Winboard Forum

Hi Vlad,

I am sorry but both new versions 0.06 and also 0.07 still show the same bug as 0.05 when playing as Black
in WB mode under Winboard. The debug still says it moves two! times as Black one time as Black and after that
it also makes Whites second move! WB ignores that move out of turn as usual and one move later Plisk
is of course out of sync and makes an invalid second move again for the White side instead of Blacks side.

Best regards,
Guenther

A debug example:

Code: Select all

recognized 'normal' (-1) as variant normal
WinBoard 4.3.15k + AtlanChess_33.exe
Reset(1, 0) from gameMode 0
recognized 'normal' (-1) as variant normal
GameEnds(0, (null), 2)
shuffleOpenings = 0
TC string = '+40/15:0'
mps=40 tc=900000 inc=0
TC string = '+40/15:0'
mps=40 tc=900000 inc=0
StartChildProcess (dir="c:\Engines\WB\AtlanChess_33") "AtlanChess_33.exe"
641 >first : xboard
protover 2
1454 <first :
1454 <first : feature sigint=0 sigterm=0 colors=0 setboard=1 usermove=1 myname="Atlanchess 3.3" variants="normal" name=0 ping=1  done=1
1454 >first : accepted sigint
1454 >first : accepted sigterm
1454 >first : accepted colors
1454 >first : accepted setboard
1454 >first : accepted usermove
1454 >first : accepted myname
1454 >first : accepted variants
1454 >first : accepted name
1454 >first : accepted ping
1454 >first : accepted done
1469 >first : new
random
1469 >first : level 40 15 0
1469 >first : post
1469 >first : hard
1469 >first : easy
1469 >first : ping 1
1469 >first : force
StartChildProcess (dir="c:\Engines\WB\Plisk_007") "Plisk_007.exe"
1985 >second: xboard
protover 2
1985 <first : pong 1
1985 <second: Welcome to Plisk (0.0.7)
1985 <second: Setup..............Config:
1985 <second:     m_iConfig_DepthSearch=40
1985 <second:     m_iConfig_UseOpeningBook=1
1985 <second:     m_iConfig_InfiniteThinkingTime=0
1985 <second:     m_iConfig_PawnHashSize=28
1985 <second:     m_iConfig_HashSize=100
1985 <second:     m_iConfig_AspirationWindow=200
1985 <second:     m_iConfig_FixedTimePerMove=1000
1985 <second:     m_iConfig_UseMP=1
1985 <second:     m_iConfig_NumberThreads=1
1985 <second:     m_iConfig_LoggingMode=1
1985 <second: ....Done.
1985 <second: feature myname="Plisk 0.0.7" name=1 setboard=1 ics=1 done=1
1985 >second: accepted myname
1985 >second: accepted name
1985 >second: accepted setboard
1985 >second: accepted ics
1985 >second: accepted done
1985 <second: -- xboard
1985 <second:
1985 <second: -- protover 2
1985 <second:
1985 <second: -- accepted myname
1985 <second:
1985 <second: -- accepted name
1985 <second:
1985 <second: -- accepted setboard
1985 <second:
1985 <second: -- accepted ics
1985 <second:
1985 <second: -- accepted done
1985 <second:
2000 >second: new
random
2000 >second: ics -
2000 >second: level 40 15 0
2000 >second: post
2000 >second: hard
2000 >second: easy
2000 >second: force
2000 >first : computer
2000 >second: computer
2000 >second: name Atlanchess 3.3
TC string = '+40/15:0'
mps=40 tc=900000 inc=0
TC string = '+40/15:0'
mps=40 tc=900000 inc=0
time odds: 1 1
2000 >first : time 90000
2000 >first : otim 90000
2000 >first : go
nps: w=-1, b=-1
2000 <second: -- new
2000 <second:
2000 <second: -- random
2000 <second:
2000 <second: -- ics -
2000 <second:
2000 <second: -- level 40 15 0
2000 <second:
2000 <second: -- post
2000 <second:
2000 <second: -- hard
2000 <second:
2000 <second: -- easy
