Did you know that one laptop can have a MAC address and an IP address at the same time without those addresses doing the same job? The MAC address identifies a network interface for local link-layer communication. An IPv4 or IPv6 address identifies an interface in an IP network and supports routing packets between networks. A Wi-Fi access point, switch and router use these kinds of information at different parts of the journey. Confusing the roles makes even a simple network diagram hard to explain.
Ministry of Education V3.0 Reference 379 explains the 2027 Singapore–Cambridge SEC G3 Computing K349 outcomes 4.2.4 and 4.2.5. It compares the purpose, bit length, degree of permanence and typical representation of MAC, IPv4 and IPv6 addresses. Using reserved documentation addresses and fictional LANs, students learn how a switch, network interface controller and router select the next part of a data path, how hexadecimal notation compresses long binary sequences, and why an address does not automatically reveal a real person’s identity.
This is an independent eduKateSG learning guide, not an official MOE, SEAB or Cambridge publication. The official 2027 K349 syllabus defines examinable scope. All addresses used as examples are documentation or fictional identifiers, not real students’ devices. Actual network configuration is more complex than these teaching diagrams, and IP addresses can change or be shared. Avoid posting a genuine personal network inventory when practising.
1. The one-minute distinction: local interface and routed destination
Imagine a laptop on a home LAN requesting a remote educational website. The laptop’s network interface participates in local communication with the access point or switch. The request also carries IP information describing source and destination at the network layer. The home router uses IP routing information to forward traffic towards the next network.
A common misunderstanding is to imagine that a packet keeps one unchanging MAC destination from the laptop all the way to the distant website. Link-layer addressing is used on individual network segments, while IP addressing serves a wider routing purpose. The link-layer destination can change at each hop; the network-level path is a different concept. Students should trace one local link at a time.
2. Read the exact 2027 syllabus boundary
K349 outcome 4.2.4 states that MAC addresses are used by network interface controllers, network switches and wireless access points to direct data within the same LAN, while IPv4 and IPv6 addresses are used by routers to direct data across different LANs. Outcome 4.2.5 asks students to compare purpose, bit length, degree of permanence and typical representation formats.
These outcomes do not require operating a public router, disclosing a real device’s address or memorising the allocation rules of every Internet provider. They reward clear conceptual distinctions, appropriate notation, and explanation of what changes when a device moves between networks. This article makes those contrasts explicit.
3. What a MAC address is
A Media Access Control address is an identifier associated with a network interface at the local link layer. A familiar form uses 48 bits, shown as six groups of two hexadecimal digits. An illustrative locally administered MAC address is 02:00:00:00:00:01. This form is useful for understanding how Ethernet and Wi-Fi interfaces are distinguished on a local network.
The ordinary 48-bit form is common, not a universal claim about every possible link-layer addressing scheme. An interface may have a manufacturer-assigned address, but modern devices can also use locally administered or randomised addresses. Thus MAC should not be described as a perfectly permanent, immutable personal serial number.
4. What IPv4 identifies
An Internet Protocol version 4 address uses 32 bits. It is commonly written as four decimal octets separated by dots, with each octet ranging from 0 to 255. The example 192.0.2.10 is drawn from a documentation range and is suitable for illustrating a device in a fictional network.
An IPv4 address can be allocated to an interface according to the network’s configuration, and it may change when a device connects elsewhere or receives a different assignment. IP addressing supports communication across routed networks. Do not assume that one observed IPv4 address proves one unique person, one physical device or one permanently fixed location.
5. What IPv6 identifies
Internet Protocol version 6 uses 128-bit addresses. The normal representation uses eight groups of four hexadecimal digits separated by colons. The example 2001:db8::1 belongs to a documentation prefix and uses a standard abbreviation for groups of zeros. It is not a genuine address to assign casually to a public Internet endpoint.
IPv6 has a much larger address space than IPv4, but that does not mean every mathematically possible address is available for arbitrary global assignment. Address types, reserved ranges and network configuration matter. At SEC level, the central contrasts are 128 bits versus 32 bits, colon-separated hexadecimal notation versus dotted decimal, and their shared role in IP routing.
6. Count the bits and bytes carefully
A typical 48-bit MAC address contains six bytes. An IPv4 address contains four bytes. An IPv6 address contains sixteen bytes. These sizes follow the definition of eight bits per byte. The number of printed characters does not directly equal the number of data bits because hexadecimal digits compress four bits each.
