Deeds on Chess Engines Testing

General discussion about computer chess...
Post Reply
botunnet
Posts: 66
Joined: Sun Jun 25, 2023 3:30 pm
Real Name: botunnet
Location: Hrvatska Republika
Contact:

when a daily tester should review his tests

Post by botunnet »

------------

:!: Notice:

This post in this topic has been reviewed and edited because it violated the OpenChess Forum Rules. This is a gentle reminder not to continue flaming this topic or others in the Forum. Further recurrent violations will not be tolerated.

Thanks,

Best regards, Moderation.

------------



Scenario : create an opening book from games of a reference player.

What people would do :

- select the games played from the start position by the reference player
- create a book with the moves played as white by the reference player
- create a book with the moves played as black by the reference player
- merge the two books

Code: Select all

polyglot.exe make-book -pgn depth4_startposition_white.pgn -bin depth4_startposition_white.bin -only-white
PolyGlot 2.0.2 by Fabien Letouzey.
inserting games ...
251 games.
13892 entries.
filtering entries ...
437 entries.
sorting entries ...
saving entries ...
all done!

polyglot.exe make-book -pgn depth4_startposition_black.pgn -bin depth4_startposition_black.bin -only-black
PolyGlot 2.0.2 by Fabien Letouzey.
inserting games ...
251 games.
12226 entries.
filtering entries ...
531 entries.
sorting entries ...
saving entries ...
all done!

polyglot.exe merge-book -in1 "depth4_startposition_white.bin" -in2 "depth4_startposition_black.bin" -out "depth4_startposition.bin"
PolyGlot 2.0.2 by Fabien Letouzey.
done!
Image

As we can see, the book moves were really played by the reference player, and their weights correspond to their percentages.
botunnet
Posts: 66
Joined: Sun Jun 25, 2023 3:30 pm
Real Name: botunnet
Location: Hrvatska Republika
Contact:

Re: when a daily tester should review his tests

Post by botunnet »

------------

:!: Notice:

This post in this topic has been reviewed and edited because it violated the OpenChess Forum Rules. This is a gentle reminder not to continue flaming this topic or others in the Forum. Further recurrent violations will not be tolerated.

Thanks,

Best regards, Moderation.

------------


- select any games played by the reference player
- create a book with any moves from the games
- arbitrarily assigned the book's filename to the reference player

Code: Select all


polyglot.exe make-book -pgn depth4_openings_white.pgn -bin depth4_openings_white.bin -only-white
PolyGlot 2.0.2 by Fabien Letouzey.
inserting games ...
allocating 1.25MB ...
allocating 2.5MB ...
501 games.
54046 entries.
filtering entries ...
779 entries.
sorting entries ...
saving entries ...
all done!

polyglot.exe make-book -pgn depth4_openings_black.pgn -bin depth4_openings_black.bin -only-black
PolyGlot 2.0.2 by Fabien Letouzey.
inserting games ...
allocating 1.25MB ...
allocating 2.5MB ...
501 games.
47189 entries.
filtering entries ...
809 entries.
sorting entries ...
saving entries ...
all done!

polyglot.exe merge-book -in1 "depth4_openings_white.bin" -in2 "depth4_openings_black.bin" -out "depth4_openings.bin"
PolyGlot 2.0.2 by Fabien Letouzey.
skipped 419 entries.
done!
Image

As we can see, the book moves weren't played by the reference player, and their weights don't correspond to their percentages.
Sedat Canbaz
Posts: 2289
Joined: Wed Jun 21, 2023 6:29 am

Re: when a daily tester should review his tests

Post by Sedat Canbaz »

------------

:!: Notice:

This post in this topic has been reviewed and edited because it violated the OpenChess Forum Rules. This is a gentle reminder not to continue flaming this topic or others in the Forum. Further recurrent violations will not be tolerated.

Thanks,

Best regards, Moderation.

------------


Why I say like this,
Depth4=Deeds book's lines is not yours?

