Bikjump V3.0 is now available for download

[Bart Weststrate]

Aart something wrong in Cutechess cli, it's about the pv sent:

And with that goes my reputation for stability

Explanation: for my own understanding, I started to include quiescence moves for the PV after a non-standard % symbol, but since this is non UCI compliant, I should of course not ship that version. Let me push a non-version changing fix for that.

Are you most interested in Windows or Linux?

Windows please. Thanks in advance. Keep up the good work Maestro.

[abik]

I posted a new 64-bit BikJump V3.0 (Windows and Linux) on my website that no longer includes non-standard quiescence moves in the PV. Since this is a build change only (no source code changes), I kept the release version the same.

With gratitude to Bart Weststrate for reporting the issue!

The following can be used to ensure you have the right version.

Old version:

Code: Select all

position fen r3k2r/Pppp1ppp/1b3nbN/nP6/BBP1N3/q43/1p3NPP/3QRRK1 w kq - 0 1
go
info depth 1 seldepth 22 nodes 1396 tbhits 0 time 0 score cp 946 pv b4a3  g6e4 %a3b2 %
info depth 1 seldepth 26 nodes 3194 tbhits 0 time 2 score mate 2 pv e4f6  e8d8 %d1d7 %
bestmove e4f6  ponder e8d8

New version:

Code: Select all

position fen r3k2r/Pppp1ppp/1b3nbN/nP6/BBP1N3/q43/1p3NPP/3QRRK1 w kq - 0 1
go
info depth 1 seldepth 22 nodes 1396 tbhits 0 time 2 score cp 946 pv b4a3
info depth 1 seldepth 26 nodes 3194 tbhits 0 time 4 score mate 2 pv e4f6
bestmove e4f6

[Bart Weststrate]

Thanks Aart. No issues here anymore.

[Bart Weststrate]

First 500 games finished, 60/20+1

C:\0_cutechess>cutechess-cli -concurrency 10 -srand 1942127704 -openings file=IM_4mvs.pgn order=sequential -engine name=SM_2182b cmd=00_2182b.exe proto=uci -engine name=Bikjump cmd=bikjump.exe proto=uci -each restart=on tc=60/20+1 -draw movenumber=100 movecount=5 score=70 -resign movecount=5 score=999 -rounds 250 -pgnout sm_bikjump_4_moves_book.pgn -repeat 2 -games 2

Score of SM_2182b vs Bikjump: 320 - 128 - 52 [0.692] 500
... SM_2182b playing White: 162 - 68 - 20 [0.688] 250
... SM_2182b playing Black: 158 - 60 - 32 [0.696] 250
... White vs Black: 222 - 226 - 52 [0.496] 500
Elo difference: 140.6 +/- 31.0, LOS: 100.0 %, DrawRatio: 10.4 %
SPRT: llr 0 (0.0%), lbound -inf, ubound inf

Player: SM_2182b
"Draw by 3-fold repetition": 26
"Draw by adjudication": 10
"Draw by fifty moves rule": 10
"Draw by insufficient mating material": 6
"Loss: Black mates": 4
"Loss: Black wins by adjudication": 64
"Loss: White mates": 8
"Loss: White wins by adjudication": 52
"Win: Black wins by adjudication": 158
"Win: White mates": 1
"Win: White wins by adjudication": 161
Player: Bikjump
"Draw by 3-fold repetition": 26
"Draw by adjudication": 10
"Draw by fifty moves rule": 10
"Draw by insufficient mating material": 6
"Loss: Black wins by adjudication": 158
"Loss: White mates": 1
"Loss: White wins by adjudication": 161
"Win: Black mates": 4
"Win: Black wins by adjudication": 64
"Win: White mates": 8
"Win: White wins by adjudication": 52
Finished match

SM (Single Malt) is a basic (language is Powerbasic CC) engine, rating +-2.300. Just a beancounter, eval is PST + some pawn scores.

If i look at the games my Gut feeling says that Bikjump should play stronger than it does, it has more than enough depth and speed....

Next match started, same conditions, only time 60/60+1.

