ChessUSA.com TalkChess.com
Hosted by Your Move Chess & Games
 
 FAQFAQ   SearchSearch   MemberlistMemberlist   UsergroupsUsergroups   RegisterRegister 
 ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in 

Senpai 2.0
Goto page 1, 2, 3, 4, 5, 6  Next
 
Post new topic       TalkChess.com Forum Index -> Computer Chess Club: General Topics Threaded
View previous topic :: View next topic  
Author Message
Fabien Letouzey



Joined: 22 Jan 2011
Posts: 125
Location: Lille, France

PostPosted: Fri Nov 10, 2017 6:07 am    Post subject: Senpai 2.0 Reply to topic Reply with quote

Hi all,

After a few years in computer draughts, I am back to chess for a little while. I see too many draws at top level, and am afraid that it will be too late in a couple of years. Big hardware and large opening books are also not helping IMO.

So I'm here to announce Senpai 2.0. The code looks all different because I started from my draughts program, but it's actually still Senpai inside. I prefer having a consistent codebase for multiple games (also Shogi and Othello).

As a rewrite it's hard to list changes, but here is an attempt:
- Chess960 is baked in, not an add-on; Senpai 2 plays chess only as a special case!
- search: restricted singular extensions
- search: additional reduction/pruning of "losing" moves (SEE < 0)
- eval: tempo bonus (it seems I forgot in 1.0 ...)
- eval: "space" (glorified pawn-chain PST); this changes playing style a lot
- eval: scoring by logistic regression; this gives me more freedom for eval features
- board: copy/make (customary in games with fewer piece types than chess)
- board: optional PEXT bitboards (BMI2)

As I recall, no single change brought more than 10 Elo. So it's more a sum of small (IMO) things.

This release has been rushed a little, so it's still lacking UCI options and multi-PV; sorry about that.

Official website: http://www.amateurschach.de/main/_senpai.htm
Direct link: http://www.amateurschach.de/download/senpai_20.zip
Expected level: Texel/Protector (again)

Thanks to everybody who helped with this release:
- Graham Banks
- Michael Byrne
- Wilhelm Hudetz
- Daniel Jose
- Steve Maughan (https://www.chessprogramming.net)
- Frank Quisinsky (http://www.amateurschach.de)
- XXX (<- this line for people I forgot, sorry in advance).

That's all for now,

Fabien.
Back to top
View user's profile Send private message
Dann Corbit



Joined: 08 Mar 2006
Posts: 8072
Location: Redmond, WA USA

PostPosted: Fri Nov 10, 2017 6:20 am    Post subject: Re: Senpai 2.0 Reply to topic Reply with quote

Fabulous news.

Perhaps it will be world champion soon
_________________
Taking ideas is not a vice, it is a virtue. We have another word for this. It is called learning.
But sharing ideas is an even greater virtue. We have another word for this. It is called teaching.
Back to top
View user's profile Send private message Visit poster's website
Brendan J. Norman



Joined: 07 Feb 2016
Posts: 625

PostPosted: Fri Nov 10, 2017 6:30 am    Post subject: Re: Senpai 2.0 Reply to topic Reply with quote

Xann wrote:
Hi all,

After a few years in computer draughts, I am back to chess for a little while. I see too many draws at top level, and am afraid that it will be too late in a couple of years. Big hardware and large opening books are also not helping IMO.

So I'm here to announce Senpai 2.0. The code looks all different because I started from my draughts program, but it's actually still Senpai inside. I prefer having a consistent codebase for multiple games (also Shogi and Othello).

As a rewrite it's hard to list changes, but here is an attempt:
- Chess960 is baked in, not an add-on; Senpai 2 plays chess only as a special case!
- search: restricted singular extensions
- search: additional reduction/pruning of "losing" moves (SEE < 0)
- eval: tempo bonus (it seems I forgot in 1.0 ...)
- eval: "space" (glorified pawn-chain PST); this changes playing style a lot
- eval: scoring by logistic regression; this gives me more freedom for eval features
- board: copy/make (customary in games with fewer piece types than chess)
- board: optional PEXT bitboards (BMI2)

As I recall, no single change brought more than 10 Elo. So it's more a sum of small (IMO) things.

This release has been rushed a little, so it's still lacking UCI options and multi-PV; sorry about that.

