1...e5 is better than 1...e6 and 1...c5, prove me wrong

Sort:
PedroG1464
Chessflyfisher wrote:
Ultimate-trashtalker wrote:

C5 does not allow symmetry and that's the main reason it's played

You never heard of 2 c4 transposing into the Botvinnik English?

white neglects development, allows a backward d-pawn, and adopts an abnormal, inflexible structure

bro just play the open sicilian why are people starting to throw games just to avoid it 😭😭😭😭

MARattigan
AhmedAryan wrote:
MARattigan wrote:
AhmedAryan wrote:
MARattigan wrote:
AhmedAryan wrote:
MARattigan wrote:
TheSampson wrote:
... only at absolute engine-level play with 100% accuracy ...

We haven't got any engines that play with 100% accuracy yet.

How do you know? We actually have an engine that can guarantee 100% accuracy, Stockfish. However, you need a high depth to 100% guaranteed. This also applies to any other minimax-based engine if they can calculate the game out to it's conclusion.

No it doesn't. If it tries to calculate to the conclusion of the game it loses on time before making any moves. the percentage of accurate moves is then 100 x 0/0 which is indeterminate, not 100.

I didn't account for loss of time because I thought we were talking about untimed chess.

Same thing. The machine on which it's running probably have crumbled into dust long before it makes a move. Indeed so might the whole universe.

In any case SF is designed to play FIDE competition rules - it's actually designed to not play basic rules accurately (which also have no 50 move or triple repetition rules). In fact it doesn't even play simple K+R v K positions accurately with realistic time controls unless you give it a tablebase.

The Stockfish I use accounts for thrice fold repetition and 50-move rule.

That's what I said. You can't get it to ignore those rules in working out the moves. You may be able to get the GUI to ignore the 50 move rule at least, but that doesn't affect SF's recommended moves. Accurate moves with the rules in effect don't necessarily correspond with accurate moves without the rules.

Also I really don't know how you got that it doesn't play king and rook accurately. Depth 18 is enough for king and rook and depth 18 takes a very short time to get to. For an example, Stockfish checkmated me in 8 seconds with depth 22.

I got it from playing it a few times at 40 moves in 2 hrs. repeating. It will usually mate, but not accurately.

AhmedAryan
IronSteam1 wrote:
AhmedAryan wrote:

This isn't supposed to be about endgames?

Fill in the rest of the pieces, and the outcome will still be the same.

We just aren't technologically capable of having 32-man tablebases yet (if ever).

How in the world does the outcome be the same??? I've explained the reasons why it won't work, here:

"Table base explanations don't work because they exclude the other pieces. Chess is not a 4-piece table base, there are 32 pieces. Some of these table bases require to push the pawns and promote, and I know you can't just do that in a regular game. Various other reasons like attacks on the king, excluding castling, and defended squares can also support this."

AhmedAryan
MARattigan wrote:
AhmedAryan wrote:
MARattigan wrote:
AhmedAryan wrote:
MARattigan wrote:
AhmedAryan wrote:
MARattigan wrote:
TheSampson wrote:
... only at absolute engine-level play with 100% accuracy ...

We haven't got any engines that play with 100% accuracy yet.

How do you know? We actually have an engine that can guarantee 100% accuracy, Stockfish. However, you need a high depth to 100% guaranteed. This also applies to any other minimax-based engine if they can calculate the game out to it's conclusion.

No it doesn't. If it tries to calculate to the conclusion of the game it loses on time before making any moves. the percentage of accurate moves is then 100 x 0/0 which is indeterminate, not 100.

I didn't account for loss of time because I thought we were talking about untimed chess.

Same thing. The machine on which it's running probably have crumbled into dust long before it makes a move. Indeed so might the whole universe.

In any case SF is designed to play FIDE competition rules - it's actually designed to not play basic rules accurately (which also have no 50 move or triple repetition rules). In fact it doesn't even play simple K+R v K positions accurately with realistic time controls unless you give it a tablebase.

The Stockfish I use accounts for thrice fold repetition and 50-move rule.

That's what I said. You can't get it to ignore those rules in working out the moves. You may be able to get the GUI to ignore the 50 move rule at least, but that doesn't affect SF's recommended moves. Accurate moves with the rules in effect don't necessarily correspond with accurate moves without the rules.

Also I really don't know how you got that it doesn't play king and rook accurately. Depth 18 is enough for king and rook and depth 18 takes a very short time to get to. For an example, Stockfish checkmated me in 8 seconds with depth 22.

I got it from playing it a few times at 40 moves in 2 hrs. repeating. It will usually mate, but not accurately.

Idk man, you said in the first post it avoids the rules and now you say its hard to get it to not follow those rules so Imma just ignore that until you clear it up.

