The August 2019 issue of Chess lists the top twenty openings compiled from a list of 2507 June games where both players were rated over 2400 Elo. One can not take position on this list too seriously because it is greatly influenced by how the openings are grouped. For example, all the Retis are grouped together, while English is separated into 1...c5, 1...e5, etc. Nevertheless, for what it is worth, some of the list entries are: 106 King's Indians, 100 Slavs, 96 declined Queen's Gambits, 83 Nimzo-Indians, 59 Queen's Indians, and 44 Classical Gruenfelds
1. e4 or 1. d4?
Sort:
RussellWestbr00k έγραψε:
d4 has large kingside attacks generally. If not, a more positional game arises. There are less good responses to 1. d4 then to 1. e4, so the games can be generalized easier with less theory needing to be memorized, so 1. d4 is good for starting out, and intermediate play. (Obviously d4 is played at top levels, It's a top level move as well as a scholastic level move). e4 has a more tactical game, with more trades and lines needing to be memorized generally. Black can sort of choose what opening/pawn structure leading into endgame which will be set up (i.e. 1. e4 e5, we'll call this opening A, 1. e4 c5 which can be called opening B, 1. e4 d6, opening C, 1. e4 d5 opening D, 1. e4 g6 opening E, the list goes on. On move one there are about ten different main lines with good strong positions for black, 6 or 7 against d4, and on move 2 there are more than 100 different positions at least where black is completely fine, as opposed to maybe 30 against d4. By move 10 with e4, there are tons of possibilities all leading to about equal positions (or at least close enough), with not as many for 1. d4) So with this being said, e4 is extremely interesting but also very complicated, while d4 is quite straightforward with more positional ideas. Bishops are e-pawn pieces*, knights are d pawn pieces. I personally play 1. e4, but I played d4 for 4 years so I figured I'd insert my input. Please give me feedback.
Good luck in your games!
*spoken in generallities
I am just curious what the heck you are using as a counter.
Almost every statement you have made is wrong, but I guess you know that, already.
All reasonable openings 1. d4 1. e4 1. Nf3 1. c4 1. e3 1. g3 and many more lead to a draw with optimum play by both sides.
ponz111 wrote:
All reasonable openings 1. d4 1. e4 1. Nf3 1. c4 1. e3 1. g3 and many more lead to a draw with optimum play by both sides.
Of course, chess is draw!
But many inexperienced player wont agreed.( I wont response/waste my time to defened this statement)
Likewise in soccer, initial touch advantage wont lead to a goal, opponent must make mistakes.
And the goal of chess is "you still need to choose an opening line to put pressure on opponent to make them mistakes."
Sep 7, 2019
0
#6
ponz111 wrote:
All reasonable openings 1. d4 1. e4 1. Nf3 1. c4 1. e3 1. g3 and many more lead to a draw with optimum play by both sides.
So you didn't believe A0's self-play that prefers c4 IIRC?
fidetrainerNET wrote:
ponz111 wrote:
All reasonable openings 1. d4 1. e4 1. Nf3 1. c4 1. e3 1. g3 and many more lead to a draw with optimum play by both sides.
So you didn't believe A0's self-play that prefers c4 IIRC?
Alpha Zero and Leela Zero prefer Catalan. This means that they have best winning statistics with Catalan. As they are trained at extremely fast games ( engines are allowed to search only 800 positions per move) to get millions of sample games to train Neural Network. So, one side makes mistakes in extremely fast time control . For example, KRB vs KR is theoretical draw but KR player will lose a few % of games in extremely fast time control. And also to let engines explore in diversity of opening, they are forced to play a few percentage of sub-optimal moves or blunders during training.
That is why, the winning percentage of white doent become 50% in initial opening. Analysis of Leela in initial position will show Catalan as 53-55% (depending on depth)
BTW, A0 is an initial NN learning prototype, Leela surpassed A0 level since 2018 Dec, and currently they are training much bigger neural network than A0 .
Sep 7, 2019
0
#8
pfren wrote:
RussellWestbr00k έγραψε:
d4 has large kingside attacks generally. If not, a more positional game arises. There are less good responses to 1. d4 then to 1. e4, so the games can be generalized easier with less theory needing to be memorized, so 1. d4 is good for starting out, and intermediate play. (Obviously d4 is played at top levels, It's a top level move as well as a scholastic level move). e4 has a more tactical game, with more trades and lines needing to be memorized generally. Black can sort of choose what opening/pawn structure leading into endgame which will be set up (i.e. 1. e4 e5, we'll call this opening A, 1. e4 c5 which can be called opening B, 1. e4 d6, opening C, 1. e4 d5 opening D, 1. e4 g6 opening E, the list goes on. On move one there are about ten different main lines with good strong positions for black, 6 or 7 against d4, and on move 2 there are more than 100 different positions at least where black is completely fine, as opposed to maybe 30 against d4. By move 10 with e4, there are tons of possibilities all leading to about equal positions (or at least close enough), with not as many for 1. d4) So with this being said, e4 is extremely interesting but also very complicated, while d4 is quite straightforward with more positional ideas. Bishops are e-pawn pieces*, knights are d pawn pieces. I personally play 1. e4, but I played d4 for 4 years so I figured I'd insert my input. Please give me feedback.
Good luck in your games!
*spoken in generallities
I am just curious what the heck you are using as a counter.
Almost every statement you have made is wrong, but I guess you know that, already.
That's TOUGH.
Prefer Nf3 (Reti). I have opened with Nf3 for all games except for four. Those 4? e4 . . .