Depth4=Deeds bin is NOT my 1st creation, ok ? I work over
Sedat Canbaz
Posts: 2289
Joined: Wed Jun 21, 2023 6:29 am

Re: when a daily tester should review his tests

Post by Sedat Canbaz »

------------

:!: Notice:

This post in this topic has been reviewed and edited because it violated the OpenChess Forum Rules. This is a gentle reminder not to continue flaming this topic or others in the Forum. Further recurrent violations will not be tolerated.

Thanks,

Best regards, Moderation.

------------


https://pixeldrain.com/u/eq7nExgA

Note also that,
My created Depth4_Deeds book is based on the above games!
In other words, the book is tuned to prefer Depth4 EXP's lines

And to be more clear,
Before starting 'Depth4_Deeds bin'' creation...
All doubles are removed.. there were plenty of same copy games!

By the way, via using SCID program,
Everyone can check is true or not and for anyone missed e.g
Depth4_Deeds bin book is still available (directly at this link):
https://mega.nz/file/voYhnaZB#dDpkHPgP3 ... a3Kz0PWrQs

And one thing more, after the end of tour,
https://sites.google.com/site/computers ... t-bn-cs-14
All the played tour games will be shared...I mean anyone who interested
Can compare...these lines are played by Depth4_ Deeds book or not...
Sedat Canbaz
Posts: 2289
Joined: Wed Jun 21, 2023 6:29 am

Re: when a daily tester should review his tests

Post by Sedat Canbaz »

------------

:!: Notice:

This post in this topic has been reviewed and edited because it violated the OpenChess Forum Rules. This is a gentle reminder not to continue flaming this topic or others in the Forum. Further recurrent violations will not be tolerated.

Thanks,

Best regards, Moderation.

------------


Well, here are two games, which may help...
In other words,

I realized to check just two games e.g
One game played during SCCT B/N CS XIV
And one game during Deeds's EXP VS BOOK

By the way,
A screenshot, where c5 is playing as GOLDEN KEY,
Regarding that the last C5 move belongs to Deeds book!
And later is out of book..I mean in both games c5 is played:
Image

And here the both games, where everything is clear:

E.g here is the recent game, which is
Played by Depth4_Deeds in SCCT B/N XIV:

[Event "SCCT- Book/NNUE CS XIV "]
[Site "Antalya"]
[Date "2023.07.05"]
[Round "19"]
[White "CachoUPD2"]
[Black "Depth4_Deeds"]
[Result "1-0"]
[ECO "E94"]
[PlyCount "108"]
[EventDate "2023.??.??"]
[TimeControl "120+1"]

