Return Match for Komodo

Discussion of computer chess matches and engine tournaments.

Moderators: hgm, Rebel, chrisw

lkaufman
Posts: 5960
Joined: Sun Jan 10, 2010 6:15 am
Location: Maryland USA

Re: Return Match for Komodo

Post by lkaufman »

Laskos wrote:
lkaufman wrote:
Lyudmil Tsvetkov wrote:Indeed, missing f2 and c2 pawns is the worse possible handicap combination at 2 pawns.
Score rises very quickly to 200 and then 400-500cps, but the most unpleasant thing for the side giving the handicap is that it does not have counterplay at all. Black places an outpost in the center, exchanges and wins.

I played some games with this handicap against Komodo 8, and it did not manage to even attack my king side a single time, while with the remaining 4 positions at 2 pawns Komodo frequently attacks me, and sometimes even mates me.

Reason for GM Neumann's success has been the handicap position, and although I naturally wish him success on August 1st, this match will be much much tougher for him.
Do you really think c2f2 handicap is worse for White than f2g2? Aside from that one, why should c2f2 be worse than d2f2? I'm trying to determine whether there is something about the c2f2 combination that could be identified as bad, or is it just that f2 plus almost any other pawn is bad?
f2g2 is shown correctly by Komodo as by far the worse in both eval and self-play.

But:
e2f2 shows an eval of -1.64 to depth 34 and -610 ELO points in self-play at blitz time control.
c2f2 shows an eval of -1.61 to depth 34 and -700 ELO points.

Is Komodo eval a bit off here?
I suspect so. But I'm not sure what principle is missing or mis-weighted. Something to think about.
Komodo rules!
Lyudmil Tsvetkov
Posts: 6052
Joined: Tue Jun 12, 2012 12:41 pm

Re: Return Match for Komodo

Post by Lyudmil Tsvetkov »

A bit off-topic, those handicap positions make us all crazy.

Ran a short 20 games shoot-out at 1' with Komodo 8 in Fritz with positions arising after 1.f3 and 1.g4.

Code: Select all

f3 stats 16 draws, 4 black wins  white scores 40%

g4 stats 1 white win, 2 draws, 17 black wins white scores 10%
As expected, seemingly 1.f3 is a draw, while 1.g4 loses.
User avatar
Guenther
Posts: 4605
Joined: Wed Oct 01, 2008 6:33 am
Location: Regensburg, Germany
Full name: Guenther Simon

Re: Return Match for Komodo

Post by Guenther »

Lyudmil Tsvetkov wrote:A bit off-topic, those handicap positions make us all crazy.

Ran a short 20 games shoot-out at 1' with Komodo 8 in Fritz with positions arising after 1.f3 and 1.g4.

Code: Select all

f3 stats 16 draws, 4 black wins  white scores 40%

g4 stats 1 white win, 2 draws, 17 black wins white scores 10%
As expected, seemingly 1.f3 is a draw, while 1.g4 loses.
You are really a funny case...
Jesse Gersenson
Posts: 593
Joined: Sat Aug 20, 2011 9:43 am

Re: Return Match for Komodo

Post by Jesse Gersenson »

User avatar
Laskos
Posts: 10948
Joined: Wed Jul 26, 2006 10:21 pm
Full name: Kai Laskos

Re: Return Match for Komodo

Post by Laskos »

First game at e2b2 odds finished in draw by 3-fold repetition. GM Petr Neuman controlled the game, having kept advantage to the end, and drew at his pleasure.

[pgn][Event "?"]
[Site "?"]
[Date "2016.08.01"]
[Round "1"]
[White "Komodo 10.1"]
[Black "GM Petr Neuman"]
[Result "1/2-1/2"]
[BlackElo "2816"]
[WhiteElo "3200"]
[TimeControl "2700+15"]
[SetUp "1"]
[FEN "rnbqkbnr/pppppppp/8/8/8/8/P1PP1PPP/RNBQKBNR w KQkq - 0 1"]

