Address Resolution Protocol (RFC 826) is how an IPv4 device discovers which MAC address corresponds to a given IP address on its local network segment, by broadcasting a request and caching the response.

No — IPv6 replaces ARP entirely with the Neighbor Discovery Protocol (NDP, RFC 4861), which performs the equivalent IP-to-MAC resolution using ICMPv6 messages, along with additional functions like router discovery that ARP never handled.

Can a MAC address be changed?

Yes — while factory-assigned (universally-administered) MAC addresses are meant to be globally unique and permanent, most operating systems allow software to present a different, locally-administered MAC address, a feature used both for privacy (MAC randomization on mobile devices) and, potentially, for spoofing.

What does the OUI portion of a MAC address tell you?

The first three octets (24 bits) are the Organizationally Unique Identifier, assigned by the IEEE to a specific hardware manufacturer — often revealing which company made a given network interface, sometimes used in network device-identification tooling.

Is MAC filtering a strong security measure?

Not on its own — MAC addresses can be spoofed by software relatively easily, so MAC filtering is better understood as a basic access-control convenience layer rather than a robust security boundary against a determined attacker.

Why can't MAC addresses be used to route across the internet?

MAC addresses were designed for local network segment delivery (Layer 2) with no built-in concept of network topology or location, unlike IP addresses (Layer 3), which are specifically structured (via hierarchical subnetting) to support routing across many interconnected networks.

MAC Address vs IP Address

In-Depth Technical Comparison & Architecture Guide

A MAC address identifies a specific physical network interface and never changes; an IP address identifies a device's logical position in a network and routinely does — the protocol that bridges the two, letting an IP-addressed packet actually reach the right physical hardware on a local network, is exactly what most comparisons of these two addressing systems skip over.

Quick Reference Matrix

Feature	MAC Address	IP Address
OSI Layer	Layer 2 (Data Link)	Layer 3 (Network)
Assignment	Burned-in at manufacture (or locally-administered/randomized)	Assigned dynamically (DHCP) or statically configured
Length	48 bits (six hex octets)	32 bits (IPv4) or 128 bits (IPv6)
Scope	Local network segment only — not internet-routable	Routable across networks, including the internet
Resolution Protocol	N/A — is the target of resolution	ARP (IPv4, RFC 826) or NDP (IPv6, RFC 4861) resolves IP to MAC

Technology Overview

A MAC (Media Access Control) address is a 48-bit identifier burned into a network interface card at manufacture, operating at OSI Layer 2 (Data Link) — it identifies a specific physical piece of hardware and has no concept of location or network topology at all; a laptop's MAC address is the same whether it's on a home network or halfway across the world.

An IP address operates at OSI Layer 3 (Network), and is a logical address reflecting where a device sits within a network's addressing scheme — assigned dynamically (commonly via DHCP) or statically configured, and changing routinely as a device moves between networks, renews its DHCP lease, or reconnects. Unlike a MAC address, an IP address is inherently routable across the wider internet; a MAC address is not designed to be.

These two addressing systems have to work together for any local network communication to actually happen: when a device wants to send a packet to an IP address on its local network segment, it needs to know which physical MAC address that IP currently corresponds to — resolving that mapping is the specific job of the Address Resolution Protocol (ARP) on IPv4 networks, and its conceptual successor, the Neighbor Discovery Protocol (NDP), on IPv6 networks.

How IP Addresses Get Mapped to MAC Addresses: ARP and NDP

On an IPv4 local network, ARP (Address Resolution Protocol, RFC 826) is how a device discovers which MAC address corresponds to a given IP address on the same segment: it broadcasts an ARP request ('who has this IP address?') and the device holding that IP responds with its MAC address, which the requester then caches for a period of time to avoid repeating the lookup for every packet.

IPv6 replaces ARP entirely with the Neighbor Discovery Protocol (NDP, RFC 4861), which performs the equivalent IP-to-MAC resolution (along with several other functions ARP never handled, like router discovery and duplicate address detection) using ICMPv6 messages instead of ARP's own separate protocol. This is a genuine, sometimes-overlooked architectural difference: IPv6 didn't just get a bigger address space, it also replaced this specific piece of IPv4's local-network machinery with a more capable, differently-designed protocol.

MAC Address Structure: More Than a Random Number

A 48-bit MAC address is conventionally written as six hexadecimal octets (e.g. 00:1A:2B:3C:4D:5E). The first three octets (24 bits) are the Organizationally Unique Identifier (OUI), assigned by the IEEE to the specific hardware manufacturer — meaning a MAC address's first half often reveals which company manufactured the network interface, a detail sometimes used in network inventory and security tooling to identify device types on a network.

