It may depend on the GUI, but I would assume that it will simply send a "result" command, telling the engine that the game is over, and the engine should stop searching (if it was) and clean up.OliverBr wrote: ↑Sat Jun 27, 2020 6:44 amHi Sergio,Sergio Martinez wrote: ↑Sun Jun 14, 2020 11:03 pm- Resign is set to -5 for 3 consecutive moves where able.
what command will be send to the engine in order to let it resign?
I would like to implement it (winboard/xboard protocol)
Thanks!
Open Swiss Edition III
Moderators: hgm, Dann Corbit, Harvey Williamson
Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
Re: Open Swiss Edition III
Re: Open Swiss Edition III
Maybe I need to say a little more, in case I misunderstood your question. The ability to resign is entirely up to the engine author. If you want your engine to resign in hopeless situations, then the engine just needs to send the "resign" command to the GUI.JVMerlino wrote: ↑Sat Jun 27, 2020 4:06 pmIt may depend on the GUI, but I would assume that it will simply send a "result" command, telling the engine that the game is over, and the engine should stop searching (if it was) and clean up.OliverBr wrote: ↑Sat Jun 27, 2020 6:44 amHi Sergio,Sergio Martinez wrote: ↑Sun Jun 14, 2020 11:03 pm- Resign is set to -5 for 3 consecutive moves where able.
what command will be send to the engine in order to let it resign?
I would like to implement it (winboard/xboard protocol)
Thanks!
If the person running the match wants to force games to end prematurely, on the assumption that the result will be obvious, then that is called adjudication, and is not the same thing as resigning. CCRL does this, as Sergio has stated, although I don't know if all CCRL testers use the same parameters.
jm
-
- Posts: 1975
- Joined: Fri Oct 05, 2012 12:35 pm
- Location: Spain
- Contact:
Re: Open Swiss Edition III
Hi OlivierOliverBr wrote: ↑Sat Jun 27, 2020 6:44 amHi Sergio,Sergio Martinez wrote: ↑Sun Jun 14, 2020 11:03 pm- Resign is set to -5 for 3 consecutive moves where able.
what command will be send to the engine in order to let it resign?
I would like to implement it (winboard/xboard protocol)
Thanks!
John has explained it perfectly. If the resing option is not implemented then in my case the GUI Arena will adjudicate the games. But do not worry, as I say, if the engine does not give up, it will award Arena. This is not a problem, either.

Member of the CCRL Group. Write me if you want I test your engine.
-
- Posts: 1975
- Joined: Fri Oct 05, 2012 12:35 pm
- Location: Spain
- Contact:
Re: Open Swiss Edition III
Code: Select all