2000 <second:
2000 <second: -- force
2000 <second:
2000 <second: -- computer
2000 <second:
2000 <second: -- name Atlanchess 3.3
2000 <second:
2000 <first : move e4
machine move 0, castling = 7 0 4 7 0 4
move to parse: e4
Disambiguate in:  0(4,-1)-(4,3) = 0 (-)
Disambiguate out: 0(4,1)-(4,3) = 0 (-)
7 0 4 7 0 4 Legality test? e2e4
(7,0) (0,0) (4,0) (7,7) (0,7) (4,7) castling rights
TC string = '+40/15:0'
mps=40 tc=900000 inc=0
CoordsToAlgebraic, piece=0 (4,1)-(4,3) -
7 0 4 7 0 4 Legality test? e2e4
movetype=33, promochar=0=-
move: e2e4
, parse: e4 (
)
repeat test fmm=1 bmm=0 ep=-4, reps=6
1 ep=-3
0 ep=-4
time odds: 1 1
2000 >second: time 90000
2000 >second: otim 90000
2000 >second: e2e4
2000 >second: black
2000 >second: go
2063 <second: -- time 90000
2063 <second:
2079 <second: -- otim 90000
2079 <second:
2079 <second: -- e2e4
2079 <second:
2079 <second:  e2e4
2079 <second:  |a b c d e f g h
2079 <second: -------------------
2079 <second: 8|r n b q k b n r
2079 <second:  |
2079 <second: 7|p p p p p p p p
2079 <second:  |
2079 <second: 6|. . . . . . . .
2079 <second:  |
2079 <second: 5|. . . . . . . .
2079 <second:  |
2079 <second: 4|. . . . P . . .
2079 <second:  |
2079 <second: 3|. . . . . . . .
2079 <second:  |
2079 <second: 2|P P P P . P P P
2079 <second:  |
2079 <second: 1|R N B Q K B N R
2079 <second:  |
2079 <second: -------------------
2094 <second:
2094 <second: rnbqkbnr/pppppppp/8/8/4P3/8/PPPP1PPP/RNBQKBNR b KQkq -
2094 <second: (1) Allowing 30000 ms for move (0, 900000, 0)
2094 <second:
2094 <second:  |a b c d e f g h
2094 <second: -------------------
2094 <second: 8|r n b q k b n r
2094 <second:  |
2094 <second: 7|p p . p p p p p
2094 <second:  |
2094 <second: 6|. . . . . . . .
2094 <second:  |
2094 <second: 5|. . p . . . . .
2094 <second:  |
2094 <second: 4|. . . . P . . .
2094 <second:  |
2094 <second: 3|. . . . . . . .
2094 <second:  |
2094 <second: 2|P P P P . P P P
2094 <second:  |
2094 <second: 1|R N B Q K B N R
2094 <second:  |
2094 <second: -------------------
2094 <second:
2094 <second: rnbqkbnr/pp1ppppp/8/2p5/4P3/8/PPPP1PPP/RNBQKBNR w KQkq -
2094 <second: move c7c5
machine move 1, castling = 7 0 4 7 0 4
move to parse: c7c5
7 0 4 7 0 4 Legality test? c7c5
7 0 4 7 0 4 Legality test? c7c5
(7,0) (0,0) (4,0) (7,7) (0,7) (4,7) castling rights
TC string = '+40/15:0'
mps=40 tc=900000 inc=0
CoordsToAlgebraic, piece=22 (2,6)-(2,4) -
7 0 4 7 0 4 Legality test? c7c5
movetype=33, promochar=0=-
move: c7c5
, parse: c5 (
)
repeat test fmm=2 bmm=0 ep=-4, reps=6
2 ep=-3
1 ep=-3
0 ep=-4
time odds: 1 1
2094 >first : time 90000
2094 >first : otim 89990
2094 >first : usermove 2094 >first : c7c5
2094 <second: -- black
2094 <second:
2094 <second: -- go
2094 <second:
2094 <second: (2) Allowing 30000 ms for move (0, 900000, 0)
2094 <second:
2094 <second:  |a b c d e f g h
2094 <second: -------------------
2094 <second: 8|r n b q k b n r
2094 <second:  |
2094 <second: 7|p p . p p p p p
2094 <second:  |
2094 <second: 6|. . . . . . . .
2094 <second:  |
2094 <second: 5|. . p . . . . .
2094 <second:  |
2094 <second: 4|. . . . P . . .
2094 <second:  |
2094 <second: 3|. . . . . N . .
2094 <second:  |
2094 <second: 2|P P P P . P P P
2094 <second:  |
2094 <second: 1|R N B Q K B . R
2094 <second:  |
2094 <second: -------------------
2094 <second:
2094 <second: rnbqkbnr/pp1ppppp/8/2p5/4P3/5N2/PPPP1PPP/RNBQKB1R b KQkq -