[abik]

Score of SM_2182b vs Bikjump: 320 - 128 - 52 [0.692] 500

Thanks for reporting! Just curious, do you have results under the same conditions of the previous BikJump 2.0.1 and/or the results for much faster games with BikJump 3? The former will say a bit about improvements (if any), the latter are more typical conditions I have optimized for. It is a bit disappointing to get defeated by a beancounter

[abik]

Oh, and if it is not too much trouble, can you copy and paste a game where BikJump is defeated here as annotated PGN, so I can have a look at the scores and progressions. And, many thanks again for your testing!

[Bart Weststrate]

Score of SM_2182b vs Bikjump: 320 - 128 - 52 [0.692] 500

Thanks for reporting! Just curious, do you have results under the same conditions of the previous BikJump 2.0.1 and/or the results for much faster games with BikJump 3? The former will say a bit about improvements (if any), the latter are more typical conditions I have optimized for. It is a bit disappointing to get defeated by a beancounter

I don't have played against Bikjump 2.0.1 yet but i can do that. Next match like the first (60/20+1 and sequential openings 8 ply). If i can download it somewhere....

[Bart Weststrate]

Oh, and if it is not too much trouble, can you copy and paste a game where BikJump is defeated here as annotated PGN, so I can have a look at the scores and progressions. And, many thanks again for your testing!

[pgn][Event "?"]
[Site "?"]
[Date "2026.04.10"]
[Round "233"]
[White "SM_2182b"]
[Black "Bikjump"]
[Result "1-0"]
[ECO "D10"]
[GameDuration "00:00:56"]
[GameEndTime "2026-04-10T07:47:39.167 W. Europe Daylight Time"]
[GameStartTime "2026-04-10T07:46:42.798 W. Europe Daylight Time"]
[Opening "Queen's Gambit declined"]
[PlyCount "56"]
[Termination "adjudication"]
[TimeControl "60/20+1"]
[Variation "Exchange variation"]

1.d4
{book}
1...d5
{book}
2.c4
{book}
2...c6
{book}
3.cxd5
{book}
3...cxd5
{book}
4.Nc3
{book}
4...Nc6
{book}
5.g3
{+0.21/11 0.87s}
5...Nf6
{+0.02/12 1.3s}
6.Bg2
{+0.11/13 0.77s}
6...Bf5
{+0.15/11 0.92s}
7.Nf3
{+0.19/12 0.94s}
7...e6
{+0.19/12 1.1s}
8.O-O
{+0.07/11 0.63s}
8...Qb6
{+0.31/11 0.97s}
9.Na4
{+0.07/10 1.1s}
9...Qb4
{+0.31/12 1.1s}
10.Bf4
{+0.34/10 0.87s}
10...b5
{+0.27/12 1.4s}
11.Rc1
{+0.52/14 0.86s}
11...Ne7
{-0.22/12 1.4s}
12.Nc5
{+0.86/12 0.70s}
12...Qxb2
{-0.44/12 0.96s}
13.Na6
{+1.34/13 0.68s}
13...Kd8
{-1.23/15 3.4s}
14.Bc7+
{+1.61/10 0.72s}
14...Ke8
{-2.22/15 4.4s}
15.Ba5
{+2.22/14 0.88s}
15...Rc8
{-2.43/15 1.3s}
16.Nc7+
{+2.06/15 0.80s}
16...Rxc7
{-2.48/16 1.3s}
17.Rxc7
{+1.87/15 0.70s}
17...Qxa2
{-2.84/13 1.3s}
18.Rxa7
{+2.29/15 0.82s}
18...Nc6
{-3.00/14 0.99s}
19.Ra8+
{+2.40/14 0.85s}
19...Kd7
{-3.72/15 1.3s}
20.Ne5+
{+3.33/14 0.82s}
20...Nxe5
{-5.07/14 1.3s}
21.dxe5
{+4.62/15 1.3s}
21...Ne8
{-8.63/12 1.3s}
22.e4
{+5.60/11 0.89s}
22...Bxe4
{-7.21/12 1.3s}
23.Bxe4
{+6.65/12 1.7s}
23...Bb4
{-8.31/12 0.93s}
24.Bxd5
{+9.57/12 0.77s}
24...Qxa5
{-10.79/13 1.1s}
25.Be4+
{+10.35/13 1.1s}
25...Ke7
{-10.83/17 1.3s}
26.Rxa5
{+10.34/16 1.3s}
26...Bxa5
{-11.42/16 0.93s}
27.Bc6
{+10.55/16 0.66s}
27...g6
{-11.86/15 0.99s}
28.Qb3
{+11.37/15 2.1s}
28...Nd6
{-12.72/13 1.3s, White wins by adjudication}
1-0


