I can't make it move under Arena

[SzotsGabor]

I installed it as a batch file in which there is this line: java -Xmx1024M -jar duch-1.0.0.jar

However, when out of book, it does not move. Here is Arena debug:

2022-03-02 14:24:21,6241------------------Starting engine 1 DuchChess 1.0.0 64-bit------------------
2022-03-02 14:24:21,6241Configured Engine 1 Type: UCI
2022-03-02 14:24:21,6241Engine 1 dir: E:\Sakk\UCI-engines\DuchChess
2022-03-02 14:24:21,6241Engine 1 commandline: E:\Sakk\UCI-engines\DuchChess\DuchChess_1.0.0-x64.bat
2022-03-02 14:24:21,7441Child Process Prio Adj: PID 14072 conhost.exe
2022-03-02 14:24:21,7441Child Process Prio Adj: PID 12272 java.exe
2022-03-02 14:24:21,7441Engine 1 ProcessID: 11804
2022-03-02 14:24:21,7441Engine 1 Prio:32 ThreadPrio:0
2022-03-02 14:24:21,782<--1:E:\Sakk\UCI-engines\DuchChess>java -Xmx1024M -jar duch-1.0.0.jar
2022-03-02 14:24:21,782-->1:uci
2022-03-02 14:24:21,965<--1:uci id duch 0.0.1
2022-03-02 14:24:21,965<--1:uci author Sebastian Pravda
2022-03-02 14:24:21,965<--1:uciok
2022-03-02 14:24:21,9651Child Process Prio Adj: PID 14072 conhost.exe
2022-03-02 14:24:21,9651Child Process Prio Adj: PID 12272 java.exe
2022-03-02 14:24:21,965-->1:isready
2022-03-02 14:24:22,043<--1:readyok
2022-03-02 14:24:27,785**----------New game---2022-03-02 14:24:27,785 Sze -------------
2022-03-02 14:24:27,793**Loading book: E:\Sakk\Arena1\Books\ShortLines.abk
2022-03-02 14:24:32,7291Start calc, move no: 1
2022-03-02 14:24:32,7291Main Book Move ShortLines.abk: e7e5
2022-03-02 14:24:35,4051Start calc, move no: 3
2022-03-02 14:24:35,4051Main Book Move ShortLines.abk: g8f6
2022-03-02 14:24:37,7141Start calc, move no: 5
2022-03-02 14:24:37,714-->1:ucinewgame
2022-03-02 14:24:37,714-->1:isready
2022-03-02 14:24:37,843<--1:readyok
2022-03-02 14:24:37,863-->1:position startpos moves e2e4 e7e5 f1c4 g8f6 c4f7
2022-03-02 14:24:37,863-->1:go wtime 118969 btime 122000 winc 1000 binc 1000

[rwbc]

Same here. (CuteChess GUI)

I checked the sources and uci output and currently it only understands 'go depth' and 'go movetime',
thus only exact movetime and exact depth play is possible for now.

It would need support for wtime/btime and winc/binc and adapted time management.