2094 <second: move g1f3
Ignoring move out of turn by second, gameMode 5, forwardMost 2
2110 <first : move c3
machine move 2, castling = 7 0 4 7 0 4
move to parse: c3
Disambiguate in:  0(2,-1)-(2,2) = 0 (-)
Disambiguate out: 0(2,1)-(2,2) = 0 (-)
7 0 4 7 0 4 Legality test? c2c3
(7,0) (0,0) (4,0) (7,7) (0,7) (4,7) castling rights
TC string = '+40/15:0'
mps=40 tc=900000 inc=0
CoordsToAlgebraic, piece=0 (2,1)-(2,2) -
7 0 4 7 0 4 Legality test? c2c3
movetype=33, promochar=0=-
move: c2c3
, parse: c3 (
)
repeat test fmm=3 bmm=0 ep=-4, reps=6
3 ep=-3
2 ep=-3
1 ep=-3
0 ep=-4
time odds: 1 1
Interrupting second
2125 >second: time 89990
2125 >second: otim 89998
2125 >second: c2c3
2125 <second: -- time 89990
2125 <second:
2125 <second: -- otim 89998
2125 <second:
2125 <second: -- c2c3
2125 <second:
2125 <second:  c2c3
2125 <second:  |a b c d e f g h
2125 <second: -------------------
2125 <second: 8|r n b q k b n r
2125 <second:  |
2125 <second: 7|p p . p p p p p
2125 <second:  |
2125 <second: 6|. . . . . . . .
2125 <second:  |
2125 <second: 5|. . p . . . . .
2125 <second:  |
2125 <second: 4|. . . . P . . .
2125 <second:  |
2125 <second: 3|. . K . . N . .
2125 <second:  |
2125 <second: 2|P P P P . P P P
2125 <second:  |
2125 <second: 1|R N B Q K B . R
2125 <second:  |
2125 <second: -------------------
2125 <second:
2125 <second: rnbqkbnr/pp1ppppp/8/2p5/4P3/2K2N2/PPPP1PPP/RNBQKB1R w KQkq -
2125 <second: (4) Allowing 29996 ms for move (0, 899900, 0)
2125 <second:  2    142      0      255    d4   Nf6
2125 <second:  3    342      0     2307   Ba6   Qa5    b4  Qxa6
2141 <second:  3    490      1     3746   Ke2   Qa5   Kd3   Nf6
2219 <second:  4    573      9    20671    h3   Qa5   Kd3   Qa6   Kc3   Qf6
2329 <second:  5    565     20    49562    d4   Nf6   Ng5   Qa5    b4  Qxb4
2860 <second:  6    482     73   219989   Ne5   Qa5   Kd3    d5   Re3  dxe4  Rxe4   Bf5
5063 <second:  7    478    293   975750    h3   Nf6   Qe2   Nc6   Qc4   Qa5    b4  cxb4
22047 <second:  8    475   1992  6698175    d4   Qa5   Kd3    d5   Rh4   Nf6    e5   Bf5   Ke2   Qb5
32141 <second:  9    675   3001 10420307    d4   Qa5   Kd3    d5   Rh4  dxe4  Rxe4   Bf5   Ke3  Bxe4  Kxe4
32141 <second:
32141 <second:  |a b c d e f g h
32141 <second: -------------------
32141 <second: 8|r n b q k b n r
32141 <second:  |
32141 <second: 7|p p . p p p p p
32141 <second:  |
32141 <second: 6|. . . . . . . .
32141 <second:  |
32141 <second: 5|. . p . . . . .
32141 <second:  |
32141 <second: 4|. . . P P . . .
32141 <second:  |
32141 <second: 3|. . K . . N . .
32141 <second:  |
32141 <second: 2|P P P . . P P P
32141 <second:  |
32141 <second: 1|R N B Q K B . R
32141 <second:  |
32141 <second: -------------------
32141 <second:
32141 <second: rnbqkbnr/pp1ppppp/8/2p5/3PP3/2K2N2/PPP2PPP/RNBQKB1R b KQkq -
32141 <second: move d2d4
machine move 3, castling = 7 0 4 7 0 4
move to parse: d2d4
7 0 4 7 0 4 Legality test? d2d4
GameEnds(37, Xboard: Forfeit due to invalid move: d2d4 (d2d4) res=0, 4)
32141 >first : result 1-0 {Xboard: Forfeit due to invalid move: d2d4 (d2d4) res=0}
Interrupting second
32141 >second: result 1-0 {Xboard: Forfeit due to invalid move: d2d4 (d2d4) res=0}
32250 >first : quit
32860 >second: quit
Exiting: Match Atlanchess 3.3 vs. Plisk 0.0.7: final score 1-0-0
ExitEvent() during GameEnds(), wait
GameEnds() seems stuck, proceed exiting


