engines have Problems to find the right mate way. Engines Need 30 moves for mate in 5 or game ended with draw after 50-moves rules (engine will not give mate). Engine lost on time 2 moves before mate ... I will see that. I will see the correct mate.
Main Problem today ...
Not checked by users, not checked by programmers because all using resing-mode for get more games!
I remember me on join, longer time not used.
I will try out it with your tools and join.
Thanks!
Best
Frank
-
Frank Quisinsky
- Posts: 7520
- Joined: Wed Nov 18, 2009 7:16 pm
- Location: Gutweiler, Germany
- Full name: Frank Quisinsky
Re: I need a new tool ... maybe for Ferdinand / Norm!?
Hi Ferdinand,
no, game ended with Shredder Classic GUI after 300 moves.
The reason that I replayed games, ended with 300 moves.
Short stat for yourself:
I believe (in my brain) after 69400 45-minutes FCP games 11 games ended after 300 moves. All 11 games are replayed. I will search the games and added it in my next database update.
Best
Frank
no, game ended with Shredder Classic GUI after 300 moves.
The reason that I replayed games, ended with 300 moves.
Short stat for yourself:
I believe (in my brain) after 69400 45-minutes FCP games 11 games ended after 300 moves. All 11 games are replayed. I will search the games and added it in my next database update.
Best
Frank
-
Frank Quisinsky
- Posts: 7520
- Joined: Wed Nov 18, 2009 7:16 pm
- Location: Gutweiler, Germany
- Full name: Frank Quisinsky
Re: Works perfect ... additional wish to the ECO output
Hi Ferdinand,
perfect output file!
All is fine, thats what I need!
---
For the stats to ECO codes ...
1:0, 0:1
I think a short dialog: Give number of moves for short 1-0, 0-1 games:
I give: 49
And the tool create a table as *.txt file:
move 21: 1 game, B56
move 26: 2 games, C37, 2x A87
up to move 49
Here it would be perfect if the games are collected in two files.
Means that I can find two databases ...
White database ... all games won 1-0 up to move number 49, games sorted by game number, first game in the database the earliest won game, in our example ... mate after 21 moves.
Black database ... all games won 0-1 up to move number 49
It would be great if the tool can generate for this ECO information a second output, *.txt file.
First output *.txt file ... what you have done with move stats.
Second output *.txt file ... the ECO stats.
After all ...
Nice for replaying fast won games.
Nice for checked opening books for killer lines.
Nice for creating stats to fast won games.
Best
Frank
perfect output file!
All is fine, thats what I need!
---
For the stats to ECO codes ...
1:0, 0:1
I think a short dialog: Give number of moves for short 1-0, 0-1 games:
I give: 49
And the tool create a table as *.txt file:
move 21: 1 game, B56
move 26: 2 games, C37, 2x A87
up to move 49
Here it would be perfect if the games are collected in two files.
Means that I can find two databases ...
White database ... all games won 1-0 up to move number 49, games sorted by game number, first game in the database the earliest won game, in our example ... mate after 21 moves.
Black database ... all games won 0-1 up to move number 49
It would be great if the tool can generate for this ECO information a second output, *.txt file.
First output *.txt file ... what you have done with move stats.
Second output *.txt file ... the ECO stats.
After all ...
Nice for replaying fast won games.
Nice for checked opening books for killer lines.
Nice for creating stats to fast won games.
Best
Frank
-
Frank Quisinsky
- Posts: 7520
- Joined: Wed Nov 18, 2009 7:16 pm
- Location: Gutweiler, Germany
- Full name: Frank Quisinsky
Re: Komodo vs. Stockfish ... example!
Hi Ferdinand,
with your work in step 1 I can do this one.
Example:
Stockfish and Komodo played 1.600 games vs. same opponents.
I can lay on top of each other (template) with Excel.
Move stats ...
We can compare in perhaps 5-move-steps ...
Example:
Komodo won 45 games between move number 40-45
Stockfish won 35 games between move number 40-45
and so on ...
So I can create a very nice stat we can see a lot
Can compare it with results of preview versions.
To draw ... won, lost games (lost games made no sense, to strong are this engines).

Best
Frank
In my opinion the secret is to do more with the games we have.
Your older sgbp script I havent forgot ... thinking all the time we can I used it. Have now different ideas for it and try out a lot.
with your work in step 1 I can do this one.
Example:
Stockfish and Komodo played 1.600 games vs. same opponents.
I can lay on top of each other (template) with Excel.
Move stats ...
