Move 22 discussion: Everyone vs stockfish

MooMoo2 · 2016-02-20, 02:31

The poll for move 21 just closed, and the most popular move was Nd4. Stockfish responded with 20...Re4, and the FEN is:
r5k1/1p3p1p/1p3p2/3p4/2nNrP2/8/PPP3PP/R2R2K1 w - - 3 22

Game so far: 1.e4 e6 2. d4 d5 3. Nd2 c5 4. Ngf3 Nf6 5. exd5 exd5 6. Bb5+ Bd7 7. Bxd7+ Nbxd7 8. O-O Be7 9. dxc5 Nxc5 10. Nb3 Nce4 11. Nfd4 O-O 12. Nf5 Re8 13. Nxe7+ Rxe7 14. f3 Nd6 15. Bg5 Nc4 16. Bxf6 gxf6 17. Qd4 Qb6 18. Qxb6 axb6 19. Rfd1 Re5 20. f4 Re2 21. Nd4 Re4

gd_barnes · 2016-02-20, 08:55

Everything has proceeded according to plan so I'd like to advocate for 22. Nf5.

See this analysis:
http://www.mersenneforum.org/showpos...8&postcount=14

The line from that posting is:
22. Nf5 Nxb2
23. Rxd5 Rxf4
24. g3 Rc4

Continuing maybe we have:
25. Rd7 Rxc2
26. Rxb7

A quote from that posting:
"Here we can chase the black king around all day and/or put single or double threats on his b7 and f7 pawns and there is nothing that he can do. Eventual moves like Nh6+, Nf5+, and Rd7. One key is that in several lines here we can put a double attack on the f7 pawn threatening to potentially pick off 3 pawns in 3 moves."

In other words, we have an easy draw from here.

If we play 22. b3, the lines become more murky and so I feel that we have to play much more exactly. See the analysis for that here:
http://www.mersenneforum.org/showpos...8&postcount=18

I think it's better for us against SF to play the more forced lines resulting from 22. Nf5.

WMHalsdorf · 2016-02-20, 15:34

Quote:

Originally Posted by gd_barnes

Everything has proceeded according to plan so I'd like to advocate for 22. Nf5.

See this analysis:
http://www.mersenneforum.org/showpos...8&postcount=14

The line from that posting is:
22. Nf5 Nxb2
23. Rxd5 Rxf4
24. g3 Rc4

Continuing maybe we have:
25. Rd7 Rxc2
26. Rxb7

A quote from that posting:
"Here we can chase the black king around all day and/or put single or double threats on his b7 and f7 pawns and there is nothing that he can do. Eventual moves like Nh6+, Nf5+, and Rd7. One key is that in several lines here we can put a double attack on the f7 pawn threatening to potentially pick off 3 pawns in 3 moves."

In other words, we have an easy draw from here.

If we play 22. b3, the lines become more murky and so I feel that we have to play much more exactly. See the analysis for that here:
http://www.mersenneforum.org/showpos...8&postcount=18

I think it's better for us against SF to play the more forced lines resulting from 22. Nf5.

I would play 25.Rd6 which could lead to freeing up our rook on a1.

LaurV · 2016-02-20, 16:05

All those moves look drawish. They may be better, but I feel ok with b3, as it was the one I looked into it.

gd_barnes · 2016-02-21, 09:16

Quote:

Originally Posted by WMHalsdorf

I would play 25.Rd6 which could lead to freeing up our rook on a1.

Yes, that might be (probably is) better. Perhaps black responds with 25...b5 or Rxc2 or Na4. We may even be able to fairly quickly gain our pawn back with a decent position.

It looks live it will be an interesting vote between 22. Nf5 and b3 here. I'm OK with either but prefer Nf5. I think we may have to work a bit harder with 22. b3 but perhaps not. The lines become more numerous with the latter move so it's hard to go through so many.

richs · 2016-02-21, 15:00

I prefer 22. Nf5 per Mr. Barnes' previous analysis.

LaurV · 2016-02-22, 01:34

I will be lost somewhere in China this week (biz trip). Assuming I can't get to a net pipe, please consider I voted for b3. No time for analysis, and I will not take a board with me.

Brian-E · 2016-02-23, 21:03

I've been absent for a while, but looking at the posted analysis now I agree with 22.Nf5. If it really is as clearcut as it seems (and I cannot fault Gary's analysis) I find it amazing that we've achieved such a safe looking position against a program which is far stronger than any human player.

gd_barnes · 2016-02-25, 10:09

Quote:

Originally Posted by Brian-E

I've been absent for a while, but looking at the posted analysis now I agree with 22.Nf5. If it really is as clearcut as it seems (and I cannot fault Gary's analysis) I find it amazing that we've achieved such a safe looking position against a program which is far stronger than any human player.

The combined efforts of several humans looking over positions for hours at a time over several days is frequently enough to overcome the strongest program looking over the position for a mere minutes per move. This is especially true in a more "forced" line like we have intentionally been playing. Have you noticed how many moves in this game we have had to think very little about? The key is to never let up on any move and don't miss any good lines 4-5 moves out, which we have each other to avoid. We can also think very long term strategically unlike a computer. We aimed for a draw as white and we have been relentless in our pursuit even at the expense of a pawn.

As black, I think we stand a reasonable chance (10-20%) to defeat SF using other programs and our own combined efforts and strategical thinking. I am very much looking forward to that game. Of course a draw is the most likely result. I seriously doubt that we will lose. SF is not that much better than some of the other excellent programs out there. We will easily overcome that difference with our own combined thinking and analysis efforts and I hope that we can actually surpass it.

