mmt Vs. Ovyron (G4 D5 BG2)

Discussion of computer chess matches and engine tournaments.

Moderator: Ras

Zenmastur
Posts: 919
Joined: Sat May 31, 2014 8:28 am

Re: mmt Vs. Ovyron (G4 D5 BG2)

Post by Zenmastur »

Ovyron wrote: Wed Feb 05, 2020 9:58 pm Yeah, even Stockfish in some variations is showing -0.80 scores at depth 38, but jumps to -1.90 after feeding it the winning lines for black. I have stopped using Leela, as I think refuting all her variations has been a big waste of time.
What hardware are you running Leela on?
Only 2 defining forces have ever offered to die for you.....Jesus Christ and the American Soldier. One died for your soul, the other for your freedom.
jp
Posts: 1480
Joined: Mon Apr 23, 2018 7:54 am

Re: mmt Vs. Ovyron (G4 D5 BG2)

Post by jp »

Zenmastur wrote: Wed Feb 05, 2020 10:02 pm What hardware are you running Leela on?
And what depths and numbers of nodes are you getting from Leela?
User avatar
Ovyron
Posts: 4558
Joined: Tue Jul 03, 2007 4:30 am

Re: mmt Vs. Ovyron (G4 D5 BG2)

Post by Ovyron »

It's the most adorable Leela around, I'm dubbing her "Baby Leela", she hasn't learned how to walk yet, and crawls around at 23 nodes per second. Probably 10 times slower than in your phone. Awww!

But despite this, I was able to see mmt's Leela line for 2.Bg2 of 800 million nodes (I found it with this Leela back in October), and in general have an idea of what's going on in her mind, because if a variation is at -1.70 and going towards -1.80, faster hardware is only going to make her go there faster, some magic drawing move isn't going to appear there anymore, so refuting lines can be done at this "microspeed".

And thus I can predict that soon mmt is going to report a score worse than -1.40 for Leela, because I don't need more than 11111 nodes to see it.
zullil
Posts: 6442
Joined: Tue Jan 09, 2007 12:31 am
Location: PA USA
Full name: Louis Zulli

Re: mmt Vs. Ovyron (G4 D5 BG2)

Post by zullil »

mmt wrote: Wed Feb 05, 2020 1:06 pm 1. g4 d5 2. Bg2 Bxg4 3. c4 c6 4. Qb3 e6 5. Qxb7 Nd7 6. Nc3 Ne7 7. cxd5 exd5 8. d4 Rb8 9. Qa6 Rb6 10. Qd3 Ng6 11. h3 Be6 12. Nf3 Bd6 13. h4 h5 14. b3 Nf6 15. Bg5 O-O 16. e3 Re8 17. Kf1 Bg4 18. Ne1 Bb4 19. Na4

[d]3qr1k1/p4pp1/1rp2nn1/3p2Bp/Nb1P2bP/1P1QP3/P4PB1/R3NK1R b - - 3 2
Komodo at depth 44 had:

19. Na4 (-1.14)
19. Qc2 (-1.38)
19. Nc2 (-1.50)
Zenmastur
Posts: 919
Joined: Sat May 31, 2014 8:28 am

Re: mmt Vs. Ovyron (G4 D5 BG2)

Post by Zenmastur »

zullil wrote: Wed Feb 05, 2020 10:38 pm
mmt wrote: Wed Feb 05, 2020 1:06 pm 1. g4 d5 2. Bg2 Bxg4 3. c4 c6 4. Qb3 e6 5. Qxb7 Nd7 6. Nc3 Ne7 7. cxd5 exd5 8. d4 Rb8 9. Qa6 Rb6 10. Qd3 Ng6 11. h3 Be6 12. Nf3 Bd6 13. h4 h5 14. b3 Nf6 15. Bg5 O-O 16. e3 Re8 17. Kf1 Bg4 18. Ne1 Bb4 19. Na4

[d]3qr1k1/p4pp1/1rp2nn1/3p2Bp/Nb1P2bP/1P1QP3/P4PB1/R3NK1R b - - 3 2
Komodo at depth 44 had:

19. Na4 (-1.14)
19. Qc2 (-1.38)
19. Nc2 (-1.50)
I had 19. Na4 at -2.28 depth 55. I don't know what Komodo is thinking. It's scores seem so low. :( :( :(
Only 2 defining forces have ever offered to die for you.....Jesus Christ and the American Soldier. One died for your soul, the other for your freedom.
zullil
Posts: 6442
Joined: Tue Jan 09, 2007 12:31 am
Location: PA USA
Full name: Louis Zulli