Official website: http://www.amateurschach.de/main/_senpai.htm
Direct link: http://www.amateurschach.de/download/senpai_20.zip
Expected level: Texel/Protector (again)

Thanks to everybody who helped with this release:
- Graham Banks
- Michael Byrne
- Wilhelm Hudetz
- Daniel Jose
- Steve Maughan (https://www.chessprogramming.net)
- Frank Quisinsky (http://www.amateurschach.de)
- XXX (<- this line for people I forgot, sorry in advance).

That's all for now,

Fabien.


Great news from a computer chess legend.

Will test Senpai A.S.A.P.

*Hoping the playing style is lovely!* Laughing
_________________
Check my site for engine reviews and other chess stuff Smile

www.chessncognac.com
Back to top
View user's profile Send private message Send e-mail
Frank Quisinsky



Joined: 18 Nov 2009
Posts: 4755
Location: Trier, Germany

PostPosted: Fri Nov 10, 2017 6:36 am    Post subject: Re: Senpai 2.0 Reply to topic Reply with quote

Hi Brendan,

and on other legend made the graphics (Wilhelm Hudetz, Austria).

Both, Fabien & John on one site ...
A good day for computer chess in my humble opinion!

Of course, not important that it's the own site!

After a longer time slower downloads on www.amateurschach.de it's solved since yesterday. My provider found & fixed a defect patch cable.

Downloads now with full speed ...

Best
Frank

I had here again a lot of fun with Fabien ...
Fabien is really a very interesting person.

I wish me to have 0.1% from his programming knowledge only and I am an expert. But unfortunately ... maybe 0.001% ...
_________________
I like computer chess!


Last edited by Frank Quisinsky on Fri Nov 10, 2017 6:46 am; edited 3 times in total
Back to top
View user's profile Send private message Visit poster's website
Michael B



Joined: 09 Mar 2006
Posts: 2342
Location: Pen Argyl, Pennsylvania

PostPosted: Fri Nov 10, 2017 6:42 am    Post subject: Re: Senpai 2.0 Reply to topic Reply with quote

Frank Quisinsky wrote:
Hi Brendan,

and on other legend made the graphics (Wilhelm Hudetz, Austria).

Both, Fabien & John on one site ...
A good day for computer chess!

Of course, not important that it's the own site!

After a longer time slower downloads on www.amateurschach.de it's solved since yesterday. My provider found & fixed a defect patch cable.

Downloads now with full power ...

Best
Frank


Good job Frank, you do nice work! 👊
Back to top
View user's profile Send private message
Evert Glebbeek



Joined: 21 Jan 2011
Posts: 2823
Location: NL

PostPosted: Fri Nov 10, 2017 6:43 am    Post subject: Re: Senpai 2.0 Reply to topic Reply with quote

Xann wrote:
The code looks all different because I started from my draughts program, but it's actually still Senpai inside. I prefer having a consistent codebase for multiple games (also Shogi and Othello).

Good man.

I have a (private) version of Senpai 1.0 that plays Makruk; as I recall the changes were not too hard to make, so hopefully I can port those over easily.
Back to top
View user's profile Send private message Visit poster's website
Fabien Letouzey



Joined: 22 Jan 2011
Posts: 125
Location: Lille, France

PostPosted: Fri Nov 10, 2017 6:53 am    Post subject: Re: Senpai 2.0 Reply to topic Reply with quote

Evert wrote:
I have a (private) version of Senpai 1.0 that plays Makruk; as I recall the changes were not too hard to make, so hopefully I can port those over easily.

More painful, I guess, but doable. However reusing eval parts would be more difficult with this version, maybe much more. The weights are probably very specific.

And (say) with your own eval, switching to version 2.0 doesn't make much sense.
Back to top
View user's profile Send private message
Guenther Simon



Joined: 01 Oct 2008
Posts: 2029
Location: Regensburg, Germany

PostPosted: Fri Nov 10, 2017 7:29 am    Post subject: Re: Senpai 2.0 Reply to topic Reply with quote

Xann wrote:
Hi all,

After a few years in computer draughts, I am back to chess for a little while. I see too many draws at top level, and am afraid that it will be too late in a couple of years. Big hardware and large opening books are also not helping IMO.

So I'm here to announce Senpai 2.0.

...

Fabien.


Thanks a lot Fabien and all who helped with this release!

Guenther
_________________
http://rwbc-chess.de/
Back to top
View user's profile Send private message Send e-mail Visit poster's website
Gabor Szots



Joined: 10 Mar 2006
Posts: 2324
Location: Szentendre, Hungary

PostPosted: Fri Nov 10, 2017 7:41 am    Post subject: Re: Senpai 2.0 Reply to topic Reply with quote

Thank you Fabien, that's marvellous news.

