Chess RU

Part I · First steps Chapter 2 of 40

Rook, bishop, queen

Three pieces that travel along the lines of the board as far as they like. We build the queen from a rook and a bishop, count the squares each one attacks, and see where the values 5, 3 and 9 come from.

Novice 35 minutes
  • rules
  • pieces

What to know first: 1 · The board and the pieces

What you will learn

  • move and capture with the rook, bishop and queen, reading each line up to the first blocker
  • explain why the queen is worth nine pawns, the rook five and the bishop three
  • spot long-range captures and set two pieces up as a battery

A rook on a1 can take a piece on a8 in one move, straight across the board. No king, knight or pawn can do that. In Paris in 1858, White's last five moves in a famous game were all made by pieces with the same long reach, two rooks, a bishop and a queen, and the last of them was checkmate. We come to that game at the end of the chapter.

In the first chapter we marked out the board: eight files, eight ranks and the slanting diagonals. So far they were lines on a map. Now the pieces travel along them. The rook, bishop and queen follow one rule: pick a line and go along it for as many squares as you like, until you run into another piece. That is why they are called line pieces. The king, knight and pawn work differently, and the next chapters are about them.

We will take the rook and the bishop one at a time, then add them together to get the queen. The addition is literal: the queen moves exactly as a rook and a bishop would if you put them on the same square. Along the way we will count how many squares each piece attacks, and the count will almost give us their values by itself.

The rook: in a straight line

The rook moves along a file or a rank: forward, back, right or left, for as many free squares as it likes. In one move it picks one direction and does not turn. Put a rook on d4 of an empty board and mark everywhere it can go.

We say that the rook attacks these squares, or keeps them under attack: if an enemy piece stood there, the rook could take it. For line pieces the squares they can move to and the squares they attack are the same, with one exception. A rook also attacks a square holding a piece of its own side, even though it cannot move there: that is how it defends that piece. If the opponent captures a defended piece, the rook captures back. A pawn moves to different squares from the ones it attacks, but that is the subject of the chapter on the pawn.

Now a question that trips up almost every beginner.

How many squares does a rook on a1 attack if the board is empty?

Seven squares up the a-file and seven to the right along rank 1. On an empty board the rook does not care where it stands: one file and one rank pass through every square, each with eight squares, and the rook occupies one of them. It is always 7 + 7 = 14.

You know the piece letters and signs from the first chapter: rook R, bishop B, queen Q, king K, knight N, and no letter for a pawn. Rd8 means "the rook moved to d8", x is a capture, + is check, # is checkmate, ! is a strong move and ? is a mistake.

In the starting position the line pieces have no moves at all. The rooks stand in the corners behind the pawns and knights, and the bishops and the queen are behind the pawns. All three need someone to open lines for them first. The rook waits longest: while its own pawns stand in front, it cannot go forward, and it usually joins the game last. A file with no pawns left on it is called open, and a rook there is like a car on an empty motorway. The fight for such lines is the subject of chapter 31.

Blockers and captures

On an empty board a line runs to the edge. In a game there is almost always someone standing on it, and then three rules apply to every line piece.

  • You cannot jump over pieces, your own or the opponent's.
  • A piece of your own on the line is a wall. You cannot land on its square, or go beyond it.
  • An enemy piece is a target. You can land on its square and remove it from the board. The move ends there: you capture only one piece in a move.

A move onto a square with an enemy piece is called a capture. You never have to capture. If a capture is bad, you can play something else.

Which black pieces can the white rook on a1 capture in one move?

Along rank 1 the path b1–e1 is empty, and the rook takes the bishop: Rxf1. Up the a-file it gets only as far as a3, where its own pawn stands, and the knight on a7 is safe behind it.

In games players more often miss the distant pieces than the near ones: the eye slides over the nearby squares and never reaches the edge of the board. Find the capture yourself. Click on the rook and the board will show every square it can go to.

