JavaScript Typed Arrays

❮ Previous Next ❯

Example

const myArr = new Int8Array(10);

Try it Yourself »

Typed Arrays

Typed arrays was designed for handling raw binary data.

Unlike standard arrays, typed arrays are array buffers of fixed length.

Typed arrays store elements of fixed types like 8-bit integers or 32-bit numbers.

Typed Array Benefits

Typed Arrays were designed to provide an efficient way to handle binary data, unlike traditional JavaScript arrays which can hold elements of mixed data types.

Typed arrays are raw memory, so JavaScript can pass them directly to any function without converting the data to another representation.

Typed arrays are seriously faster than normal arrays for passing data to functions that can use raw binary data. Typed Arrays are highly suitable for:

WebGL and Canvas:
Fast graphics rendering and image processing.
File APIs:
Fast reading and writing of local files.
Media APIs:
Fast handling of audio and video data.
WebSockets:
Efficient binary data transfer over network.

Typed Arrays were introduced to JavaScript as part of the ECMAScript 2015 (ES6) specification, released in June 2015.

Differences from Regular Arrays

Fixed Length:
Typed Arrays cannot be dynamically resized using methods like push() or pop().
Type Restriction:
Elements must adhere to the specified data type of the typed array.
Underlying Buffer:
Typed Arrays are views into an ArrayBuffer, allowing direct manipulation of binary data.

Typed Array Types

Name	Min	Max	Bytes	Type
Int8Array	-128	127	1	byte
Uint8Array	0	255	1	octet
Uint8ClampedArray	0	255	1	octet
Int16Array	-32768	32767	2	short
Uint16Array	0	65535	2	unsigned short
Int32Array	-2147483648	2147483647	4	long
Uint32Array	0	4294967295	4	unsigned long
BigInt64Array	-2⁶³	2⁶³ - 1	8	bigint
BigUint64Array	0	2⁶⁴ - 1	8	unsigned bigint

Float16Array	-65504	65504	2	unrestricted half
Float32Array	-3.4e38	3.4e38	4	unrestricted float
Float64Array	-1.8e308	1.8e308	8	unrestricted double

8 Bit Integers

Name	Data Type	Range
Int8Array	Signed integer (byte)	-128/127
Uint8Array	Unsigned integer (octet)	0/255
Uint8ClampedArray	Unsigned integer (octet)	0/255

Examples

Create a typed array of 10 signed 8-bit integers (byte format):

const myArr = new Int8Array(10);

Try it Yourself »

Create a typed array of 10 unsigned 8-bit integers (octet format):

const myArr = new Uint8Array(10);

Try it Yourself »

Create a typed array of 10 usigned 8-bit integers (clamped format):

const myArr = new Uint8ClampedArray(10);

Try it Yourself »

Uint8Array vs Uint8ClampedArray

The difference between an Uint8Array and an Uint8ClampedArray is how values are added.

If you set one element in an Uint8ClampedArray to a value outside the 0-255 range, it will default to 0 or 255.

A typed array will just take the first 8 bits of the value.

Note

Typed arrays are not arrays.

isArray() on a typed array returns false.

Many array methods (like push and pop) are not supported by typed arrays.

16-Bits Integers

Name	Data Type	Range
Int16Array	Short integer	-32768/32767
Uint16Array	Unsigned short integer	0/65535

Examples

Create a typed array of 10 signed 16-bit integers (short format):

const myArr = new Int16Array(10);

Try it Yourself »

Create a typed array of 10 unsigned 16-bit integers (unsigned short format):

const myArr = new Uint16Array(10);

Try it Yourself »

32-Bit Integers

Name	Data Type	Range
Int32Array	Signed long integer	-2147483648 / 2147483647
Uint32Array	Unsigned long integer	0 / 4294967295

Examples

Create a typed array of 10 signed 32-bit integers (long format):

const myArr = new Int32Array(10);

Try it Yourself »

Create a typed array of 10 unsigned 32-bit integers (unsigned long format):

const myArr = new Uint32Array(10);

Try it Yourself »

64-Bit Integers

Name	Data Type	Range
BigInt64Array	Big signed integer	-2⁶³/2⁶³-1
BigUint64Array	Big unsigned integer	0/2⁶⁴

Examples

Create a typed array of 10 signed 64-bit integers (bigint format):

const myArr = new Bigint64Array(10);

Try it Yourself »

Create a typed array of 10 unsigned 64-bit integers (bigint format):

const myArr = new Biguint64Array(10);

Try it Yourself »

Floating Point Numbers

Name	Description	Range
Float16Array	Half precision - 3 significant decimal digits	-65504 / 65504
Float32Array	Normal precision - 7 significant decimal digits	-3.4e38 / 3.4e38
Float64Array	Double precision- 15 significant decimal digits	-1.8e308 / 1.8e308