LEARNING SESSION 02

ECO code, last opening move, start position Most likely
endgame ELO scores after
500, 1000, 1500, 2000
games

a00 g2g4 rnbqkbnr/pppppppp/8/8/6P1/8/PPPPPP1P/RNBQKBNR b KQkq g3 0 1 KRNPPvKRB 46,8 42,6 41,2 41,7

a46 e8g8 r1bq1rk1/1ppnbpp1/p2pp2p/8/3PP3/2N5/PPPQBPPP/2KRN2R w - - 6 11 KRBPvKRRP 43,3 49,0 43,5 47,2

a75 a4a5 r4rk1/1p1nqpbp/p2p1np1/P1pP4/4PB2/2N2B2/1P3PPP/R2QR1K1 b - - 0 14 KRPPvKRPP 0,0 0,7 0,9 0,2

b06 f6d7 r1bq1rk1/1ppn1pb1/p1npp1pp/4P3/2BP4/2N1BN2/PPP1QPPP/2KR3R w - - 1 11 KRPPvKRPP 0,0 1,4 1,6 2,8

b06 f8e8 r1bqr1k1/ppp2pbp/2np1np1/8/4PP2/2N1B2P/PPPQB1P1/2KR2NR w - - 3 11 KQBPvKQPP -5,6 -23,3 -28,3 -29,1

b08 h7h5 r1bq1rk1/pp1n1pb1/2p2np1/4p1Bp/4P2P/2NB1N2/PPPQ1PP1/2KR3R w - h6 0 11 KRBPvKBBP 8,3 -2,1 -7,2 -10,1

b16 e8c8 2kr1b1r/ppqnpp1p/2p2p2/8/2PP2b1/5N2/PP2BPPP/R1BQ1RK1 w - - 1 11 KRPPPvKRP
KRNPPvKRB -10,4 -0,3 4,2 7,6

b18 g8f6 2kr1b1r/ppqn1ppp/2p1pnb1/8/2BP4/2P2NN1/PP2QPPP/R1B2RK1 w - - 1 11 KRPPvKRPP -2,1 5,9 9,0 7,6

b34 a7a5 r2q1rk1/1p2ppbp/2npbnp1/p7/4P3/1NN1BP2/PPPQ2PP/2KR1B1R w - a6 0 11 KRPPPvKRP 35,6 29,6 23,2 9,9

b63 a7a6 r1b2rk1/1p2bppp/pqnppn2/6B1/4P3/1NN5/PPPQBPPP/2KR3R w - - 0 11 KRPPvKRPP -15,3 -29,6 -14,4 -4,2

b88 d8b8 rq3rk1/pp1bbppp/2nppn2/8/3NP3/1BN1B3/PPP1QPPP/2KR3R w - - 8 11 KRPPvKRPP 1,4 0,0 -3,0 3,5

b89 c6a5 r1b2rk1/ppq1bppp/3ppn2/n7/2BNP3/2N1B3/PPP1QPPP/2KR2R1 w - - 8 11 KRPPvKRPP 47,5 54,3 50,6 47,2

b89 d8e8 r1b1qrk1/1p2bppp/p1nppn2/8/2BNP3/2N1B3/PPP1QPPP/2KR2R1 w - - 2 11 KRPPPvKRP 13,2 20,5 19,0 18,6

b99 h1g1 r1b1kb1r/2q2p1p/p2ppP2/1pn5/3NP3/2N2Q2/PPP4P/2KR1BR1 b kq - 2 16 KRNPPvKRN 2,1 -2,4 -6,7 -7,8

c18 e2c3 r3k1r1/ppqbnp2/2n1p3/3pP3/5P2/P1NQ4/2P3PP/R1B1KB1R b KQq - 0 13 KRPPvKRPP -3,5 -2,4 -2,1 -1,9

c51 b3c2 r2q1knr/pppb1Bpp/1b1p4/n7/3PP3/2N2N2/P1Q2PPP/R1B2RK1 b - - 2 13 KRPPvKRNP 6,9 -2,1 -8,1 -9,7

c55 e8c8 2kr3r/ppp2ppp/1bn1bPq1/6N1/2ppNPP1/8/PPP4P/R1BQR1K1 w - - 1 14 KRNPvKRPP 31,4 25,1 25,8 25,8

d17 b8a6 r3k2r/pp3ppp/n1p1p3/3qN3/P1p5/8/1PK1Q1PP/R4B1R w kq - 3 15 KQBPvKQPP
KRBPvKRPP 1,4 -4,9 -5,6 -4,5

d31 e8g8 r2q1rk1/1b1n1ppp/4pn2/p7/1pPP4/3BPN2/1B3PPP/R2Q1RK1 w - - 6 15 KRPPvKRPP 17,4 13,9 9,7 8,5

d68 d7b6 r1b2rk1/pp2qppp/1np5/8/2BP4/2R2N2/PPQ2PPP/5RK1 w - - 1 15 KRPPvKRPP 1,4 1,4 0,9 0,9