ROUND 2
=========
Ceibo 0.5 64-bit 1.0 -- 0.0 Raven 0.90 64-bit
FrankWalter 2.4.0 64-bit 1.0 -- 0.0 Honzovy Sachy 2 64-bit
Tao 5.6 1.0 -- 0.0 Abbess 2018.02.07
Myrddin 0.87 64-bit 4CPU 1.0 -- 0.0 Merlin 2.3.1
FoxSEE 3.3.4 64-bit 0.0 -- 1.0 AnMon 5.75
Isa 2.0.64 64-bit 1.0 -- 0.0 Rebel 6
Betsabe II 1.96 1.0 -- 0.0 Waxman 2017
SOS 5.1 1.0 -- 0.0 Anubisx 0.052
Pseudo 0.7c 0.5 -- 0.5 Yace 0.17
The Crazy Bishop 0052 0.5 -- 0.5 Tinker T1062 64-bit
Amyan 1.72 0.5 -- 0.5 OliThink 5.4.7 64-bit
CT800 1.40 64-bit 0.0 -- 1.0 Ruffian 2.1.0
Delocto 0.61n 64bit 0.5 -- 0.5 Giraffe 20161023 64-bit
NoC 4.20 64-bit 0.0 -- 1.0 Movei 00.8.438 (10 10 10)
Nejmet 3.07 0.0 -- 1.0 Orion 0.6 64-bit
GreKo 2020.03 64-bit 1.0 -- 0.0 Quark 2.35
Cerebro 4.03 64-bit 0.0 -- 1.0 Karballo 1.8 64-bit
Halogen 4 64-bit 4CPU 0.0 -- 1.0 Invictus.r305.64-bit 4CPU
Chezzz 1.0.3 0.0 -- 1.0 Wasp 4.00 64-bit 4CPU
Pharaon 3.5.1 0.0 -- 1.0 Marvin 3.6.0 64-bit 4CPU
SpaceDog 0.97.7 64-bit 0.0 -- 1.0 Winter 0.8 64-bit 4CPU
Phalanx XXV 0.0 -- 1.0 iCE 4.0 64-bit 4CPU
Floyd 0.9 64-bit 0.0 -- 1.0 Igel 2.5.0 64-bit 4CPU
Delfi 5.4 0.0 -- 1.0 Houdini 1.5a 64-bit 4CPU
ChessFighter 3.4 RTX2080 1.0 -- 0.0 Danasah 8,3
Arasan 22.0 64-bit 4CPU 1.0 -- 0.0 Fruit 2.3.1
RuyDos 1.1.11 64-bit 0.5 -- 0.5 RubiChess 1.7.3 64-bit 4CPU
tomitankChess 4.0 64-bit 0.0 -- 1.0 Booot 6.4 64-bit 4CPU
Daydreamer 2.0.0-pre2 64-bit 0.0 -- 1.0 Stoofvlees II a14 RTX2080
RofChade 2.3 64-bit 4CPU 1.0 -- 0.0 Glaurung 2.2 64-bit 4CPU
Francesca MAD 0.29 64-bit 0.0 -- 1.0 Fritz 17 64-bit 4CPU
Komodo 14 64-bit 4CPU 1.0 -- 0.0 Godel 7.0 64-bit
Combusken 1.2.0 64-bit 4CPU 0.0 -- 1.0 Honey XI-r2 64-bit 4CPU
SmarThink 1.98 64-bit 0.0 -- 1.0 Stockfish 11 64-bit 4CPU
Lc0 0.25.1 t40-1541 RTX2080 1.0 -- 0.0 Amoeba 3.1 64-bit 4CPU
Code: Select all
Rank Engine ELO 1 2 Score Tie White ELO
-------------------------------------------------------------------------------
1: Lc0 0.25.1 t40-1541 RTX2080 3600 36B+ 28W+ 2.0 2.0 1 0
2: Stockfish 11 64-bit 4CPU 3490 37W+ 29B+ 2.0 2.0 1 +1
3: Komodo 14 64-bit 4CPU 3421 38W+ 30W+ 2.0 2.0 2 +1
4: Booot 6.4 64-bit 4CPU 3298 39W+ 31B+ 2.0 2.0 1 +1
5: ChessFighter 3.4 RTX2080 3200 40B+ 32W+ 2.0 2.0 1 +1
6: Houdini 1.5a 64-bit 4CPU 3183 42W+ 33B+ 2.0 2.0 1 +1
7: Igel 2.5.0 64-bit 4CPU 3134 41W+ 34B+ 2.0 2.0 1 +1
8: iCE 4.0 64-bit 4CPU 3123 43W+ 35B+ 2.0 2.0 1 +1
9: Honey XI-r2 64-bit 4CPU 3450 51W+ 24B+ 2.0 1.5 1 +1
10: Fritz 17 64-bit 4CPU 3385 52W+ 26B+ 2.0 1.5 1 +1
11: RofChade 2.3 64-bit 4CPU 3318 53B+ 25W+ 2.0 1.5 1 +1
12: Stoofvlees II a14 RTX2080 3300 54W+ 27B+ 2.0 1.5 1 +1
13: Arasan 22.0 64-bit 4CPU 3220 59B+ 21W+ 2.0 1.0 1 +1