Hi Guenther,

Yes you are right. I apologize for that. I actually fixed the problem in 0.0.8. It was basically due to Plisk not following the "force" mode and immediately answering to a move with a move when it should wait for the "go" command. Also it happened when xboard would launch Plisk as a second engine (using scp). This is fixed
in 0.0.8, which I hope to upload soon enough. I found the real problem the second day after I've put up 0.0.7.

I apologize again for the troubles you encounter and I appreciate reporting the problem.

Regards,
Vlad.

Hi Guenther (and everyone),

Plisk 0.0.8 has been uploaded to the website (http://sites.google.com/site/pliskchess/). I think this time I fixed the xboard issue for good. I would appreciate if you give Plisk another run.

I am planing to upload a Mac OSX version today and a Linux build tomorrow.

Regards,
Vlad.

Plisk 0.0.8 for MacOSX (Intel CPUs) has been uploaded on the website. I will do the Linux version tomorrow.

Regards,
Vlad.

vladstamate wrote:Hi Guenther (and everyone),

Plisk 0.0.8 has been uploaded to the website (http://sites.google.com/site/pliskchess/). I think this time I fixed the xboard issue for good. I would appreciate if you give Plisk another run.

I am planing to upload a Mac OSX version today and a Linux build tomorrow.

Regards,
Vlad.

Hi Vlad,

I may happily report that the xboard issue with playing as Black is gone!
More later after the current tournament.

Regards,
Guenther

Guenther Simon wrote:

vladstamate wrote:Hi Guenther (and everyone),

Plisk 0.0.8 has been uploaded to the website (http://sites.google.com/site/pliskchess/). I think this time I fixed the xboard issue for good. I would appreciate if you give Plisk another run.

I am planing to upload a Mac OSX version today and a Linux build tomorrow.

Regards,
Vlad.

Hi Vlad,

I may happily report that the xboard issue with playing as Black is gone!
More later after the current tournament.

