Super Tournament XXVIII 8CPU

Discussion of computer chess matches and engine tournaments.

Moderators: bob, hgm, Harvey Williamson

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
Dann Corbit
Posts: 10118
Joined: Wed Mar 08, 2006 7:57 pm
Location: Redmond, WA USA
Contact:

Re: Games 25-30

Post by Dann Corbit » Thu Oct 12, 2017 10:42 pm

arunsoorya1309 wrote:I think you can stop the tournament and start it with another book. I believe its waste of your energy :)
Not my intention.
This was for "next time."

And just a suggestion. Graham does a good job picking books.
But with ultra strong programs on very strong hardware, it is good to take measures to prevent horrific draw-fests.
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.

User avatar
Graham Banks
Posts: 33149
Joined: Sun Feb 26, 2006 9:52 am
Location: Auckland, NZ

Re: Games 25-30

Post by Graham Banks » Thu Oct 12, 2017 10:46 pm

arunsoorya1309 wrote:I think you can stop the tournament and start it with another book. I believe its waste of your energy :)
Well I'm perfectly happy with my GBSelect2017-2 opening book.
There have been many interesting games despite them all being draws so far. :)
My email addresses:
gbanksnz at gmail.com
gbanksnz at yahoo.co.nz

arunsoorya1309
Posts: 204
Joined: Tue Nov 03, 2015 3:55 pm

Re: Games 25-30

Post by arunsoorya1309 » Thu Oct 12, 2017 10:47 pm

Graham is doing an amazing job, but in this case its really waste of his resources as too many draws. There is no harm in stopping it for now and restarting with another book

User avatar
Graham Banks
Posts: 33149
Joined: Sun Feb 26, 2006 9:52 am
Location: Auckland, NZ

Re: Games 25-30

Post by Graham Banks » Thu Oct 12, 2017 10:54 pm

arunsoorya1309 wrote:Graham is doing an amazing job, but in this case its really waste of his resources as too many draws. There is no harm in stopping it for now and restarting with another book
I'll carry on as is. Wouldn't matter what 'fair' book you used with these four engines, you'd still get mainly draws. :)
My email addresses:
gbanksnz at gmail.com
gbanksnz at yahoo.co.nz

User avatar
Graham Banks
Posts: 33149
Joined: Sun Feb 26, 2006 9:52 am
Location: Auckland, NZ

Re: Games 25-30

Post by Graham Banks » Thu Oct 12, 2017 10:55 pm

Dann,

have you got a pgn file in standard format?

Graham.
My email addresses:
gbanksnz at gmail.com
gbanksnz at yahoo.co.nz

Michael Sherwin
Posts: 3041
Joined: Fri May 26, 2006 1:00 am
Location: WY, USA
Full name: Michael Sherwin

Re: Games 25-30

Post by Michael Sherwin » Fri Oct 13, 2017 12:42 am

Dann Corbit wrote:Try this book next time:
http://rybkaforum.net/cgi-bin/rybkaforu ... pid=575689
There is a great need for a well balanced book with lots of play for both sides. However, after seeing these positions in this book it is not for me.

[D]rnbqkbnr/pppppppp/8/8/8/P7/1PPPPPPP/RNBQKBNR b KQkq - 0 1
[D]rnbqkbnr/pppppppp/8/8/8/2P5/PP1PPPPP/RNBQKBNR b KQkq - 0 1
[D]rnbqkbnr/pppppppp/8/8/8/4P3/PPPP1PPP/RNBQKBNR b KQkq - 0 1

And that is just from the first 10 positions.

If I have to go through and vet all 400 positions myself then I might as well start from scratch and create my own book.

Actually about 10 years ago I made a pgn of all the columns of Chess Openings the Easy Way by Nick De Firmian. It was over 900 lines. It was downloadable from WBEC for years. Unfortunately, I lost my copy of the pgn file after I got sick and suffered a hard disk failure or two. It never caught on.