d69 e7e5 r1b2rk1/pp3ppp/2p5/4q3/2B5/2R1P3/PP3PPP/3Q1RK1 w - - 0 15 KRPPvKRPP -0,7 -0,3 -0,2 -0,2

d89 e6a2 r2q1rk1/pp2ppbp/6p1/n7/3PP3/3BBP2/b3N1PP/2RQ1RK1 w - - 0 15 KRPPvKRPP 0,0 0,3 0,2 0,2

d89 g7a1 r2q1rk1/pp2pp1p/4b1p1/n2P4/4P3/3BBP2/P3N1PP/b2Q1RK1 w - - 0 15 KBPPvKNPP 0,0 0,3 -0,2 -0,2

e21 e6e5 rnb2rk1/pp1p1ppp/6q1/2pPp3/1bP1N3/2NQ4/PP2PPPP/2KR1B1R w - - 0 11 KRPPPvKRP 39,1 67,5 64,9 63,0

elo no doubt, there are the best moves and the most effective !

b16 e8c8 :

When all engines have a learning feature :

- they penalize their best moves when they lose => they improve their immediate defenses => their opponents need more and more moves to win
- they favor their best moves when they win => they improve the "long term" effectiveness of their attacks

get by pgnToWinLength.exe, contact me to get your personal copy of pgnToWinLength

b34 a7a5 :

When all engines have a learning feature, they start by playing their best moves.

After a few games :
- if they lost several times, their respective best moves are sufficiently penalized to try their next best moves from their evaluations
- if they won several times, they continue by playing their respective best moves

[Eman experience] start by playing :
11...Rc8 (score @ +1.02/40, white's win rate @ 78%) then
11...Nh5 (score @ +1.74/40, white's win rate @ 86%) then
11...Ne8 (score @ +1.44/45, white's win rate @ 88%) then
11...Qc8 (score @ +1.04/40, white's win rate @ 65%)

[BrainLearn] start by playing :
11...Ne8 (score @ +0.74/32, white's win rate @ 88%) or
11...Nh5 (score @ +0.80/33, white's win rate @ 86%) then
11...Qc8 (score @ +0.57/35, white's win rate @ 65%)

get by pgnToMovesFreq.exe, contact me to get your personal copy of pgnToMovesFreq

Statistics

005 players                     : avg. D30, 001 613 ms/move, 048 000 games, 04 958 153 moves

brainlearn 13.1 nn-6762d36ad265 : avg. D24, 001 508 ms/move, 016 000 games, 01 060 006 moves
brainlearn 13.1 nn-ac07bd334b62 : avg. D30, 001 431 ms/move, 014 000 games, 00 530 305 moves
brainlearn 13.1 nn-cdf1785602d6 : avg. D28, 001 455 ms/move, 010 000 games, 00 521 758 moves
brainlearn 15.2 nn-ac07bd334b62 : avg. D30, 001 599 ms/move, 008 000 games, 00 364 988 moves
eman experience                 : avg. D33, 001 731 ms/move, 048 000 games, 02 481 096 moves

Averages :
0 167 sec/game (3 min/game)
131 plies/game (103 played plies/game)

Game stages :
- opening @ avg. D27, 001 984 ms/move
- middleg @ avg. D29, 001 578 ms/move
- endgame @ avg. D35, 001 353 ms/move

Ordo

# PLAYER                             :  RATING  ERROR   POINTS  PLAYED   (%)     W      D     L  D(%)  OppAvg  OppN
1 brainlearn 13.1 nn-cdf1785602d6    :       9      5   5129.5   10000  51.3   853   8553   594  85.5       0     1
2 Eman experience                    :       0   ----  24620.5   48000  51.3  7530  34181  6289  71.2      -9     4
3 brainlearn 15.2 nn-ac07bd334b62    :      -2      5   3977.0    8000  49.7  1233   5488  1279  68.6       0     1
4 brainlearn 13.1 nn-ac07bd334b62    :      -7      4   6858.5   14000  49.0   735  12247  1018  87.5       0     1
5 brainlearn 13.1 nn-6762d36ad265    :     -26      4   7414.5   16000  46.3  3468   7893  4639  49.3       0     1

