Computer Chess Club Archives


Search

Terms

Messages

Subject: Fruitbat 0.1 Re: endgame position

Author: Eelco de Groot

Date: 16:20:21 11/02/05

Go up one level in this thread


On November 02, 2005 at 11:25:53, Will Singleton wrote:

>[d]5b2/5k1r/7P/R1Ppp1B1/7P/5p2/1K1R2b1/8 w - - bm Ra6
>
>This proves difficult for many top programs, while others do ok (Ruffian,
>Crafty, Zappa).  The theme seems to be to control the 6th rank, keeping the
>black king bottled up.  Failing that, black has chances with its advanced pawns.
> (from an ICC game that black won.)
>
>The line is Ra6 d4 Kc1! or h5!
>
>Will

Inspired by Thorsten Czub's settings for Toga II, and a post below from Franz
Hagra about batchfiles for a name I created some settings Fruit 2.2.1. and
called it

Fruitbat 0.1 Experimental

King Safety 160
History Treshold 50
Passed Pawns 110

There are not that many options possible yet in Fruit 2.2.1 so that was quickly
done...

Fruitbat 0.1 Experimental is a bit quicker in this position than standard Fruit,
about five minutes or one ply:

00:00:00.0	1,93	1	10	Rxd5
00:00:00.0	1,27	2	615	Rxd5 f2 Rd7+ Ke6
00:00:00.0	1,76	2	694	Ra7+ Kg8 Rxh7 Kxh7 Rxd5
00:00:00.0	1,76	3	1302	Ra7+ Kg8 Rxh7 Kxh7 Rxd5
00:00:00.0	1,04	4	2783	Ra7+ Kg8 Rxh7 Kxh7 c6 d4
00:00:00.0	1,77	4	7326	c6 Kg6 Rc2 e4 Rxd5 Bxh6 Bxh6 Rxh6
00:00:00.0	1,83	5	14359	c6 Kg6 Rc2 Bd6 Rxd5
00:00:00.1	1,23	6	33397	c6 Ke6 Rdxd5 f2 Rxe5+ Kd6 c7 Rxc7
00:00:00.2	1,62	7	79805	c6 Ke6 Raxd5 Bb4 Rd1 f2 Rd7 Rxd7 cxd7
00:00:00.3	1,54	8	141066	c6 Ke6 Raxd5 Bb4 Rd1 f2 Rd7 Bf3 Rf1
00:00:00.7	1,65	9	429630	c6 Kg8 h5 Rc7 Rc2 Rf7 Rf2 Rc7 Rxd5 Rxc6 Rxe5
00:00:01.1	1,65	10	759469	c6 Kg8 h5 Rc7 Rc2 Rf7 Rf2 Rc7 Rxd5 Rxc6 Rxe5
00:00:02.2	1,65	11	1538903	c6 Kg8 h5 Rc7 Rc2 Rf7 Rf2 Rc7 Rxd5 Rxc6 Rxe5
00:00:05.3	1,49	12	3851699	c6 Kg8 Rc2 Rc7 Bf6 e4 Rxd5 Bxh6 Be5 Rc8 c7 Kf7 Rd7+
Ke6
00:00:09.4	1,52	13	7026412	c6 Kg8 Rc2 Rc7 Bf6 Rf7 c7 Bh3 Ra6 Bc8 Rb6 Bxh6 Bxe5
f2 Rg6+ Rg7
00:00:32.6	1,66	14	25472294	c6 Kg6 Raxd5 f2 Rxf2 Bxd5 Rxf8 Rc7 Rf6+ Kh7 Rf5 Rxc6
Rxe5 Rb6+ Kc3 Rb3+ Kd4
00:00:58.5	1,78	15	46414484	c6 Kg8 Rc2 Rc7 Rb5 d4 Rb7 Bd6 Rb8+ Bf8 Bf6 d3 Bxe5
dxc2 Bxc7 f2 h7+ Kxh7 Rxf8 f1Q Rxf1 Bxf1 Kxc2
00:01:24.2	1,78	16	67036025	c6 Kg8 Rc2 Rc7 Rb5 d4 Rb7 Bd6 Rb8+ Bf8 Bf6 d3 Bxe5
dxc2 Bxc7 f2 h7+ Kxh7 Rxf8 f1Q Rxf1 Bxf1 Kxc2
00:02:32.6	1,78	17	122855802	c6 Kg8 Rc2 Rc7 Rb5 d4 Rb7 Bd6 Rb8+ Bf8 Bf6 d3 Bxe5
dxc2 Bxc7 f2 h7+ Kxh7 Rxf8 f1Q Rxf1 Bxf1 Kxc2
00:05:12.9	2,07	17	257342344	Ra6 d4 h5 Bh3 Rf6+ Ke8 Rxf3 Bg4 Rg3 Bxh5 c6 Bd6 Rb3
Bg6 Rf2 Be4 Rf6 Bc7
00:08:34.5	2,55	18	429146184	Ra6 d4 Kc1 Kg8 Rg6+ Kf7 Rf6+ Kg8 Rb2 Rf7 Rxf7 Kxf7
Rb7+ Ke6 h7 f2 Rb6+ Kd5 Rf6 Bg7 Rxf2 Be4 Rf7 Bh8 Rf8


