What to play against 1. d4 ?

Discussion of anything and everything relating to chess playing software and machines.

Moderator: Ras

Henk
Posts: 7261
Joined: Mon May 27, 2013 10:31 am

Re: What to play against 1. d4 ?

Post by Henk »

Reti played e4, d4, c4, Nf3. Sometimes you trust mathematics.
corres
Posts: 3657
Joined: Wed Nov 18, 2015 11:41 am
Location: hungary

Re: Poor Stockfish

Post by corres »

[quote="Lyudmil Tsvetkov"]
[quote="corres"]
For your shake I looked through my databases and I have found 3120 white games of Kasparov from which 1123(!) start 1.d4
[/quote]

then 2 times as much start with e4 and c4. :)
[/quote]

White games of Kasparov with 1.e4 are 1597 but with 1.c4 226(!) only.
I love facts you love dreams, I think.
Henk
Posts: 7261
Joined: Mon May 27, 2013 10:31 am

Re: What to play against 1. d4 ?

Post by Henk »

Lyudmil Tsvetkov wrote:
Henk wrote:Capablanca played d4.
that is why Capablanca lost to Alekhine, who in turn, even if on par with Botvinnik, was weaker than Fischer and Kasparov.

Really, looking at Capablanca games, they are so simple, nothing special, later players play much more refined chess, in general, with each new generation players get stronger and stronger, until Carlsen best players were Fischer and Kasparov, we will see what Carlsen will play when his repertoire matures in a decade or so.
Not so sure if Aljechin at his best was weaker than Fischer. Maybe he drank too much.

Maybe best (human) play is playing simple moves.
Lyudmil Tsvetkov
Posts: 6052
Joined: Tue Jun 12, 2012 12:41 pm

Re: Poor Stockfish

Post by Lyudmil Tsvetkov »

on above position, white gets considerable advantage after b6 d3 Bb7 f4, or b6 d3 Na6 (Re8) Nf3.
Lyudmil Tsvetkov
Posts: 6052
Joined: Tue Jun 12, 2012 12:41 pm

Re: Poor Stockfish

Post by Lyudmil Tsvetkov »

corres wrote:
Lyudmil Tsvetkov wrote:
corres wrote: For your shake I looked through my databases and I have found 3120 white games of Kasparov from which 1123(!) start 1.d4
then 2 times as much start with e4 and c4. :)
White games of Kasparov with 1.e4 are 1597 but with 1.c4 226(!) only.
I love facts you love dreams, I think.
1597 (e4) > 1123 (d4);
1597 (e4) + 226 (c4) ~= 2*1123 (d4);

For Fischer, the numbers are 600 (e4) = 200*3 (d4);

Kasparov and Fischer belong to the old generation, the players who will demonstrate c4 is even better than e4 still have to come along :)

and yes, I love dreams, life without dreams is like popcorn without salt.
PK
Posts: 913
Joined: Mon Jan 15, 2007 11:23 am
Location: Warsza

Re: Poor Stockfish

Post by PK »

Frankly, I don't know how the position would play out. With that many pawns on the board, I'm not too concerned with the bishop pair. 7.Qb3 looks interesting, but a bit provocative. I'd reply 7...d6.

Later in some lines Black can happily play ...Be6, sacrificing b7 pawn - it certainly works against 8.Ne2?!. In other cases Qb3 and Ba3 might justify playing c6-c5, rendering white queenside pieces awkwardly placed. In this case, if I'm not mistaken, Black manages to pull off sacrificial ...b5 before White gets to play f4 or perhaps immediately after f4. I envisage something like:

1. c4 e5 2. Nc3 Bb4 3. g3 Nf6 4. Bg2 O-O 5. e4 Bxc3 6. bxc3 c6 7. Qb3 d6 8.
Ba3 c5 9. d3 Nc6 10. Ne2 a6 11. O-O b5 *
Lion
Posts: 539
Joined: Fri Mar 31, 2006 1:26 pm
Location: Switzerland

Re: Poor Stockfish

Post by Lion »

Lyudmil Tsvetkov wrote:
Lion wrote:
Lyudmil Tsvetkov wrote:
Lion wrote:Hello,

Are you kidding?

1.c4 e5 is a complete = for black.

