CanvasRenderingContext2D: setTransform() method

The CanvasRenderingContext2D.setTransform() method of the Canvas 2D API resets (overrides) the current transformation to the identity matrix, and then invokes a transformation described by the arguments of this method. This lets you scale, rotate, translate (move), and skew the context.

Note: See also the transform() method; instead of overriding the current transform matrix, it multiplies it with a given one.


setTransform(a, b, c, d, e, f)

The transformation matrix is described by: [ a c e b d f 0 0 1 ] \left[ \begin{array}{ccc} a & c & e \ b & d & f \ 0 & 0 & 1 \end{array} \right]

This transformation matrix gets multiplied on the left of a column vector representing each point being drawn on the canvas, to produce the final coordinate used on the canvas.


setTransform() has two types of parameter that it can accept. The older type consists of several parameters representing the individual components of the transformation matrix to set:

a (m11)

The cell in the first row and first column of the matrix.

b (m12)

The cell in the second row and first column of the matrix.

c (m21)

The cell in the first row and second column of the matrix.

d (m22)

The cell in the second row and second column of the matrix.

e (m41)

The cell in the first row and third column of the matrix.

f (m42)

The cell in the second row and third column of the matrix.

Alternatively, you can pass a single parameter which is an object containing the values above as properties. The parameter names are the property keys, and if two synonymous names are both present (e.g. m11 and a), they must be the same number value, or a TypeError will be thrown. Using the object form allows omitting some parameters — a and d default to 1, while the rest default to 0.

If a point originally had coordinates ( x , y ) (x, y) , then after the transformation it will have coordinates ( a x + c y + e , b x + d y + f ) (ax + cy + e, bx + dy + f) . This means:

  • e and f control the horizontal and vertical translation of the context.
  • When b and c are 0, a and d control the horizontal and vertical scaling of the context.
  • When a and d are 1, b and c control the horizontal and vertical skewing of the context.

Return value

None (undefined).


Skewing a shape

This example skews a rectangle both vertically (.2) and horizontally (.8). Scaling and translation remain unchanged.


<canvas id="canvas"></canvas>


const canvas = document.getElementById("canvas");
const ctx = canvas.getContext("2d");

ctx.setTransform(1, 0.2, 0.8, 1, 0, 0);
ctx.fillRect(0, 0, 100, 100);


Retrieving and passing a DOMMatrix object

In the following example, we have two <canvas> elements. We apply a transform to the first one's context using the first type of setTransform() and draw a square on it, then retrieve the matrix from it using CanvasRenderingContext2D.getTransform().

We then apply the retrieved matrix directly to the second canvas context by passing the DOMMatrix object directly to setTransform() (i.e. the second type), and draw a circle on it.


<!-- First canvas (ctx1) -->
<canvas width="240"></canvas>
<!-- Second canvas (ctx2) -->
<canvas width="240"></canvas>


canvas {
  border: 1px solid black;


const canvases = document.querySelectorAll("canvas");
const ctx1 = canvases[0].getContext("2d");
const ctx2 = canvases[1].getContext("2d");

ctx1.setTransform(1, 0.2, 0.8, 1, 0, 0);
ctx1.fillRect(25, 25, 50, 50);

let storedTransform = ctx1.getTransform();

ctx2.arc(50, 50, 50, 0, 2 * Math.PI);



HTML Standard
# dom-context-2d-settransform-dev

Browser compatibility

BCD tables only load in the browser

See also