Gauntlet in progress although I don't think I can reach this week's update.
_________________
Gabor Szots

CCRL testing group
Back to top
View user's profile Send private message
Gabor Szots



Joined: 10 Mar 2006
Posts: 2324
Location: Szentendre, Hungary

PostPosted: Fri Nov 10, 2017 7:51 am    Post subject: Re: Senpai 2.0 Reply to topic Reply with quote

First game, first surprise: instead of agreeing to a draw Senpai goes into a lost endgame.

[Event "40 Moves in 2 min"]
[Site "Senpai 2.0 64-bit - Nov 10, LUCIFER"]
[Date "2017.11.09"]
[Round "1.1"]
[White "Senpai 2.0 64-bit"]
[Black "Arasan 20.3 64-bit"]
[ECO "E20"]
[Result "0-1"]

1. d4 {book 0s} Nf6 {book 0s} 2. c4 {book 0s} e6 {book 0s}
3. g3 {book 0s} d5 {book 0s} 4. Nf3 {book 0s} Bb4+ {book
0s} 5. Nc3 {book 0s} O-O {book 0s} 6. Bg2 {book 0s} c6
{book 0s} 7. O-O {0.00/15 2s} Nbd7 {book 0s (dxc4)} 8. Qd3
{+0.23/16 2s} b6 {book 0s} 9. cxd5 {+0.13/17 3s} cxd5 {book
0s (exd5)} 10. Bf4 {+0.45/17 2s} Bxc3 {+0.51/21 3s}
11. Qxc3 {+0.73/17 3s} Ba6 {+0.62/22 3s (Ne4)} 12. Rfe1
{+0.44/18 1s} Qe7 {+0.57/23 3s} 13. Qc6 {+0.46/20 1s} Rfc8
{+0.69/25 3s} 14. Qa4 {+0.34/19 1s} Bb7 {+0.49/24 3s}
15. Rac1 {+0.33/20 3s} a5 {+0.46/23 3s} 16. Rxc8+ {+0.31/19
2s} Rxc8 {+0.71/24 3s} 17. Rc1 {+0.29/20 1s} Rxc1+
{+0.70/22 3s (Qd8)} 18. Bxc1 {+0.41/19 0s} h6 {+0.71/22 3s
(Qd8)} 19. h4 {+0.36/19 4s} Ne4 {+0.63/19 3s (Qd8)} 20. a3
{+0.51/20 7s} Ndf6 {+0.58/19 3s (Ba6)} 21. Ne5 {+0.48/22
4s} Nd6 {+0.61/21 3s (Qc7)} 22. Qc2 {+0.55/19 1s} Nf5
{+0.46/21 3s (Nfe4)} 23. e3 {+0.54/20 6s} Ne4 {+0.56/19 3s
(Nd7)} 24. Bxe4 {+0.37/20 5s} dxe4 {+0.18/22 3s} 25. Bd2
{+0.11/20 4s} f6 {+0.04/22 3s (Kh7)} 26. Ng6 {+0.46/18 1s}
Qd7 {+0.18/23 3s} 27. h5 {+0.20/20 5s} Nd6 {+0.18/24 3s}
28. Be1 {+0.21/22 3s} Bd5 {+0.18/24 3s} 29. Kg2 {+0.21/23
6s} a4 {+0.18/27 3s (Kf7)} 30. Bb4 {+0.40/20 2s} Nc4
{+0.18/29 3s} 31. Ne7+ {+0.40/22 3s} Kh7 {+0.18/30 3s
(Kf7)} 32. Qd1 {+0.41/23 2s} f5 {+0.18/29 3s} 33. Ng6
{+0.41/25 2s} Qe8 {0.00/29 3s} 34. Qc2 {+0.41/24 3s} Qb5
{+0.18/31 3s (Kg8)} 35. Kg1 {+0.32/24 12s} Qe8 {+0.18/33
3s} 36. Nf4 {+0.31/24 2s} Qd7 {0.00/30 3s (Kg8)} 37. Kh2
{+0.31/24 2s} Kg8 {+0.18/29 3s (Qd8)} 38. Kg2 {+0.32/24 1s}
Kh7 {+0.18/31 3s} 39. Kg1 {+0.31/23 3s} Kg8 {+0.18/28 3s
