Mini shogi engines and rating list

[Ferdy]

I think Evert cured this recently. It was the consequence of Sjaak considering any repetition a loss for the side causing it. This was based on the idea that if the side that wins by repetition could not get the opporunity to repeat 4 times if the side that loses by repetition would avoid any repetition. But the current game shows this idea to be false: the 4 occurrences of the final position are reached by 4 different moves from 4 positions that occurred for the first time (Rb5-a5, Rd5-a5, Be2-c4, Bd5-c4).

The latest released version so far is 1.4.1.

I updated the minishogi site regarding specific problem to each engine.
https://sites.google.com/view/minishogi/engines/sjaak

[hgm]

The fix was reported here: http://www.talkchess.com/forum/viewtopi ... 4&start=20 . Pehaps it has not been officially released yet.

[Ferdy]

https://sites.google.com/view/minishogi ... t/standard

[Ferdy]

I am running an RR tour of bullet TC 1m+1s inc/move using WB tour manager and I got this deadlock. It seems to me that Sjaak is involved. In the task manager, all six Sjaak are consuming cpu time. I was forced to terminate winboard and engines.

[hgm]

It seems Sjaak does have an issue that can make it hang. I had to stop using it as an opponent when testing CrazyWa for Crazyhouse, for that reason.

When Sjaak hangs it forfeits on time, but when WinBoard then tries to start it for the next game, Sjaak will not respond to ping, and WinBoard will wait forever. The problem is that there is little that can bedone for error recovery from WinBoard. I could never get killing of engine processes (through the WinAPI function TerminateProcess()) to work reliably on Windows). Sjaak could be thinking on a move, so it is not obvious what a reasonable timeout would be after which to attempt a restart. Installing the engine as /xreuse would force a restart before every game, but if the old engine process keeps hanging using CPU, because the killing did not work, it still wrecks the remaining games. (And when I tried this, WinBoard actually crashed when Sjaak was hanging, and I still have not been able to figure out why.)

WinBoard has a option -afterGame that could be used to issue a command, and there exists a utility 'pv.exe' that can reliably kill processes. Problem is that it kills them by name of the executable. Which is unacceptable when you use concurrency, and multiple instances of Sjaak are running.

It is a bit suspect that all 6 crashes occurred in the same position. Perhaps this is reproducible, which should allow Evert to debug it.

[Ferdy]

It seems Sjaak does have an issue that can make it hang. I had to stop using it as an opponent when testing CrazyWa for Crazyhouse, for that reason.

When Sjaak hangs it forfeits on time, but when WinBoard then tries to start it for the next game, Sjaak will not respond to ping, and WinBoard will wait forever. The problem is that there is little that can bedone for error recovery from WinBoard. I could never get killing of engine processes (through the WinAPI function TerminateProcess()) to work reliably on Windows). Sjaak could be thinking on a move, so it is not obvious what a reasonable timeout would be after which to attempt a restart. Installing the engine as /xreuse would force a restart before every game, but if the old engine process keeps hanging using CPU, because the killing did not work, it still wrecks the remaining games. (And when I tried this, WinBoard actually crashed when Sjaak was hanging, and I still have not been able to figure out why.)

WinBoard has a option -afterGame that could be used to issue a command, and there exists a utility 'pv.exe' that can reliably kill processes. Problem is that it kills them by name of the executable. Which is unacceptable when you use concurrency, and multiple instances of Sjaak are running.

It is a bit suspect that all 6 crashes occurred in the same position. Perhaps this is reproducible, which should allow Evert to debug it.

This is the pgn in the opening suite where it would not play.

Code: Select all

[Event "Edited game"]
[Site "MAKRO-PC"]
[Date "2013.08.06"]
[Round "7"]
[White "-"]
[Black "-"]
[Result "*"]
[Variant "shogi"]
[FEN "rbsgk/4p/5/P4/KGSBR[-] w 0 1"]
[SetUp "1"]