0.000: - Starting server: java -jar DuchChess100.jar
0.062: - Priority class set to normal
0.062: - Thread priority set to normal
2.231: > uci
2.231: < uci id duch 0.0.1
2.231: < uci author Sebastian Pravda
2.231: < uciok
3.697: > isready
3.697: < readyok
16.723: > position startpos moves e2e4 e7e5 f1c4 g8f6 c4f7
21.263: > go depth 12
21.278: < info depth 1 score cp 218 nodes 5 nps 416 time 12 pv e8f7 
21.278: < info depth 2 score cp 189 nodes 77 nps 4812 time 16 pv e8f7 g1f3 
21.294: < info depth 3 score cp 218 nodes 258 nps 8322 time 31 pv e8f7 g1f3 b8c6 
21.294: < info depth 4 score cp 189 nodes 552 nps 14153 time 39 pv e8f7 g1f3 b8c6 b1c3 
21.310: < info depth 5 score cp 204 nodes 1233 nps 23711 time 52 pv e8f7 g1f3 b8c6 b1c3 f8c5 
21.325: < info depth 6 score cp 200 nodes 2384 nps 36121 time 66 pv e8f7 g1f3 b8c6 b1c3 f8c5 d1e2 
21.356: < info depth 7 score cp 210 nodes 5947 nps 65351 time 91 pv e8f7 g1f3 b8c6 b1c3 f8c5 f3g5 f7g6 
21.388: < info depth 8 score cp 204 nodes 12160 nps 96507 time 126 pv e8f7 g1f3 b8c6 b1c3 f8c5 f3g5 f7g6 g5f3 
21.434: < info depth 9 score cp 210 nodes 27714 nps 161127 time 172 pv e8f7 g1f3 b8c6 b1c3 f8c5 d2d3 h8f8 c1g5 d8e7 
21.497: < info depth 10 score cp 202 nodes 57707 nps 243489 time 237 pv e8f7 g1f3 b8c6 b1c3 f8d6 d2d3 d6c5 c1e3 c5e3 f2e3 
21.559: < info depth 11 score cp 203 nodes 92121 nps 307070 time 300 pv e8f7 g1f3 b8c6 b1c3 f8d6 d2d3 d6b4 a1b1 b4c5 c1e3 c5e3 
21.653: < info depth 12 score cp 201 nodes 143732 nps 360230 time 399 pv e8f7 g1f3 b8c6 b1c3 f8d6 d2d3 h8f8 c1e3 f7g6 e1e2 f6g4 e3g5 
21.653: < bestmove e8f7

0.000: - Starting server: java -jar DuchChess100.jar
0.063: - Priority class set to normal
0.063: - Thread priority set to normal
2.699: > uci
2.699: < uci id duch 0.0.1
2.699: < uci author Sebastian Pravda
2.699: < uciok
4.150: > isready
4.150: < readyok
10.390: > position startpos moves e2e4 e7e5 f1c4 g8f6 c4f7
15.725: > go movetime 10000
15.741: < info depth 1 score cp 218 nodes 5 nps 384 time 13 pv e8f7 
15.741: < info depth 2 score cp 189 nodes 77 nps 4529 time 17 pv e8f7 g1f3 
15.756: < info depth 3 score cp 218 nodes 258 nps 8062 time 32 pv e8f7 g1f3 b8c6 
15.756: < info depth 4 score cp 189 nodes 552 nps 14153 time 39 pv e8f7 g1f3 b8c6 b1c3 
15.772: < info depth 5 score cp 204 nodes 1233 nps 22418 time 55 pv e8f7 g1f3 b8c6 b1c3 f8c5 
15.803: < info depth 6 score cp 200 nodes 2384 nps 33111 time 72 pv e8f7 g1f3 b8c6 b1c3 f8c5 d1e2 
15.819: < info depth 7 score cp 210 nodes 5947 nps 66077 time 90 pv e8f7 g1f3 b8c6 b1c3 f8c5 f3g5 f7g6 
15.850: < info depth 8 score cp 204 nodes 12160 nps 101333 time 120 pv e8f7 g1f3 b8c6 b1c3 f8c5 f3g5 f7g6 g5f3 
15.897: < info depth 9 score cp 210 nodes 27714 nps 167963 time 165 pv e8f7 g1f3 b8c6 b1c3 f8c5 d2d3 h8f8 c1g5 d8e7 
15.959: < info depth 10 score cp 202 nodes 57707 nps 251995 time 229 pv e8f7 g1f3 b8c6 b1c3 f8d6 d2d3 d6c5 c1e3 c5e3 f2e3 
16.022: < info depth 11 score cp 203 nodes 92121 nps 309130 time 298 pv e8f7 g1f3 b8c6 b1c3 f8d6 d2d3 d6b4 a1b1 b4c5 c1e3 c5e3 
16.115: < info depth 12 score cp 201 nodes 143732 nps 360230 time 399 pv e8f7 g1f3 b8c6 b1c3 f8d6 d2d3 h8f8 c1e3 f7g6 e1e2 f6g4 e3g5 
16.349: < info depth 13 score cp 203 nodes 255712 nps 414444 time 617 pv e8f7 g1f3 b8c6 b1c3 f8e7 d2d3 h8f8 c1g5 e7b4 d1d2 d8e7 e1e2 b4c5 
17.332: < info depth 14 score cp 201 nodes 579487 nps 358815 time 1615 pv e8f7 g1f3 b8c6 b1c3 f8d6 d2d3 h8f8 c1e3 d8e7 d1d2 d6c5 e1e2 f7g6 h1g1 
17.675: < info depth 15 score cp 201 nodes 802457 nps 412786 time 1944 pv e8f7 g1f3 b8c6 b1c3 f8d6 d2d3 h8f8 c1e3 d8e7 d1d2 d6c5 e1e2 f7g6 h1g1 g6h5 
18.096: < info depth 16 score cp 201 nodes 1114145 nps 470103 time 2370 pv e8f7 g1f3 b8c6 b1c3 f8d6 d2d3 h8f8 c1e3 d8e7 d1d2 d6c5 e1e2 c5e3 d2e3 e7b4 a1b1 
25.725: < info depth 17 score cp 201 nodes 8853234 nps 885146 time 10002 pv e8f7 g1f3 b8c6 b1c3 f8c5 d2d3 d7d6 c1g5 c8e6 g5h4 c5b4 f3g5 f7e7 g5e6 b4c3 e1e2 e7e6 
25.725: < bestmove e8f7

