The Chess Vacuum
Sort:
Jun 10, 2016
0
#2
Your argument hinges on the idea that the bishop is better in more positions than the knight (if I'm not mistaken). I not sure about that.
Jun 10, 2016
0
#3
blastforme wrote:
Your argument hinges on the idea that the bishop is better in more positions than the knight (if I'm not mistaken). I not sure about that.
That was what I expected people would say. While it can't be proven, I defintely think it's true, because the knight is less agile and flexible. But I guess it could go either way. Actually, I belive that chess is too dynamic to assign a value to the pieces. The "points system" is kind of silly, but it's useful for beginners, so I suppose it has its place.
Jun 10, 2016
0
#4
It's definitely not something worth arguing about, but I would compare them like this:
The bishop has more range and can influence more squares at once than the knight in a given position.
The knight is harder to block because it can move in 8 "directions" vice the bishop's 4, and can go around (or over?) other pieces. But more importantly, the Knight can be positioned to attack/defend any square on the board, whereas the bishop only sees 1/2 of the squares. This is important in general, and particularly (for example) when trying to advance/promote passed pawns in an endgame.
I'm not saying the knight is better, I think the question is like comparing apples/oranges.
The bishop has more range and can influence more squares at once than the knight in a given position.
The knight is harder to block because it can move in 8 "directions" vice the bishop's 4, and can go around (or over?) other pieces. But more importantly, the Knight can be positioned to attack/defend any square on the board, whereas the bishop only sees 1/2 of the squares. This is important in general, and particularly (for example) when trying to advance/promote passed pawns in an endgame.
I'm not saying the knight is better, I think the question is like comparing apples/oranges.
A bishop cannot checkmate.
A knight cannot checkmate.
A bishop and a knight can checkmate.
A knight and a knight cannot checkmate.
Therefore
B + N > N + N
- N - N
B > N
QED
Jun 10, 2016
0
#6
An interesting idea xman, but didn't we already establish that the K vs K vacuum is no good for judging piece value?
Jun 10, 2016
0
#7
Strong players don't answer that way because it is simple and ends the conversation. "It depends on the position" is an extremely useful answer, especially to a beginner. It challenges them to think more deeply about the positions they are in. The most dangerous thing about the pieces "points" system is that it locks a beginner into not thinking fully about relative value. Which can be hard to break.
If a strong player ever answered "bishop" or "knight" the beginner might think that is doctrine for far too long. It would be a much more damaging answer than "it depends."
Please try to realize that Minor piece nuances are one of the most complex and interesting parts of the game, and even GM's struggle to examine them perfectly. Your fight to answer this simply is missing that in this case the answer just isn't simple, and shouldn't be.
If a strong player ever answered "bishop" or "knight" the beginner might think that is doctrine for far too long. It would be a much more damaging answer than "it depends."
Please try to realize that Minor piece nuances are one of the most complex and interesting parts of the game, and even GM's struggle to examine them perfectly. Your fight to answer this simply is missing that in this case the answer just isn't simple, and shouldn't be.
Jun 10, 2016
0
#8
Good points. I agree - my post only explored a tiny, tiny facet of the whole issue. It's very hard to take the 'long view' of the issue when the bigger picture is just SO big.
I've thought a lot about this.
The only real "vacuum" that applies to minor pieces would be an endgame with opposing kings plus the minor pieces in question.
As another poster properly noted, it is possible to mate with K+R, K+B+B, but NOT with K+N+N. However, it IS possible to mate with K+N+N+Tempo (if the opposing king also has a pawn he can push...).
This is a VERY minor "edge" given to the Bishop.
I would personally evaluate it as:
K = Infinite
Q = 9
R = 5
B = 3.25
N = 3
P = 1
I have personally found that the above values are pretty spot-on for real-world play.
Jun 10, 2016
0
#10
Indeed, Bawker, that's a very common system for piece evaluation. As far as I know, the most widely used one is that, but with bishops being 3 as well. I like your version better, though, because I do like bishops at the moment. I say "at the moment" because I'm still investigating the topic, and I've probably already changed my mind a hundred times.
Jun 10, 2016
0
#11
DoctorKraken42 wrote:
blastforme wrote:
Your argument hinges on the idea that the bishop is better in more positions than the knight (if I'm not mistaken). I not sure about that.
That was what I expected people would say. While it can't be proven, I defintely think it's true, because the knight is less agile and flexible. But I guess it could go either way. Actually, I belive that chess is too dynamic to assign a value to the pieces. The "points system" is kind of silly, but it's useful for beginners, so I suppose it has its place.
IIRC:
Taking 300,000 games from players rated at least 2300, Kauffman calculated an average difference of 1/50th of a pawn between bishop and knight.
---
In any case, that's a difficulty when trying to answer a beginner's question, because they don't ask useful questions. If you answer the question in the context of what will be useful for them, "it depends on the position" is one of the best (if not the best). If you try to just answer it in general with as much detail as possible, you'll probably just confuse them and prompt a subsequent non-useful question.