Re: mmt Vs. Ovyron (G4 D5 BG2)

Post by zullil »

Zenmastur wrote: Thu Feb 06, 2020 8:39 am
zullil wrote: Wed Feb 05, 2020 10:38 pm
mmt wrote: Wed Feb 05, 2020 1:06 pm 1. g4 d5 2. Bg2 Bxg4 3. c4 c6 4. Qb3 e6 5. Qxb7 Nd7 6. Nc3 Ne7 7. cxd5 exd5 8. d4 Rb8 9. Qa6 Rb6 10. Qd3 Ng6 11. h3 Be6 12. Nf3 Bd6 13. h4 h5 14. b3 Nf6 15. Bg5 O-O 16. e3 Re8 17. Kf1 Bg4 18. Ne1 Bb4 19. Na4

[d]3qr1k1/p4pp1/1rp2nn1/3p2Bp/Nb1P2bP/1P1QP3/P4PB1/R3NK1R b - - 3 2
Komodo at depth 44 had:

19. Na4 (-1.14)
19. Qc2 (-1.38)
19. Nc2 (-1.50)
I had 19. Na4 at -2.28 depth 55. I don't know what Komodo is thinking. It's scores seem so low. :( :( :(
Looks like your engine is off by a factor of 2. :wink:

Maybe Komodo simply isn't looking deeply enough. After all, I'm running MultiPV = 3, and so it seems that I'm more or less limited to depths ~45.

Perhaps I'll simply switch to MultiPV = 1 after Black's reply.
User avatar
Ovyron
Posts: 4558
Joined: Tue Jul 03, 2007 4:30 am

Re: mmt Vs. Ovyron (G4 D5 BG2)

Post by Ovyron »

Zenmastur wrote: Thu Feb 06, 2020 8:39 am I had 19. Na4 at -2.28 depth 55. I don't know what Komodo is thinking. It's scores seem so low. :( :( :(
My evals are nowhere close to that, but I suspect it's mostly because of lack of 6men TBs.

My top move choices (19...Rb8 and 19...Rb7) are really close so that's why I've been taking my time for this move. Assuming only one of them wins then this is the hardest position I've faced in the game. If I could win with both then we've reached the point where I could mostly flip a coin to decide my moves, but I have no way to know if I'm on that situation yet.
jp
Posts: 1480
Joined: Mon Apr 23, 2018 7:54 am

Re: mmt Vs. Ovyron (G4 D5 BG2)

Post by jp »

Ovyron wrote: Thu Feb 06, 2020 4:08 pm My evals are nowhere close to that, but I suspect it's mostly because of lack of 6men TBs.
If TBs make all the difference, then this really is a matter of endgame ability. Maybe someone can give us evals with and without TBs to see what difference it makes.
mmt
Posts: 343
Joined: Sun Aug 25, 2019 8:33 am
Full name: .

Re: mmt Vs. Ovyron (G4 D5 BG2)

Post by mmt »

Ovyron wrote: Thu Feb 06, 2020 4:08 pm My top move choices (19...Rb8 and 19...Rb7) are really close so that's why I've been taking my time for this move. Assuming only one of them wins then this is the hardest position I've faced in the game. If I could win with both then we've reached the point where I could mostly flip a coin to decide my moves, but I have no way to know if I'm on that situation yet.
I think it's close to this point.
User avatar
Ovyron
Posts: 4558
Joined: Tue Jul 03, 2007 4:30 am

Re: mmt Vs. Ovyron (G4 D5 BG2)

Post by Ovyron »

1. g4 d5 2. Bg2 Bxg4 3. c4 c6 4. Qb3 e6 5. Qxb7 Nd7 6. Nc3 Ne7 7. cxd5 exd5 8. d4 Rb8 9. Qa6 Rb6 10. Qd3 Ng6 11. h3 Be6 12. Nf3 Bd6 13. h4 h5 14. b3 Nf6 15. Bg5 O-O 16. e3 Re8 17. Kf1 Bg4 18. Ne1 Bb4 19. Na4 Rb8

[d]1r1qr1k1/p4pp1/2p2nn1/3p2Bp/Nb1P2bP/1P1QP3/P4PB1/R3NK1R w - -