1. d4 (Nf3 Nf6) d5 2. Nf3 (Nf3 e6 Bd3 c5 c3 Bd6 dxc5 Bxc5 Nbd2 Nf6 c4 b6
cxd5 O-O O-O Qxd5 Bc4 Qf5 Qe2 Rd8 Nb3 Bb7 Nxc5 Bxf3 Qxf3 Qxf3 gxf3 bxc5 Be3
Nc6 Kg2 Nd4 Rg1 Rab8 Rad1) {-0.53/29 133} Nf6 3. c4 (c4 e6 cxd5 Nxd5 Bd3
Be7 O-O b6 Bb2 O-O Nc3 Ba6 Rc1 Bxd3 Qxd3 Nc6 Rfd1 Ncb4 Qe2 a5 g3 Qe8 Ne5
Bd6 a3 Nxc3 Rxc3 Nd5 Rc2 Bxe5 dxe5 c5 h3 Rc8 Bc1 Rd8) {-0.55/30 249} e6 4.
cxd5 (cxd5 Nxd5 Bd3 Be7 O-O b6 Qa4 Bd7 Qb3 O-O Nc3 Bc6 Bc4 Nf6 Be2 a6 Be3
Bb7 Ne5 Nc6 Rad1 Nb4 Rde1 Nbd5 Bf3 b5 h3 Rc8 a4 b4 Nxd5 Nxd5 Bd2 c5)
{-0.63/31 50} exd5 5. Bd3 (Bd3 Bb4 Bd2 Qe7 Ne5 Nfd7 O-O Nxe5 dxe5 O-O Nc3
Bxc3 Bxc3 c5 f4 Nc6 Bb5 d4 Be1 Bf5 Rc1 a6 Bxc6 bxc6 a4 Rab8 Qe2 d3 Qf2 Bg4
Bc3 Be2 Rfe1 Rfd8 f5 d2 Qxe2) {-0.66/30 121} Bb4+ 6. Bd2 (Bd2 Qe7 Ne5 Nfd7
O-O Nxe5 dxe5 O-O Nc3 Rd8 Qb1 Nc6 Bxh7 Kh8 Bd3 d4 Ne4 Bxd2 Nxd2 Nxe5 f4
Nxd3 Qxd3 Qe3 Qxe3 dxe3 Nc4 e2 Rfe1 Be6 Ne3 Rd2 Kf2 g6 Rxe2 Rad8 g4 Kg7 f5
Bc8 h3 b6 Rae1 Ba6 Rxd2) {-0.85/33 34} Bxd2+ 7. Qxd2 (Qxd2 Qe7 Be2)
{-0.72/32 177} O-O 8. O-O (O-O Re8 Nc3) {-0.53/29 73} b6 9. Nc3 (Nc3 c5
Rad1) {-0.65/29 76} Bb7 10. Rab1 (Rab1 Nbd7 h3 c5 Rfe1 Re8 Bf5 Rxe1 Rxe1 h6
Qf4 Qb8 Qd2 Qf8 Bxd7 Nxd7 dxc5 Nxc5 Nb5 Ne4 Qb2 Qe7 Nbd4 Qf6 Rc1 Rc8 Rxc8
Bxc8 Qc1 Nc5 Qe3 Ne6 Ne5 Nxd4 Qxd4 Qd6) {-0.72/27 210} Nc6 11. Qf4 (Qf4 Nh5
Qd2 Re8 Ne2 Nf6 Ng3 Ne4 Qb2 Nd6 Qd2 Na5 Qf4 Nac4 Rbe1 Rxe1 Rxe1 g6 Ne5 f5
Ne2 Qf6 Nc3 Re8 Re2 Qg7 h3 g5 Qc1 Nxe5 dxe5 Rxe5 Rxe5 Qxe5) {-0.62/28 89}
Ne7 12. h4 (h4 c6 a4 Ng6 Qh2 Qb8 g3 Bc8 a5 Bg4 Qg2 Qd6 Rfc1 Rfe8 Nh2 b5
Nxg4 Nxg4 Qf3 h5 Bf5 Nf8 Kg2 Nf6 Na2 Ne6 Qd1 g6 Bxe6 Rxe6 Nb4) {-0.75/28
120} Ng6 13. Qh2 (Qh2 Re8 Nb5 Re7 Rfc1 Ne8 h5 Nf8 h6 g6 a4 Ne6 Qg3 c6 Nc3
Qd6 Qh4 Nf4 Bf1 Qf6 Qxf6 Nxf6 a5 Nd7 Na4 b5 a6 Bc8 Nb2 Nb8 Nd3 Nxd3 Bxd3
Bd7 Ne5 f6 Nxd7 Rxd7) {-0.69/30 69} Ng4 14. Qg3 (Qg3 h5 Nb5) {-0.18/31 63}
Nf6 15. Qh2 (Qh2 Re8 Nb5 Re7 Rfc1 Ne8 h5 Nf8 h6 g6 a4 Ne6 Re1 a6 Nc3 Nf6
Qg3 Nh5 Qh2 Nef4 Bf1 Rxe1 Rxe1 Qf6 Ne2 Ne6 a5 Re8 axb6 cxb6 Qh3 Qe7 Ne5 Nf6
Qb3 b5 g3 Ne4) {-0.79/30 142} Ng4 16. Qg3 (Qg3 h5 Nb5 Ba6 Rfc1 Bxb5 Rxb5
Re8 a4 Nf8 Qxc7 Qxc7 Rxc7 Ne6 Rc2 Rac8 Rxc8 Rxc8 Bf1 Kf8 Rxd5 Nf6 Re5 Rd8
a5 Nxd4 Nxd4 Rxd4 axb6 axb6 g3 Rd2 Rb5 Ne4 Rxb6 Nxf2 Rb8 Ke7 Rb5 Kf6 Rxh5
Ne4 g4 Rd1 Ra5) {-0.22/32 40} Nf6 17. Qh2 (Qh2) {0.00/50 117 3-fold
repetition}