For instance, each two-digit MAC group represents one byte, while each four-digit IPv6 group represents sixteen bits. Delimiters such as colons and dots are representation aids, not extra bits in the underlying address length. A student should be able to explain the counted groups and units independently of a memorised slogan.
Typical local and IP address forms
MAC (48 bit) 02:00:00:00:00:01 6 groups × 8 bits
IPv4 (32 bit) 192.0.2.10 4 octets × 8 bits
IPv6 (128 bit) 2001:db8::1 eight 16-bit groups when expanded
These are illustrative locally administered or documentation identifiers.7. Explain hexadecimal without guessing
Hexadecimal is base 16, using digits 0–9 and letters A–F to represent values from zero through fifteen. One hexadecimal digit represents four binary bits. Therefore two hex digits represent eight bits and four hex digits represent sixteen bits. This is why a 48-bit address can appear as twelve hexadecimal digits grouped into pairs.
A simple check is the byte FF in hexadecimal, equal to binary 11111111, which is decimal 255. Students do not need to assume a printed MAC address is a decimal number. The syllabus includes binary, denary and hexadecimal conversions elsewhere, so those representations reinforce the same underlying data-encoding ideas.
8. Understand dotted decimal IPv4
An IPv4 address such as 192.0.2.10 consists of four decimal octets. The value 192 corresponds to binary 11000000; 0 corresponds to 00000000; 2 to 00000010; and 10 to 00001010. The four 8-bit values form 32 bits in total. The dots separate the four parts in human-readable notation.
An octet cannot hold a decimal value of 300 because eight bits can represent only 0 through 255. A student who sees 192.0.2.300 should recognise that it is not a valid ordinary IPv4 dotted-octet address. Correct-looking separators do not make an out-of-range component acceptable.
9. Understand colon-separated IPv6
An unabbreviated IPv6 address has eight 16-bit groups, each displayed as four hexadecimal digits. For example 2001:0db8:0000:0000:0000:0000:0000:0001 uses eight groups. It can be shortened to 2001:db8::1 using the defined omission of leading zeros and one compression of a consecutive run of all-zero groups.
The double colon can be used once in an address to compress one contiguous run of zero groups. Using it twice would make the expanded length ambiguous. The common 2001:db8 prefix is intended for examples and documentation, so readers should not treat it as evidence of a real endpoint.
10. MAC is local; IP is for routing across networks
A switch or wireless access point uses link-layer information to direct traffic among appropriate local interfaces. A router uses network-layer information to forward packets between networks. The two activities can occur as part of the same end-to-end communication, but they should not be described as one identical lookup.
When a laptop sends traffic to a remote service, its local link-layer destination is normally an appropriate local next hop such as the default gateway, not the MAC address of the distant server across the Internet. The route depends on network configuration and protocols; students should emphasise the distinct roles rather than drawing one direct physical wire to the remote destination.
11. Addresses can change under different conditions
A manufacturer-provided hardware interface identity may remain stable through ordinary use, but software can assign a different local MAC identity or randomise it in suitable contexts. An IP address may be assigned dynamically and can change when the same device reconnects or moves to a new network. That makes permanence a matter of degree and configuration, not a binary truth.
A table comparing typical permanence should therefore avoid saying that all MAC addresses are always permanent or all IP addresses always change. A device can have a stable IP assignment and a randomised Wi-Fi MAC. The appropriate answer describes typical patterns and acknowledges exceptions.
12. A device can have multiple interfaces and addresses
One laptop may include both a wired Ethernet NIC and a wireless Wi-Fi NIC. Each interface can have its own MAC address, and each can receive IP configuration appropriate to the connected network. An endpoint might also hold more than one IPv6 address or support both IPv4 and IPv6.
This is why the question ‘What is the computer’s one address?’ is often underspecified. The answer depends on the interface, layer, network and time of observation. For a basic SEC diagram, clearly label the interface being discussed and the type of address shown.
13. The MAC address is not a password
A MAC address can be visible or inferable to devices participating in a suitable local networking context. It is an identifier used in communication, not a secret authentication credential by itself. Using a MAC address as though it proves someone’s identity creates a weak security assumption.
Likewise, knowing an IP address does not entitle a user to access an endpoint. Actual permission depends on authentication, services and security controls. The network can use addressing to deliver packets without making them automatically authorised or confidential.