{--------------
r b s g k
. . . . p
. . . . .
P . . . .
K G S B R
white to play
--------------}
1. Gc2 Bc4 {+0.68/18 5:51} 2. Gb2 {-0.24/18 7:20} Sd4 {+0.20/17 1:55} 3.
Bc2 {-0.36/18 4:39} Rb5 {+0.12/17 4:04} 4. Sd2 {+0.16/16 2:31} Ra5
{-0.08/16 4:07} 5. Rc1 {+0.16/16 2:29} Rb5 {-0.20/16 1:53} 6. Ga3
{+0.80/16 3:46} Be2 {-0.44/16 2:30} 7. Bb3 {+0.44/17 3:50} Gc5
{-0.32/17 4:59} 8. Rxc5+ {+0.28/16 1:52} Rxc5 {-0.20/17 2:09} 9. Gb4
{+0.76/16 3:12} Bd1+ {-0.68/16 4:10} 10. Bxd1 {+1.20/16 4:57} R@c1
{-1.28/16 3:33} 11. G@b1 {+1.16/17 5:29} Rxd1+ {-1.16/16 3:34} 12. Gxc5
{+1.16/15 3:15} B@b2 {-1.16/14 2:20} 13. Kxb2 {+1.72/1} +Rxd2
{-1.12/14 4:28} 14. Ka1 {+1.04/14 4:11} Sxc5 {-1.08/13 2:18} 15. B@b2
{+1.12/13 7:55} S@d4 {-1.16/13 4:00} 16. R@b5 {+1.12/13 2:06} G@d5
{-0.96/15 3:58} 17. B@a5 {+1.24/13 1:58} +Rd3 {-1.08/15 3:15} 18. Be1+
{+0.84/17 4:08} Sc4 {-0.88/16 2:23} 19. Rb4+ {+0.96/16 6:05} +Re3
{-0.60/16 4:21} 20. +Bd1 {-0.20/16 4:29}
*

Restarting the tournament does not work (it would return to "waiting for first chess program"), so I just remove Sjaak in the participant list, and continue the RR tour.

[Evert]

It is a bit suspect that all 6 crashes occurred in the same position. Perhaps this is reproducible, which should allow Evert to debug it.

Sortof. If I cross-compile for Windows and run this position through there, SjaakII hangs. If I try it under Linux/OS X, everything seems to be fine.

So at least I can reproduce it, but debugging non-native code is a bit of a pain...

[Evert]

It is a bit suspect that all 6 crashes occurred in the same position. Perhaps this is reproducible, which should allow Evert to debug it.

Sortof. If I cross-compile for Windows and run this position through there, SjaakII hangs. If I try it under Linux/OS X, everything seems to be fine.

So at least I can reproduce it, but debugging non-native code is a bit of a pain...

Actually, I spoke too soon. It doesn't hang, Wine just stops displaying output unless I force it to by pressing a key. Not sure what's up with that, it shouldn't be buffering output. Then again, console applications are not the core focus.

I cannot reproduce the problem with either the position, or the sequence of moves.

I'd like to fix this for the next release, but I need a way to reproduce the problem first. It looks like the problem should reproduce for you guys when entering that particular position. Is that true? Does it only depend on the position, or also on the move sequence leading to it?

[Ferdy]

It is a bit suspect that all 6 crashes occurred in the same position. Perhaps this is reproducible, which should allow Evert to debug it.

Sortof. If I cross-compile for Windows and run this position through there, SjaakII hangs. If I try it under Linux/OS X, everything seems to be fine.

So at least I can reproduce it, but debugging non-native code is a bit of a pain...

Actually, I spoke too soon. It doesn't hang, Wine just stops displaying output unless I force it to by pressing a key. Not sure what's up with that, it shouldn't be buffering output. Then again, console applications are not the core focus.

I cannot reproduce the problem with either the position, or the sequence of moves.

I'd like to fix this for the next release, but I need a way to reproduce the problem first. It looks like the problem should reproduce for you guys when entering that particular position. Is that true? Does it only depend on the position, or also on the move sequence leading to it?

I could no longer reproduce it. But here is the later part of the debug when I tried to enable the debug in the *.trn file when I was still able to reproduce it.

132 0132132132132132 0132
132 0132132132132132 0132
132 0132132132132132 0132
132 0132132132132132 0132
132 0132132132132132 0132
LegalDrop: 2 @ 5,3)
Parsed game start '[Event "Edited game"]' (39)
Parsed PGNTag: [Event "Edited game"]
Parsed PGNTag: [Site "MAKRO-PC"]
Parsed PGNTag: [Date "2013.08.06"]
Parsed PGNTag: [Round "7"]
Parsed PGNTag: [White "-"]
Parsed PGNTag: [Black "-"]
Parsed PGNTag: [Result "*"]
Parsed PGNTag: [Variant "shogi"]
recognized 'shogi' (-1) as variant shogi
Parsed PGNTag: [FEN "rbsgk/4p/5/P4/KGSBR[-] w 0 1"]
shuffleOpenings = 0
Parsed PGNTag: [SetUp "1"]
Parsed PositionDiagram: {--------------

r b s g k

. . . . p

. . . . .

