Modified Schola here...

http://cap.connx.com/chess-engines/new-approach/schola.zip

I turned buffering off for stdin and stdout and also added a draw for insufficient material. Probably, the auther will repair my clumsy surgery at some point but at least it doesn't dance around for 2 hours with KBk on the board.

I added tag names for the structs and enums also, so that Doxygen can say something sensible about them.

The zip has everything in it needed to run (the randomx.bin files are also important for book loading) as well as source, etc.

I have only a 64 bit binary in the zip, but I built it to use direct linking instead of DLL for CRT, so I hope this one will not be troublesome to load.

I included Dann's changes in the latest version on my website. I don't have a 64 bit version there though since the engine is 0x88 based I suspect the difference in performance is minimal.

http://mysite.verizon.net/the_platts/schola.html

The above version also fixes some nasty problems that occurred during pondering.

Andy.

64 bit version in here:
http://cap.connx.com/chess-engines/new-approach/schola.zip

Does anyone have a rough idea of the strength of Schola?

Best,

George Speight wrote:Does anyone have a rough idea of the strength of Schola?

Best,

An extremely rough idea here..

43 Sunsetter C10 JA : 2110 121 111 52 77.9 % 1891 1.9 %
44 InmiChess 3.05 : 2108 119 110 45 75.6 % 1912 8.9 %
45 Gnuchess 5.07 JA : 2102 117 108 51 76.5 % 1897 3.9 %
46 Heracles 0.5.14 : 2099 205 176 21 78.6 % 1873 4.8 %
47 LaDameBlanche 2.0c : 2099 182 163 21 78.6 % 1873 14.3 %
48 Schola 1.0.0 : 2098 137 129 30 66.7 % 1978 6.7 %
49 Plywood 1.7.3 : 2097 147 141 23 63.0 % 2004 13.0 %
50 Belzebub 0.67 : 2095 114 116 32 45.3 % 2128 15.6 %
51 Mint 2.3 : 2089 155 146 24 64.6 % 1985 4.2 %
52 Embracer_112 : 2082 120 114 37 66.2 % 1966 8.1 %
53 EnginMax 5.11c : 2082 117 108 53 76.4 % 1878 1.9 %
54 Ghost_v0_14 : 2079 117 112 38 64.5 % 1975 7.9 %
55 Sharper 0.17 : 2074 147 139 25 64.0 % 1974 8.0 %
56 TheDuke3 : 2069 110 104 45 68.9 % 1931 8.9 %

"Rough" but appreciated. Thanks, Tony


Best,

Tony Thomas wrote:

George Speight wrote:Does anyone have a rough idea of the strength of Schola?

Best,

An extremely rough idea here..

43 Sunsetter C10 JA : 2110 121 111 52 77.9 % 1891 1.9 %
44 InmiChess 3.05 : 2108 119 110 45 75.6 % 1912 8.9 %
45 Gnuchess 5.07 JA : 2102 117 108 51 76.5 % 1897 3.9 %
46 Heracles 0.5.14 : 2099 205 176 21 78.6 % 1873 4.8 %
47 LaDameBlanche 2.0c : 2099 182 163 21 78.6 % 1873 14.3 %
48 Schola 1.0.0 : 2098 137 129 30 66.7 % 1978 6.7 %
49 Plywood 1.7.3 : 2097 147 141 23 63.0 % 2004 13.0 %
50 Belzebub 0.67 : 2095 114 116 32 45.3 % 2128 15.6 %
51 Mint 2.3 : 2089 155 146 24 64.6 % 1985 4.2 %
52 Embracer_112 : 2082 120 114 37 66.2 % 1966 8.1 %
53 EnginMax 5.11c : 2082 117 108 53 76.4 % 1878 1.9 %
54 Ghost_v0_14 : 2079 117 112 38 64.5 % 1975 7.9 %
55 Sharper 0.17 : 2074 147 139 25 64.0 % 1974 8.0 %
56 TheDuke3 : 2069 110 104 45 68.9 % 1931 8.9 %

In some tests I was running recently, LaDameBlanche was significantly ahead of Schola but Heracles was behind. I guess it depends quite a bit on time conditions, opening book, etc.. Also Heracles and Schola would repeat about 95% of their games so that wasn't a very good statistic!

In general it's weakest part is pawn evaluation and king safety - at least by looking at how it loses. I am working on both of those (actually, not right now because I'm on vacation without a compiler for two weeks!) and I hope to significantly improve it's position.

Andy.