Fruit 2.2.1 Athlon 2009 MHz, 200 Mb HT, ChessPartner interface needs about ten
minutes

00:00:00.0	1,77	2	731	Ra7+ Kg8 Rxh7 Kxh7 Rxd5
00:00:00.0	1,77	3	1365	Ra7+ Kg8 Rxh7 Kxh7 Rxd5
00:00:00.0	1,06	4	2846	Ra7+ Kg8 Rxh7 Kxh7 Rxd5 f2 Rd7+ Kg6
00:00:00.0	1,79	4	6853	c6 Kg6 Kb3 Rc7 Raxd5 Rxc6 Rxe5
00:00:00.0	1,79	5	11825	c6 Kg6 Kb3 Rc7 Raxd5 Rxc6 Rxe5
00:00:00.1	1,18	6	32616	c6 Ke6 Rdxd5 f2 Rxe5+ Kd6 c7 Rxc7
00:00:00.2	1,50	7	72207	c6 Ke6 Raxd5 Bb4 Rd1 f2 Rd7 Rxd7 cxd7
00:00:00.2	1,40	8	141668	c6 Ke6 Raxd5 Bb4 Rd1 f2 Rd7 Bf3 Rf1
00:00:00.5	1,77	9	335343	c6 Ke6 Raxd5 Bb4 Rd1 f2 Rd7 Bf3 Rf1 Rxd7 cxd7
00:00:01.4	1,69	10	970123	c6 Kg8 Rc2 Rc7 h5 Rf7 Rf2 Rc7 Rxd5 Rxc6 Rxe5
00:00:03.1	1,49	11	2217061	c6 Kg6 Raxd5 f2 Rxf2 Bxd5 Rxf8 Rc7 Rf6+ Kh7 Rf5 Rxc6
Rxe5
00:00:06.5	1,67	12	4750730	c6 Kg8 Rb5 Rc7 Rb7 Rxc6 h7+ Kh8 Rb8 Kxh7 Rxf8 Kg7 Ra8
d4
00:00:10.1	1,86	13	7628834	c6 Kg8 Rc2 Rf7 Be3 Rc7 Bc5 Rc8 Bxf8 Rxf8 c7 Bh3 Rf2
Be6 Ra7
00:00:22.0	1,68	14	16977878	c6 Kg6 Raxd5 f2 Rxf2 Bxd5 Rxf8 Rc7 Rf6+ Kh7 Rf5 Rxc6
Rxe5 Rb6+ Kc3 Rb3+ Kd4
00:00:41.2	1,66	15	32554291	c6 Kg8 Rc2 Rc7 Rb5 d4 Rb7 Bd6 Rb8+ Kh7 Rd8 Bb4 Rd7+
Rxd7 cxd7 Ba5 Kb3 e4
00:01:06.1	1,67	16	52773090	c6 Kg8 Rc2 Rc7 Rb5 Rf7 Rb7 Bh3 Rxf7 Kxf7 c7 Bc8 Rc6
e4 Kc3 Kg8 Be3 Kf7
00:02:01.0	1,81	17	97802562	c6 Kg8 Rc2 Rc7 Rb5 Rf7 Rb7 Bh3 Rxf7 Kxf7 c7 Bc8 Rc6
e4 Kc3 Kg8 Be3 Kh7 Rb6
00:06:33.1	1,86	18	322321110	c6 Kg6 Raxd5 f2 Rxf2 Bxd5 Rxf8 Rc7 Rf6+ Kh7 Rf5
Bxc6 Rxe5 Rd7 Be3 Bd5 Kc3 Kg6 h5+ Kf6
00:10:48.4	2,37	18	537973507	Ra6 d4 Kc1 Kg8 Rg6+ Kf7 Rf6+ Kg8 Rb2 Rf7 Rxf7 Kxf7
Rb7+ Ke6 h7 f2 Rb6+ Kd5 Rf6 Bg7 Rxf2 Be4 Rf7 Bh8 Rf8
00:14:45.5	2,86	19	742598464	Ra6 d4 Kc1 Kg8 Rg6+ Kf7 Rf6+ Kg8 Rb2 Rf7 Rxf7 Kxf7
Rb7+ Ke6 h7 f2 Rb6+ Kd5 Rf6 Bg7 Rxf2 Be4 Rf7 Bh8 Rf8 Bg7
00:20:54.2	2,75	20	1055431667	Ra6 d4 Kc1 Kg8 Rg6+ Kf7 Rf6+ Kg8 Rb2 Rf7 Rxf7 Kxf7
Rb7+ Ke6 h7 f2 Rb6+ Kd5 Rf6 Bg7 Rxf2 Be4 Rf8 Bxh7 Rf7 Be4 Rxg7 Kxc5

I'm not seriously planning to do much testing with 'Fruitbat 0.1' , but it might
be a nice experiment to see how Fruit reacts to changed King Safety settings and
Uri's suggested History Treshold change in a few testpositions.