1. d4 Nf6 2. c4 g6 3. Nc3 Bg7 4. e4 d6 5. Nf3 O-O 6. Be2 e5 7. O-O Nbd7 8. Be3
c6 9. d5 c5 10. Kh1 Kh8 {-1.01/23 1.8s} 11. Bg5 a6 {-1.00/27 5.6s} 12. a4 Rb8 {
-0.93/26 7.1s} 13. g3 Qe8 {-0.87/24 1.4s} 14. Bd2 Qe7 {-0.93/25 2.4s} 15. Ng5
b6 {-1.09/27 6.0s} 16. Qc2 Re8 {-1.21/26 4.1s} 17. Kg2 Nf8 {-1.04/22 2.1s} 18.
h4 {+1.50/24 7.0s} Ng8 {-1.22/26 8.7s} 19. Rh1 {+1.66/25 3.1s} h5 {
-1.22/24 1.6s} 20. Raf1 {+1.70/28 11s} Nh7 {-1.25/26 5.5s} 21. Qc1 {
+1.75/27 6.2s} Rf8 {-1.37/25 1.9s} 22. f3 {+1.91/27 2.2s} Rb7 {-1.25/27 2.8s}
23. Nh3 {+1.91/27 7.9s} Qe8 {-0.95/25 3.0s} 24. Nf2 {+1.57/28 9.9s} f5 {
-0.92/26 2.7s} 25. Rh2 {+1.50/26 1.9s} Rbf7 {-0.95/25 4.4s} 26. Kh1 {
+1.44/26 6.8s} Ngf6 {-0.94/30 15s} 27. Rg1 {+1.61/23 3.7s} Bd7 {-1.07/25 2.0s}
28. Qc2 {+1.43/28 5.3s} Qe7 {-0.86/25 6.7s} 29. Qb3 {+1.42/26 6.1s} Qd8 {
-0.96/28 1.4s} 30. Qd1 {+1.43/25 3.7s} Qe7 {-0.98/27 10s} 31. exf5 {
+1.44/25 6.1s} gxf5 {-0.71/25 1.2s} 32. Nh3 {+1.59/21 1.5s} Qe8 {-0.77/26 2.4s}
33. Qb1 {+1.76/26 2.8s} Kg8 {-0.99/30 6.9s} 34. b4 {+1.61/27 4.7s} e4 {
-1.05/27 2.8s} 35. Nf4 {+1.64/28 6.1s} Qc8 {-0.95/27 2.7s} 36. Rhg2 {
+1.47/27 3.9s} exf3 {-1.17/26 4.7s} 37. Bxf3 {+1.48/27 2.7s} cxb4 {
-1.33/29 4.4s} 38. Qxb4 {+1.91/27 2.1s} Ng4 {-1.28/26 2.0s} 39. Ne6 {
+2.23/26 2.0s} Re8 {-1.67/31 8.2s} 40. Nxg7 {+2.40/25 2.5s} Kxg7 {-1.67/24 1.3s
} 41. Be2 {+2.32/25 3.4s} Nhf6 {-1.62/23 1.4s} 42. Qxb6 {+2.55/29 6.7s} Rxe2 {
-1.78/29 6.5s} 43. Nxe2 {+2.37/26 1.9s} Qxc4 {-1.70/27 1.4s} 44. Bc3 {
+2.50/26 3.4s} Qxd5 {-1.55/25 1.6s} 45. Rc1 {+2.35/25 2.2s} Kh7 {-1.83/26 4.6s}
46. Bxf6 {+2.75/27 1.8s} Rxf6 {-1.98/24 1.3s} 47. Nf4 {+2.74/24 2.2s} Qe4 {
-2.37/29 7.3s} 48. a5 {+3.23/22 2.0s} Bb5 {-2.33/22 1.8s} 49. Qc7+ {
+3.54/22 1.8s} Kg8 {-3.03/25 2.9s} 50. Qc8+ {+3.69/25 1.9s} Be8 {-2.87/23 0.94s
} 51. Qxa6 {+4.10/25 2.6s} d5 {-3.00/22 2.5s} 52. Qb7 {+4.37/22 1.8s} Rc6 {
-2.63/22 0.80s} 53. Ra1 {+5.23/20 2.0s} Qe3 {-5.14/21 5.3s} 54. Qb2 {
+5.92/22 2.6s} Rc8 {-5.71/21 1.9s, White wins by adjudication} 1-0

------------------------------------------------------------------------

And the other game.. played in Deeds's EXP VS BOOK:

[Event "EXP VS BOOK"]
[Site "dual xeon e5-2660v3"]
[Date "2022.07.13"]
[Round "1"]
[White "Solista-010722 (Eman 8.10)"]
[Black "Depth4_050522 (Eman 8.10)"]
[Result "1/2-1/2"]
[ECO "E94"]
[PlyCount "116"]
[EventDate "2022.??.??"]
[Source "SC"]
[TimeControl "60+1"]