P . . . .

K G S B R

white to play

--------------}
Load Game
Parsed Gc2 into b1c2 virgin=0,0
Parsed Bc4 into b5c4 virgin=8,0
Parsed Comment: {+0.68/18 5:51}
Append: in='{+0.68/18 5:51}' 0
Parsed Gb2 into c2b2 virgin=8,8
Parsed Comment: {-0.24/18 7:20}
Append: in='{-0.24/18 7:20}' 0
Parsed Sd4 into c5d4 virgin=8,8
Parsed Comment: {+0.20/17 1:55}
Append: in='{+0.20/17 1:55}' 0
Parsed Bc2 into d1c2 virgin=8,18
Parsed Comment: {-0.36/18 4:39}
Append: in='{-0.36/18 4:39}' 0
Parsed Rb5 into a5b5 virgin=28,18
Parsed Comment: {+0.12/17 4:04}
Append: in='{+0.12/17 4:04}' 0
Parsed Sd2 into c1d2 virgin=28,1c
Parsed Comment: {+0.16/16 2:31}
Append: in='{+0.16/16 2:31}' 0
Parsed Ra5 into b5a5 virgin=38,1c
Parsed Comment: {-0.08/16 4:07}
Append: in='{-0.08/16 4:07}' 0
Parsed Rc1 into e1c1 virgin=38,1c
Parsed Comment: {+0.16/16 2:29}
Append: in='{+0.16/16 2:29}' 0
Parsed Rb5 into a5b5 virgin=78,1c
Parsed Comment: {-0.20/16 1:53}
Append: in='{-0.20/16 1:53}' 0
Parsed Ga3 into b2a3 virgin=78,1c
Parsed Comment: {+0.80/16 3:46}
Append: in='{+0.80/16 3:46}' 0
Parsed Be2 into c4e2 virgin=78,1c
Parsed Comment: {-0.44/16 2:30}
Append: in='{-0.44/16 2:30}' 0
Parsed Bb3 into c2b3 virgin=78,1c
Parsed Comment: {+0.44/17 3:50}
Append: in='{+0.44/17 3:50}' 0
Parsed Gc5 into d5c5 virgin=78,1c
Parsed Comment: {-0.32/17 4:59}
Append: in='{-0.32/17 4:59}' 0
Parsed Rxc5+ into c1c5+ virgin=78,3c
Parsed Comment: {+0.28/16 1:52}
Append: in='{+0.28/16 1:52}' 0
Parsed Rxc5 into b5c5 virgin=78,3c
Parsed Comment: {-0.20/17 2:09}
Append: in='{-0.20/17 2:09}' 0
Parsed Gb4 into a3b4 virgin=78,3c
Parsed Comment: {+0.76/16 3:12}
Append: in='{+0.76/16 3:12}' 0
Parsed Bd1+ into e2d1+ virgin=78,3c
Parsed Comment: {-0.68/16 4:10}
Append: in='{-0.68/16 4:10}' 0
Parsed Bxd1 into b3d1 virgin=78,3c
Parsed Comment: {+1.20/16 4:57}
Append: in='{+1.20/16 4:57}' 0
LegalDrop: 69 @ 4,0)
LegalDrop: 69 @ 4,0)
Parsed R@c1 into R@c1
Parsed Comment: {-1.28/16 3:33}
Append: in='{-1.28/16 3:33}' 0
LegalDrop: 9 @ 3,0)