A favourite testposition:

Fruitbat 0.1 Experimental in Ferret-Gandalf position goes to Nf5 or e5 fairly
quickly where standard Fruit plays Bxb4 a lot longer. These moves don't change
the bad position White is in but have some fighting value. The line after Nf5
looks a bit better than e5 dxe5 Bxb4 to my taste

[D]rr4k1/5p2/3p2p1/3P3p/pnP1P2P/Bn2NR2/5PP1/qB1Q2K1 w - - 1 1

00:00:00.0	-0,91	1	3	Bxb4 Rxb4
00:00:00.0	-1,21	2	401	Bxb4 Rxb4 Qd3
00:00:00.0	0,01	3	1752	Bxb4 Nd4 Qe1 Nxf3+ gxf3
00:00:00.0	-0,13	4	2792	Bxb4 Nd4 Qe1 Nxf3+ gxf3 a3
00:00:00.0	-0,58	5	6285	Bxb4 Nd4 Qe1 Nxf3+ gxf3 Qb2 Nc2
00:00:00.1	-0,73	6	13808	Bxb4 Nd4 Qe1 Nxf3+ gxf3 Qb2 Nc2 a3
00:00:00.1	-0,90	7	32592	Bxb4 Nd4 Qe1 Qb2 Nc2 Nxc2 Bxc2 Rxb4
00:00:00.1	-0,90	8	67472	Bxb4 Nd4 Qe1 Qb2 Nc2 Nxc2 Bxc2 Rxb4
00:00:00.2	-1,14	9	138787	Bxb4 Nd4 Qe1 Qb2 Nc2 Nxc2 Bxc2 Rxb4 c5
00:00:00.5	-1,11	10	315628	Bxb4 Nd4 Qe1 Qb2 Nc2 Nxc2 Bxc2 Rxb4 c5 dxc5
00:00:00.9	-1,60	11	608072	Bxb4 Nd4 Qe1 Qb2 Nc2 Nxc2 Bxc2 Rxb4 Rc3 a3 e5
00:00:02.0	-1,79	12	1404787	Bxb4 Nd4 Qe1 Qb2 Nc2 Nxc2 Bxc2 Rxb4 c5 Rc4 Bd3 Rxc5
00:00:04.4	-2,09	13	3204405	Bxb4 Nd4 Qe1 Qb2 Nc2 Nxc2 Bxc2 Rxb4 c5 Rc4 Bd1 Rxc5
Be2
00:00:09.3	-2,24	14	6843245	Bxb4 Nd4 Qe1 Qb2 Nc2 Nxc2 Bxc2 Rxb4 c5 Rc4 Bd1 Rxc5
Be2 a3
00:00:24.7	-2,21	15	17890858	Bxb4 Nd4 Qe1 Qb2 Nc2 Nxc2 Bxc2 Rxb4 c5 Rc4 Bd1 Rxc5
Rg3 Rc1 Rd3
00:01:51.9	-3,35	16	80494739	Bxb4 Nd4 Qe1 a3 Kh2 Nxf3+ gxf3 Qe5+ Kg2 a2 Bxa2
Rxa2 Nd1 Ra1 Bc3 Rxd1 Bxe5 Rxe1 Bxd6 Rb2 c5
00:03:44.1	-2,87	16	159456054	e5 dxe5 Bxb4 Nd4 Qe1 Nxf3+ gxf3 Qb2 Nc2 a3 Bxa3
Qxb1 Qxb1 Rxb1+ Kg2 Ra4 c5 Rc4 Ne3 Rxh4
00:07:41.8	-2,76	17	323265043	e5 dxe5 Bxb4 Nd4 Qe1 Nxf3+ gxf3 Qb2 Nc2 a3 Bxa3
Qxb1 Qxb1 Rxb1+ Kg2 Ra4 Bd6 f6 Ne3 Ra2
00:16:32.0	-2,24	17	689266726	Nf5 Ra7 Nh6+ Kg7 Nf5+ gxf5 Rxf5 Qxa3 Qf3 Kf8 Qxh5
Qc1+ Kh2 Rab7 Qh8+ Ke7 Qf6+ Kd7 Qxf7+ Kd8 Qf8+ Kc7 Rf7+ Kb6 Qxd6+ Ka5 Rxb7 Rxb7
Qd8+ Rb6 Qa8+ Na6

I think this can be improved! Fabien, we definitely could use some more options
to test!

Maybe some more later..

 Eelco



This page took 0 seconds to execute

Last modified: Thu, 15 Apr 21 08:11:13 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.