Interesting times lie ahead!

Brian-E · 2016-02-25, 10:37

Don't count the chickens yet, Gary. I fully expect us to lose this game, although I am gratified that we are apparently holding our own at the moment. I am waiting to see what we have missed.

The "combined effort" effect which you mention might help a bit, but I have the impression that this is limited and can in some circumstances work adversely (though that adverse effect has not shown itself in this game as far as I can see). The opinion of the analytically strongest player or handful of players (and I'm defintely looking at you now) tends to prevail in the choice of move.

I agree that with software help in the next game it will be another matter. I also have the feeling that we will not really be playing that game ourselves, so I am rather less enthusiastic about it. That is not intended to dampen anyone else's enthusiasm though.

axn · 2016-02-25, 11:10

Quote:

Originally Posted by Brian-E

I agree that with software help in the next game it will be another matter. I also have the feeling that we will not really be playing that game ourselves, so I am rather less enthusiastic about it. That is not intended to dampen anyone else's enthusiasm though.

Perhaps this piece will help to put in context what human(s) can bring to the table.

2016-02-20, 02:31	#1
MooMoo2 "Michael Kwok" Mar 2006 2²×5×59 Posts	Move 22 discussion: Everyone vs stockfish The poll for move 21 just closed, and the most popular move was Nd4. Stockfish responded with 20...Re4, and the FEN is: r5k1/1p3p1p/1p3p2/3p4/2nNrP2/8/PPP3PP/R2R2K1 w - - 3 22 Game so far: 1.e4 e6 2. d4 d5 3. Nd2 c5 4. Ngf3 Nf6 5. exd5 exd5 6. Bb5+ Bd7 7. Bxd7+ Nbxd7 8. O-O Be7 9. dxc5 Nxc5 10. Nb3 Nce4 11. Nfd4 O-O 12. Nf5 Re8 13. Nxe7+ Rxe7 14. f3 Nd6 15. Bg5 Nc4 16. Bxf6 gxf6 17. Qd4 Qb6 18. Qxb6 axb6 19. Rfd1 Re5 20. f4 Re2 21. Nd4 Re4 Attached Thumbnails

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Stockfish-assisted game vs Stockfish, move 8 discussion	MooMoo2	Other Chess Games	11	2016-10-21 15:47
Move 33 discussion: Everyone vs stockfish	MooMoo2	Other Chess Games	8	2016-04-23 14:07
Move 30 discussion: Everyone vs stockfish	MooMoo2	Other Chess Games	2	2016-04-03 19:55
Move 29 discussion: Everyone vs stockfish	MooMoo2	Other Chess Games	8	2016-03-29 10:27
Move 5 discussion: Everyone vs stockfish	MooMoo2	Other Chess Games	5	2015-11-13 13:39

2016-02-20, 08:55	#2
gd_barnes "Gary" May 2007 Overland Park, KS 30007₈ Posts	Everything has proceeded according to plan so I'd like to advocate for 22. Nf5. See this analysis: http://www.mersenneforum.org/showpos...8&postcount=14 The line from that posting is: 22. Nf5 Nxb2 23. Rxd5 Rxf4 24. g3 Rc4 Continuing maybe we have: 25. Rd7 Rxc2 26. Rxb7 A quote from that posting: "Here we can chase the black king around all day and/or put single or double threats on his b7 and f7 pawns and there is nothing that he can do. Eventual moves like Nh6+, Nf5+, and Rd7. One key is that in several lines here we can put a double attack on the f7 pawn threatening to potentially pick off 3 pawns in 3 moves." In other words, we have an easy draw from here. If we play 22. b3, the lines become more murky and so I feel that we have to play much more exactly. See the analysis for that here: http://www.mersenneforum.org/showpos...8&postcount=18 I think it's better for us against SF to play the more forced lines resulting from 22. Nf5.

2016-02-20, 16:05	#4
LaurV Romulan Interpreter "name field" Jun 2011 Thailand 41·251 Posts	All those moves look drawish. They may be better, but I feel ok with b3, as it was the one I looked into it.

2016-02-21, 15:00	#6
richs "Rich" Aug 2002 Benicia, California 1718₁₀ Posts	I prefer 22. Nf5 per Mr. Barnes' previous analysis.

2016-02-22, 01:34	#7
LaurV Romulan Interpreter "name field" Jun 2011 Thailand 2833₁₆ Posts	I will be lost somewhere in China this week (biz trip). Assuming I can't get to a net pipe, please consider I voted for b3. No time for analysis, and I will not take a board with me.

2016-02-23, 21:03	#8
Brian-E "Brian" Jul 2007 The Netherlands 110011001110₂ Posts	I've been absent for a while, but looking at the posted analysis now I agree with 22.Nf5. If it really is as clearcut as it seems (and I cannot fault Gary's analysis) I find it amazing that we've achieved such a safe looking position against a program which is far stronger than any human player.

2016-02-25, 10:37	#10
Brian-E "Brian" Jul 2007 The Netherlands 2×11×149 Posts	Don't count the chickens yet, Gary. I fully expect us to lose this game, although I am gratified that we are apparently holding our own at the moment. I am waiting to see what we have missed. The "combined effort" effect which you mention might help a bit, but I have the impression that this is limited and can in some circumstances work adversely (though that adverse effect has not shown itself in this game as far as I can see). The opinion of the analytically strongest player or handful of players (and I'm defintely looking at you now) tends to prevail in the choice of move. I agree that with software help in the next game it will be another matter. I also have the feeling that we will not really be playing that game ourselves, so I am rather less enthusiastic about it. That is not intended to dampen anyone else's enthusiasm though.