{White Time: 21:24 Black Time: 19:44} *[/pgn]
lkaufman
Posts: 5960
Joined: Sun Jan 10, 2010 6:15 am
Location: Maryland USA

Re: Return Match for Komodo

Post by lkaufman »

Apparently he was worried about an attack if he avoided the draw. Win pays 3x as much as a draw, so he won't take a draw if he feels he has kept his advantage. Maybe he misjudged here.
Komodo rules!
User avatar
Laskos
Posts: 10948
Joined: Wed Jul 26, 2006 10:21 pm
Full name: Kai Laskos

Re: Return Match for Komodo

Post by Laskos »

Second game, this time at e2c2 odds, was won by Komodo, which steadily improved its position and in the endgame overwhelmed the GM.

[pgn][Event "?"]
[Site "?"]
[Date "2016.08.01"]
[Round "2"]
[White "Komodo 10.1 (23 cores)"]
[Black "GM Petr Neuman"]
[Result "1-0"]
[BlackElo "2816"]
[WhiteElo "3200"]
[TimeControl "2700+15"]
[SetUp "1"]
[FEN "rnbqkbnr/pppppppp/8/8/8/8/PP1P1PPP/RNBQKBNR w KQkq - 0 1"]

1. d4 d5 2. Nf3 c6 3. Qc2 (Qc2 Nf6 Bd3 Bg4 Ne5 Bh5 Bg5 e6 O-O Qb6 Be3 Bg6
Nc3 Bd6 g3 Bxd3 Nxd3 Nbd7 f3 O-O Na4 Qc7 Nac5 Rac8 Kg2 h6 Rac1 Rfe8 Rcd1
Nf8 Rfe1) {-0.42/29 101} Nf6 4. Bd3 (Bd3 Bg4 Ne5 Bh5 O-O e6 Bg5 Nbd7 Nxd7
Qxd7 Nc3 Bg6 Rac1 Be7 Bf4 O-O Na4 Bxd3 Qxd3 Nh5 Be3 b6 Nc3 c5 Qe2 Nf6 Bf4
Rac8 Rfd1 cxd4 Rxd4 Bc5 Rdd1 Bd6 Be5) {-0.59/30 92} Bg4 5. Ne5 (Ne5 Bh5 O-O
e6 Bg5 Nbd7 Nxd7 Qxd7 Nd2 Bd6 g3 Qc7 Rae1 Bg6 f4 O-O Bxf6 gxf6 Kh1 f5 Nb3
a5 a3 Bh5 Be2 Bxe2 Qxe2 a4 Qh5 Qd8 Nc5 Bxc5 dxc5 d4 Qf3) {-0.68/31 43} Bh5
6. Bg5 (Bg5 Nbd7 Nd2 e6 O-O Bd6 Rfe1 Qc7 Bf4 Bg6 Bxg6 hxg6 Ndf3 O-O Rac1
Rac8 Qe2 Nxe5 Bxe5 Ne4 Bxd6 Qxd6 Ne5 c5 f3 cxd4 Rxc8 Rxc8 Nxf7 Qb6 Qd3 Qxb2
fxe4 Rc1 Qg3 Rxe1 Qxe1 Kxf7 exd5 exd5) {-0.72/32 294} Bg6 7. Nxg6 (Nxg6
hxg6 Nc3) {-0.59/29 160} hxg6 8. Nc3 (Nc3 Nbd7 O-O-O e6 Kb1 Rc8 h4 Bb4 Na4
Be7 g3 b5 Nc5 Nxc5 dxc5 a5 Bf4 d4 Be5 Qd5 f4 a4 Bxf6 gxf6 Be4 Qxc5 Qxc5
Bxc5 Rc1 f5 Bf3 Bd6) {-0.64/28 36} e6 9. O-O-O (O-O-O Nbd7 Kb1) {-0.63/31
185} Be7 10. h4 (h4 Nbd7 Kb1) {-0.65/28 79} Nbd7 11. Kb1 (Kb1 c5 dxc5)