[/pgn]

[pgn][Event "?"]
[Site "?"]
[Date "2026.04.10"]
[Round "153"]
[White "Bikjump"]
[Black "SM_2182b"]
[Result "0-1"]
[ECO "B24"]
[GameDuration "00:01:01"]
[GameEndTime "2026-04-10T07:10:00.338 W. Europe Daylight Time"]
[GameStartTime "2026-04-10T07:08:58.957 W. Europe Daylight Time"]
[Opening "Sicilian"]
[PlyCount "65"]
[Termination "adjudication"]
[TimeControl "60/20+1"]
[Variation "Closed"]

1.e4
{book}
1...c5
{book}
2.Nc3
{book}
2...Nc6
{book}
3.g3
{book}
3...Rb8
{book}
4.f4
{book}
4...g6
{book}
5.Nf3
{+0.25/11 1.3s}
5...Bg7
{+0.10/12 0.87s}
6.Bc4
{+0.11/11 1.4s}
6...d6
{+0.28/12 0.63s}
7.a3
{+0.09/10 1.1s}
7...Nf6
{+0.72/12 0.63s}
8.d3
{+0.13/11 1.1s}
8...O-O
{+0.69/12 0.65s}
9.Qd2
{+0.03/10 1.4s}
9...Nd4
{+1.22/14 1.2s}
10.Nxd4
{-0.20/12 1.2s}
10...cxd4
{+1.27/14 0.92s}
11.Nd5
{-0.24/12 1.0s}
11...Nh5
{+1.36/15 0.77s}
12.Qf2
{-0.16/10 1.2s}
12...e6
{+2.87/16 1.2s}
13.Nb4
{-1.02/14 1.4s}
13...a5
{+2.90/15 0.76s}
14.g4
{-1.02/13 1.4s}
14...Nf6
{+3.59/14 1.0s}
15.g5
{-1.46/13 1.4s}
15...Ng4
{+3.63/14 0.74s}
16.Qe2
{-2.10/13 1.0s}
16...Ne3
{+2.90/15 0.71s}
17.Na2
{-2.23/14 1.4s}
17...b5
{+2.75/16 0.71s}
18.c3
{-3.11/14 1.4s}
18...bxc4
{+4.40/16 0.69s}
19.cxd4
{-3.11/14 1.2s}
19...Bxd4
{+4.45/16 0.76s}
20.Bxe3
{-3.87/14 1.3s}
20...Rxb2
{+4.53/16 0.91s}
21.Bxd4
{-4.32/15 1.1s}
21...Rxe2+
{+4.56/16 0.91s}
22.Kxe2
{-4.79/15 1.1s}
22...cxd3+
{+4.47/13 0.86s}
23.Kxd3
{-4.95/15 1.4s}
23...e5
{+4.98/14 1.2s}
24.Be3
{-5.03/14 0.92s}
24...Ba6+
{+5.34/14 0.93s}
25.Kc2
{-5.08/16 1.4s}
25...Qc8+
{+5.82/14 0.81s}
26.Nc3
{-5.53/15 1.1s}
26...Qh3
{+6.96/14 0.77s}
27.Rhe1
{-5.69/13 1.2s}
27...exf4
{+8.69/13 0.84s}
28.Bd2
{-8.98/13 1.2s}
28...Qd3+
{+9.33/14 1.1s}
29.Kc1
{-10.31/16 1.4s}
29...f3
{+9.83/15 0.82s}
30.Rh1
{-10.63/14 1.2s}
30...Rc8
{+15.50/17 0.70s}
31.Ra2
{-16.98/17 1.4s}
31...Rxc3+
{+16.54/17 1.1s}
32.Bxc3
{-18.65/18 1.4s}
32...Qxc3+
{+17.49/17 1.9s}
33.Rc2
{-20.36/18 1.4s, Black wins by adjudication}
0-1[/pgn]