14. IP addresses are not precise personal-location records
An IP address can be associated with a service provider, network or publicly routable endpoint, but it does not automatically reveal an individual’s exact physical location or identity. Network address translation, shared access, dynamic allocation, VPNs and provider architecture can complicate interpretation.
For study, use documentation addresses. No class exercise requires a student to post their home’s genuine public IP address. A responsible explanation distinguishes the logical purpose of IP addressing from claims about a specific person’s whereabouts, which would need other evidence and safeguards.
15. Route to a local peer versus a remote website
Suppose two authorised laptops on the same suitable LAN communicate directly through local switching or wireless infrastructure. Their frames can be delivered locally without the traffic first visiting a remote Internet server. The endpoints still may use IP addresses for their applications, but the local link-layer path can remain within the LAN.
Now suppose one laptop requests a website on a different network. Its route generally involves the router and onward network devices. The link-layer addressing used to reach the local router is distinct from the IP destination of the website. A strong student can draw these as two different arrows with labels.
16. Routers need more than a printed address
A router’s forwarding decision depends on an IP destination together with configured routing information and the network’s addressing arrangement. An IPv4 address alone, without the relevant network context and route, does not explain every forwarding detail. Similar principles apply to IPv6.
The SEC learning target is a comparison of address purpose, format and common use; it is not necessary to construct a full global routing table. However, avoiding the myth that routers simply search a permanent list of all Internet devices helps students explain network scalability more honestly.
17. Switch forwarding and wireless access have local scope
A network switch generally forwards link-layer traffic through the appropriate local interface based on its learned forwarding information and local rules. A wireless access point bridges communication with compatible radio clients into the relevant LAN arrangement. Both can operate with MAC addresses in that local context.
This is not a guarantee that every packet reaches its destination or that every device is authorised. Local isolation policies, disconnected links and other conditions can prevent delivery. The addresses help identify interfaces, while actual access depends on a functioning and appropriately configured network.
18. Address representation is a human convention
A computer handles address information as bit patterns. Dotted decimal and colon-separated hexadecimal make those patterns manageable for human readers. A shortened IPv6 address still denotes a complete 128-bit value; the missing printed zeros have not vanished from the represented binary address.
It helps to compare 2001:0db8:0000:0000:0000:0000:0000:0001 and 2001:db8::1 as different textual forms of the same IPv6 address. The shortening rules reduce clutter but do not reduce the number of bits in IPv6.
19. Why IPv6 is not simply ‘faster IPv4’
The main distinction in this syllabus comparison is address length, representation, purpose and permanence, not a guarantee that IPv6 makes every Internet connection faster. Real performance depends on path, implementation, equipment, traffic conditions and service support. The much larger address space is an architectural difference.
A question asking students to choose the ‘best’ address without a scenario is oversimplified. IPv4 and IPv6 can coexist, and network transitions take time. Students should identify compatibility and addressing requirements rather than claim that one address format automatically improves physical wireless coverage.
20. An exam-ready comparative answer
A concise response can say: a common MAC identifier uses 48 bits and hexadecimal byte pairs and is used by NICs, switches and WAPs for local LAN communication; IPv4 uses 32 bits and dotted decimal; IPv6 uses 128 bits and colon-separated hexadecimal; the IP versions support routing between LANs. MAC assignments are often more interface-associated, while IP assignments are more network-dependent, with exceptions.
Add a real explanation to the table: when the laptop moves from one home network to another, its IP assignment is likely to depend on the new network, while its interface identifier may remain the same or may be randomised. This contrasts typical behaviour without claiming that a device always has exactly one unchanging address.
21. Twelve-stage MAC–IP comparison
This is an independent eduKateSG classroom routine rather than an official marking rubric. It aims to make addresses understandable through bit lengths, notation, forwarding scope and realistic permanence. Use documentation addresses and synthetic diagrams, not real household logs.
Stage 1 — Name the scope
Ask whether the communication remains within a LAN or requires movement across networks.
Stage 2 — Identify the network interface
Determine which wired or wireless NIC is sending the local frame.
Stage 3 — Recognise the notation
Classify hexadecimal MAC pairs, dotted IPv4 octets or colon-separated IPv6 groups.
Stage 4 — Count bits
Typical comparison: 48 bits for MAC, 32 bits for IPv4, 128 bits for IPv6.
Stage 5 — Validate groups
Check six MAC bytes, four IPv4 octets, and eight expanded IPv6 16-bit groups.
Stage 6 — Check notation limits
An IPv4 octet cannot exceed 255; IPv6 compression has defined constraints.