{-0.65/28 0} Qb6 12. Rhe1 (Rhe1 Kf8 Bf1) {-0.48/27 109} O-O-O 13. g3 (g3
Qc7 Rc1 a6 Na4 Kb8 Qb3 Ka7 Bd2 Rhe8 Qc3 Ng4 Re2 Bd6 Bg5 Ngf6 Nc5 e5 Nxd7
Rxd7 dxe5 Rxe5 Be3 Kb8 Qb3 Rde7 Bb6 Qd7 Rxe5 Bxe5 a3 Bc7 Bd4 Ne4 Qc2 Be5
Bb6 Qg4) {-0.50/30 107} Kb8 14. Bf4+ (Bf4 Ka8 Na4) {-0.40/29 65} Ka8 15.
Na4 (Na4 Qa5 a3) {-0.46/31 110} Qa5 16. Rc1 (Rc1 Nh5 Bg5 Rde8 a3 Qc7 Bxe7
Rxe7 Nc5 Nxc5 Qxc5 Nf6 Qc3 Rhe8 f4 Rd7 Bc2 Kb8 Ka2 Red8 Qe3 Ne4 Red1 Qa5
Rb1 Qb5 Bd3 Qb6 Rbc1 f5 Re1 Re8 Rc2 Rc8 Rcc1) {-0.54/31 111} Rc8 17. Bf1
(Bf1 c5 Bd2) {-0.85/30 197} c5 18. Bd2 (Bd2 Qd8) {-0.85/30 0} Qd8 19. dxc5
(dxc5 Nxc5 Bf4 Qa5 Nxc5 Bxc5 Bd3 Rc6 Qe2 Bd4 Bb5 Rxc1 Rxc1 e5 Bd2 Qb6 Be3
Bxe3 fxe3 Kb8 Bd3 Qe6 a3 Rc8 Rxc8 Qxc8 Bc2 Qg4 Qd2 Qc4 Bd3 Qc5 Qf2 Qc7 Qg2
Qb6 Qf2 a6 Bc2 Qc7 Qd2 Qc4 Bd3) {-1.15/32 22} Nxc5 20. Bf4 (Bf4 Qa5 Nxc5
Bxc5 Bd3 Rc6 Qe2 Bd4 Red1 Qb6 Rd2 Rxc1 Kxc1 Rc8 Kb1 Nh5 Be5 Bxe5 Qxe5 Qc7
Qe1 Qd6 Rd1 Nf6 Qe2 Kb8 a3 a6 Bc2 Qc5 Rd2 Qb6 Rd1 Qc7 Rd2 Qc5 Rd1 Rc7 Rd2
Rc6 Rd1 Ka7) {-1.19/35 159} Nfd7 21. Nxc5 (Nxc5 Nxc5 Red1) {-0.93/30 136}
Nxc5 22. Red1 (Red1 g5 hxg5 Bxg5 Bxg5 Qxg5 b4 Na6 Qxc8 Rxc8 Rxc8 Nb8 Bd3
Qg4 Kc1 e5 Rg8 Qf3 Bc2 Qa3 Kb1 Qxb4 Kc1 a6 Rxd5 Ka7 Rxe5 Nc6 Re3 Qc4 Rc8
Qf1 Kb2 Qxf2 Rc3 Nd4 Rd3 Qf6 a3 Nb5 Kc1 Qa1 Kd2 Nxa3 Rc7 Nxc2 Kxc2 Qf1 Rdd7
Qe2 Kc3 Qe3) {-1.01/31 162} Ne4 23. Qa4 (Qa4 Qb6 Qd7 Bc5 Rd3 Rhd8 Qxf7 Bd6
Qxe6 Rxc1 Bxc1 Nc5 Qe3 Nxd3 Qxb6 axb6 Bxd3 Rf8 f4 Bb4 Bxg6 Be1 g4 Bxh4 Bf5
Re8 Kc2 Kb8 a4 Kc7 Kd3 Kc6 Bd2 Bf2 Bc3 Re3 Kc2 Rf3 Bxg7 Rxf4 b3 Bc5 Bf6)
{-0.78/31 45} Nc5 24. Qb4 (Qb4 b6 Be3) {-0.58/31 91} Bd6 25. Bg5 (Bg5 Qd7
Be3 Qe7 Bg2 Rc7 Rc2 Rhc8 Rdc1 b6 Qd2 Qd7 f4 f5 Bxc5 Rxc5 a3 Rxc2 Rxc2 Rxc2
Qxc2 Kb7 Qc3 Qe7 Qd2 a6 Bf3 Qf7 b4 Be7 Ka2 Bf6 Qd3 Ka7) {-0.62/28 69} Qb6
26. Qc3 (Qc3 Rhg8 Be3 Rc6 Qe1 Qd8 Bb5 Rc7 Bd4 Qe7 b4 Ne4 Rxc7 Qxc7 Rc1 Qe7
a3 e5 Bb2 Qf6 Qd1 Qe6 Rc2 Rd8 Bf1 f5 Bb5 Kb8 Qd3 d4 Bc4 Qe8 Bd5 Nf6 Bb3
Rc8) {-0.64/29 29} f6 27. Be3 (Be3 Qc6 Qa3 b6 b4 Qa4 Qxa4 Nxa4 Bd2 Be5 Bh3