Please provide a little deeper lines as to how white can gain an advantage?

rgds
Why should I be kidding?

1.c4 is simply the best move.
Until now I have sometimes claimed that 1.e4 is the best move, simply because people think either 1.e4 or 1.d4 is best, and a very big scandal will arise, when I claim 1.c4 is actually best. Better avoid that.

I can not of course go through all the possible lines, but, just a casual game with new SF 8 until I prove my case.

[pgn][Event "Blitz 2m+1s"]
[Site "Microsoft"]
[Date "2016.11.03"]
[Round "?"]
[White "Lyudmil Tsvetkov, owner"]
[Black "Stockfish 8 64 POPCNT"]
[Result "*"]
[ECO "A25"]
[Annotator "owner"]
[PlyCount "36"]
[EventDate "2016.??.??"]
[TimeControl "120+1"]

{512MB, OWNER-PC} 1. c4 {0} e5 {0.07/22 5} 2. Nc3 {2} Bb4 {0.05/22 8} 3. g3 {9}
Nf6 {-0.06/20 5} 4. Bg2 {4} O-O {0.06/22 1} 5. e4 {3} Nc6 {-0.12/19 2} 6. Nge2
{2} d6 {-0.11/22 1} 7. h3 {6} Bc5 {-0.09/20 5} 8. O-O {4} Be6 {-0.11/23 22} 9.
d3 {2} a6 {-0.07/22 2} 10. Kh2 {18} Nd4 {-0.08/18 4} 11. f4 {5} b5 {-0.09/20 10
} 12. f5 {14} Bd7 {-0.19/20 2} 13. b3 {10} c6 {-0.28/19 2} 14. g4 {1} h6 {-0.
25/20 5} 15. Bd2 {24} Nxe2 {-0.20/20 8} 16. Qxe2 {12} Bd4 {-0.20/18 2} 17. Rad1
{19} b4 {-0.07/20 2} 18. Na4 {12} c5 {0.00/28 0} *

[/pgn]

I was looking at eval while the game was played, and stopped the game as soon as eval started largely favouring white, with this I have proven my case, no interest in winning the game.

As you see from eval, up until move 18...c5, SF eval is either 0.0, or, more often, favours black by a sligth margin. On move 19, however, I saw SF show +50cps white advantage from the 0.0 score it showed the previous move.

[d]r2q1rk1/3b1pp1/p2p1n1p/2p1pP2/NpPbP1P1/1P1P3P/P2BQ1BK/3R1R2 w - - 0 19

that is the end position, if you want, analyse it, but most probably white is winning.

White has been winning throughout, it is only that current top engines have very primitive eval, and search without good eval in complex middlegame positions means nothing.
Well, the best order move IMHO for black to = is to play Nf6 instead of Bb4.
So how does white gain an advantage after 2... Nf6 ?
You are not reading this thread.
Mainline is 1.c4 e5 2.Nc3 Nf6 3.e4
OK, but still, after 3.e4 Bc5 and what next for white? g3 is not immediately possible... I would even say that 3.e4 is ?!
Lyudmil Tsvetkov
Posts: 6052
Joined: Tue Jun 12, 2012 12:41 pm

Re: Poor Stockfish

Post by Lyudmil Tsvetkov »

Carlsen record:


With the White pieces:
Sicilian (232)
B90 B30 B40 B33 B51
Ruy Lopez (150)
C65 C78 C67 C80 C77
Slav (69)
D15 D17 D12 D10 D11
Nimzo Indian (64)
E32 E21 E20 E54 E36
Queen's Pawn Game (61)
A45 E10 D02 A40 A46
French Defense (57)
C00 C11 C18 C02 C03

e4 (Sicilian+Ruy Lopez+French) = 439 games
d4 (Slav + Nimzo-Indian+Queen's pawn) = 194

ratio >2 in favour of e4


With the Black pieces:
Sicilian (232)
B33 B30 B22 B31 B90
Ruy Lopez (148)
C67 C95 C65 C78 C69
Queen's Indian (88)
E15 E12 E17 E16 E14
Nimzo Indian (58)
E32 E34 E21 E20 E55
Queen's Gambit Declined (57)
D37 D38 D31 D30 D36
Queen's Pawn Game (55)
A45 A46 E00 E10 A40

on e4 232 Sicilians (c5) vs 148 (Ruy Lopez)
ratio 3/2 in favour of 1...c5

Not a single French! (what SF would consider best)

So, Carlsen joins the club of Kasparov and Fischer too, what concerns if e4 is better than d4.

