CSVN Programmer's Tournament is underway

[jdart]

Currently playing round 2, although I don't see the pairings. Round 1 results are here: http://www.computerschaak.nl/

Arasan is playing thanks to Richard Pijl, who offered to have his daughter operate Arasan via network to my local server.

--Jon

[jdart]

The Baron by Richard Pijl has won the tournament:

http://www.computerschaak.nl/pt46-round-1.

Congrats to Richard! Dirty came in 2nd.

Arasan finished 4th, behind Rookie. Generally it did ok but mysteriously dropped a Rook in the 2nd round game against DBP/PuppetMaster - not sure what happened there.

--Jon

[Dr.Wael Deeb]

The Baron by Richard Pijl has won the tournament:

http://www.computerschaak.nl/pt46-round-1.

Congrats to Richard! Dirty came in 2nd.

Arasan finished 4th, behind Rookie. Generally it did ok but mysteriously dropped a Rook in the 2nd round game against DBP/PuppetMaster - not sure what happened there.
--Jon

Symptoms of a nasty bug I assume somewhere in the evaluation function....
Dr.D

[Andres Valverde]

The Baron by Richard Pijl has won the tournament:

http://www.computerschaak.nl/pt46-round-1.

Congrats to Richard! Dirty came in 2nd.

Arasan finished 4th, behind Rookie. Generally it did ok but mysteriously dropped a Rook in the 2nd round game against DBP/PuppetMaster - not sure what happened there.

--Jon

I see Arasan 2nd ...

[Harvey Williamson]

The Baron by Richard Pijl has won the tournament:

Congratulations to Richard on his 1st tournament win!! The Baron was helped a little by running remotely on 1 of my machines.

[jdart]

Symptoms of a nasty bug I assume somewhere in the evaluation function....
Dr.D

I don't think so. The bad move is .. Rd2 here:

[D] 4r1k1/ppq2pbp/2p3p1/8/3r2b1/BP2N3/P1P2PB1/2R1QRK1 b - -

I cannot reproduce this. I am wondering actually if Rd2 was transmitted to the opponent instead of Rd7 (the tournament is not automated but requires a human operator). Rd7 is maybe not the best move but is ok. However when I tried an offline search Arasan seems to prefer Bf3 in this position. Game follows.

[pgn]
[Event "CSVN PT46"]
[Site "Leiden"]
[Date "2014.05.17"]
[Round "2"]
[White "DBP/Puppetmaster"]
[Black "Arasan"]
[Result "1-0"]
[ECO "A05"]
[Annotator "1. -0.02"]
[PlyCount "76"]
[EventDate "2014.05.17"]
[TimeControl "3600"]

1. Nf3 Nf6 2. Nc3 d5 3. d4 g6 4. e3 Bg7 5. Bd3 Nc6 6. O-O O-O 7. b3 Bg4 8. Ba3
Qd7 9. h3 Bh5 10. Kh1 Rfe8 11. g4 Nxg4 12. hxg4 Bxg4 13. Be2 e5 14. Nh2 Bf5 15.
Bf3 exd4 16. Nxd5 Rad8 17. Rc1 Ne5 18. Bg2 Ng4 19. exd4 Nxh2 20. Kxh2 Bg4 21.
Qd2 c6 22. Ne3 Qc7+ 23. Kg1 Rxd4 24. Qe1 Rd2 25. Qxd2 Rd8 26. Qb4 Bh5 27. Rfe1
h6 28. Qe7 Qxe7 29. Bxe7 Rd7 30. Nc4 Bc3 31. Re3 Bd4 32. Re4 Bc3 33. Bc5 Kg7
34. Re3 Bf6 35. Rce1 Bg4 36. Ne5 Bxe5 37. Rxe5 b6 38. Bb4 Rd8 1-0
[/pgn]

[jdart]

> I see Arasan 2nd ..

Yes, that seems to be correct. I don't know what page I was looking at but it was evidently not the final standings.

--Jon

[Graham Banks]

Symptoms of a nasty bug I assume somewhere in the evaluation function....
Dr.D

I don't think so. The bad move is .. Rd2 here:

[D] 4r1k1/ppq2pbp/2p3p1/8/3r2b1/BP2N3/P1P2PB1/2R1QRK1 b - -

I cannot reproduce this. I am wondering actually if Rd2 was transmitted to the opponent instead of Rd7 (the tournament is not automated but requires a human operator). Rd7 is maybe not the best move but is ok. However when I tried an offline search Arasan seems to prefer Bf3 in this position. Game follows.

[pgn]
[Event "CSVN PT46"]
[Site "Leiden"]
[Date "2014.05.17"]
[Round "2"]
[White "DBP/Puppetmaster"]
[Black "Arasan"]
[Result "1-0"]
[ECO "A05"]
[Annotator "1. -0.02"]
[PlyCount "76"]
[EventDate "2014.05.17"]
[TimeControl "3600"]

1. Nf3 Nf6 2. Nc3 d5 3. d4 g6 4. e3 Bg7 5. Bd3 Nc6 6. O-O O-O 7. b3 Bg4 8. Ba3
Qd7 9. h3 Bh5 10. Kh1 Rfe8 11. g4 Nxg4 12. hxg4 Bxg4 13. Be2 e5 14. Nh2 Bf5 15.
Bf3 exd4 16. Nxd5 Rad8 17. Rc1 Ne5 18. Bg2 Ng4 19. exd4 Nxh2 20. Kxh2 Bg4 21.
Qd2 c6 22. Ne3 Qc7+ 23. Kg1 Rxd4 24. Qe1 Rd2 25. Qxd2 Rd8 26. Qb4 Bh5 27. Rfe1
h6 28. Qe7 Qxe7 29. Bxe7 Rd7 30. Nc4 Bc3 31. Re3 Bd4 32. Re4 Bc3 33. Bc5 Kg7
34. Re3 Bf6 35. Rce1 Bg4 36. Ne5 Bxe5 37. Rxe5 b6 38. Bb4 Rd8 1-0
[/pgn]

Strange.
I do a quick check through all of the 1000+ games that are played on my computers each week and I've never seen Arasan play such an obvious blunder.

[hgm]

Arasan finished 4th, behind Rookie. Generally it did ok but mysteriously dropped a Rook in the 2nd round game against DBP/PuppetMaster - not sure what happened there.

I was not watching that game, as I had to play my own, but this is what they told me later:

Arena is what happened! It mysteriously removed a Queen from the position somewhere during the game, and so probably was feeding Arasan nonsense positions. (Arasan was running as remote engine through ssh on a Linux machine, under Arena, which was running locally on the laptop. In the game of Arasan against Spartacus also a Queen suddenly disappeared when it was moved, but there we noticed it immediately, and could recover the situation through takebacks before any damage was done.)

[jdart]

That is certainly strange! Arena 3.0 (with a non-remote engine) has been pretty reliable for me. I tried to use Winboard but as reported in the Winboard forum, I could not get it to work with a remote engine through a ssh connection. Maybe it is possible but I ran out of time and had to give up.

--Jon