The Baron - Latest Version Released

[pijl]

I did try version 2.23 some time ago, but it did not seem any better than 1.7.0 and did not have a native linux version like 1.7.0
A 32 bit Windows version and a 64 bit linux version of the new engine would both be useful to me if the Syzygy problem is fixed.

Enjoy it. By the way, 2.23 is tested by CCRL as well as and they have a different finding. Perhaps 3.29 is something for you then. At least a 32-bit windows version is part of that package (and significantly stronger than 1.7.0).

[Guenther]

Now after I specified the same folder for The Baron for DTZ too, it still shows just 8.x scores here in the given K+NB - K endgame. This is in Winboard.
Do I need them in a separate folder Richard?

The szdtzpath is actually ignored. syzygypath is used for both wdl and dtz files.
Note that the 8.x scores are normal until it finds the mate. Handling of tb positions is not really as it should be, but it should be able to play winning moves.

Thanks for the info Richard. Nice to see you here again :)

Guenther

[Werner]

Note that the 8.x scores are normal until it finds the mate. Handling of tb positions is not really as it should be, but it should be able to play winning moves.

Are you sure?

[pgn][Event "40 Züge in 2 min"]
[Site "Engine Match"]
[Date "2017.10.24"]
[Round "1"]
[White "The Baron 3.41uRP x64 1CPU"]
[Black "Arasan 20.1 x64 1CPU"]
[Result "0-1"]
[SetUp "1"]
[FEN "4k3/4p3/8/8/3p4/8/4N3/4KB2 b - -"]

1... e5 {+M35/24 3s} 2. Kd2 {+100.98/19 18s} Kd7 {+M35/25
3s (Ke7)} 3. Ng3 {+100.98/18 7s} Kd6 {+M35/24 3s (Ke7)}
4. Bc4 {+100.98/18 10s} Kc6 {+M35/26 3s (Kd7)} 5. Nf5
{+100.98/17 3s} Kc5 {+M35/27 3s} 6. Bf7 {+100.98/17 2s} e4
{+M35/25 3s (Kb4)} 7. Be6 {+100.98/14 2s} d3 {+M35/26 3s}
8. Ke3 {+100.98/15 5s} Kb6 {+M35/29 3s (Kc6)} 9. Ne7
{0.00/18 0s} Kc5 {+M35/28 3s (Ld5)} 10. Nc8 {+100.98/15 3s}
d2 {+M35/35 3s (Kb4)} 11. Na7 {0.00/30 0s} d1=Q {-130.62/18
3s (Kxd2)} 12. Kf4 {-100.98/11 4s} Qd6+ {-130.62/16 3s}
13. Ke3 {-100.99/11 0s} Qxe6 {-130.81/17 2s time} 0-1 [/pgn]

The computer thinks about its move
Time Management: Time Left 0:00:59.688, Moves to play 27, start --:--:--.---, min 0:00:00.568, max 0:00:02.490, easy 0:00:00.113, depth 79
Switching to SyZyGy DTZ bases, clearing hashtable
Starting search ...
Error (go): No legal moves found after search

[pgn][Event "Endgames"]
[Site "I56600K"]
[Date "2017.10.24"]
[Round "1"]
[White "The Baron 3.41wb x64 1CPU"]
[Black "Stockfish 8.0 x64 1CPU"]
[Result "1/2-1/2"]
[BlackElo "2200"]
[Time "19:42:28"]
[WhiteElo "2200"]
[TimeControl "40/120:40/120:40/120"]
[SetUp "1"]
[FEN "4k3/4p3/8/8/3p4/8/4N3/4KB2 w - - 0 1"]
[Termination "adjudication"]
[PlyCount "9"]
[WhiteType "program"]
[BlackType "program"]

1. Nc3 {+M5/48 0} dxc3 {0.00/127 2} 2. Kd1 {0.00/31 2} Kf7 {0.00/34 2} 3.
Ba6 {0.00/38 2} Kf6 {0.00/35 1} 4. Bc8 {0.00/37 1} Ke5 {0.00/31 1} 5. Bb7
{0.00/37 1 Benutzer Entscheidung} 1/2-1/2[/pgn]

[pijl]

Are you sure?

Apparently I should not be. I'll check this a bit more carefully.

[Colin-G]

BTW I do not have any dtz bases on my computers.
...

Are you sure? I have checked my Syzygy 5 men folder and in fact it contains DTZ too, which I did not know before (145 WDL and 145 DTZ files ~940MB together), because all programs worked with them so far.
Guenther

Perhaps I do have the DTZ files after all.
My Syzygy folder contains 145 ".rtbz" and 145 ".rtbw" files totaling 984MB.
I had assumed one set was for white to move, and the other was for black to move, like the Nalimov files.