Stage 7 — Explain MAC use
Describe local interface identification and link-layer switching or Wi-Fi delivery.
Stage 8 — Explain IP use
Describe addressing for packets that routers send between LANs.
Stage 9 — Trace the local next hop
Distinguish the remote IP destination from the MAC used on one local link.
Stage 10 — Qualify stability
Static IP assignments and randomised MAC addresses are real possibilities.
Stage 11 — Respect privacy
Neither address alone establishes a person or exact physical location.
Stage 12 — State all four comparisons
Conclude with purpose, bit length, typical format and degree of permanence.
22. Twenty-four original address cases
Case 1 — Common MAC length
02:00:00:00:00:01 consists of six hexadecimal groups of two digits each. Six bytes correspond to 48 bits, not to the number of punctuation symbols.
Case 2 — IPv4 length
192.0.2.10 represents four octets. At eight bits per octet, the complete address has 32 bits.
Case 3 — IPv6 length
2001:db8::1 is a compressed notation for a 128-bit IPv6 address, not a shorter data type.
Case 4 — Invalid IPv4 value
192.0.2.300 cannot be a standard IPv4 dotted-decimal address because 300 exceeds the eight-bit maximum 255.
Case 5 — One hex byte
Hexadecimal FF represents 11111111 in binary, or decimal 255. Two hex digits express the eight bits of one byte.
Case 6 — Hex digits and MAC bits
Twelve hex digits represent 48 bits because each digit encodes four binary bits. Separators do not count as address bits.
Case 7 — Expanded IPv6
2001:0db8:0000:0000:0000:0000:0000:0001 and 2001:db8::1 identify one address when expanded under IPv6 rules.
Case 8 — Ambiguous compression
Two independent double-colon abbreviations would leave the number of omitted zero groups unclear and do not form ordinary valid IPv6 notation.
Case 9 — Ethernet NIC
An Ethernet interface may use a link-layer MAC identity on the local medium even though its IP address is assigned separately.
Case 10 — Wi-Fi NIC
A wireless network interface can use a locally administered or randomised MAC identity for a particular network.
Case 11 — Two interfaces
A single laptop can have both a wired and wireless NIC; there is no necessary one-to-one relation between one computer and one MAC.
Case 12 — Local printing
A laptop and printer permitted on the same LAN may exchange traffic locally without first visiting an Internet website.
Case 13 — Remote browsing
A request for a website outside the local LAN needs an appropriate routed path using network-layer IP addressing.
Case 14 — Next-hop frame
A locally transmitted frame to a gateway normally uses suitable local link-layer destination information, not the distant website’s MAC.
Case 15 — Routing decision
A router forwards packets based on IP-level information and routing configuration, not by preserving one MAC destination across the world.
Case 16 — Moving networks
A device moving between LANs can receive a different IP assignment, so a previous address might no longer be usable.
Case 17 — Fixed IP option
IP addresses can be deliberately stable in some configurations, so describing all IP assignments as constantly changing is false.
Case 18 — Randomised MAC option
Modern Wi-Fi clients can use changing or locally administered MACs, so a MAC is not necessarily fixed forever.
Case 19 — Dual IP families
An interface may support IPv4 and IPv6 concurrently while the two address types retain different lengths and representations.
Case 20 — Not a password
An IP or MAC address is an identifier, not a valid substitute for authentication or explicit access permission.
Case 21 — Not a personal ID
Knowing a MAC does not identify a verified human and can be complicated by virtual or randomised interfaces.
Case 22 — Not a street address
An IP address alone cannot reliably establish a unique person’s exact current physical location.
Case 23 — Documentation examples
The IPv4 192.0.2.0/24 and IPv6 2001:db8::/32 documentation spaces allow practice without exposing users’ real addresses.
Case 24 — A diagram needs two layers
A good sketch shows both local link-layer forwarding and IP routing across LANs instead of depicting every address as interchangeable.
23. Three short investigations
An endpoint changes Wi-Fi networks
A fictional laptop travels between two independent wireless LANs. Its IP configuration may change with the new network. The MAC identity used on Wi-Fi may remain interface-associated or be randomised depending on the device’s privacy settings. This case reveals why an exam answer should describe a degree of permanence rather than make an absolute promise about every address.
The student should label which address is used locally and which supports inter-network routing. Nothing in the bit pattern proves the identity of the laptop’s owner. The network layer and the physical person are different concepts.
