A printer can keep the same address for years while the laptop beside it gets an address automatically whenever it joins the network. Both devices may sit on the same subnet and use the same router. The difference is not where they are on the network. It is how their addresses are assigned and whether anyone is expected to keep those addresses stable.
A static IP address is deliberately kept stable for a host, interface, service, or connection. That stability may come from manual configuration or from a provider or network policy designed to return the same address. A dynamic IP address is allocated automatically from a pool without a promise that the assignment will remain permanent.
Neither term tells you whether the address is public or private. A printer can have a static private address that never leaves the local network. A residential router can have a dynamic public address that remains unchanged for months. Public/private and static/dynamic describe different properties, and mixing them together is the source of much of the confusion around this topic.
If you need the addressing basics first, see What Is an IP Address? and Public vs Private IP Address. This guide focuses on assignment behavior: what keeps an address stable, what can make it change, and what that means for DHCP, CGNAT, DDNS, remote access, gaming, security, and IPv6.
01
The short version
A static address is kept stable on purpose. A dynamic address is assigned automatically and may change, although it may also remain unchanged for a long time. Static does not mean public, and dynamic does not mean private. A static address does not make a connection faster, lower latency, or improve security by itself. Under CGNAT, the address shown on a router’s WAN interface may not even be the public IPv4 address seen by the Internet, so the question “Is my WAN IP static?” can be the wrong question to ask.
| Property | Static IP | Dynamic IP |
|---|---|---|
| Assignment | Fixed by configuration or a policy intended to preserve the address | Allocated automatically from a pool |
| Expected stability | Intended to remain stable until an administrator or operator changes it | May remain stable, but there is no guarantee |
| Can be private? | Yes | Yes |
| Can be public? | Yes | Yes |
| Typical examples | Servers, gateways, printers, cameras, business Internet endpoints | Laptops, phones, guest devices, many residential WAN services |
| Guarantees Internet reachability? | No | No |
| Improves speed? | No | No |
| DNS implication | A record can usually remain unchanged | DDNS or another naming mechanism may be useful if the address must be reachable by name |
02
What is a static IP address?
A static IP address is an address deliberately kept stable for a particular host, interface, service, or connection. The important word is stable, not permanent.
Static addresses can change. An ISP may renumber a block. A company may redesign its network. A service can move to a different access technology. Equipment may be replaced, an address plan may be consolidated, or a customer may migrate to another product. None of those events make the original address “not static.” They simply show that static addressing is an operational policy, not an immutable property of the protocol.
There are two common ways to create that stability.
Manual static configuration
The address is configured directly on the endpoint. A server, for example, may be given its IP address, prefix length, default gateway, and DNS settings in the operating system or network appliance configuration. The host does not consult a DHCP server to obtain that address.
That can be useful for infrastructure that must come up independently of DHCP, but it also shifts responsibility to the administrator. If the manually configured address overlaps an active DHCP pool, another device may eventually be offered the same address. Central IPAM or monitoring systems may discover the address, but they do not automatically become the source of truth unless the configuration is documented there.
Reserved or provider-controlled stable assignment
The endpoint still requests an address automatically, but the system answering the request is configured to return a specific one. On a LAN, that is often a DHCP reservation. At an ISP, the same outcome may be produced by subscriber provisioning tied to a circuit, session, account, or other operator policy.
From the user’s perspective, both methods can look identical: the device has the same address today that it had last month. Operationally, they are not the same, and the difference matters when troubleshooting.
03
What is a dynamic IP address?
A dynamic IP address is assigned from a pool through an automated allocation mechanism without a guarantee that the assignment will remain fixed.
That does not mean the address changes whenever the device reconnects. Many dynamic addresses remain unchanged for weeks or months. A broadband subscriber can reconnect several times and still receive the same public IPv4 address. A laptop can repeatedly renew the same private address from the home router. The address is still dynamic if the system is free to assign a different one later.
Several factors influence how long a dynamic assignment appears stable: lease or session duration, how renewal works, how long the client is offline, the client’s identity, subscriber-management policy, pool utilization, and the design of the access network. There is no universal ISP rotation interval.
The useful question is not “Has the address changed?” but “Is there a policy that guarantees it will stay?” If the answer is no, observed stability alone does not make the assignment static.
04
Static/dynamic and public/private are separate dimensions
Public/private describes address scope and routability. Static/dynamic describes assignment and expected stability. The two dimensions are independent.
Two independent dimensions
Predictable local address. Not reachable from the Internet by itself.
Simplifies client onboarding. Not reachable from the Internet by itself.
Stable remote target, subject to routing and filtering.