Regards,
Guenther

Hi Vlad,

I noticed that the original Plisk 0.08(I see that you released another one 3 days later)
doesn't unlock the cpu on my Quadcore machine when playing with ponder off!
(It does not calculate after its move but still using one cpu at full power, probably still polling?)
I have not tested yet if this also happens when all cpus are busy. So far it had no influence on my testing and it will
get a first rating soon, but if I run more tournaments simultaneously again there will be a problem and Plisk
cannot be used then.(BTW I also did not check yet if it happens on my single core computer too).
Other things: 2 time losses so far in very long games in drawish endgames.(debugs available)

Regards,
Guenther

Hi Guenther,

Thanks for reporting that. I think you are right, it is still polling. I will try to fix that soon.

Regards,
Vlad.

vladstamate wrote:Hi Guenther,

Thanks for reporting that. I think you are right, it is still polling. I will try to fix that soon.

Regards,
Vlad.

Thanks for the quick response! I just want to add that the 2 time losses were in 50 games
so far, which means things have improved a lot meanwhile

Regards,
Guenther

Hi Guenther (and everyone),

Plisk 0.0.9 is out now at the usual place (http://sites.google.com/site/pliskchess/). It has quite a number of improvements over 0.0.8 and plays stronger. I also think I have fixed the 100% CPU utilization that you reported. I would appreciate some feedback and confirmation on that though.

Regards,
Vlad.

vladstamate wrote:Hi Guenther (and everyone),

Plisk 0.0.9 is out now at the usual place (http://sites.google.com/site/pliskchess/). It has quite a number of improvements over 0.0.8 and plays stronger. I also think I have fixed the 100% CPU utilization that you reported. I would appreciate some feedback and confirmation on that though.

Regards,
Vlad.

Thanks Vlad
In a couple of days I'll start testing Plisk 0.0.9 on my Q9650 overclocked to 4.0 GHz....
I hope the MP feature of the program works fine now....
Please buzz e if you need the games to have a look at....
Dr.D

Dr.Wael Deeb wrote:I hope the MP feature of the program works fine now....

Hi,

What do you mean by that? Unfortunately the MP support is off for now, since it still has some issues. At the moment it only uses one thread.

Or do you mean the 100% CPU utilization after a move?

Regards,
Vlad.

vladstamate wrote:

Dr.Wael Deeb wrote:I hope the MP feature of the program works fine now....

Hi,

What do you mean by that? Unfortunately the MP support is off for now, since it still has some issues. At the moment it only uses one thread.

Or do you mean the 100% CPU utilization after a move?

Regards,
Vlad.

Hi Vlad,
No,I mean using all the CPUs while running on my quad....So the MP support is off by now,sad to say the least
I hope that you fix it soon....
Dr.D

vladstamate wrote:Hi Guenther (and everyone),

Plisk 0.0.9 is out now at the usual place (http://sites.google.com/site/pliskchess/). It has quite a number of improvements over 0.0.8 and plays stronger. I also think I have fixed the 100% CPU utilization that you reported. I would appreciate some feedback and confirmation on that though.

Regards,
Vlad.

Hi again Vlad (I nearly had written Vas now...),

Finally I had the time to check the problem again, when we are already at Plisk 0.11 and I can tell you the problem with cpu usage
on my Quad is gone!
One thing I noticed already in my current test. It seems Plisk has a tendency to allow 50 moves draws/repetitions in positions
it thinks it is still quite better(I must look closer if they are winning positions at all, but at least Plisks eval showed a big plus)

Regards,
Guenther

Hi Guenther,

Guenther Simon wrote:Finally I had the time to check the problem again, when we are already at Plisk 0.11 and I can tell you the problem with cpu usage
on my Quad is gone!

Excellent!

Guenther Simon wrote:One thing I noticed already in my current test. It seems Plisk has a tendency to allow 50 moves draws/repetitions in positions
it thinks it is still quite better(I must look closer if they are winning positions at all, but at least Plisks eval showed a big plus)

Interesting. I shall look into this.

Regards,
Vlad.