A request goes beyond the LAN
A student laptop connects to its local gateway. The link-layer frame uses a suitable next-hop MAC destination, while the IP-level destination refers to a remote service. The switch or access point helps with local communication, and routers handle inter-network forwarding. A distant website’s MAC is not copied unchanged onto every hop.
Draw each LAN as a separate boundary, then indicate the IP route between boundaries. This is a more accurate mental model than one long cable with a single eternal destination MAC address.
A privacy claim needs evidence
Someone sees a public IP string and claims it proves exactly where a student lives. That conclusion is unsupported from the address alone. Shared or dynamic allocation, service providers and network routing complicate location and identity inferences. A MAC is also not a personal identity credential, and it can be randomised.
Learning exercises should use reserved documentation examples rather than classmates’ device details. Clear and privacy-respecting examples focus attention on understanding the technology.
24. Frequently asked questions
Does MAC mean 48 bits in every possible system?
The common form taught for comparison is 48 bits; other technical identifiers can exist.
Why is IPv4 32 bits?
It uses four eight-bit octets, usually displayed as dotted decimal numbers.
Why is IPv6 128 bits?
Its address contains eight 16-bit groups in the expanded hexadecimal representation.
Do switches use IP instead of MAC for every local frame?
The syllabus distinguishes local link-layer MAC forwarding from routers’ inter-network IP role.
Is every MAC address permanently fixed?
No. Randomisation and locally administered values are relevant exceptions.
Is every IP address temporary?
No. Some addresses are assigned statically or otherwise remain stable.
Can one device have multiple IP addresses?
Yes. Multiple interfaces or both IPv4 and IPv6 can create multiple assignments.
What does an IPv6 double colon do?
It abbreviates one consecutive run of zero groups, without changing the 128-bit value.
Does an address prove access rights?
No. Addressing and authentication are separate issues.
What does a good SEC comparison include?
Purpose, bit length, typical representation and degree of permanence, with the local versus routed distinction.
25. Official syllabus and connected reading
The official source is the 2027 SEC G3 Computing K349 syllabus, especially outcomes 4.2.4–4.2.5. For illustrative prefixes see RFC 5737 and RFC 3849. Continue with Reference 378 — Home Networks, the MOE V3.0 library and the immutable Clementi teaching standard.
Reviewed 31 October 2026, Singapore. All address examples are documentation or fictional identifiers, never personal device records. Compare the networking layer, format, bit length, typical stability and meaning of a forwarded destination.
26. A complete four-part address comparison exercise
Draw two imaginary home LANs joined through routing equipment. Place an Ethernet computer and a Wi-Fi laptop in LAN A and a remote server in LAN B. Give the local interfaces fictional MAC identifiers, give the devices documentation-style IP addresses appropriate to their respective families, and label the router on the path between LANs. The challenge is not to make every identifier the same, but to show which one matters at which connection.
First count the lengths: a typical MAC is 48 bits, IPv4 is 32 bits, and IPv6 is 128 bits. Second describe local MAC use by the NIC, switch and WAP. Third describe how routers use IP addressing to forward traffic towards another LAN. Fourth discuss whether the laptop’s displayed address could change when it reconnects to a different wireless network. A correct answer gives a qualified permanence comparison rather than inventing certainty.
Now replace the Ethernet link with a Wi-Fi link in the imaginary diagram. The medium changes, but the need for an appropriate local interface and IP routing remains. The final test asks why an observer of one public IP cannot conclude the exact identity and location of the person using the laptop. The address has a technical network function; a personal claim requires additional reliable evidence and appropriate safeguards.
27. Correcting four common exam misconceptions
Misconception one: an IPv6 address is shorter than an IPv4 address because it can be written with ::. Actually IPv6 is 128 bits and its text can be compressed. Misconception two: an IPv4 octet can contain any positive number. In fact, each octet is limited to zero through 255.
Misconception three: every MAC is permanent and every IP changes automatically. Real networks include randomised or locally administered MACs and stable IP assignments, so the comparison should use typical patterns with exceptions. Misconception four: the MAC of a remote website is the destination on the laptop’s local Wi-Fi frame. The local next-hop address and remote IP destination belong to different forwarding scopes.
A student who corrects these misconceptions with a diagram, not only a memorised definition, has built a transferable model of networking. The best answer identifies the interface, layer, bit length, representation and network boundary for every example, then says what the evidence does and does not establish.