If the best 3 players of all time say e4 is better than d4, maybe it is.
And if the best 3 players of all time say after 1.e4 1...c5 is best, maybe it is.
Henk
Posts: 7261
Joined: Mon May 27, 2013 10:31 am

Re: Poor Stockfish

Post by Henk »

Steinitz said d4 was better than e4. No one knows. After 100 years they all play f4 or b3 who knows.
Lyudmil Tsvetkov
Posts: 6052
Joined: Tue Jun 12, 2012 12:41 pm

Re: Poor Stockfish

Post by Lyudmil Tsvetkov »

Lion wrote:
Lyudmil Tsvetkov wrote:
Lion wrote:
Lyudmil Tsvetkov wrote:
Lion wrote:Hello,

Are you kidding?

1.c4 e5 is a complete = for black.

Please provide a little deeper lines as to how white can gain an advantage?

rgds
Why should I be kidding?

1.c4 is simply the best move.
Until now I have sometimes claimed that 1.e4 is the best move, simply because people think either 1.e4 or 1.d4 is best, and a very big scandal will arise, when I claim 1.c4 is actually best. Better avoid that.

I can not of course go through all the possible lines, but, just a casual game with new SF 8 until I prove my case.

[pgn][Event "Blitz 2m+1s"]
[Site "Microsoft"]
[Date "2016.11.03"]
[Round "?"]
[White "Lyudmil Tsvetkov, owner"]
[Black "Stockfish 8 64 POPCNT"]
[Result "*"]
[ECO "A25"]
[Annotator "owner"]
[PlyCount "36"]
[EventDate "2016.??.??"]
[TimeControl "120+1"]

{512MB, OWNER-PC} 1. c4 {0} e5 {0.07/22 5} 2. Nc3 {2} Bb4 {0.05/22 8} 3. g3 {9}
Nf6 {-0.06/20 5} 4. Bg2 {4} O-O {0.06/22 1} 5. e4 {3} Nc6 {-0.12/19 2} 6. Nge2
{2} d6 {-0.11/22 1} 7. h3 {6} Bc5 {-0.09/20 5} 8. O-O {4} Be6 {-0.11/23 22} 9.
d3 {2} a6 {-0.07/22 2} 10. Kh2 {18} Nd4 {-0.08/18 4} 11. f4 {5} b5 {-0.09/20 10
} 12. f5 {14} Bd7 {-0.19/20 2} 13. b3 {10} c6 {-0.28/19 2} 14. g4 {1} h6 {-0.
25/20 5} 15. Bd2 {24} Nxe2 {-0.20/20 8} 16. Qxe2 {12} Bd4 {-0.20/18 2} 17. Rad1
{19} b4 {-0.07/20 2} 18. Na4 {12} c5 {0.00/28 0} *

[/pgn]

I was looking at eval while the game was played, and stopped the game as soon as eval started largely favouring white, with this I have proven my case, no interest in winning the game.

As you see from eval, up until move 18...c5, SF eval is either 0.0, or, more often, favours black by a sligth margin. On move 19, however, I saw SF show +50cps white advantage from the 0.0 score it showed the previous move.

[d]r2q1rk1/3b1pp1/p2p1n1p/2p1pP2/NpPbP1P1/1P1P3P/P2BQ1BK/3R1R2 w - - 0 19

that is the end position, if you want, analyse it, but most probably white is winning.

White has been winning throughout, it is only that current top engines have very primitive eval, and search without good eval in complex middlegame positions means nothing.
Well, the best order move IMHO for black to = is to play Nf6 instead of Bb4.
So how does white gain an advantage after 2... Nf6 ?
You are not reading this thread.
Mainline is 1.c4 e5 2.Nc3 Nf6 3.e4
OK, but still, after 3.e4 Bc5 and what next for white? g3 is not immediately possible... I would even say that 3.e4 is ?!
why not. precisely 4.g3