We can compare in perhaps 5-move-steps ...
Example:
Komodo won 45 games between move number 40-45
Stockfish won 35 games between move number 40-45
and so on ...
So I can create a very nice stat we can see a lot
Can compare it with results of preview versions.
To draw ... won, lost games (lost games made no sense, to strong are this engines).
Best
Frank
In my opinion the secret is to do more with the games we have.
Your older sgbp script I havent forgot ... thinking all the time we can I used it. Have now different ideas for it and try out a lot.
-
Frank Quisinsky
- Posts: 7520
- Joined: Wed Nov 18, 2009 7:16 pm
- Location: Gutweiler, Germany
- Full name: Frank Quisinsky
Re: FCP stats ...
Correction:
13 games ...
Stats:
After 70.000 games 10 double games produced = 0,014%
After 70.000 games 221 short draw games up to move number 20 produced = 0,315%
After 70.000 games 13 draw games ended with number 300 produced = 0,017%
= 10 + 221 + 13 = 244 games = 0.348%
All these games are replayed!
I added the databases in version 2.07.s6a on my site.
Best
Frank
13 games ...
Stats:
After 70.000 games 10 double games produced = 0,014%
After 70.000 games 221 short draw games up to move number 20 produced = 0,315%
After 70.000 games 13 draw games ended with number 300 produced = 0,017%
= 10 + 221 + 13 = 244 games = 0.348%
All these games are replayed!
I added the databases in version 2.07.s6a on my site.
Best
Frank
-
Ferdy
- Posts: 4856
- Joined: Sun Aug 10, 2008 3:15 pm
- Location: Philippines
Re: I need a new tool ... maybe for Ferdinand / Norm!?
Now with eco counts. Looks to me not easy to read because there are a lot of eco codes, from a00 to a99 ... e00 to e99.
Sample.
This also takes a lot of time, because you have to look in every eco code per move number length. I plan to issue this as version 2, so it will output the game counts with eco codes.
It is not completed yet, but just showing the typical draws output.
The longest from the drawn games.
Sample.
Code: Select all
A. Drawn (1/2-1/2) games:
20 moves = 75 game(s), 2x A03, 1x A09, 1x A11, 1x A25, 1x A34, 1x A51, 1x A61, 1x A80, 1x A90, 1x B14, 1x B18, 1x B33, 2x B40, 1x B41, 1x B47, 1x B51, 1x B56, 1x B60, 1x B67, 1x B74, 1x B80, 1x B83, 1x B90, 1x B99, 1x C07, 1x C12, 1x C18, 1x C26, 1x C35, 1x C41, 1x C44, 1x C45, 2x C67, 1x C68, 1x C83, 1x C84, 1x C89, 1x C90, 2x C99, 1x D00, 1x D02, 1x D09, 1x D14, 2x D31, 1x D39, 1x D52, 1x D58, 2x D73, 1x D75, 1x D78, 1x D83, 1x D84, 3x D85, 2x D94, 1x E05, 1x E07, 1x E10, 1x E11, 1x E15, 1x E16, 1x E35, 1x E68, 1x E83, 1x E90, 1x E91, 1x E92,
[...] more even longer than the above
It is not completed yet, but just showing the typical draws output.
The longest from the drawn games.