14: Winter 0.8 64-bit 4CPU 3108 60W+ 23B+ 2.0 1.0 1 0
15: Marvin 3.6.0 64-bit 4CPU 3075 61W+ 22B+ 2.0 1.0 1 +1
16: RubiChess 1.7.3 64-bit 4CPU 3229 44B+ 17B= 1.5 2.5 0 -4
17: RuyDos 1.1.11 64-bit 2787 62B+ 16W= 1.5 1.5 1 +5
18: Wasp 4.00 64-bit 4CPU 3131 48W= 49B+ 1.5 1.0 1 -5
19: Invictus.r305.64-bit 4CPU 2878 49W= 50B+ 1.5 1.0 1 -4
20: Karballo 1.8 64-bit 2730 50W= 48B+ 1.5 1.0 1 -2
21: Fruit 2.3.1 2763 46W+ 13B- 1.0 3.0 1 0
22: Pharaon 3.5.1 2688 47W+ 15W- 1.0 3.0 2 -1
23: SpaceDog 0.97.7 64-bit 2243 45W+ 14W- 1.0 3.0 2 +10
24: Combusken 1.2.0 64-bit 4CPU 3071 55B+ 9W- 1.0 2.5 1 -1
25: Glaurung 2.2 64-bit 4CPU 2917 56W+ 11B- 1.0 2.5 1 0
26: Francesca MAD 0.29 64-bit 2904 57B+ 10W- 1.0 2.5 1 0
27: Daydreamer 2.0.0-pre2 64-bit 2896 58B+ 12W- 1.0 2.5 1 0
28: Amoeba 3.1 64-bit 4CPU 3073 63B+ 1B- 1.0 2.0 0 0
29: SmarThink 1.98 64-bit 3072 64W+ 2W- 1.0 2.0 2 -1
30: Godel 7.0 64-bit 2975 65W+ 3B- 1.0 2.0 1 0
31: tomitankChess 4.0 64-bit 2820 66B+ 4W- 1.0 2.0 1 0
32: Danasah 8,3 2690 67W+ 5B- 1.0 2.0 1 0
33: Delfi 5.4 2681 68B+ 6W- 1.0 2.0 1 0
34: Floyd 0.9 64-bit 2659 69B+ 7W- 1.0 2.0 1 0
35: Phalanx XXV 2646 70B+ 8W- 1.0 2.0 1 0
36: Orion 0.6 64-bit 2633 1W- 61B+ 1.0 2.0 1 +2
37: Movei 00.8.438 (10 10 10) 2627 2B- 64B+ 1.0 2.0 0 +2
38: Ruffian 2.1.0 2611 3B- 65B+ 1.0 2.0 0 +2
39: SOS 5.1 2576 4B- 66W+ 1.0 2.0 1 +2
40: AnMon 5.75 2540 5W- 67B+ 1.0 2.0 1 +1
41: Tao 5.6 2483 7B- 69W+ 1.0 2.0 1 +1
42: Myrddin 0.87 64-bit 4CPU 2483 6B- 68W+ 1.0 2.0 1 +1
43: FrankWalter 2.4.0 64-bit 2463 8B- 70W+ 1.0 2.0 1 +1
44: Betsabe II 1.96 2563 16W- 62W+ 1.0 1.5 2 +2
45: GreKo 2020.03 64-bit 2716 23B- 63W+ 1.0 1.0 1 -8
46: Isa 2.0.64 64-bit 2314 21B- 59W+ 1.0 1.0 1 +7
47: Ceibo 0.5 64-bit 2130 22B- 60W+ 1.0 1.0 1 +8
48: Cerebro 4.03 64-bit 2472 18B= 20W- 0.5 3.0 1 +3
49: Chezzz 1.0.3 2323 19B= 18W- 0.5 3.0 1 +5
50: Halogen 4 64-bit 4CPU 2300 20B= 19W- 0.5 3.0 1 +4
51: Delocto 0.61n 64bit 2626 9B- 55W= 0.5 2.5 1 -3
52: Amyan 1.72 2599 10B- 56W= 0.5 2.5 1 -3
53: Tinker T1062 64-bit 2588 11W- 57B= 0.5 2.5 1 -3
54: Pseudo 0.7c 2577 12B- 58W= 0.5 2.5 1 -3
55: Giraffe 20161023 64-bit 2406 24W- 51B= 0.5 1.5 1 +3
56: OliThink 5.4.7 64-bit 2374 25B- 52B= 0.5 1.5 0 +3
57: The Crazy Bishop 0052 2365 26W- 53W= 0.5 1.5 2 +3
58: Yace 0.17 2350 27W- 54B= 0.5 1.5 1 +3
59: Rebel 6 2550 13W- 46B- 0.0 3.0 1 -8
60: Raven 0.90 64-bit 2429 14B- 47B- 0.0 3.0 0 -9
61: Nejmet 3.07 2427 15B- 36W- 0.0 3.0 1 -2
62: Waxman 2017 2315 17W- 44B- 0.0 2.5 1 -2
63: Quark 2.35 2424 28W- 45B- 0.0 2.0 1 -2
64: NoC 4.20 64-bit 2416 29B- 37W- 0.0 2.0 1 -2
65: CT800 1.40 64-bit 2391 30B- 38W- 0.0 2.0 1 -2