What do you mean it won't mate accurately? You said it was repeating and I don't think Stockfish would do that so either your Stockfish has a problem or I don't know. How are you judging the accuracy of the mates? Are you judging by similarity to human techniques or if it took the fastest path?

JogoReal

Wrong. Pontos/Rate is seen from whites perspective. Of the main answers to 1.e4, Sicilian Defense is the one who gives whites a lower rating. (From my collection of 4.805.055 chess games in Scid vs PC software).

MARattigan
AhmedAryan wrote:
MARattigan wrote:
AhmedAryan wrote:
MARattigan wrote:
AhmedAryan wrote:
MARattigan wrote:
AhmedAryan wrote:
MARattigan wrote:
TheSampson wrote:
... only at absolute engine-level play with 100% accuracy ...

We haven't got any engines that play with 100% accuracy yet.

How do you know? We actually have an engine that can guarantee 100% accuracy, Stockfish. However, you need a high depth to 100% guaranteed. This also applies to any other minimax-based engine if they can calculate the game out to it's conclusion.

No it doesn't. If it tries to calculate to the conclusion of the game it loses on time before making any moves. the percentage of accurate moves is then 100 x 0/0 which is indeterminate, not 100.

I didn't account for loss of time because I thought we were talking about untimed chess.

Same thing. The machine on which it's running probably have crumbled into dust long before it makes a move. Indeed so might the whole universe.

In any case SF is designed to play FIDE competition rules - it's actually designed to not play basic rules accurately (which also have no 50 move or triple repetition rules). In fact it doesn't even play simple K+R v K positions accurately with realistic time controls unless you give it a tablebase.

The Stockfish I use accounts for thrice fold repetition and 50-move rule.

That's what I said. You can't get it to ignore those rules in working out the moves. You may be able to get the GUI to ignore the 50 move rule at least, but that doesn't affect SF's recommended moves. Accurate moves with the rules in effect don't necessarily correspond with accurate moves without the rules.

Also I really don't know how you got that it doesn't play king and rook accurately. Depth 18 is enough for king and rook and depth 18 takes a very short time to get to. For an example, Stockfish checkmated me in 8 seconds with depth 22.

I got it from playing it a few times at 40 moves in 2 hrs. repeating. It will usually mate, but not accurately.

Idk man, you said in the first post it avoids the rules and now you say its hard to get it to not follow those rules so Imma just ignore that until you clear it up.

What do you mean it won't mate accurately? You said it was repeating and I don't think Stockfish would do that so either your Stockfish has a problem or I don't know. How are you judging the accuracy of the mates? Are you judging by similarity to human techniques or if it took the fastest path?

Sorry if you found my first post unclear.

SF calculates incorporating both rules mentioned and doesn't have an option to turn them off. So, for example, if you give it king and rook against lone king in a position with ply count greater than 100 it will happily offer you its rook - which is not accurate play under basic rules (or even best play under competition rules).

An accurate move under either set of rules is a move that either reduces the distance to mate by one ply if the the position is theoretically won for one side or the other or, if the position is theoretically drawn results in a position that is theoretically drawn.

Those moves can be different under the two sets of rules (actually will usually be if you consider all legal positions rather than positions that occur in games, and maybe even then).

The percentage accuracy in any sequence of moves by one side can then be taken as 100 x #accurate moves/#moves.

When I said it doesn't mate accurately I meant usually its moves were not 100% accurate. (That is it didn't take the fastest path with the rook or slowest with the lone king.)

I didn't say it repeated - though SF will happily repeat whether it thinks it's losing or not (just not enough to allow a triple repetition claim if it thinks it's winning).

AhmedAryan
MARattigan wrote:
AhmedAryan wrote:
MARattigan wrote:
AhmedAryan wrote:
MARattigan wrote:
AhmedAryan wrote:
MARattigan wrote:
AhmedAryan wrote:
MARattigan wrote:
TheSampson wrote:
... only at absolute engine-level play with 100% accuracy ...

We haven't got any engines that play with 100% accuracy yet.

How do you know? We actually have an engine that can guarantee 100% accuracy, Stockfish. However, you need a high depth to 100% guaranteed. This also applies to any other minimax-based engine if they can calculate the game out to it's conclusion.

No it doesn't. If it tries to calculate to the conclusion of the game it loses on time before making any moves. the percentage of accurate moves is then 100 x 0/0 which is indeterminate, not 100.

I didn't account for loss of time because I thought we were talking about untimed chess.

Same thing. The machine on which it's running probably have crumbled into dust long before it makes a move. Indeed so might the whole universe.