f5 Bf1 Nc3 Bxc3 Rxc3 Re1 Rxc1 Kxc1 Bd4 Rxe6 g5 Bg2 gxh4 Bxd5 Kb8 gxh4 Bf6
Kc2 Rxh4 Kb3 Rd4 Bc4 Kc7 Bb5 Rd5 a4 Rd2 Re2 Rxe2 Bxe2 Kd6 f4 g5) {-0.35/28
50} Qd8 28. Qa3 (Qa3 Qe7 f4 b6 Bxc5 bxc5 Bb5 Rc7 Qa6 Rhc8 Rc2 c4 Re2 c3 Rc2
g5 hxg5 fxg5 Re1 Rc5 Rxc3 Rxc3 bxc3 Qc7 Rxe6 Qxc3 Qxc8 Qxc8 Re8 Qxe8 Bxe8
gxf4 Bc6 Kb8 gxf4 Bxf4 Bxd5 Kc7 Kc2 Kd6 Bf3 Ke5 Kd3 g5 Bg4) {-0.35/28 49}
Qe7 29. Rc2 (Rc2 b6 Bxc5) {-0.50/30 79} Na6 30. Qa4 (Qa4 Nc5 Qa3) {0.00/36
85} Rxc2 31. Qxc2 (Qxc2 Qe8 a3 Nb8 Rc1 e5 Bg2 d4 Bd2 Nc6 f4 Ne7 fxe5 Bxe5
Bxb7 Kxb7 Qb3 Ka8 Qf3 Kb8 Qb3 Ka8) {0.00/33 39} Nb4 32. Qxg6 (Qxg6 Nc6 Qc2
Qd7 Qd3 Ne5 Qd4 Nc6 Qa4 Bb8 a3 Rd8 Bg2 a6 f4 Ba7 Bxa7 Kxa7 Qb3 d4 Bh3 Na5
Qb4 Qd5 Bxe6 Qe4 Ka1 Qxe6) {+0.08/26 18} Rc8 33. a3 (a3 Nc6 Rc1 f5 Be2 Qd7
g4 f4 Bd2 Bb8 Bb5 a6 Bxc6 Rxc6 Rxc6 Qxc6 Qxg7 Qc4 h5 Qd3 Kc1 Qf1 Kc2 Qxf2
h6 Qe2 h7 Qe4 Kc1 Qh1 Kc2 Qe4) {0.00/31 40} Nc6 34. Rc1 (Rc1 Bb8 h5 a6 f4
Ba7 Bd2 Bd4 Qg4 Ka7 Bg2 Rd8 Bh3 f5 Qe2 Qf6 Bf1 Rc8 Be3 Rh8 Rxc6 bxc6 Bxd4
Qxd4 Qxa6 Kb8 Qxc6 Qd1 Ka2 Qxf1 Qb6 Ka8 Qc6 Ka7 Qc7 Ka6 Qc6 Ka7) {0.00/32
71} Bc7 35. Qd3 (Qd3 Rd8 Qc2 e5 Bb5 Qe6 f3 Rc8 Bxc6 bxc6 Qxc6 Qxc6 Rxc6 d4
Bd2 Kb7 Rc4 f5 f4 d3 fxe5 Bxe5 Rxc8 Kxc8 Bf4 Bf6 Kc1 Kd7 a4 Ke6 Be3 a6 Kd2
Kd5 Kxd3 Bxb2 Bf4 Bd4 h5 Bb6 Be3) {+0.18/31 49} Bb6 36. Bxb6 (Bxb6 axb6 Bh3
Ka7 Re1 Ne5 Qd1 Rc6 f4 Nc4 f5 e5 Qxd5 Qd6 Rd1 Qxd5 Rxd5 e4 Rd7 b5 Bg4 Rd6
Rxg7 Rd2 Re7 e3 Kc1 Ne5 Bh5 Nd3 Kb1 Ne5 Kc1) {0.00/32 29} axb6 37. Bh3 (Bh3
Ka7 Re1 Ne5 Qd1 Rc6 f4 Nc4 f5 e5 Qxd5 Kb8 Rd1 Nxa3 bxa3 Qxa3 Qg8 Rc8 Rd8
Qc1 Ka2 Qc2 Ka1 Qc1 Ka2) {0.00/37 18} Rd8 38. Qe3 (Qe3 Rd6 Qxb6 Kb8 Bg2 Qd8
Qe3 Ne5 Qd4 Qb6 Qxb6 Rxb6 Bf1 Nf3 Ka2 Rc6 Rxc6 bxc6 b4 Kb7 Kb2 c5 Bh3 cxb4
axb4 Nd4 Kc3 e5 Bd7 Kc7 Be8 Kb6 Ba4 Kc7 g4) {+0.29/27 23} e5 39. Qxb6 (Qxb6
Rd6 Qc5 Kb8 h5 Rd8 Qb6 Rd6 Bf5 Qd8 Qe3 Ne7 Bc2 Rb6 Qf3 Nc6 Qd3 Nd4 Ba4 Ra6
Bd1 Rc6 Rxc6 bxc6 Qh7 Qf8 h6 gxh6 Qd7 f5 Ba4 Qc8 Qd6 Qc7 Qxh6) {+0.17/30
47} Rd6 40. Bf5 (Bf5 Kb8 Qc5 Qd8 Qe3 Ne7 Bc2 Qa5 b4 Qb5 Kb2 d4 Qd3 Qd7 Bb3