66: Anubisx 0.052 2320 31W- 39B- 0.0 2.0 1 -2
67: FoxSEE 3.3.4 64-bit 2152 32B- 40W- 0.0 2.0 1 -1
68: Merlin 2.3.1 2095 33W- 42B- 0.0 2.0 1 -1
69: Abbess 2018.02.07 2065 34W- 41B- 0.0 2.0 1 -1
70: Honzovy Sachy 2 64-bit 2042 35W- 43B- 0.0 2.0 1 -1
Code: Select all
------------------ ROUND 3. PAIRINGS -------------------
1: NoC 4.20 64-bit ---- Honzovy Sachy 2 64-bit
2: Quark 2.35 ---- Abbess 2018.02.07
3: Merlin 2.3.1 ---- Waxman 2017
4: FoxSEE 3.3.4 64-bit ---- Nejmet 3.07
5: Raven 0.90 64-bit ---- Anubisx 0.052
6: Rebel 6 ---- CT800 1.40 64-bit
7: Yace 0.17 ---- Amyan 1.72
8: Delocto 0.61n 64bit ---- The Crazy Bishop 0052
9: OliThink 5.4.7 64-bit ---- Halogen 4 64-bit 4CPU
10: Chezzz 1.0.3 ---- Giraffe 20161023 64-bit
11: Cerebro 4.03 64-bit ---- Pseudo 0.7c
12: Tinker T1062 64-bit ---- Ceibo 0.5 64-bit
13: Isa 2.0.64 64-bit ---- Delfi 5.4
14: Danasah 8,3 ---- GreKo 2020.03 64-bit
15: tomitankChess 4.0 64-bit ---- Betsabe II 1.96
16: FrankWalter 2.4.0 64-bit ---- Godel 7.0 64-bit
17: Myrddin 0.87 64-bit 4CPU ---- SmarThink 1.98 64-bit
18: Amoeba 3.1 64-bit 4CPU ---- Tao 5.6
19: AnMon 5.75 ---- Daydreamer 2.0.0-pre2 64-bit
20: Francesca MAD 0.29 64-bit ---- SOS 5.1
21: Ruffian 2.1.0 ---- Glaurung 2.2 64-bit 4CPU
22: Movei 00.8.438 (10 10 10) ---- Combusken 1.2.0 64-bit 4CPU
23: Orion 0.6 64-bit ---- SpaceDog 0.97.7 64-bit
24: Phalanx XXV ---- Pharaon 3.5.1
25: Floyd 0.9 64-bit ---- Fruit 2.3.1
26: RuyDos 1.1.11 64-bit ---- Karballo 1.8 64-bit
27: RubiChess 1.7.3 64-bit 4CPU ---- Invictus.r305.64-bit 4CPU
28: Wasp 4.00 64-bit 4CPU ---- Marvin 3.6.0 64-bit 4CPU
29: Igel 2.5.0 64-bit 4CPU ---- Winter 0.8 64-bit 4CPU
30: Houdini 1.5a 64-bit 4CPU ---- Arasan 22.0 64-bit 4CPU
31: ChessFighter 3.4 RTX2080 ---- Stoofvlees II a14 RTX2080
32: RofChade 2.3 64-bit 4CPU ---- Booot 6.4 64-bit 4CPU
33: Fritz 17 64-bit 4CPU ---- Komodo 14 64-bit 4CPU
34: Honey XI-r2 64-bit 4CPU ---- Stockfish 11 64-bit 4CPU
35: iCE 4.0 64-bit 4CPU ---- Lc0 0.25.1 t40-1541 RTX2080
Broadcast now:
http://ccrl.chessdom.com/sergio/pgn4web/live1.html
Member of the CCRL Group. Write me if you want I test your engine.
-
- Posts: 1975
- Joined: Fri Oct 05, 2012 12:35 pm
- Location: Spain
- Contact:
Re: Open Swiss Edition III
Some games of Round 2:
Member of the CCRL Group. Write me if you want I test your engine.
Re: Open Swiss Edition III
Hi Sergio.
I watched the broadcast today and got into some very long games. Maybe you should think about some draw adjudication e.g. using 5-piece tablebase. There were at least two games with > 100 moves in an endless KRvKR or similar endgame.