(Qe8)} 40. Kh2 {+0.25/22 4s} Kh7 {+0.18/30 3s (Qe8)}
41. Kg2 {+0.32/26 3s} Qc6 {0.00/31 3s (Qe8)} 42. Qc3
{+0.25/22 2s} Qd7 {+0.18/32 3s (Qb5)} 43. b3 {+0.32/24 2s}
axb3 {0.00/27 3s} 44. Qxb3 {+0.32/25 3s} Na5 {+0.18/28 3s
(Nd2)} 45. Qc3 {+0.32/26 4s} Nc4 {+0.18/31 3s} 46. Qc2
{+0.32/27 3s} Qb5 {+0.18/28 3s (Qe8)} 47. Kg1 {+0.32/25 2s}
Kg8 {0.00/28 3s (Qc6)} 48. Kh2 {+0.32/23 3s} Kh7 {0.00/30
3s (Qd7)} 49. Kg2 {+0.32/23 2s} Qc6 {+0.18/31 3s (Qe8)}
50. Qe2 {+0.14/24 8s} Qa4 {0.00/28 3s (e5)} 51. g4
{+0.32/21 2s} Qd7 {+0.22/23 3s (Qc6)} 52. Ng6 {+0.42/20 6s}
Qa4 {+0.18/28 3s (Qf7)} 53. Kh2 {+0.45/19 2s} Nxa3
{+0.18/26 3s (Qc6)} 54. Bxa3 {+0.57/21 2s} Qxa3 {+0.18/30
3s} 55. gxf5 {+0.20/22 2s} exf5 {+0.15/29 10s} 56. Qb5
{+0.16/24 7s} Qb3 {+0.18/28 2s} 57. Qd7 {0.00/24 2s} Bg8
{+0.25/28 2s} 58. Nf8+ {0.00/26 2s} Kh8 {+0.49/2 0s}
59. Qxf5 {0.00/26 2s} Qf7 {+0.18/30 2s} 60. Ng6+ {0.00/25
1s} Kh7 {+0.18/29 2s} 61. Nf8+ {0.00/28 2s} Kh8 {0.00/2 0s}
62. Ng6+ {0.00/30 1s} Kh7 {+0.18/31 3s} 63. Qf4 {0.00/26
1s} Qxf4+ {+0.14/27 2s (b5)} 64. Nxf4 {-0.56/23 7s} Bf7
{-0.27/28 2s (Bb3)} 65. Kg1 {-0.45/24 3s} Bc4 {-0.25/30 2s
(b5)} 66. d5 {-1.17/22 3s} Kg8 {-0.62/32 2s} 67. Ng6
{-1.07/24 2s} b5 {-0.53/32 2s} 68. d6 {-1.16/25 1s} Be6
{-0.79/32 2s} 69. Kf1 {-1.22/23 1s} Kf7 {-1.00/31 2s}
70. Ke1 {-1.50/23 2s} Bd7 {-1.18/30 2s} 71. Kd2 {-1.59/24
2s} Ke6 {-1.67/30 2s} 72. Kc3 {-2.44/21 4s} Be8 {-1.53/33
2s (Kxd6)} 73. d7 {-1.90/22 2s} Bxd7 {-2.18/27 2s (Kxd7)}
74. Nf8+ {-2.66/18 1s} Kd6 {-3.26/30 2s} 75. Ng6 {-3.83/22
6s} Be8 {-3.21/33 2s} 76. Nf4 {-4.45/19 3s} Ke5 {-3.28/30
2s} 77. Kd2 {-4.54/18 1s} Kf5 {-3.76/31 2s} 78. Ne2
{-4.62/17 0s} Kg4 {-3.51/29 2s} 79. Nd4 {-4.55/17 0s} Kxh5
{-4.03/29 2s} 80. Nc2 {-4.97/17 0s} Bf7 {-4.98/29 2s (Kg4)}
81. Nd4 {-9.60/24 58s} Bc4 {-5.42/30 3s} 82. f4 {-9.75/20
17s} exf3 {-5.83/29 3s} 0-1
_________________
Gabor Szots

CCRL testing group
Back to top
View user's profile Send private message
Display posts from previous:   
Post new topic       TalkChess.com Forum Index -> Computer Chess Club: General Topics All times are GMT
Goto page 1, 2, 3, 4, 5, 6  Next
Threaded
Page 1 of 6

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum




Powered by phpBB © 2001, 2005 phpBB Group
Enhanced with Moby Threads