[rwbc]

Here is an example with a working game in CuteChess with fixed time per move setup.
(This was with an allowed time margin of 50ms, because I noticed it tries to use exactly all time,
which would very often lead to a time loss, sometimes it oversteps by may be 20ms here)

The game also showed an interesting bug, which I will report in another issue.
(see R blunder out of the blue in a drawn endgame)

[Event "My Tournament"]
[Site "RWBC-CAPPUCCINO Win7U64 Q8200 2.33Ghz"]
[Date "2022.03.05"]
[Round "1.1.1"]
[White "DuchChess_100"]
[Black "Vice_10-64"]
[Result "0-1"]
[ECO "D06"]
[GameDuration "00:07:58"]
[GameEndTime "2022-03-05T14:04:15.556 Mitteleuropäische Zeit"]
[GameStartTime "2022-03-05T13:56:16.826 Mitteleuropäische Zeit"]
[Opening "Queen's Gambit declined"]
[PlyCount "102"]
[TimeControl "5/move"]
[Variation "Symmetrical (Austrian) defense"]

1. d4 {book} d5 {book}
2. c4 {book} c5 {book}
3. Nf3 {book} cxd4 {book}
4. cxd5 {+0.19/17 5.01} Qxd5 {-0.05/9 4.96}
5. Nc3 {+0.28/17 5.00} Qd7 {-0.10/9 4.96}
6. Qxd4 {+0.37/16 5.00} Nc6 {-0.25/9 4.96}
7. Qa4 {+0.27/18 5.00} e5 {-0.15/8 4.95}
8. Bg5 {+0.53/16 5.00} Be7 {-0.20/8 4.96}
9. Rd1 {+0.49/14 5.00} Qg4 {+0.05/9 4.96}
10. Qxg4 {+0.42/16 5.00} Bxg4 {+0.10/10 4.96}
11. Nd5 {+0.31/17 5.00} Rc8 {+0.15/9 4.96}
12. Nxe7 {+0.25/16 5.00} Ngxe7 {+0.15/10 4.96}
13. b3 {+0.28/18 5.00} f6 {+0.50/9 4.96}
14. Be3 {+0.24/16 5.00} Nb4 {+0.55/9 4.96}
15. Bxa7 {+0.15/18 5.00} Nxa2 {+0.55/9 4.96}
16. Ra1 {+0.10/17 5.00} Nb4 {+0.70/10 4.96}
17. Ra4 {0.00/20 5.00} Rc1+ {+0.80/9 4.96}
18. Kd2 {0.00/26 5.00} Rc2+ {+0.55/10 4.96}
19. Ke1 {0.00/25 5.00} Nec6 {+0.55/9 4.96}
20. Bb6 {-0.33/16 5.00} O-O {+1.10/10 4.96}
21. Kd1 {-0.54/19 5.00} Rc3 {+1.25/9 4.96}
22. Ra3 {-0.62/20 5.00} Bf5 {+1.35/8 4.96}
23. Nh4 {-0.87/17 5.00} Bc2+ {+1.55/9 4.96}
24. Kd2 {-0.93/19 5.00} Nd5 {+1.55/10 4.95}
25. e3 {-0.96/16 5.00} Bxb3 {+1.45/9 4.96}
26. Bd3 {-0.88/18 5.00} Ncb4 {+1.25/8 4.96}
27. Bf5 {-0.71/20 5.00} g6 {+1.40/8 4.95}
28. Rb1 {-0.19/20 5.00} Ba2 {+0.75/9 4.96}
29. Rxa2 {-0.09/22 5.00} gxf5 {+1.05/10 4.96}
30. Rab2 {-0.29/24 5.00} Rfc8 {+1.05/10 4.96}
31. Rxb4 {-0.17/25 5.00} Nxb4 {+1.05/9 4.96}
32. Rxb4 {-0.17/23 5.00} Rc2+ {+0.85/11 4.96}
33. Ke1 {-0.17/21 5.00} Rc1+ {+0.80/11 4.95}
34. Ke2 {+0.07/24 5.00} R8c2+ {+0.50/11 4.95}
35. Kf3 {+0.06/23 5.00} Rf1 {+0.55/11 4.96}
36. e4 {+0.12/23 5.00} f4 {+0.40/10 4.96}
37. Rb3 {+0.16/19 5.00} Rd1 {+0.50/10 4.95}