1. d4 {book} Nf6 {book} 2. c4 {book} g6 {book} 3. Nc3 {book} Bg7 {book} 4. e4 {
book} d6 {book} 5. Nf3 {book} O-O {book} 6. Be2 {book} e5 {book} 7. O-O {book}
Nbd7 {book} 8. Be3 {book} c6 {book} 9. d5 {+1.67/22 1.8s} c5 {+1.35/28 6.2s}
10. a4 Qe7 {+1.31/24 0.69s} 11. a5 {+1.72/31 8.1s} Ne8 {+1.39/22 1.3s} 12. Qb1
{+1.57/29 4.9s} Kh8 {+1.30/23 1.4s} 13. Bd3 {+1.55/28 4.3s} Nc7 {+1.35/29 3.7s}
14. Bg5 {+1.44/25 1.8s} Bf6 {+1.36/24 1.4s} 15. Bd2 {+1.60/25 1.4s} Na6 {
+1.25/27 4.3s} 16. Qe1 {+1.61/26 1.9s} Nb4 {+1.18/27 2.3s} 17. Bb1 {
+1.55/25 1.5s} b6 {+1.22/25 1.7s} 18. Nd1 {+1.74/24 1.9s} a6 {+1.56/27 8.0s}
19. Ra3 {+1.74/25 3.3s} bxa5 {+1.51/24 2.0s} 20. Rxa5 {+1.63/29 10s} Bg7 {
+1.28/27 0.93s} 21. Bxb4 {+1.52/23 2.3s} cxb4 {+1.29/29 1.0s} 22. Qxb4 {
+1.37/28 1.3s} Nc5 {+1.45/25 1.2s} 23. Qa3 {+1.52/24 1.1s} Qa7 {+1.31/29 1.1s}
24. Qc3 {+1.45/29 2.3s} Nb7 {+1.28/26 1.9s} 25. Ra2 {+1.38/27 1.5s} a5 {
+1.39/26 1.8s} 26. Bc2 {+1.44/29 9.7s} Qb6 {+1.12/27 1.9s} 27. b3 {
+1.49/24 0.71s} Bd7 {+1.09/30 7.7s} 28. Nb2 {+1.25/25 2.4s} f5 {+1.23/24 1.2s}
29. Na4 {+1.14/24 1.5s} Qd8 {+1.16/26 1.9s} 30. Rfa1 {+1.12/25 4.1s} Rb8 {
+0.97/25 4.4s} 31. Rb1 {+1.31/23 2.0s} Bf6 {+1.05/23 1.1s} 32. h3 {
+1.23/22 2.9s} Qc7 {+0.84/22 1.8s} 33. Rab2 {+1.12/24 0.64s} Rg8 {+0.75/23 1.6s
} 34. Ra2 {+1.00/27 7.5s} Rgf8 {+0.83/24 2.5s} 35. Rab2 {+1.01/24 1.4s} Rg8 {
+0.81/25 4.5s} 36. Ra1 {+1.12/24 2.6s} Bd8 {+0.73/27 8.6s} 37. Rd1 {
+0.75/21 2.8s} Rc8 {+0.62/25 2.5s} 38. Rdb1 {+0.84/21 1.1s} Rg7 {+0.46/21 0.71s
} 39. Nd2 {+0.55/21 1.3s} f4 {+0.60/22 1.4s} 40. Kf1 {+0.61/21 0.69s} Bxa4 {
+0.38/22 1.3s} 41. bxa4 {+0.59/17 0.14s} Nc5 {+0.17/22 0.70s} 42. h4 {
+0.56/23 1.4s} Qd7 {+0.15/24 2.0s} 43. Nf3 {+0.21/22 1.3s} h6 {+0.10/21 0.48s}
44. Rb5 {+0.49/20 0.50s} Qe7 {+0.17/21 0.84s} 45. Kg1 {+0.41/19 0.57s} Kh7 {
+0.21/22 0.91s} 46. Qa3 {+0.42/23 0.68s} Qa7 {+0.14/25 2.4s} 47. Qc3 {
+0.31/24 1.2s} Qe7 {+0.24/23 1.3s} 48. Qa3 {+0.21/23 0.79s} Qa7 {+0.15/25 1.6s}
49. Qb2 {+0.25/23 1.1s} Qe7 {+0.20/23 0.64s} 50. Qc3 {+0.27/21 0.27s} Qd7 {
+0.23/25 1.8s} 51. Qb2 {+0.24/24 2.3s} Na6 {+0.07/24 0.95s} 52. Rb7 {
+0.30/29 0.57s} Qxb7 {+0.06/28 0.36s} 53. Qxb7 {+0.12/23 0.33s} Rxb7 {
0.00/34 0.64s} 54. Rxb7+ {0.00/29 0.31s} Kg8 {0.00/41 0.66s} 55. Kf1 {
0.00/35 0.81s} Nc5 {0.00/42 0.71s} 56. Rb5 {0.00/33 0.61s} Kg7 {0.00/42 0.66s}
57. Ke2 {0.00/40 2.5s} Kf7 {0.00/33 0.77s} 58. Nd2 {0.00/33 0.23s} Kg7 {
0.00/42 0.76s, Draw by adjudication} 1/2-1/2
Last edited by Sedat Canbaz on Mon Jul 10, 2023 9:42 am, edited 1 time in total.
User avatar
deeds
Posts: 1482
Joined: Wed Oct 20, 2021 9:24 pm
Location: France
Contact:

Re: when a daily tester lies to people...