Jun 10, 2016
0
#12
By the way an elementary way to talk about the values of the pieces is to point out how many squares they control in the middle of an empty board and how quickly they can reach new squares.
Even in infinite moves, bishops can't reach 1/2 of the squares on the board. The knight is slower and controls fewer squares in the middle of an open board, but can eventually reach every square. Rooks control slightly more than bishops, but can go anywhere.
To add to your endgame considerations, a knight can't lose a tempo, so in some cases you can't play for zugzwang.
Jun 10, 2016
0
#13
By the way, a king is often evaluated as being between a minor piece and a rook e.g. a value of 4.
If you haven't heard that before, that might give you something to think about 
Jun 10, 2016
0
#14
Interesting. Yeah, I remember researching the king's value as an attacker before. I really don't understand how it's valued at more than a minor piece, though, considering its slowness.
Jun 10, 2016
0
#15
Good point on the slowness, hmm.
Maybe it synergizes really well. For example something as simple as attacking a moving pawn, either minor piece will struggle. When a king needs to combine with another piece for attack or defense on a complex of (local) squares, it does extremely well.
A king is way better than a minor piece.
Imagine king & pawn vs. King. It's very close and depends on oposition, right?
Now imagine king & pawn vs. knight.
King & pawn vs. bishop
The knight can't even draw and knights can get checkmated by queens.
The bishop cannot draw, it's terrible.
Unfortunatly chess.com isn't letting me make game scores where black doesn't have a king.
But trust me, if your opponent has a pawn, you want a king. A rook is also useful. As long as the pawn is pawn is not past the fourth rank and the king is behind the pawn, then a rook can draw by itself.
A queen can draw by itself but you need at least a knight to checkmate.
The saving grace of king & pawn vs. Bishop ending is that king and queen isn't enough to checkmate a bishop as far as I know, in which case all endings are drawn but for the wrong reason.
Jun 11, 2016
0
#17
King and pawn vs Knight or Bishop is a draw 99% of the time... the piece sacs itself for the pawn...
but if they got a queen, both are easy wins for the queen vs the minor piece lol...
A knight cannot stop a king and pawn. It can't even waste a move.
We are talking about whether a king or a knight is a more valuable piece. Against a king and pawn, I would rather have a king than a knight. I would take a bishop but the bishop can't stop the promotion, it just can't be checkmated by king and queen.
A knight can't sac itself for the pawn... that would be like the king saccing itself for the pawn. It doesn't make sense. The whole point is to talk about ONLY a piece.
Jun 11, 2016
0
#19
Even a king can checkmate a knight, but it can't be forced. Can a queen force mate against a knight?
As we all know, there are many, many, many, many threads on chess.com that go like this:
"Which is better, the bishop or the knight?"
"It depends."
"It depends."
"KNites!!!!!!!1!"
"It depends."
Even the strongest of players will almost invariably say this. That it depends on the position seems to be the most universally agreed upon answer to the B vs N question. And I'm not going to dispute that it is true - it clearly does depend on the position. However, I AM going to dispute the idea that it is an adequate answer to the B vs N question. Here's why.
When asked which is better, a rook or a minor piece, one can easily reply, just like one would to the BVN question, that it depends on the position. This is obvious to even the most dim-witted individual with an advanced degree in hyperbolic topology: just look at Petrosian's games, for crying out loud.
Yet, the rook is considered better.
You might say that the rook is better in more positions than the minor piece is, and you're probably right. HOWEVER, and this is key - most strong players would probably agree that there are, at the very least, ever-so-slightly more positions where the bishop is favoured than there are positions where the knight is. Thus, if you argue that the rook is generally superior to the minor based on the number of positions in which it is superior, than you must also concede that the bishop is generally slightly better than the knight. No, the rook must be better than the minor for a different reason. But what?
When the beginner asks which is better, a bishop or a knight, I'm sure they KNOW that it depends on the position. Only an idiot would not be able to figure that out after two seconds of thought. I'm sure what the asker REALLY means is, which piece is better in a vaccum?
I have concluded that the generally accepted view on the value of pieces in a vacuum is as such: rooks are better than minors in a vacuum, and the minors are equal to each other in a vacuum. Most players would probably agree with this sentiment.
But what is this vacuum?
The most pure vacuum one could think of would be a simple king vs king position. However, this does not prove to be an adequate vacuum, or at least not one that aligns with any kind of conventional wisdom relating to the value of the chessmen, as pawns now take on an extremely fluid value, and rooks and queens are worth infinitely more than minor pieces.
Without a vacuum to place the pieces in, it's very difficult to quantify their value, as you must do it ONLY on a case-by-case basis. However, either such a vacuum does not exist, or the king vs king position is the vacuum. (If somebody has an alternate suggestion, I'd love to hear it).
In the first case (there is no vacuum), then ultimately, bishops are slightly better than knights (see gray section of text). In the second, like I said, majors become infinitely better, so the vacuum is useless in quantification of piece value anyway.
Thus, no matter what you conclude, it is true that if rooks can be said to be better than minors "in general", than bishops can be said to be slightly better than knights "in general".