White to move. One of the black pieces is attacked by the rook and has no defender. Take it.

Rxa7: the rook runs the whole length of the a-file and takes the knight. The bishop on e6 is not on the rook's lines, neither on the a-file nor on rank 1. If you do not take the knight now, it will escape to b5 or c6.

The bishop: along diagonals, on one colour

The bishop moves along a diagonal for as many free squares as it likes, with the same rules about blockers and captures. This has a consequence that is easy to miss. Take one step along a diagonal and you are still on a square of the same colour. Take another and the colour has not changed. A bishop that starts the game on a dark square stays on dark squares for the whole game and can never reach a light one.

So each side has two different bishops. The one on f1 (for Black, on c8) is the light-squared bishop, and the bishop on c1 (for Black, on f8) is the dark-squared bishop. Exactly half the board is closed to each of them: 32 squares.

The bishop has a second quirk. In the centre four long rays are open to it, near the edge two, and in a corner a single diagonal. A rook attacks 14 squares from anywhere, but a bishop attacks from 7 to 13, so where it stands matters. From the centre it attacks almost twice as many squares as from a corner.

That gives you a piece of advice that will help in your first games. Your own pawns standing on squares of your bishop's colour block its diagonals. The more of them there are, the worse the bishop: it turns into a big pawn. So experienced players, when they decide which squares to put their pawns on, keep in mind the colour of their bishop. There is more on this in the chapter "Bishop against knight".

The bishop is on c1. Which of the marked squares can it reach if you give it several moves in a row?

The square c1 is dark, so the bishop moves on dark squares for the whole game. Of the marked squares only h6 is dark, and the bishop can even reach it in one move: c1–d2–e3–f4–g5–h6, if the path is clear.

White to move. The knight on e5 has attacked the bishop on d3. You could simply move the bishop away, but there is a better move.

Bxa6: the bishop leaves the attack and captures a rook on the way. The square a6 is light, like d3, and the diagonal d3–c4–b5–a6 is clear. The bishop cannot take the knight on e5: e5 is a dark square, and it is not on the light-squared bishop's lines. White had a bishop against a rook and a knight, and now it is a bishop against a knight, so the forces are almost level.

The queen: rook plus bishop

Now we add them together. The queen moves both as a rook and as a bishop: along files, ranks and diagonals, for as many free squares as it likes. In one move it picks one of eight directions and goes along it. Put it on d4 and colour the squares: the green ones it attacks "like a bishop", and the other marked ones "like a rook".

From a corner the queen attacks 14 + 7 = 21 squares, from the centre 14 + 13 = 27. It is the most mobile piece on the board. But it too moves only in straight lines: it cannot jump, or turn in the middle of a move.

The queen's power has a flip side. If a cheaper piece attacks it, the queen almost always has to run, because trading it for a bishop or a knight would be a huge loss. So an opponent chases a queen that came out too early all over the board, and gains time to bring out their own pieces. Chapter 10 shows how to avoid this.

Can the queen go from d1 to e3 in one move?

The move from d1 to e3 is a knight's move: one square sideways and two up. It is beyond the queen. It can reach e3 in two moves, for example via d3 or e2.

Long range works both ways. Your queen attacks from far away, but an enemy bishop also attacks across the whole board. Before you put a piece on a square, run your eye along all eight rays from it: the file, the rank and both diagonals, in each direction. Is there an enemy rook, bishop or queen on any of them? Try it yourself.

The queen on d3 wants to attack the rook on h5. From which of the marked squares can it do so without coming under attack?

From e2 the queen attacks h5 along the diagonal e2–f3–g4–h5, and it stands on none of Black's lines. Black attacks each of the other three squares, and you can only see that by running your eye along every ray from the square.

Now the same position, but White rushes. The queen went to f3 to attack the rook on h5 and missed the bishop on b7. Ask the engine what it makes of the move.