White advantage = 54.50 +/- 1.13
Draw rate (equal opponents) = 50.00 % +/- 0.00

ECO code, last opening move, start position		Most likely endgame	ELO scores after 500, 1000, 1500, 2000 games
a00 g2g4	rnbqkbnr/pppppppp/8/8/6P1/8/PPPPPP1P/RNBQKBNR b KQkq g3 0 1	KRNPPvKRB	46,8	42,6	41,2	41,7
a46 e8g8	r1bq1rk1/1ppnbpp1/p2pp2p/8/3PP3/2N5/PPPQBPPP/2KRN2R w - - 6 11	KRBPvKRRP	43,3	49,0	43,5	47,2
a75 a4a5	r4rk1/1p1nqpbp/p2p1np1/P1pP4/4PB2/2N2B2/1P3PPP/R2QR1K1 b - - 0 14	KRPPvKRPP	0,0	0,7	0,9	0,2
b06 f6d7	r1bq1rk1/1ppn1pb1/p1npp1pp/4P3/2BP4/2N1BN2/PPP1QPPP/2KR3R w - - 1 11	KRPPvKRPP	0,0	1,4	1,6	2,8
b06 f8e8	r1bqr1k1/ppp2pbp/2np1np1/8/4PP2/2N1B2P/PPPQB1P1/2KR2NR w - - 3 11	KQBPvKQPP	-5,6	-23,3	-28,3	-29,1
b08 h7h5	r1bq1rk1/pp1n1pb1/2p2np1/4p1Bp/4P2P/2NB1N2/PPPQ1PP1/2KR3R w - h6 0 11	KRBPvKBBP	8,3	-2,1	-7,2	-10,1
b16 e8c8	2kr1b1r/ppqnpp1p/2p2p2/8/2PP2b1/5N2/PP2BPPP/R1BQ1RK1 w - - 1 11	KRPPPvKRP KRNPPvKRB	-10,4	-0,3	4,2	7,6
b18 g8f6	2kr1b1r/ppqn1ppp/2p1pnb1/8/2BP4/2P2NN1/PP2QPPP/R1B2RK1 w - - 1 11	KRPPvKRPP	-2,1	5,9	9,0	7,6
b34 a7a5	r2q1rk1/1p2ppbp/2npbnp1/p7/4P3/1NN1BP2/PPPQ2PP/2KR1B1R w - a6 0 11	KRPPPvKRP	35,6	29,6	23,2	9,9
b63 a7a6	r1b2rk1/1p2bppp/pqnppn2/6B1/4P3/1NN5/PPPQBPPP/2KR3R w - - 0 11	KRPPvKRPP	-15,3	-29,6	-14,4	-4,2
b88 d8b8	rq3rk1/pp1bbppp/2nppn2/8/3NP3/1BN1B3/PPP1QPPP/2KR3R w - - 8 11	KRPPvKRPP	1,4	0,0	-3,0	3,5
b89 c6a5	r1b2rk1/ppq1bppp/3ppn2/n7/2BNP3/2N1B3/PPP1QPPP/2KR2R1 w - - 8 11	KRPPvKRPP	47,5	54,3	50,6	47,2
b89 d8e8	r1b1qrk1/1p2bppp/p1nppn2/8/2BNP3/2N1B3/PPP1QPPP/2KR2R1 w - - 2 11	KRPPPvKRP	13,2	20,5	19,0	18,6
b99 h1g1	r1b1kb1r/2q2p1p/p2ppP2/1pn5/3NP3/2N2Q2/PPP4P/2KR1BR1 b kq - 2 16	KRNPPvKRN	2,1	-2,4	-6,7	-7,8
c18 e2c3	r3k1r1/ppqbnp2/2n1p3/3pP3/5P2/P1NQ4/2P3PP/R1B1KB1R b KQq - 0 13	KRPPvKRPP	-3,5	-2,4	-2,1	-1,9
c51 b3c2	r2q1knr/pppb1Bpp/1b1p4/n7/3PP3/2N2N2/P1Q2PPP/R1B2RK1 b - - 2 13	KRPPvKRNP	6,9	-2,1	-8,1	-9,7
c55 e8c8	2kr3r/ppp2ppp/1bn1bPq1/6N1/2ppNPP1/8/PPP4P/R1BQR1K1 w - - 1 14	KRNPvKRPP	31,4	25,1	25,8	25,8
d17 b8a6	r3k2r/pp3ppp/n1p1p3/3qN3/P1p5/8/1PK1Q1PP/R4B1R w kq - 3 15	KQBPvKQPP KRBPvKRPP	1,4	-4,9	-5,6	-4,5
d31 e8g8	r2q1rk1/1b1n1ppp/4pn2/p7/1pPP4/3BPN2/1B3PPP/R2Q1RK1 w - - 6 15	KRPPvKRPP	17,4	13,9	9,7	8,5
d68 d7b6	r1b2rk1/pp2qppp/1np5/8/2BP4/2R2N2/PPQ2PPP/5RK1 w - - 1 15	KRPPvKRPP	1,4	1,4	0,9	0,9
d69 e7e5	r1b2rk1/pp3ppp/2p5/4q3/2B5/2R1P3/PP3PPP/3Q1RK1 w - - 0 15	KRPPvKRPP	-0,7	-0,3	-0,2	-0,2
d89 e6a2	r2q1rk1/pp2ppbp/6p1/n7/3PP3/3BBP2/b3N1PP/2RQ1RK1 w - - 0 15	KRPPvKRPP	0,0	0,3	0,2	0,2
d89 g7a1	r2q1rk1/pp2pp1p/4b1p1/n2P4/4P3/3BBP2/P3N1PP/b2Q1RK1 w - - 0 15	KBPPvKNPP	0,0	0,3	-0,2	-0,2
e21 e6e5	rnb2rk1/pp1p1ppp/6q1/2pPp3/1bP1N3/2NQ4/PP2PPPP/2KR1B1R w - - 0 11	KRPPPvKRP	39,1	67,5	64,9	63,0
elo	no doubt, there are the best moves and the most effective !