38. Nf5 {+0.19/18 5.00} Rdd2 {+0.35/10 4.96}
39. Ne7+ {+0.19/22 5.00} Kf7 {+0.50/9 4.96}
40. Nd5 {+0.17/20 5.00} f5 {+0.30/9 4.96}
41. Ba7 {+0.25/16 5.00} b6 {+0.50/9 4.95}
42. Bxb6 {+0.23/25 5.00} Rc4 {+0.40/10 4.96}
43. Nxf4 {+0.23/27 5.00} exf4 {+0.40/10 4.96}
44. exf5 {+0.23/26 5.01} Rd5 {+0.45/12 4.96}
45. Kg4 {+0.23/24 5.00} Kf6 {+0.45/10 4.96}
46. Ra3 {+0.25/21 5.00} h5+ {+0.60/10 4.95}
47. Kxh5 {-0.09/20 5.00} Kxf5 {+0.60/9 4.96}
48. g4+ {-0.20/21 5.00} fxg3 {+3.50/10 4.96}
49. Rd3 {+1.11/22 5.00} Rxd3 {+9.75/11 4.96}
50. f3 {-9.88/22 5.00} gxh2 {+299.93/9 4.95}
51. Bd4 {-327.51/24 5.00} h1=Q# {+299.99/8 4.95}
{Black mates}
0-1

[starvy]

I see, I will take a look at it when I have little time.
Thanks a lot for your feedback

[starvy]

@rwbc What makes chess engine very strong (like top 100 in CCRL)? I think I've implemented many search optimalization techniques but the engine is still computing a lot of nodes. Is it the evaluation that makes the difference? Is it value tuning? Or is it something else?

[rwbc]

@rwbc What makes chess engine very strong (like top 100 in CCRL)? I think I've implemented many search optimalization techniques but the engine is still computing a lot of nodes. Is it the evaluation that makes the difference? Is it value tuning? Or is it something else?

I suggest to come to the talkchess forum (programmers section) for a discussing such things.
http://talkchess.com/forum3/viewforum.php?f=7&sid=19be091e076a02f561c2e799dea7caec

Before adding a goal like top 100 in CCRL you should first concentrate on existent bugs.

It seems DuchChess reaches high depths already, probably due to lots of reductions, but may be some of them are unsound
or not correctly implemented.
I cannot help here much, as I am not an active developer anymore since very long (also never developed chess engines, but still knowing a few things about their internals, but I am not in the position to give advice here).

With your release and after fixing the time control thing it will be tested anyway and you'll get informations about
current bugs soon. The talkchess community is still very helpful (at least in the programmers section) with a new
growing engine. Just try it.

Edit:
Note that registering at talkchess might last a while, but if you are really interested I could possibly enforce
a way for a faster process. You should be warned though that chess programming could be very time
consuming and addictive!