LegalDrop: 9 @ 3,0)
Parsed G@b1 into G@b1
Parsed Comment: {+1.16/17 5:29}
Append: in='{+1.16/17 5:29}' 0
Parsed Rxd1+ into c1d1+ virgin=78,3c
Parsed Comment: {-1.16/16 3:34}
Append: in='{-1.16/16 3:34}' 0
Parsed Gxc5 into b4c5 virgin=78,3c
Parsed Comment: {+1.16/15 3:15}
Append: in='{+1.16/15 3:15}' 0
LegalDrop: 68 @ 3,1)
LegalDrop: 68 @ 3,1)
Parsed B@b2 into B@b2
Parsed Comment: {-1.16/14 2:20}
Append: in='{-1.16/14 2:20}' 0
Parsed Kxb2 into a1b2 virgin=78,3c
Parsed Comment: {+1.72/1}
Append: in='{+1.72/1}' 0
Parsed +Rxd2 into d1d2 virgin=7c,3c
Parsed Comment: {-1.12/14 4:28}
Append: in='{-1.12/14 4:28}' 0
Parsed Ka1 into b2a1 virgin=7c,3c
Parsed Comment: {+1.04/14 4:11}
Append: in='{+1.04/14 4:11}' 0
Parsed Sxc5 into d4c5 virgin=7c,3c
Parsed Comment: {-1.08/13 2:18}
Append: in='{-1.08/13 2:18}' 0
LegalDrop: 2 @ 3,1)
LegalDrop: 2 @ 3,1)
Parsed B@b2 into B@b2
Parsed Comment: {+1.12/13 7:55}
Append: in='{+1.12/13 7:55}' 0
LegalDrop: 71 @ 5,3)
LegalDrop: 71 @ 5,3)
Parsed S@d4 into S@d4
Parsed Comment: {-1.16/13 4:00}
Append: in='{-1.16/13 4:00}' 0
LegalDrop: 3 @ 3,4)
LegalDrop: 3 @ 3,4)
Parsed R@b5 into R@b5
Parsed Comment: {+1.12/13 2:06}
Append: in='{+1.12/13 2:06}' 0
LegalDrop: 75 @ 5,4)
LegalDrop: 75 @ 5,4)
Parsed G@d5 into G@d5
Parsed Comment: {-0.96/15 3:58}
Append: in='{-0.96/15 3:58}' 0
LegalDrop: 2 @ 2,4)
LegalDrop: 2 @ 2,4)
Parsed B@a5 into B@a5
Parsed Comment: {+1.24/13 1:58}
Append: in='{+1.24/13 1:58}' 0
Parsed +Rd3 into d2d3 virgin=7c,3c
Parsed Comment: {-1.08/15 3:15}
Append: in='{-1.08/15 3:15}' 0
Parsed Be1+ into a5e1+ virgin=7c,3c
Parsed Comment: {+0.84/17 4:08}
Append: in='{+0.84/17 4:08}' 0
Parsed Sc4 into c5c4 virgin=7c,3c
Parsed Comment: {-0.88/16 2:23}
Append: in='{-0.88/16 2:23}' 0
Parsed Rb4+ into b5b4+ virgin=7c,3c
Parsed Comment: {+0.96/16 6:05}
Append: in='{+0.96/16 6:05}' 0
Parsed +Re3 into d3e3 virgin=7c,3c
Parsed Comment: {-0.60/16 4:21}
Append: in='{-0.60/16 4:21}' 0
Parsed +Bd1 into e1d1 virgin=7c,3c
Parsed game end: {-0.20/16 4:29}