Code: Select all
82 moves = 234 game(s), 4x A00, 1x A02, 3x A03, 3x A04, 2x A05, 3x A06, 2x A07, 1x A08, 3x A09, 1x A10, 3x A11, 3x A12, 1x A14, 5x A17, 3x A20, 1x A21, 1x A22, 2x A23, 3x A25, 2x A26, 3x A27, 4x A28, 4x A30, 2x A31, 1x A32, 1x A33, 6x A34, 4x A36, 3x A37, 6x A38, 6x A39, 4x A40, 1x A41, 1x A42, 4x A43, 1x A44, 5x A45, 6x A46, 3x A47, 8x A48, 1x A50, 1x A51, 2x A52, 2x A53, 3x A54, 1x A55, 1x A57, 1x A58, 1x A59, 4x A60, 2x A61, 1x A62, 2x A63, 2x A65, 1x A66, 2x A68, 1x A69, 5x A70, 1x A72, 1x A73, 1x A75, 1x A77, 1x A78, 1x A79, 2x A80, 1x A81, 1x A82, 1x A83, 1x A84, 2x A85, 1x A86, 1x A87, 2x A88, 2x A89, 4x A90, 4x A91, 2x A93, 2x A96, 1x A97, 3x A98, 2x A99, 3x B00, 1x B01, 1x B02, 2x B04, 3x B05, 6x B06, 8x B07, 5x B08, 3x B09, 1x B10, 3x B11, 5x B12, 4x B13, 2x B14, 2x B15, 1x B18, 1x B19, 3x B20, 2x B21, 6x B22, 2x B23, 4x B26, 1x B27, 4x B28, 1x B29, 7x B30, 2x B31, 2x B32, 8x B33, 1x B34, 3x B36, 2x B37, 1x B38, 1x B39, 1x B40, 2x B41, 5x B42, 1x B44, 1x B45, 3x B46, 5x B47, 5x B48, 2x B49, 3x B50, 4x B51, 2x B52, 2x B54, 1x B55, 1x B56, 2x B57, 1x B58, 1x B59, 2x B60, 1x B62, 3x B64, 1x B65, 1x B66, 5x B67, 1x B70, 2x B71, 1x B72, 2x B74, 1x B75, 2x B76, 1x B77, 1x B78, 7x B80, 3x B81, 2x B82, 5x B84, 3x B85, 1x B86, 4x B87, 1x B88, 2x B89, 3x B90, 4x B91, 3x B92, 1x B93, 1x B94, 4x B96, 1x B97, 2x B98, 2x B99, 6x C02, 3x C03, 2x C04, 4x C06, 1x C07, 1x C08, 1x C09, 1x C11, 3x C13, 1x C14, 2x C16, 1x C18, 1x C19, 1x C26, 2x C27, 1x C28, 1x C31, 1x C32, 1x C34, 1x C35, 1x C37, 6x C41, 7x C42, 4x C44, 8x C45, 1x C46, 6x C48, 2x C49, 3x C50, 2x C54, 1x C55, 1x C57, 2x C60, 1x C61, 1x C62, 4x C64, 7x C65, 13x C67, 7x C68, 1x C69, 1x C71, 1x C72, 1x C75, 6x C76, 7x C77, 7x C78, 3x C79, 1x C80, 2x C81, 6x C82, 6x C84, 2x C85, 2x C86, 5x C87, 3x C88, 4x C89, 5x C90, 5x C91, 5x C92, 2x C93, 9x C95, 5x C96, 4x C97, 8x C98, 4x C99, 12x D00, 4x D01, 8x D02, 3x D03, 2x D04, 3x D05, 1x D06, 1x D08, 3x D09, 8x D10, 2x D11, 6x D12, 1x D13, 1x D14, 4x D15, 1x D16, 3x D17, 1x D18, 2x D19, 1x D21, 3x D22, 1x D23, 2x D26, 4x D27, 1x D29, 7x D30, 9x D31, 1x D32, 5x D34, 4x D35, 3x D36, 1x D37, 5x D38, 2x D39, 3x D43, 1x D44, 4x D45, 1x D46, 1x D47, 1x D48, 1x D49, 2x D51, 1x D53, 4x D56, 1x D57, 2x D59, 7x D61, 1x D62, 3x D63, 2x D64, 1x D65, 1x D67, 2x D70, 2x D71, 3x D73, 1x D74, 1x D75, 3x D76, 2x D77, 3x D78, 6x D79, 5x D80, 1x D82, 1x D83, 4x D85, 1x D86, 1x D87, 2x D88, 1x D89, 2x D91, 2x D93, 7x D94, 1x D95, 2x D96, 2x D97, 3x D98, 2x D99, 1x E00, 2x E01, 2x E02, 7x E04, 4x E05, 4x E06, 3x E07, 3x E08, 4x E10, 10x E11, 8x E12, 1x E13, 3x E14, 9x E15, 10x E16, 1x E18, 1x E19, 7x E20, 3x E21, 1x E22, 1x E23, 1x E24, 2x E25, 1x E26, 2x E29, 2x E30, 2x E31, 7x E32, 1x E33, 1x E34, 1x E35, 3x E36, 1x E38, 1x E39, 1x E40, 1x E41, 1x E42, 1x E48, 1x E50, 1x E51, 1x E53, 1x E54, 1x E55, 1x E56, 8x E60, 7x E61, 5x E62, 2x E63, 2x E64, 2x E65, 1x E66, 1x E68, 3x E69, 3x E70, 1x E71, 5x E73, 2x E75, 3x E77, 1x E78, 1x E80, 2x E81, 2x E82, 1x E83, 1x E84, 1x E86, 1x E87, 1x E88, 7x E90, 4x E91, 1x E92, 1x E93, 5x E94, 2x E97, 1x E99,