This game may well break me. The AI is so bad. I think I may be getting worse at chess by playing it. It really does have to be seen to be believed. I am in the unusual position of being bad at chess, but also moderately well-read on chess theory. So when I make a bad move, I know it almost immediately, and I can usually understand why it's a bad move (and yet somehow I can never translate this knowledge to effective strategy - if only there were some way to train my foresight to be as perceptive as my hindsight). And yet for all of that, the Chess 2 AI makes mistakes that stagger even me. I'll make an impulsive move, grabbing some flashy, but unnecessary capture, and after the fact I'll see clear as day that I left an opening for my opponent to win in 2-3 moves, but then the AI will do something completely baffling, like retreat their king, or move a rook into a threatened square when it could move one square more, capture one my pieces, and wind up unthreatened.
I'm beginning to suspect that the AI moves pieces more or less at random. At the very least, it doesn't understand threatened or protected squares. You can't play chess against opponent who lacks that knowledge. I'm not sure what I've been doing for the last 10 hours, but it hasn't been chess.
The culprit is probably the "duel" rule. It sucks the life completely out of the game and I don't think the AI can handle it. How it works is that you have a number of stones - up to 6, but you start with 3 and gain more by capturing enemy pawns. Whenever the opponent captures one of your pieces, you have the option of declaring a duel. When you do, you and your opponent secretly bid 0, 1, or 2 stones. Then you reveal your bids simultaneously and if the defending player bid more stones than the attack, then the attacking piece is removed from the board (the defending piece is always captured, regardless of the outcome of the bids). There are some additional nuances, but that's the gist of it.
It's my working theory that when the AI calculates its next move, if you have stones and the AI doesn't (easy enough when you aggressively hunt pawns), it counts capturing a piece as sacrificing one of its own. It's gotten so that I'll deliberately move into threatened squares in order to set up a capture 2 moves down the line. And while this has occasionally backfired on me, it has proven safe more often than not. I already have a problem with overlooking my opponent's defense. This is only going to make my habit worse.
The other major rules change is the midline invasion alternate victory condition. It's difficult to say exactly how I feel about it, because I've not yet faced an opponent who has used it well. I've lost a couple of times, thanks to it, but that was before I learned to defend against it. It's clear in retrospect that those losses were pure chance. I've seen the AI, on multiple occasions, pass up the opportunity to clear or obstruct my defense and then win on the following move. It certainly doesn't have the wherewithal to make a concerted attack on a hardened position.
But how would I imagine it might work, were I to face a competent adversary? That's where it gets hard to pin down. To win with a midline invasion, you have to move your king across the halfway point on the board, to any space in row 5. The one caveat is that you cannot move your king into check. If a square on row 5 is threatened, you have to remove the obstruction before you can move into it.
My experience so far has been that mainline invasion is a perfunctory victory condition, one you pursue when the enemy is substantially broken, and it's easier to advance your king 4 times in a row than maneuver into a checkmate. But if I were playing against someone who could attack effectively, who would capture as many of my pieces as I did theirs, I can see how it might change the endgame. The question is "into what?"
The endgame in chess is, ideally, a battle of wits. An attempt to use the rules of the game to maneuver your opponent into a losing position. And while victory of one side or the other is often a foregone conclusion, a canny player can play defensively and force a draw.
Chess 2 removes stalemates from the equation. If you have no legal move, you lose. A losing player can theoretically win by moving their king across mid line, but it's hard to see how they could force it. What's more likely is that the leading player will advance their king as an alternative to seeking a checkmate. This is faster and more efficient than the traditional endgame, but vitiates the puzzle aspect of traditional chess.
I'd have to see to see what it looks like when two good players go up against each other. I know that when it's a bad player versus a terrible player midline invasion tends to make matches short and anticlimactic, but it's possible that two properly matched players could turn it into a tense battle of wits.
It's nice to think that's a possibility. From my position, Chess 2: The Sequel is just a series of pointless 7-minute exercises in navigating the geometry of a chessboard. I'm not being challenged at all and while I normally appreciate easy wins in games, this is different. It feels like I'm disrespecting the very game of chess itself.
No comments:
Post a Comment