[pgn][Event "?"]
[Site "?"]
[Date "2026.04.10"]
[Round "52"]
[White "SM_2182b"]
[Black "Bikjump"]
[Result "1-0"]
[ECO "B05"]
[GameDuration "00:01:02"]
[GameEndTime "2026-04-10T06:23:47.794 W. Europe Daylight Time"]
[GameStartTime "2026-04-10T06:22:45.209 W. Europe Daylight Time"]
[Opening "Alekhine's defense"]
[PlyCount "66"]
[Termination "adjudication"]
[TimeControl "60/20+1"]
[Variation "Modern variation, 4...Bg4"]

1.e4
{book}
1...Nf6
{book}
2.e5
{book}
2...Nd5
{book}
3.d4
{book}
3...d6
{book}
4.Nf3
{book}
4...Bg4
{book}
5.Be2
{+0.44/12 0.87s}
5...Nc6
{-0.09/11 1.1s}
6.c4
{+0.54/14 0.91s}
6...Ndb4
{-0.31/12 1.3s}
7.a3
{+0.96/13 0.75s}
7...Na6
{-0.39/13 1.3s}
8.exd6
{+1.06/13 1.6s}
8...Qxd6
{-0.36/11 1.1s}
9.Nbd2
{+1.02/12 0.85s}
9...Bf5
{-0.21/10 1.4s}
10.d5
{+0.99/13 1.1s}
10...Ne5
{-0.55/14 1.3s}
11.Nxe5
{+1.11/12 1.7s}
11...Qxe5
{-0.52/14 1.3s}
12.Nf3
{+0.97/12 0.81s}
12...Qf6
{-0.53/13 1.3s}
13.O-O
{+0.90/12 1.5s}
13...Nc5
{-0.40/12 1.2s}
14.Nd4
{+0.99/11 0.62s}
14...Bd7
{-0.48/12 1.4s}
15.Be3
{+1.03/12 0.73s}
15...a6
{-0.57/11 1.4s}
16.b4
{+1.36/13 1.0s}
16...Na4
{-0.77/11 1.1s}
17.Qc2
{+1.60/10 0.87s}
17...Qe5
{-0.79/11 1.1s}
18.c5
{+2.12/12 0.73s}
18...b5
{-1.04/12 1.3s}
19.c6
{+2.26/11 0.80s}
19...Bc8
{-1.03/12 1.3s}
20.Rad1
{+2.03/13 0.96s}
20...e6
{-0.89/11 1.3s}
21.dxe6
{+2.15/13 0.79s}
21...fxe6
{-1.05/13 1.1s}
22.g3
{+1.94/12 0.76s}
22...Qf6
{-1.01/12 1.2s}
23.Bg4
{+1.93/11 0.78s}
23...h5
{-0.86/12 1.2s}
24.Bh3
{+1.84/12 0.68s}
24...Kf7
{-0.72/10 1.4s}
25.f4
{+2.29/9 0.70s}
25...Qe7
{-1.87/11 3.2s}
26.f5
{+5.25/14 1.2s}
26...Qe8
{-4.39/12 1.3s}
27.fxe6+
{+7.51/13 1.1s}
27...Kg8
{-5.40/14 1.3s}
28.Rf7
{+7.90/15 1.2s}
28...Rh7
{-6.94/11 0.96s}
29.Rdf1
{+12.03/14 1.4s}
29...Bd6
{-12.27/13 1.3s}
30.Qg6
{+18.05/15 1.1s}
30...Bxe6
{-13.28/13 1.3s}
31.Nxe6
{+1599.84/15 0.73s}
31...Kh8
{-M12/10 0.049s}
32.Bd4
{+1599.88/19 0.69s}
32...Be5
{-M10/4 0.006s}
33.Bxe5
{+1599.92/23 0.67s}
33...Qxf7
{-M8/4 0.003s, White wins by adjudication}
1-0
[/pgn]

[Punga2026]

Hello. Do you also have the Android version? Thank you

[Bart Weststrate]

This is the second run 60/60+1:

Score of SM_2182b vs Bikjump: 343 - 91 - 66 [0.752] 500
... SM_2182b playing White: 181 - 41 - 28 [0.780] 250
... SM_2182b playing Black: 162 - 50 - 38 [0.724] 250
... White vs Black: 231 - 203 - 66 [0.528] 500
Elo difference: 192.7 +/- 32.1, LOS: 100.0 %, DrawRatio: 13.2 %