White played Qf3: the queen attacks the undefended rook on h5 along the diagonal f3–g4–h5. But the bishop on b7 looks at f3 along the long diagonal a8–h1, and c6, d5 and e4 are empty.

Why the queen is worth nine

Chess players long ago agreed to measure the pieces in pawns: a pawn is 1, a knight and a bishop are 3 each, a rook is 5 and a queen is 9. The king is not valued at all: it can be neither traded nor given up, and checkmate, as you remember from the first chapter, ends the game. These numbers help you decide whether to trade, and chapter 8 covers them in detail. For now let us see where they come from. For each square of the board we count how many squares a piece attacks from it, then average over all 64 squares. The "Squares" column shows the range from a corner to the centre. For comparison we add the knight, which we meet in chapter 3.

PieceSquaresAverageValue
Bishop7–138¾3
Rook14145
Queen21–2722¾9
Knight2–85¼3

This is the piece's mobility, and it explains the values quite well. A rook attacks 14 squares on average and a bishop 8¾, which is 1.6 times fewer. The values 5 and 3 differ by almost the same factor, 1.67. The queen's mobility is exactly the sum of the rook's and the bishop's, since a queen is a rook and a bishop on one square. Yet it is valued above the sum: 9, not 5 + 3 = 8. A piece that attacks along both rook lines and bishop lines finds double attacks where a rook or a bishop cannot: one target on its rank, another on its diagonal. A rook or a bishop would need two moves for such a blow, and you only get one.

In the table the knight looks weaker than the bishop, though it is worth the same. But mobility is not the whole story. The knight jumps over pieces, so a cramped position does not bother it, while in such a position the bishop's lines are blocked by its own pawns. So the values are only an agreement among experienced players, not a law of nature, and in any given position a piece is worth a little more or a little less than its value.

The engine of this course has values of its own, which the program fitted on thousands of games. In the endgame it rates the rook clearly higher than in the middlegame, because by then the files have opened. Chapter 27 explains how the engine judges a position.

Your rook can take a bishop, but on the next move the opponent takes your rook. What is the score?

That trade costs you two points. Players call it "giving up the exchange", and in chapter 8 we will see when it is still justified.

Now the values at work. The queen has three captures to choose from. The rook and the knight are undefended, and the bishop is defended by the rook on b6.

White to move. The queen on d4 can take the rook, the bishop or the knight. Which piece should it take?

Qxb6: this is the most valuable prize, the rook (5). You cannot take the bishop: after Qxd6 the rook on b6 takes the queen along the sixth rank. The bishop on d6 and the knight on h4 are worth only 3 each. The queen takes the rook along the diagonal d4–c5–b6, "like a bishop", while it would take the bishop or the knight "like a rook".

And here is that attack on two targets at once. In a single move the queen can give the king check, that is, attack it, and attack a rook at the same time. A check has to be dealt with at once (chapter 5 explains how), and Black will not have time to save the rook.

White to move. Find the queen move after which Black loses a rook.

Qd8+: along the eighth rank the queen checks the king on g8, and along the diagonal d8–c7–b6–a5 it attacks the rook. The king steps aside and the queen takes the rook on a5. One piece works along two lines at once, a rook line and a bishop line.

Together: the battery and the ladder

Two line pieces standing one behind the other on the same line are called a battery. The front piece goes ahead, and the rear one looks at its back and supports it. If the opponent takes the front piece, the rear one takes back. A battery can be two rooks on one file or rank, a queen and a rook, or a queen and a bishop on a diagonal.

A battery of two rooks on one file is known as doubled rooks. See what that gives you.

White to move. The knight on e5 is defended by the rook on a5. Can you still win it?

Rxe5. Two white rooks attack the knight and one black rook defends it. If Black recaptures with Rxe5, the second white rook captures again: White wins the knight, is left with a rook, and Black has no pieces at all. Count it like this: two attackers, one defender, so the last capture belongs to White.