My mistake was I did not realize how lazy people were. The few that took a look wanted to use it as is. It was not meant to be used as is. It was meant as a resource for authors and testers to cull their own test suites from as it covered all the most played openings. It took me a long time to record 900+ lines into a pgn file. It should have been a wonderful resource. And yet here we are 10 years later still struggling with this issue.

If anyone has a copy of EasyWay.pgn please send it to me or supply a link where it can be downloaded from. Thanks

mjsherwin@att.net
I hate if statements. Pawns demand if statements. Therefore I hate pawns.

Dann Corbit
Posts: 10118
Joined: Wed Mar 08, 2006 7:57 pm
Location: Redmond, WA USA
Contact:

Re: Games 25-30

Post by Dann Corbit » Fri Oct 13, 2017 12:56 am

Graham Banks wrote:Dann,

have you got a pgn file in standard format?

Graham.
The PGN file observes the PGN standard.

But I guess what you want is one that has the moves that lead up to the position. Is that right?
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.

Michael Sherwin
Posts: 3041
Joined: Fri May 26, 2006 1:00 am
Location: WY, USA
Full name: Michael Sherwin

Re: Games 25-30

Post by Michael Sherwin » Fri Oct 13, 2017 1:00 am

Dann Corbit wrote:
Graham Banks wrote:Dann,

have you got a pgn file in standard format?

Graham.
The PGN file observes the PGN standard.

But I guess what you want is one that has the moves that lead up to the position. Is that right?
Yes. That is what I created to be used as a resource. :)

Edit: I read too hastily but that is also my answer for EasyWay.pgn if it exist somewhere. :oops:
I hate if statements. Pawns demand if statements. Therefore I hate pawns.

Dann Corbit
Posts: 10118
Joined: Wed Mar 08, 2006 7:57 pm
Location: Redmond, WA USA
Contact:

Re: Games 25-30

Post by Dann Corbit » Fri Oct 13, 2017 1:10 am

Michael Sherwin wrote:
Dann Corbit wrote:Try this book next time:
http://rybkaforum.net/cgi-bin/rybkaforu ... pid=575689
There is a great need for a well balanced book with lots of play for both sides. However, after seeing these positions in this book it is not for me.

[D]rnbqkbnr/pppppppp/8/8/8/P7/1PPPPPPP/RNBQKBNR b KQkq - 0 1
[D]rnbqkbnr/pppppppp/8/8/8/2P5/PP1PPPPP/RNBQKBNR b KQkq - 0 1
[D]rnbqkbnr/pppppppp/8/8/8/4P3/PPPP1PPP/RNBQKBNR b KQkq - 0 1

And that is just from the first 10 positions.

If I have to go through and vet all 400 positions myself then I might as well start from scratch and create my own book.

Actually about 10 years ago I made a pgn of all the columns of Chess Openings the Easy Way by Nick De Firmian. It was over 900 lines. It was downloadable from WBEC for years. Unfortunately, I lost my copy of the pgn file after I got sick and suffered a hard disk failure or two. It never caught on.

My mistake was I did not realize how lazy people were. The few that took a look wanted to use it as is. It was not meant to be used as is. It was meant as a resource for authors and testers to cull their own test suites from as it covered all the most played openings. It took me a long time to record 900+ lines into a pgn file. It should have been a wonderful resource. And yet here we are 10 years later still struggling with this issue.

If anyone has a copy of EasyWay.pgn please send it to me or supply a link where it can be downloaded from. Thanks

mjsherwin@att.net
Let's examine at your objection positions, one at a time:
[d]rnbqkbnr/pppppppp/8/8/8/P7/1PPPPPPP/RNBQKBNR b KQkq - acd 45; acs 58951; bm c5; c3 "e5"; cce 39; ce 0; id "es-dc-neutral.0054"; pm e5 {456} d5 {222} g6 {160} Nf6 {129} c5 {83} a6 {15} Nc6 {13} h5 {10} f5 {9} e6 {8} Na6 {7} a5 {7} b5 {7} b6 {5} c6 {5} h6 {5} g5 {4} f6 {2} Nh6 {1} d6 {1}; pv c5 e4 Nc6 Nc3 g6 Nf3 Bg7 Bb5 Nf6 O-O O-O Bxc6 bxc6 d3 d6 h3 Rb8 Rb1 Nd7 Be3 Re8 Qc1 Qa5 Bd2 Bb7 Re1 Qc7 Bh6 Bf6 Bg5 Bg7 Bh6; white_wins 401; black_wins 499; draws 227; Opening Anderssen Opening: General. 1.a3; CaxtonID: 49; ECO: A00;

Computer scores it perfectly equal, the actual outcomes from games is about +39 centipawns for black (the side to move).

