Little Endian versus Big Endian

Compare JUMP instructions

Little Endian CPU
Intel 8080

Big Endian CPU
RCA/GE 1806

Address	Data	Instruction	Comment
1000	C3	JMP 4000	; Jump to 4000
1001	00		; Low Byte
1002	40		; High Byte

Address	Data	Instruction	Comment
1000	C0	LBR 4000	; Jump to 4000
1001	40		; High Byte
1002	00		; Low Byte

16-bit as an extension of 8-bit

Little Endian CPU
Intel 8080

Big Endian CPU
RCA/GE 1806

Addr	Data	Instruction	Comment
4000	3E	MVI A,01	; Load Accu, 01
4001	01		; Data

Addr	Data	Instruction	Comment
4000	F8	LDI 01	; Load Accu, 01
4001	01		; Data

Addr	Data	Instruction	Comment
4002	21	LXI H,0001	; Load HL, 0001
4003	01		; Low Byte
4004	00		; High Byte

Addr	Data	Instruction	Comment
4002	68C0	RLDI R0,0001	; Load R0, 0001
4004	00		; High Byte
4005	01		; Low Byte

Little Endian versus Big Endian

When dealing with multibyte values (eg. 16-bit or 32-bit data) in memory a Big Endian CPU stores the most significant byte on the lowest memory address.
In datacommunication systems a Big Endian CPU will first transmit the most significant byte of a multibyte value.

A Little Endian CPU stores the least significant byte on the lowest memory address.
In datacommunication systems a Little Endian CPU will tend to transmit the least significant byte first, resulting in a presentation problem when communicating to a Big Endian CPU.

Since TCP/IP was developed using PDP-11 Big Endian CPUs, Big Endian is the standard on the network. When communicating to the network, classical TCP/IP applications (eg. Telnet, FTP) convert all data to/from Big Endian using libraries. RPC applications use a separate protocol: XDR.