In any case SF is designed to play FIDE competition rules - it's actually designed to not play basic rules accurately (which also have no 50 move or triple repetition rules). In fact it doesn't even play simple K+R v K positions accurately with realistic time controls unless you give it a tablebase.

The Stockfish I use accounts for thrice fold repetition and 50-move rule.

That's what I said. You can't get it to ignore those rules in working out the moves. You may be able to get the GUI to ignore the 50 move rule at least, but that doesn't affect SF's recommended moves. Accurate moves with the rules in effect don't necessarily correspond with accurate moves without the rules.

Also I really don't know how you got that it doesn't play king and rook accurately. Depth 18 is enough for king and rook and depth 18 takes a very short time to get to. For an example, Stockfish checkmated me in 8 seconds with depth 22.

I got it from playing it a few times at 40 moves in 2 hrs. repeating. It will usually mate, but not accurately.

Idk man, you said in the first post it avoids the rules and now you say its hard to get it to not follow those rules so Imma just ignore that until you clear it up.

What do you mean it won't mate accurately? You said it was repeating and I don't think Stockfish would do that so either your Stockfish has a problem or I don't know. How are you judging the accuracy of the mates? Are you judging by similarity to human techniques or if it took the fastest path?

Sorry if you found my first post unclear.

SF calculates incorporating both rules mentioned and doesn't have an option to turn them off. So, for example, if you give it king and rook against lone king in a position with ply count greater than 100 it will happily offer you its rook - which is not accurate play under basic rules (or even best play under competition rules).

An accurate move under either set of rules is a move that either reduces the distance to mate by one ply if the the position is theoretically won for one side or the other or, if the position is theoretically drawn results in a position that is theoretically drawn.

Those moves can be different under the two sets of rules (actually will usually be if you consider all legal positions rather than positions that occur in games, and maybe even then).

The percentage accuracy in any sequence of moves by one side can then be taken as 100 x #accurate moves/#moves.

When I said it doesn't mate accurately I meant usually its moves were not 100% accurate. (That is it didn't take the fastest path with the rook or slowest with the lone king.)

I didn't say it repeated - though SF will happily repeat whether it thinks it's losing or not (just not enough to allow a triple repetition claim if it thinks it's winning).

Alright, I think it's that we're using different Stockfishes, one will draw regardless of if it's winning, and one will win even at draw offers if it thinks it is. Since the depth is on 18 we're assuming, Rook and king take over 18 moves so it's not guaranteed 100% accuracy.

Ethan_Brollier
AhmedAryan wrote:
Ethan_Brollier wrote:
AhmedAryan wrote:

if the best move for white is e4 most likely its the same for e5

Well… the best move for White has been determined to be c4… e4 is actually the fourth best, behind d4, which is behind Nf3, which is behind c4. A0’s “primary variation” aka best chess moves for both sides is actually the Reversed Sicilian Dragon. Lc0 seems to mostly agree, but it places more stock in Nf3 than A0 did.

A0 had beaten Stockfish before it became as developed as today. Btw how do you even know what A0's recommending rn? Sure, we haven't seen A0 play against Stockfish for a while but that also means we don't know its capabilities against Stockfish either.

The A0 project was abandoned, this is all as of 2017. However Lc0 runs the same way and is stronger than A0 was, so it still holds up afaik.

Mazetoskylo
d4iscrazy wrote:
Chessflyfisher wrote:
Ultimate-trashtalker wrote:

C5 does not allow symmetry and that's the main reason it's played

You never heard of 2 c4 transposing into the Botvinnik English?

c4 does not transpose; it's a trash attempt at a maroczy bind (for reference the maroczy when u have like pawns on c4 and e4 with the bind on d5) and allows potential outposts on d4; the reason why regular maroczys aren't bad is because the d4 pawn is traded so there is no backward pawn target.

I fail to understand that "backward pawn target" thingy... the d- file is closed, isn't it.

That c4-e4-d3(d2) structure is fundamentally sound- the softness of the d4 square is compensated by other positional factors, which I'm afraid you don't understand.

This type of structure is currently very trendy, even with the c5/c4 square available for an opponent's piece. Some opening examples:

- 1.e4 c5 2.c3 e5!? is a great way to meet the Alapin, used by many top GM's.

- 1.e4 c5 2.Nf3 Nc6 3.Nc3 e5! is a rock solid equalizer against that particular move order by white.

- 1.e4 c5 2.Nf3 Nc6 3.Bb5 e5!? is currently very trendy as well against the Rossolimo - some guys like Carlsen, Caruana Nakamura, Maghsoodloo etc even played the bizarre 4.0-0 Bd6! and undestandably so: it's the computer top choice.