*
GameEnds(29, -0.20/16 4:29, 3)
ForwardInner(39), current 0, forward 39
PauseEvent(): pausing 0
LoadGame(): on exit, gameMode 7
StartChildProcess (dir="../SjaakII") SjaakII-64.exe
nice engine proc to 10
532 >first : xboard
protover 2
532 <first : Sjaak II version 1.4.1 (x86_64)
532 <first : Type 'help' for a list of commands and help topics
532 <first : 8r n b q k b n r
532 <first : 7p p p p p p p p
532 <first : 6
532 <first : 5
532 <first : 4
532 <first : 3
532 <first : 2P P P P P P P P
532 <first : 1R N B Q K B N R *
532 <first : a b c d e f g h
548 <first : #[Chess] 0w>
548 <first : feature setboard=1 time=1 sigint=0 colors=0 highlight=1 ping=1 memory=1 analyze=1 pause=1 nps=1 sjef=1 myname="Sjaak II 1.4.1" myversion="[1.4.1 (x86_64)]" variants="seirawan,shatar,makruk,shatranj,sittuyin,crazyhouse,chessgi,asean,ai-wok,super,spartan,pocketknight,kingofthehill,knightmate,berolina,losalamos,micro,capablanca,gothic,embassy,greatshatranj,courier,grand,opulent,omega,minishogi,shoshogi,shogi,torishogi,xiangqi,chess960,fischerandom,fischerrandom,nocastle,wildcastle,great,caparandom,minisho,5x5+5_shogi,tori,7x7+6_shogi,sho,9x9+0_shogi,king-of-the-hill,fairy,8x8+0_fairy,8x8+6_fairy,8x8+1_fairy,6x6+0_fairy,5x5+0_fairy,10x8+0_fairy,12x8+0_fairy,10x10+0_fairy,12x12+0_fairy,5x5+5_fairy,9x9+0_fairy,9x9+8_fairy,7x7+6_fairy,9x10+0_fairy,normal"
548 >first : accepted setboard
548 >first : accepted time
548 >first : accepted sigint
548 >first : accepted colors
548 >first : accepted highlight
548 >first : accepted ping
548 >first : accepted memory
548 >first : accepted analyze
548 >first : accepted pause
548 >first : accepted nps
548 >first : rejected sjef
548 >first : accepted myname
548 >first : rejected myversion
548 >first : accepted variants
548 <first : feature option="Variant fairy selects -combo chess (8x8+0) /// seirawan (8x8+0) /// shatar (8x8+0) /// makruk (8x8+0) /// shatranj (8x8+0) /// sittuyin (8x8+6) /// crazyhouse (8x8+6) /// chessgi (8x8+6) /// asean (8x8+0) /// ai-wok (8x8+0) /// super (8x8+0) /// spartan (8x8+0) /// pocketknight (8x8+1) /// kingofthehill (8x8+0) /// knightmate (8x8+0) /// berolina (8x8+0) /// losalamos (6x6+0) /// micro (5x5+0) /// capablanca (10x8+0) /// gothic (10x8+0) /// embassy (10x8+0) /// greatshatranj (10x8+0) /// courier (12x8+0) /// grand (10x10+0) /// opulent (10x10+0) /// omega (12x12+0) /// minishogi (5x5+5) /// shoshogi (9x9+0) /// shogi (9x9+8) /// torishogi (7x7+6) /// xiangqi (9x10+0) /// chess960 (= chess) /// fischerandom (= chess) /// fischerrandom (= chess) /// nocastle (= chess) /// wildcastle (= chess) /// great (= greatshatranj) /// caparandom (= capablanca) /// minisho (= minishogi) /// 5x5+5_shogi (= minishogi) /// tori (= torishogi) /// 7x7+6_shogi (= torishogi) /// sho (= shoshogi) /// 9x9+0_shogi (= shoshogi) /// king-of-the-hill (= kingofthehill)"
548 >first : accepted option
548 <first : feature option="Mate search -combo Disabled /// *Enabled for drop games /// Enabled"
548 >first : accepted option
548 <first : feature option="Level -combo Clueless /// Random /// Static /// *Normal"
548 >first : accepted option
548 <first : feature option="MultiPV -spin 1 1 256"
548 >first : accepted option
548 <first : feature option="Draw offer threshold -spin 0 0 1000"
548 >first : accepted option
548 <first : feature option="Moves before draw offer (0 to disable) -spin 0 0 1000"
548 >first : accepted option
548 <first : feature option="Resign threshold -spin 500 100 16000"
548 >first : accepted option
548 <first : feature option="Moves before resigning (0 to disable) -spin 0 0 1000"
548 >first : accepted option
548 <first : feature option="Randomise opening moves -spin 10 0 40"
548 >first : accepted option
548 <first : feature option="Random amplitude (0 to disable) -spin 20 0 100"
548 >first : accepted option
548 <first : feature option="Send 'piece' descriptions -check 1"
548 >first : accepted option
548 <first : feature option="Mark holes in board -check 1"
548 >first : accepted option
548 <first : feature option="List user-defined variants before buildin variants -check 1"
548 >first : accepted option