[d]rnbqkbnr/pppppppp/8/8/8/2P5/PP1PPPPP/RNBQKBNR b KQkq - acd 44; acs 40794; bm d5; c3 "Nf6"; cce 27; ce 0; id "es-dc-neutral.0126"; pm Nf6 {626} e5 {414} d5 {251} c5 {203} f5 {43} g6 {39} b6 {17} d6 {10} Nc6 {8} e6 {7} g5 {7} h5 {6} a6 {5} c6 {5} h6 {5} Na6 {4} a5 {3} f6 {1}; pv d5 Nf3 Nf6 d4 e6 Bf4 Be7 e3 O-O Nbd2 Nh5 Be5 Nf6 Bd3 Nbd7 Bf4 Nh5 Be5 Nhf6; white_wins 619; black_wins 713; draws 235; Opening Saragossa Opening: General. 1.c3; CaxtonID: 1890; ECO: A00;

Computer scores it perfectly equal.
In actual outcomes, it is about +27 centipawns for black (the side to move).

[d]rnbqkbnr/pppppppp/8/8/8/4P3/PPPP1PPP/RNBQKBNR b KQkq - acd 44; acs 79030; bm d5; c3 "Nf6"; cce 43; ce -6; id "C.A.P. 445901"; pm e5 {862} d5 {578} Nf6 {459} g6 {403} c5 {154} e6 {91} c6 {32} b6 {26} d6 {25} f5 {22} Nc6 {13} a6 {5} f6 {3} g5 {3} a5 {2} h5 {2} h6 {2} Na6 {1} Nh6 {1} b5 {1}; pv d5 Nf3 Nf6 c4 e6 Nc3 b6 b3 Bb7 Bb2 Nbd7 Rc1 a6 cxd5 exd5 d4 Bd6 g3 O-O Bg2 Ne4 O-O Nxc3 Bxc3 a5 Bb2 Re8 Qc2 c6 Rfd1 Qe7 Bf1 g6 Bd3 Nf6 Ne5 Rec8 Qb1 a4 bxa4 Rxa4 Bc3 Ne4 Qxb6 Rxa2; white_wins 772; black_wins 992; draws 510; Opening Van't Kruijs Opening: General. 1.e3; CaxtonID: 2620; ECO: A00;

The computer scores it -6 centipawns for black (the side to move).
In actual outcomes, it was about +43 centipawns for black.

Now in each of these cases, we have more than one thousand actual games played. Both a black win and a white win are more probable than a draw (at least historically speaking). Hence, we might expect that these openings would have lower draw rates than games which historically have far more draws than wins for white or black.

The fact that they are not the most popular openings is important if you want to play popular openings.

But if you want a contest with wins and losses in it, using statistics to find sharp openings is a good way to do it.

It is also possible that you may find from study of the games that some of these are desirable to play in real life.
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.

Michael Sherwin
Posts: 3041
Joined: Fri May 26, 2006 1:00 am
Location: WY, USA
Full name: Michael Sherwin

Re: Games 25-30

Post by Michael Sherwin » Fri Oct 13, 2017 1:20 am

Dann, the flaw in your thinking is these moves have a higher win lose % for humans because it takes humans out of their preparation and causes confusion that leads to less draws. Engines especially the top rated ones would not suffer from such confusion. In fact these three moves allow black equality much easier and sooner and would imho lead to more draws amongst stronger engines than normal.
I hate if statements. Pawns demand if statements. Therefore I hate pawns.

Post Reply