2.c4 is just one way out of many to play a game based on chess principles rather than variation memorization, and there's nothing wrong with it. Some strong players make a living out of it.

Serbian/Turkish GM Dragan Solak, former top-100 player is not your everyday bozo, and he has done very,very well with 2.c4 against very strong competition. And of course there's not just Solak.

AhmedAryan
Ethan_Brollier wrote:
AhmedAryan wrote:
Ethan_Brollier wrote:
AhmedAryan wrote:

if the best move for white is e4 most likely its the same for e5

Well… the best move for White has been determined to be c4… e4 is actually the fourth best, behind d4, which is behind Nf3, which is behind c4. A0’s “primary variation” aka best chess moves for both sides is actually the Reversed Sicilian Dragon. Lc0 seems to mostly agree, but it places more stock in Nf3 than A0 did.

A0 had beaten Stockfish before it became as developed as today. Btw how do you even know what A0's recommending rn? Sure, we haven't seen A0 play against Stockfish for a while but that also means we don't know its capabilities against Stockfish either.

The A0 project was abandoned, this is all as of 2017. However Lc0 runs the same way and is stronger than A0 was, so it still holds up afaik.

alr man thats a good point but theory just kinda disagrees here

1cbb

a5 is the best

MARattigan
AhmedAryan wrote:
... Since the depth is on 18 we're assuming, Rook and king take over 18 moves so it's not guaranteed 100% accuracy.

For info, if a KRvK is winning the win can always be achieved in a most 16 moves (or 32 ply). A depth of 18 reported by SF is in ply, so corresponds with 9 full moves. But that doesn't represent a full tree search to 18 ply because it doesn't search all lines to the same depth.

Mazetoskylo
Ethan_Brollier wrote:
AhmedAryan wrote:

if the best move for white is e4 most likely its the same for e5

Well… the best move for White has been determined to be c4… e4 is actually the fourth best, behind d4, which is behind Nf3, which is behind c4. A0’s “primary variation” aka best chess moves for both sides is actually the Reversed Sicilian Dragon. Lc0 seems to mostly agree, but it places more stock in Nf3 than A0 did.

Since chess is a draw (which is "almost" proven by the vast percentage of draws in modern CC) the "best move" is the one you are more familiar with.

Objectively, 1.e4 and 1.d4 and 1.c4 and 1.Nf3 have been "refuted": White cannot claim an advantage out of the opening.

Shocking, eh?

grantli

if the french defense is bad for the queen's bishop, why don't you play the caro-kann instead?

pleewo
TheSampson wrote:
Chessflyfisher wrote:
Ultimate-trashtalker wrote:

C5 does not allow symmetry and that's the main reason it's played

You never heard of 2 c4 transposing into the Botvinnik English?

white neglects development, allows a backward d-pawn, and adopts an abnormal, inflexible structure

bro just play the open sicilian why are people starting to throw games just to avoid it 😭😭😭😭

I wouldn’t exactly call the d-pawn backward 🤔

AngryPuffer
grantli wrote:

if the french defense is bad for the queen's bishop, why don't you play the caro-kann instead?

because the caro is less sound and much slower/passive for black

Talekhine09

The task was to prove you wrong and I accept it. 1. ...c5 is the best reply to 1. e4 for black. The reason is simple: in 1. e4 e5 lines white achieves what they are after: more space, better clamp on the center and more activity. Black is cramped and defending. Meanwhile, with 1. ...c5, black allows all that, but forces white to make positional concessions with the pawn structure. So while black is cramped, in this structure, black can eventually get d5! in, breaking free and counterattacking white. Also, black gets a monster c-file. Of course, white can avoid all this by not playing an open sicilian, but after any anti-sicilian except the Rossolimo, black has achieved equality, and you can avoid the Rossolimo by avoiding 2. ...Nc6. In any sicilian, black can counterattack and put pressure on white. In e4 e5 positions, most sharp, counterattacking lines are gambits or do not work (or both). Additionally, in e4 e5 endgames, white is often better due to the positional concessions black had to make in the opening, while with the sicilian, if white can't get a good positional hold in the middlegame, black has better pawn structure in the endgame. However, this is mostly a moot point for anyone below 2000 elo, and here it is just a matter of preference.

Talekhine09
d4iscrazy wrote:

idk where you all get that sicilian is cramped just stop playing d6 e6 and it's not cramped

In the main lines, black has almost no presence in the center, but this is counteracted by the c-file and the dynamic potential with d5. However, there are exceptions.

Talekhine09

Well in that line you are right, however after some more moves in that line, white has way more space, also 6. Bg5 is slightly more common, and in that line, black plays e6 and is very cramped

AngryPuffer

Bg5 is the more testing variation

Be3 is the more sound/calmer variation