Rc6 Rxc6 Qxc6 f3 Nd5 b5 Qc3 Qxc3 Nxc3 b6 Ne2 g4 Nf4 Kc2 Kc8 Ba4 Ng2 h5 Ne1
Kd2 Nxf3 Kd3 Ng5 Bb5 Kd8) {+0.16/30 10} e4 41. g4 (g4 d4 Qb5 e3 Re1 Qe5
Qxe5 Nxe5 fxe3 Nf3 Re2 Nxh4 Be4 d3 Rd2 Re6 Bh7 Rxe3 Rxd3 Rxd3 Bxd3 Nf3 Kc2
Ne5 Bf5 Kb8 Kc3 Kc7 Kd4 g6 Be4 g5 Bf5 Kc6 b3 Nf3 Ke3 Ne5 Ke4 Kd6 Kd4)
{+0.40/31 35} Kb8 42. h5 (h5 d4 Qc5 e3 Re1 Re6 Qxe7 Rxe7 fxe3 dxe3 Kc2 Nd4
Kd3 Nxf5 gxf5 Re5 Rxe3 Rxf5 Rg3 Rxh5 Rxg7 Ka7 Re7 Rh2 Kc3 Rh3 Kb4 Rh2 Kb3
Rh3 Ka2 Rf3 a4 Rf4 Ka3 Rf3 b3 Kb6 Kb4 Rf4 Kc3 Kc6 Rf7 Rf2 Kd4) {+0.29/34
47} Qe5 43. a4 (a4 Qd4 Qxd4 Nxd4 Rc8 Ka7 Rc7 Nxf5 gxf5 Rd8 Rxg7 Rh8 Rd7
Rxh5 Rxd5 Rh2 Rd6 Rxf2 Rxf6 e3 Ka2 Rf4 Re6 Rxa4 Kb3 Rf4 Re5 e2 Rxe2 Rxf5
Re6 Rf2 Kc3 b5 Rd6 Kb7 Rh6 Rf4 b3 Ka7 Kd2 Rf3 b4) {+0.19/34 75} Qd4 44.
Qxd4 (Qxd4 Nxd4 Rc8 Ka7 Re8 Nxf5 gxf5 Rc6 Re7 Rc8 Rxg7 Rh8 Rd7 Rxh5 Rxd5
Rh2 Rd6 Rxf2 Rxf6 e3 Ka2 e2 Re6 Rxf5 Ka3 Rf3 Kb4 Rf4 Kc5 Rf2 b3 Rf5 Kc4 Rf4
Kb5 Rf5 Kb4 Rf2 Re7 Kb6 Kc3 Rf3 Kc2 Rf2 Kd2 Rf3 b4 Rf4) {+0.19/34 0} Nxd4
45. Rc8+ (Rc8 Ka7 Re8 Nxf5 gxf5 Rc6 Rd8 Rc5 Rd7 d4 Rxg7 Rxf5 Re7 Rxf2 Rxe4
Rh2 Rxd4 Rxh5 b4 Rh2 Kc1 Kb6 Rf4 Ka7 Re4 Kb6 Re6 Ka7 a5 f5 Re5 f4 Kd1 Rb2
Rb5 f3 Ke1 Re2 Kf1 Rh2 Rd5 Ra2 Rd6 Rb2 Rb6) {+0.07/36 42} Ka7 46. Re8 (Re8
Nxf5 gxf5 Rc6 Re7 Rc8 Rxg7 Rh8 Rd7 Rxh5 Rxd5 Rh2 Rd4 Rxf2 Rxe4 Rxf5 Re6 Ra5
Kc2 Rxa4 Rxf6 b5 Kc3 Rc4 Kb3 Rd4 Rg6 Rh4 Re6 Ra4 Kc2 Rh4 Rg6 Rc4 Kb1 Ra4
Rd6 Rh4 Ka2 Ra4 Kb3 Rc4 Rh6) {+0.01/36 32} Nxf5 47. gxf5 (gxf5 Rc6 Re7 Rc8
Rxg7 Rh8 Rd7 Rxh5 Rxd5 Rh2 Rd4 Rxf2 Rxe4 Rxf5 Kc2 Re5 Rxe5 fxe5 b4 Kb6 Kd3
Kc6 Ke4 Kd6 a5 Kc6 Kxe5) {+0.01/36 0} d4 48. Rxe4 (Rxe4 Rd5 Rf4 b5 Rg4 Rxf5
Rxg7 Ka6 Rg4 Rxh5 Rxd4 Rf5 f4 bxa4 Ka2 Kb5 Ka3 Rc5 Rb4 Kc6 Kxa4 Rc2 f5 Rc5
Rf4 Kd6 b3 Rd5 Rf1 Re5 Rf3 Kd5 Rf2 Kc5 Rf4 Kc6 Kb4 Rb5 Kc4 Rc5 Kd4 Kb5)
{+0.42/34 16} Rd5 49. Rf4 (Rf4 b5 Rg4 Rxf5 Rxg7 Ka6 axb5 Kxb5 Rg4 Rxh5 Rxd4
Rf5 Rd2 Kc4 Kc2 Rf3 Kd1 Rb3 Ke2 Kc5 Kf1 f5 Ke2 Kc4 Kd1 Kc5 Kc1 Kc4 Kc2 Rf3
Kd1 Rb3 Ke1 Rh3 Kf1 Kc5 Kg2 Rb3 Kh2 Rf3 Kg1 Rh3) {+0.37/37 9} b5 50. Rg4