Post by deeds »

------------

:!: Notice:

This post in this topic has been reviewed and edited because it violated the OpenChess Forum Rules. This is a gentle reminder not to continue flaming this topic or others in the Forum. Further recurrent violations will not be tolerated.

Thanks,

Best regards, Moderation.

------------


book :
Image

exp file :
Image
Last edited by deeds on Mon Jul 10, 2023 10:21 am, edited 1 time in total.
User avatar
deeds
Posts: 1482
Joined: Wed Oct 20, 2021 9:24 pm
Location: France
Contact:

Re: when a daily tester should review his tests

Post by deeds »

------------

:!: Notice:

This post in this topic has been reviewed and edited because it violated the OpenChess Forum Rules. This is a gentle reminder not to continue flaming this topic or others in the Forum. Further recurrent violations will not be tolerated.

Thanks,

Best regards, Moderation.

------------

Image

As botunnet already said,
botunnet wrote: Wed Jul 05, 2023 10:35 am Image

Christophe JEAN's "D.C.S. Book 040723" polyglot bin book is private.
DeeDs "Cumulated" polyglot bin book is available since July 24, 2021.
So Sedat created an opening book which also contains GUI's moves with the weights which don't correspond to the percents assigned "Depth4_deeds" as the book's filename
Sedat Canbaz
Posts: 2289
Joined: Wed Jun 21, 2023 6:29 am

Re: when a daily tester lies to people...

Post by Sedat Canbaz »

------------

:!: Notice:

This post in this topic has been reviewed and edited because it violated the OpenChess Forum Rules. This is a gentle reminder not to continue flaming this topic or others in the Forum. Further recurrent violations will not be tolerated.

Thanks,

Best regards, Moderation.

------------


Well,
I created a book based on your published games,
Which is highly tuned to prefer your EXP ([Depth4)

Btw, what about here ?
https://sites.google.com/site/computers ... t-bn-cs-14

To be more clear,
If you was as real HONEST tester..
Then simply you had to name your exp such as
GUI's moves (as player),
I mean not as Depth4 exp player
Last edited by Sedat Canbaz on Mon Jul 10, 2023 10:59 am, edited 1 time in total.
User avatar
deeds
Posts: 1482
Joined: Wed Oct 20, 2021 9:24 pm
Location: France
Contact:

Re: when a daily tester should review his tests

Post by deeds »

------------

:!: Notice:

This post in this topic has been reviewed and edited because it violated the OpenChess Forum Rules. This is a gentle reminder not to continue flaming this topic or others in the Forum. Further recurrent violations will not be tolerated.

Thanks,

Best regards, Moderation.

------------


Depth4_xxxxxx has already played the "1.f4" move.
In the 1500 games, the "1.f4 {book}" was played 20 times by the GUI (cutechess)

Image

So Depth4_xxxxxx has never played "1 .f4"
Sedat Canbaz
Posts: 2289
Joined: Wed Jun 21, 2023 6:29 am

Re: when a daily tester lies to people...

Post by Sedat Canbaz »

------------

:!: Notice:

This post in this topic has been reviewed and edited because it violated the OpenChess Forum Rules. This is a gentle reminder not to continue flaming this topic or others in the Forum. Further recurrent violations will not be tolerated.

Thanks,

Best regards, Moderation.

------------

deeds wrote: Mon Jul 10, 2023 10:57 am In the liar's book, Sedat want to tell us that Depth4_xxxxxx has already played the "1.f4" move.
In the 1500 games, the "1.f4 {book}" was played 20 times by the GUI (cutechess)

Image

So Depth4_xxxxxx has never played "1 .f4"

Oops
What are you talking about? it's not clear in your all screenshots!
Btw, why are you afraid so much to share your current position in PGN ?

Please follow my work.. it is much more clear,...


Thank you in advance
Post Reply