In a battery of queen and bishop, the bishop stands behind and defends the square the queen goes to. Remember the rule from the first chapter: a king may not step onto a square attacked by an enemy piece, even to capture. So a piece defended only by the king can be taken if another of your pieces also attacks the square of capture. There is more on the king in chapter 3.

White to move. The queen on d3 and the bishop on b1 stand on one diagonal. How can you use that?

Qxh7+: the queen takes the knight with check. The king on g8 is right next to it but cannot take the queen: as soon as the queen left d3, the diagonal b1–h7 opened, and the bishop defends h7. The king has to retreat to f8. Then comes Qh8+ and Qxa8: check along the eighth rank and an attack on the rook, as in the problem with Qd8+.

Now here is how two rooks work when the opponent has only a king left. Each rook covers a whole rank. While one holds a rank, the other checks on the next one, and the king retreats step by step to the edge. This mate is called the ladder mate, or the rook roller.

The rook on a3 holds the third rank: the king cannot go lower. Check along the fourth rank. The king has to leave it, and it cannot go down, so it can only go up. Now the rooks have swapped roles: the rook on b4 holds the fourth rank, and the rook on a5 checks along the fifth. Check along the sixth rank. Each check takes one more rank away from the king. Check along the seventh rank. Only the eighth rank is left to the king, because the rook on b6 holds the sixth. Checkmate: the rook on b8 checks along the eighth rank, the rook on a7 covers the seventh, and there is nowhere to run. If the king comes towards the rooks, you move them to the other edge of the board. In chapter 9 you will learn how to deliver this mate against any defence.

The Opera Game: a finish on the lines

One of the most famous games in the history of chess was played in Paris in 1858. The American Paul Morphy, one of the strongest players of the 19th century, had White against two opponents who consulted each other over every move: Duke Karl of Brunswick and Count Isouard. The game was played in a box at an opera house during a performance, which is why it is called the Opera Game. We skip the first twelve moves, which hold many opening ideas from the third part of the course, and start from the position after Black's twelfth move.

The bishop on b5 looks through the knight on d7 at the king on e8: if the knight moves, the king is in check. The bishop on g5 attacks the knight on f6, with the queen on e7 behind it. Both black knights are tied down by the lines of White's bishops. Morphy gives a rook for a knight: by the values that is minus two. But in return the square d1 is freed for the second rook, and if a black rook takes on d7, it will shield the king from the bishop on b5 in the same way. The second rook goes to d1. Now a rook and a bishop attack the rook on d7, and it shields its own king from the bishop on b5. Black defends the rook with the queen and offers a queen trade. The bishop took the rook, and the knight on f6 took the bishop. Morphy has given a rook and a bishop for a rook and a knight, and now his pieces look at the eighth rank, where the king stands.

Now it is your move. In this position Morphy found a finish in two moves. A hint: White's most valuable piece sacrifices itself to open a line for the rook.

White to move. Mate in two.

Qb8+! The queen sacrifices itself: the knight on d7 has to take the queen, and the move Nxb8 clears the d-file. The rook goes to d8 and mates. It is defended by the bishop on g5 along the diagonal g5–f6–e7–d8, and the bishop also covers e7.

Once you have solved it, look at who did all the work. White's five moves in this finish were three with rooks, one with a bishop and one with the queen. There was not one move with a knight, a king or a pawn. The line pieces took turns opening lines for each other: the rooks pressed down the d-file, the bishop removed the black rook from it, and the queen's sacrifice drew away its last defender, the knight on d7.

What's next

All three pieces of this chapter share one weakness: they cannot jump. Block the line and any of them is helpless. The knight jumps over anything, but not far, and its move looks like no line at all. The king moves like a queen, but only one square (it also has a special move, castling, covered in chapter 6), and it may not step into an attack. How these two pieces move, why the knight's L-shape gets round any wall, and how to plan knight routes, you can read in the chapter "King and knight".