Two specific bits within the address carry their own meaning: the least significant bit of the first octet distinguishes a unicast address (intended for a single device) from a multicast address (intended for a group), and the second-least-significant bit distinguishes a universally-administered address (factory-assigned, globally unique per the IEEE) from a locally-administered address (manually or software-assigned, common for virtual machine network interfaces and MAC randomization privacy features on mobile devices).

MAC Address Advantages & Disadvantages

Advantages / Pros

Guaranteed globally unique per the IEEE-assigned OUI scheme (for universally-administered addresses).
Stable, permanent hardware identifier that doesn't change with network reconnection.

Disadvantages / Cons

Not routable beyond the local network segment.
Reveals manufacturer information via its OUI, a minor privacy/fingerprinting consideration MAC randomization features address.

IP Address Advantages & Disadvantages

Advantages / Pros

Routable across networks, enabling global internet communication.
Hierarchical addressing (via subnetting) supports structured, scalable network design.

Disadvantages / Cons

Can change dynamically, requiring resolution (via DNS or ARP/NDP) to reliably reach a specific device.
Provides no inherent hardware-identity guarantee the way a MAC address does.

Real-World Use Cases

MAC Address

Local network device identification

Switches forwarding frames to the correct physical port based on MAC address tables.

Network access control (MAC filtering)

Allowing or denying specific hardware devices on a local network based on their MAC address.

IP Address

Global internet routing

Addressing packets across the internet between networks that have never directly interacted before.

Subnetting and network segmentation

Structuring a large network into logical, hierarchically-addressed segments.

Developer Recommendation

Understand both as complementary, not competing — a working network connection requires an IP address for logical routing and a MAC address for actual local-segment delivery, resolved between the two via ARP (IPv4) or NDP (IPv6).

Use MAC-based controls (MAC filtering, static ARP/NDP entries) specifically for local network access control, not as a security boundary on its own — MAC addresses can be spoofed relatively easily by software, so treat MAC filtering as a convenience layer, not a strong security guarantee.

When troubleshooting local network connectivity, check the ARP/NDP mapping directly (whether a device's IP address is correctly resolving to the expected MAC address) as a diagnostic step distinct from checking DNS or broader internet routing.

Frequently Asked Questions

What is ARP?: Address Resolution Protocol (RFC 826) is how an IPv4 device discovers which MAC address corresponds to a given IP address on its local network segment, by broadcasting a request and caching the response.
Does IPv6 use ARP?: No — IPv6 replaces ARP entirely with the Neighbor Discovery Protocol (NDP, RFC 4861), which performs the equivalent IP-to-MAC resolution using ICMPv6 messages, along with additional functions like router discovery that ARP never handled.
Can a MAC address be changed?: Yes — while factory-assigned (universally-administered) MAC addresses are meant to be globally unique and permanent, most operating systems allow software to present a different, locally-administered MAC address, a feature used both for privacy (MAC randomization on mobile devices) and, potentially, for spoofing.
What does the OUI portion of a MAC address tell you?: The first three octets (24 bits) are the Organizationally Unique Identifier, assigned by the IEEE to a specific hardware manufacturer — often revealing which company made a given network interface, sometimes used in network device-identification tooling.
Is MAC filtering a strong security measure?: Not on its own — MAC addresses can be spoofed by software relatively easily, so MAC filtering is better understood as a basic access-control convenience layer rather than a robust security boundary against a determined attacker.
Why can't MAC addresses be used to route across the internet?: MAC addresses were designed for local network segment delivery (Layer 2) with no built-in concept of network topology or location, unlike IP addresses (Layer 3), which are specifically structured (via hierarchical subnetting) to support routing across many interconnected networks.

Part of the Networking & Infrastructure Developer Hub

This comparison is part of our Networking & Infrastructure topic guide, covering related tools, standards, and decision guidance.

Related Comparisons

Other technology decisions in the same topic area as Networking & Infrastructure:

IPv4 vs IPv6

IPv4 and IPv6's real difference isn't just address space — IPv6's simplified header and native autoconfiguration reflect two decades of learning from IPv4's operational pain points, though one commonly-repeated claim about IPv6 (that it mandates IPsec) has actually been out of date since 2011 and is worth correcting directly.

Read Guide

Launch Interactive Developer Tools

Put these concepts into practice. Test, format, serialize, or analyze your inputs locally with these secure, browser-only utilities:

MAC Address Generator

Generate randomized MAC addresses with optional unicast/local bits.

Open Tool

IPv4 Subnet Calculator

Calculate subnet masks, host ranges, and broadcast addresses for IPv4.

Open Tool

Wake-on-LAN Packet Generator

Generate Wake-on-LAN magic packets in hexadecimal and binary forms.

Open Tool