(Rg4 Rxf5 Rxg7 Ka6 Rg4 Rxh5 Rxd4 Rf5 axb5 Kxb5 Rd2 Kc5 Kc2 Rf3 Kc1 Rf5 Rc2
Kd4 Kd1 Rf3 Rd2 Kc5 Re2 Rb3 Kc1 Kd4 Kc2 Rf3 Kd2 Kc4 Kd1 Kc5 Rd2 Kc4 Ke2 Rb3
Kf1 Kc5 Kg2 f5 Kf1 Kc4 Ke2 Kc5 Ke1 Kc4 Kd1) {+0.39/41 0} Rxf5 51. Rxg7+
(Rxg7 Ka6 Rg4 Rxh5 Rxd4 Rf5 axb5 Kxb5 Rd2 Kc4 Kc2 Rf3 Kd1 Rb3 Rc2 Kd3 Re2
Kc4 Kc2 Rf3 Kc1 Kd4 Rd2 Kc4 Rc2 Kd4 Re2 Kd3 Kd1 Kc4 Rc2 Kd5 Rd2 Kc5 Ke1 Kc4
Ke2 Rb3 Kf1 f5 Ke2 Kc5 Ke1 f4 Kd1 Rf3 Kc1 Kb4 Rc2 Rd3 Re2) {+0.36/43 0} Kb6
52. Rg3 (Rg3 Rxh5 Rb3 Rh1 Kc2 Ka5 Rxb5 Kxa4 Rd5 Re1 b3 Kb4 Rxd4 Kb5 Rf4 Re6
b4 Kb6 Kc3 Kc7 Rf5 Kd6 Kd4 Kc7 b5 Rb6 Rc5 Kd6 Kc4 Kd7 Rd5 Kc7 Rf5 Re6 Kc5
Re2 Rxf6 Rc2 Kd5 Rd2 Kc4 Rc2 Kb3 Re2) {+1.79/33 35} Rxh5 53. Rb3 (Rb3 Rh1
Kc2 Ka5 Rxb5 Kxa4 Rd5 Rh3 Rxd4 Kb5 Rf4 Rh6 b4 Rh3 Kd2 Rh6 Kd3 Kc6 Kc4 Kc7
b5 Kd6 Kb4 Kc7 Kc5 Rh5 Kc4 Rh6 Kd5 Kd7 Kc5 Kc7 Rf5 Rg6 Kc4 Kd6 Rd5 Kc7 Rc5
Kb7 Rf5 Rh6 Kd5 Kc7 Ke6 Rh1) {+1.79/33 0} Rh1+ 54. Kc2 (Kc2 Ka5 Rxb5)
{+1.79/33 0} Ka5 55. Rxb5+ (Rxb5 Kxa4 Rd5 Rh7 Rxd4 Kb5 Kc3 Rc7 Kd3 Rb7 b4
Kc6 Kc4 Rb8 Rf4 Kd7 Rxf6 Rc8 Kd3 Ke7 Rf5 Ke6 Rf4 Ke5 Re4 Kd6 Rd4 Ke6 b5 Rb8
Kc4 Rc8 Kb4 Rb8 Rc4 Kd5 Rc6 Rh8 Rc5 Kd6 b6 Rh4 Kb5 Rh3 Rf5) {+3.00/38 11}
Kxa4 56. Rd5 (Rd5 Rh7 Rxd4 Kb5 Rf4 Rf7 Kc3 Kc6 b4 Kd6 Rd4 Kc6 Kc4 Re7 b5
Kc7 f4 Re2 Kc5 Rb2 Rd5 Rc2 Kd4 Re2 Rf5 Re6 Kc5 Kd7 Rd5 Kc7 Rd4 f5 Rd5 Rf6
Kc4 Kb7 Rc5 Rb6 Rxf5 Rh6 Kd5 Rh7 Re5) {+3.11/39 6} Rh4 57. Kd3 (Kd3 Kb3
Rxd4 Rh8 f4 Rb8 Rd6 f5 Rd5 Rf8 Rb5 Ka2 Kd4 Kb1 Ke5 Kc2 Rd5 Kb3 Rd4 Rh8 Rd3
Kc4 Rd6 Kb5 b4 Rf8 Rf6 Rh8 Rxf5 Rh2 Kd6 Kb6 Re5 Rh6 Kd5 Rh3 b5 Kb7 Rf5 Rd3
Ke6 Rc3 Re5 Rc8 f5 Re8 Kd6 Rh8 f6) {+3.99/41 20} Kb3 58. Rxd4 (Rxd4 Rh5 f4
Rf5 Ke3 Ra5 Rd6 f5 Kd4 Ra8 Ke5 Kb4 Kxf5 Rf8 Ke5 Rh8 f5 Kc5 b4 Kc4 Rb6 Rh7
b5 Kc3 Re6 Kb4 b6 Rb7 Kf6 Kb5 Kg6 Ka5 f6 Rb8 f7 Rd8 b7 Rd4 Ra6 Kb4 b8=Q Kc3
Qb3 Kd2 Qb2 Kd3 Qxd4) {+250.00/41 32} Rh2 59. f4 (f4 Rh6 Ke4 Rh5 Rd6 Rh6
Kf5 Kc4 Rxf6 Rh8 Ke5 Kb4 Rb6 Kc4 Re6 Rg8 f5 Rh8 f6 Rf8 Kf5 Kd3 Rd6 Ke3 Ra6
Rg8 f7 Rc8 Re6 Kd2 Rf6 Rh8 f8=Q Kc1 Qxh8) {+250.00/48 14}

