1: e4 d5
2: exd5 Qxd5
3: Nc3 Nc6
Does white have a wonderful game here? What I mean by that us with perfect play from both sides, can white force a mate?
Unless I’m misunderstanding the question, White is a Queen ahead, clear, which should be unassailable even for average play by both players. White should easily be able to trade down pieces until it has a Queen against Black’s empty board.
How is white a queen ahead? Both sides have lost a pawn, and black’s queen is alive and well on d5, unless I’m misunderstanding something. (I’m not a chess player.)
Black’s queen hangs with white to play.
In answer to the OP, there is a practical answer and a technical answer.
Technically, no we don’t know. Chess isn’t a solved game and hypothetically black may have a forced draw in 60+ moves or even a win; we can’t calculate all lines that far.
In real terms being a queen down is like trying to win a car race with 4 flat tires.
I suspect the OP meant to ask :
"
1: e4 d5
2: exd5 Qxd5
3: Nc3
Does white have a wonderful game here? What I mean by that us with perfect play from both sides, can white force a mate?
"
In which case, White will have a fun time developing their minor pieces while harassing Black’s Queen. White should achieve a significant lead in development. Is it an already won game? Not by a million miles, but 2. … , Qxd5 looks very questionable indeed.
Agree about Qxd5 but don’t think queen gets bounced around with correct play. Black should just send the queen back (it even rhymes).
And accepto the loss of a tempo.
…Qa5 is a more popular move than …Qd8 these days (there is also …Qd6), and if you simply count tempos, Black his not behind because White developed Black’s Queen with his 2. exd5. It has never been super popular at the grandmaster level and little more at lower levels, but Anand risked it against Kasparov in a world championship match (losing, though not because of the opening–though that is a tricky game to play), and Bent Larsen once beat Karpov with it at the height of Karpov’s success.
In short, “wonderful” seems too strong an adjective–perhaps “pleasant”.
While I don’t think anybody’s ever done it, “white up a queen” just might be exhaustively searchable to a solved state, given realistic resources. Remember, you don’t need to explore all of the winning side’s paths to prove a win, only all of the losing side’s, and with that kind of advantage, the obvious move for white is almost always going to be good enough. Even searching through all of black’s possible responses won’t be as bad as it might seem, because with that kind of firepower, white is going to be gobbling black’s pieces very quickly, leaving black with few moves available.
As already pointed out, white is a queen ahead after the next move, so, yes, a wonderful situation for white. Even with this tremendous advantage, there is no forced mate - and an average player might even lose against a grandmaster playing black from this position.
The position after 3 Nc3 is well known. Black is playing the so-called Scandinavian defence, which is considered playable but a notch below other replies to 1. e4.
If that final move by black is correct, then the knight on c3 takes the queen. I cannot imagine a case in which, with perfect play by both sides, white does not win from that position.
But how do you know that? The amount of possible moves left must be more than 10 to the 150th power. With all of these possible combinations, how do you know that black can’t force a draw. Or even a win. Maybe black played Nc3 because it’s a poison queen and after Nxd5?? white loses by force. If there is a forced win for white, I’m quite sure the best computer in the world couldn’t find it.
I never said I knew that. I said “I cannot imagine a case” where white doesn’t win with perfect play by both sides. I also imagine I am right.
While not proving anything, I’d be kind of curious to see with black set at the highest ELO rating on any given computer program (let’s say Komodo or Fritz), how far down you’d have to turn down white’s “difficulty” setting/ELO rating before you can consistently get a loss or draw.
My guess would be that a difference of around 400 in favour of black would be more than enough to equalise things. I’m basing this on my own experience of a couple of hundred games against a player 400-500 points above me. Played me blind, played me with 5 mins on his clock (me untimed), spotted me a rook, etc, and all combinations thereof – I think I managed to win once and get a couple of draws.
But in your original post, you said that there was ‘perfect’ play by both white and black, which means that white will win, based on their material advantage. In fact not only has black not opened up some potential shock mating trap, but they have to follow up their Nc6 blunder immediately with a defensive or protective move, or else they face Nxc7+ followed by losing their rook as well, so they’re not only behind in material, they’re behind in development as well.
Chess is of course a game with countless numbers of possible games from the beginning, but white - a queen up - will have vastly more winning options each move, and can force exchanges that will become more and more weighted in their favour until black is just a lonely king on the board waiting to be mated.
It’s hard to see how one can argue that one has achieved such an overwhelming positional advantage in the early stages of this opening that it justifies a queen sacrifice!
Out of curiosity I plugged the position into a very capable chess program, only going down to white’s Nc3 and omitting black’s Nc6, wanting to see what the computer did with black. I was expecting it to retreat and withdraw the queen back into the ranks, but instead, it moved the queen to a5. But it certainly didn’t allow the queen to be captured. If I’d had the time and patience I could have made the computer play itself from black’s Nc6, but I don’t think there’s much doubt about who would win.
It would probably depend on how the difficulty settings were arranged. I used to have an electronic chess set where the lower difficulty settings worked by making occasional deliberate mistakes. Once, on the setting one higher than what I could consistently beat, I actually managed to get a slightly-embellished Fool’s Mate against the computer (six moves instead of two, but the others were irrelevant). Allow the weaker computer to make a mistake bad enough, and the stronger can in principle recover from anything.
To me, it doesn’t really make that much of a difference of how the difficulty settings are established, and it would be across a range of various chess programs. I would think to lose or draw being a queen up would take one or two downright beginner-level blunders, and that type of play wouldn’t happen in anything but the lowest levels where the computer is really just picking a move at random as opposed to, say, the fifth or sixth best move down the tree.
I think you’re missing the point of the distinction between practically and technically true here, which I did try to make in my first post.
If I were to blunder like black here I would immediately resign. And I would bet my life savings that white will win with correct play from both sides.
However, that’s not the same thing as saying we know it’s a forced win. Hypothetically, mathematically, there could be a 50-move combination that starts with giving up the queen and ends with black winning. There are too many pieces on the board for us to absolutely rule that out.
It’s very bad for black (i.e. worse than being a queen down) as after the obvious Nxd5 the best move for black is to move the king into the half-open d file to avoid the fork.