Regards, Andreas
I watched the broadcast today and got into some very long games. Maybe you should think about some draw adjudication e.g. using 5-piece tablebase. There were at least two games with > 100 moves in an endless KRvKR or similar endgame.
Regards, Andreas
-
- Posts: 3951
- Joined: Thu May 15, 2008 7:57 pm
- Location: Berlin, Germany
- Full name: Sven Schüle
- Contact:
Re: Open Swiss Edition III
Actually there is no "resign" command in the WB protocol. Engines tell the GUI that they do not continue to play by sendingJVMerlino wrote: ↑Sat Jun 27, 2020 6:42 pmMaybe I need to say a little more, in case I misunderstood your question. The ability to resign is entirely up to the engine author. If you want your engine to resign in hopeless situations, then the engine just needs to send the "resign" command to the GUI.JVMerlino wrote: ↑Sat Jun 27, 2020 4:06 pmIt may depend on the GUI, but I would assume that it will simply send a "result" command, telling the engine that the game is over, and the engine should stop searching (if it was) and clean up.OliverBr wrote: ↑Sat Jun 27, 2020 6:44 amHi Sergio,Sergio Martinez wrote: ↑Sun Jun 14, 2020 11:03 pm- Resign is set to -5 for 3 consecutive moves where able.
what command will be send to the engine in order to let it resign?
I would like to implement it (winboard/xboard protocol)
Thanks!
If the person running the match wants to force games to end prematurely, on the assumption that the result will be obvious, then that is called adjudication, and is not the same thing as resigning. CCRL does this, as Sergio has stated, although I don't know if all CCRL testers use the same parameters.
jm
RESULTSTRING {COMMENT}
where RESULTSTRING is 1-0, 0-1, or 1/2-1/2, and {COMMENT} is optional. E.g. if the engine plays White and resigns it may send:
0-1 {White resigns}
The GUI terminates a game by sending:
result RESULTSTRING {COMMENT}
UCI has no way for engines to resign on their own. And the GUI will only let UCI engines know that a new game has been started by sending "ucinewgame" but it will not inform the engine about result or termination reason of a game.
Sven Schüle (engine author: Jumbo, KnockOut, Surprise)
-
- Posts: 669
- Joined: Tue Dec 18, 2007 8:38 pm
- Location: Munich, Germany
- Full name: Dr. Oliver Brausch
- Contact:
Re: Open Swiss Edition III
Actually OliThink (including 5.4.10) gets in trouble, when the result is sent after the opponent's last move, when OliThink already startet calculating. The resign feature of the polyglot adapter of works this way, which is problematical.
Resigning should be best done when being on move.
Of course, OliThink still gets trouble when somebody is adjudicating the game and this happens when it's calculation the next move, so this needs to be fixed.
Re: Open Swiss Edition III
From the CECP page:Sven wrote: ↑Mon Jun 29, 2020 7:58 amActually there is no "resign" command in the WB protocol. Engines tell the GUI that they do not continue to play by sending
RESULTSTRING {COMMENT}
where RESULTSTRING is 1-0, 0-1, or 1/2-1/2, and {COMMENT} is optional. E.g. if the engine plays White and resigns it may send:
0-1 {White resigns}
The GUI terminates a game by sending:
result RESULTSTRING {COMMENT}
====================
resign
If your engine wants to resign, it can send the command "resign". Alternatively, it can use the "RESULT {comment}" command if the string "resign" is included in the comment; for example "0-1 {White resigns}". xboard relays the resignation to the user, the ICS, the other engine in Two Machines mode, and the PGN save file as required. Note that many interfaces work more smoothly if you resign before you move.
-
- Posts: 1975
- Joined: Fri Oct 05, 2012 12:35 pm
- Location: Spain
- Contact:
Re: Open Swiss Edition III
Hi AndreasRubiChess wrote: ↑Sun Jun 28, 2020 7:17 pmHi Sergio.
I watched the broadcast today and got into some very long games. Maybe you should think about some draw adjudication e.g. using 5-piece tablebase. There were at least two games with > 100 moves in an endless KRvKR or similar endgame.
Regards, Andreas
Yes, you are right, I must change that. Now I can't change the parameters because it would interrupt the tournament, but I will for the next.
Thank you very much for your interest and I hope that the games are not too long.
And I want to apologize, I could not upload the results today due to some problems I have had.
Best regards.
Member of the CCRL Group. Write me if you want I test your engine.