{White Time: 02:52 Black Time: 57:40} *[/pgn]
lkaufman
Posts: 5960
Joined: Sun Jan 10, 2010 6:15 am
Location: Maryland USA

Re: Return Match for Komodo

Post by lkaufman »

We discovered that somehow Hyperthreading had been enabled on my 24 core machine today (and maybe for the previous match or two). It's turned off now for tomorrow's games, which makes Komodo run 6.5% faster, at least when each thread runs a separate game. So maybe Komodo will be about five elo stronger tomorrow! Not that anyone could tell the difference in two games. We always learn something from these matches, even if it's only "check to make sure hyperthreading is still disabled".
Komodo rules!
carldaman
Posts: 2283
Joined: Sat Jun 02, 2012 2:13 am

Re: Return Match for Komodo

Post by carldaman »

lkaufman wrote:We discovered that somehow Hyperthreading had been enabled on my 24 core machine today (and maybe for the previous match or two). It's turned off now for tomorrow's games, which makes Komodo run 6.5% faster, at least when each thread runs a separate game. So maybe Komodo will be about five elo stronger tomorrow! Not that anyone could tell the difference in two games. We always learn something from these matches, even if it's only "check to make sure hyperthreading is still disabled".
You mean the games will not be replayed? :P

(sorry, couldn't resist) :lol:
User avatar
Laskos
Posts: 10948
Joined: Wed Jul 26, 2006 10:21 pm
Full name: Kai Laskos

Re: Return Match for Komodo

Post by Laskos »

lkaufman wrote:We discovered that somehow Hyperthreading had been enabled on my 24 core machine today (and maybe for the previous match or two). It's turned off now for tomorrow's games, which makes Komodo run 6.5% faster, at least when each thread runs a separate game. So maybe Komodo will be about five elo stronger tomorrow! Not that anyone could tell the difference in two games. We always learn something from these matches, even if it's only "check to make sure hyperthreading is still disabled".
The games 3 and 4 were Komodo wins, I was unable to follow them live. Observing the 4 games, the difference was made mostly by 1-4 blunders per game of the sort 0.5 to 1.0 and one 4.0 pawns in four games. It seems to contradict the notion that the outcome of the game against GM is determined by accumulation of many small inaccuracies of the human.