548 <first : feature option="Report fail low -check 0"
548 >first : accepted option
548 <first : feature option="Report fail high -check 0"
548 >first : accepted option
548 <first : feature option="Claim repetitions -check 1"
548 >first : accepted option
548 <first : feature option="Send O-O/O-O-O for castling -check 1"
548 >first : accepted option
548 <first : feature option="Remember evaluation parameter file -check 0"
548 >first : accepted option
548 <first : feature option="Variant 'normal' is -string chess"
548 >first : accepted option
548 <first : feature option="Set variant alias -string "
548 >first : accepted option
548 <first : feature option="Variant configuration file -file "
548 >first : accepted option
548 <first : feature option="Evaluation parameter file -file "
548 >first : accepted option
548 <first : feature done=1
548 >first : accepted done
563 >first : memory 132
563 >first : new
random
563 >first : variant 5x5+5_shogi
563 >first : level 0 1 1
563 >first : post
563 >first : hard
563 >first : easy
563 >first : ping 1
Feeding position and moves 0 through 39 to first chess program
563 >first : force
write FEN 50-move: 0 39 0
e0. p=-1
e1. p=-6
e2. p=-6
e3. p=-6
e4. p=-6
e5. p=-6
e6. p=-6
e7. p=-6
e8. p=-6
e9. p=-6
e10. p=-6
e11. p=-6
e12. p=-6
e13. p=-6
e14. p=-6
e15. p=-2
e16. p=-2
e17. p=-6
e18. p=-6
e19. p=-2
e20. p=-6
e21. p=-6
e22. p=-2
e23. p=-2
e24. p=-6
e25. p=-2
e26. p=-2
e27. p=-6
e28. p=-2
e29. p=-6
e30. p=-6
e31. p=-6
e32. p=-6
e33. p=-6
e34. p=-6
e35. p=-6
e36. p=-6
e37. p=-6
e38. p=-6
e39. p=-6
563 >first : setboard rbsgk/4p/5/P4/KGSBR[-] w 0 1
feedMoves
563 >first : b1c2
563 >first : b5c4
563 >first : c2b2
563 >first : c5d4
563 >first : d1c2
563 >first : a5b5
563 >first : c1d2
563 >first : b5a5
563 >first : e1c1
563 >first : a5b5
563 >first : b2a3
563 >first : c4e2
563 >first : c2b3
563 >first : d5c5
563 >first : c1c5+
563 >first : b5c5
563 >first : a3b4
563 >first : e2d1+
563 >first : b3d1
563 >first : R@c1
563 >first : G@b1
563 >first : c1d1+
563 >first : b4c5
563 >first : B@b2
563 >first : a1b2
563 >first : d1d2
563 >first : b2a1
563 >first : d4c5
563 >first : B@b2
563 >first : S@d4
563 >first : R@b5
563 >first : G@d5
563 >first : B@a5
563 >first : d2d3
563 >first : a5e1+
563 >first : c5c4
563 >first : b5b4+
563 >first : d3e3
563 >first : e1d1
579 <first : # New game 'Chess'
579 <first : # New game 'minishogi'
579 <first : setup (P.BR.S...G.+.++.+Kp.br.s...g.+.++.+k) 5x5+5_shogi rbsgk/4p/5/P4/KGSBR [-] w 0 1
579 <first : piece P& fWf@4
579 <first : piece S& FfW
579 <first : piece B& B
579 <first : piece R& R
579 <first : piece G& WfF
579 <first : piece K& K
579 <first : piece +P& WfF
579 <first : piece +S& WfF
579 <first : piece +B& BW
579 <first : piece +R& RF
GameEnds(29, -0.20/16 4:29, 2)
results = '+-+---+----++--++++-+--++----++---+++--++-+----++-+--+--+++++-+-+-+--+-++--+-++----+--------+++-+--++-+--+--+--+-++++-+---+---+--------+--+-----+++---+-+++------++-----+-+--+---++-+--++-+--++++--+* +- -+ '
Shokidoki UEC9+ 32bit\012TJshogi 5x5 0.19 32bit\012Sjaak II v1.4.1 64bit\012Lima v4.0 64bit\012CrazyWa 1.0.2 32bit\012NebiyuAlien 1.46 32bit\012
pick next game from '+-+---+----++--++++-+--++----++---+++--++-+----++-+--+--+++++-+-+-+--+-++--+-++----+--------+++-+--++-+--+--+--+-++++-+---+---+--------+--+-----+++---+-+++------++-----+-+--+---++-+--++-+--++++--+ +- -+ ': 197
Reserved, next=197, nr=196
ScheduleDelayedEvent: event already scheduled
8703 >first : quit

I think Martin mentioned he found an issue with the Sjaak was that already addressed?, it could be related to this.

[Evert]

I think Martin mentioned he found an issue with the Sjaak was that already addressed?, it could be related to this.

Good point. Yes, that's fixed, but not yet released. I'll wrap up a new release over the weekend.
Would be nice if that fixes it, but the log looks different: in that case it responded to "new" but not to "variant". The issue was caused by allocating large objects on the stack, if there are more instances of that I may be able to find them by decreasing the stack space in Linux (I assume that's possible).