Provider-managed from a pool, subject to routing, NAT, CGNAT and filtering.
| Combination | Example | What it solves | Internet reachable by itself? |
|---|---|---|---|
| Private + static | Printer at 192.168.1.20 | Gives local systems a predictable address | No |
| Private + dynamic | Laptop receiving 192.168.1.137 from DHCP | Simplifies client onboarding | No |
| Public + static | Business gateway with a fixed public IPv4 | Gives remote systems a stable target | Potentially, subject to routing and filtering |
| Public + dynamic | Residential router with an ISP-assigned public IPv4 | Lets the provider manage addresses from a pool | Potentially, subject to routing, NAT, CGNAT, and filtering |
A static private address does not make a camera or NAS reachable from the Internet. It only gives that device a predictable address inside the local routing domain. Remote reachability still depends on the rest of the path: public addressing, NAT or port mapping where applicable, routing, firewalls, and the service itself.
05
How DHCP actually works
DHCP is commonly associated with dynamic addressing, but the two are not synonyms. DHCP is a protocol. Dynamic allocation is one use of it; reserved assignments are another.
In IPv4, a client that needs configuration can discover DHCP servers, receive one or more offers, request an offered address, and receive an acknowledgement containing the address, lease time, and other parameters such as the default gateway and DNS resolvers. RFC 2131 defines the core protocol, while RFC 2132 defines many of the options used with it.
The lease is often misunderstood. A lease is permission to use an address for a period of time. Before that period ends, the client normally tries to renew the existing lease. If direct renewal fails, it later enters a rebinding phase and tries more broadly. A lease reaching its expiry point does not mean the network has scheduled a new address for the client. The normal outcome is often that the same address continues.
A reboot does not change that logic. A client that retained information about its previous lease can ask to reuse the same address, and a server may grant it again if policy and availability allow. That is why a “dynamic” address can survive many restarts.
DHCP lease lifecycle
The existing lease is extended. The address continues unchanged.
Client rebinds and asks more broadly. A server may still return the same address, or a different one, depending on policy and availability.
Two DHCP concepts are worth separating:
- A pool or scope is the set of addresses the server is allowed to allocate dynamically.
- A reservation associates a particular address with a particular client identity. Whether the reserved address sits inside or outside the ordinary allocation range depends on implementation and local policy.
The practical result is simple: DHCP tells you how the address was delivered. It does not automatically tell you whether the operator intends that address to be dynamic, sticky, or reserved.
06
DHCP reservation vs manual static configuration
A manual static address and a DHCP reservation can both leave a device on the same address every day. The mechanism behind that stability is different.
With manual configuration, the endpoint holds the settings. The network does not need a DHCP server to supply the address. With a reservation, the endpoint remains a DHCP client and the DHCP server remains responsible for the assignment.
| Property | Manual static configuration | DHCP reservation | Dynamic DNS |
|---|---|---|---|
| Where the setting lives | Host or interface | DHCP server | DNS provider, updated by a client |
| What it stabilizes | Interface address | Address returned to that client | Hostname, not the address |
| Typical use | Infrastructure, servers, devices that must boot independently of DHCP | Printers, NAS devices, cameras, stable LAN clients | Remote access to a changing public address |
| During DHCP server outage | Existing configuration is unaffected | Existing clients can continue while their leases remain valid; new or expired clients need DHCP service | Not applicable |
| Central visibility | Only if documented or discovered | Yes, in DHCP records | Yes, in DNS records |
| Solves inbound reachability? | No | No | No |
For a home or small office, a reservation is often easier to maintain because addressing stays centralized. Manual configuration remains appropriate when the device must be able to start with the correct network settings even if DHCP is unavailable.
07
Why a dynamic IP can stay the same for months
A public address that has not changed for six months may still be dynamic. Providers often produce what operators informally call sticky dynamic behavior: the address comes from an automated system, but normal reconnects keep returning the same value.
That can happen because a lease keeps renewing, because the subscriber is recognized through the same CPE or session identity, because the pool manager prefers the previous address when it is free, or because the subscriber remains attached to the same part of the access network. None of those mechanisms creates a contractual guarantee.
Sticky dynamic: three observations, same address
Public subscriber addressing also does not always use DHCP. Fixed-line access may use DHCP or PPP/PPPoE. Mobile networks use packet-core subscriber mechanisms. Other access technologies may have their own provisioning systems. A DHCP lease timer on one router and an active PPP session on another describe different delivery mechanisms, even if both customers ultimately receive dynamic public addresses.
08
What can make a dynamic public IP change?
A dynamic public address may change when a lease or session is reassigned instead of renewed, when the connection stays offline long enough for the old address to be returned to the pool, when customer equipment changes, or when the provider moves the subscriber to a different access platform. Network maintenance, pool rebalancing, renumbering, aggregation changes, or migration to another service can also trigger a new assignment.
A router reboot is not a reliable trigger. A short restart may recover the same session or lease. A longer outage may make reassignment more likely, but even then the provider can still return the same address. Reboot behavior is useful only as one observation, not as proof of the assignment policy.
09
Static public IP vs static private IP
A static private address and a static public address solve different problems.
Inside a local network, stable private addressing is useful when other systems refer to a device by address. Printers, NAS devices, cameras, NVRs, backup jobs, firewall rules, host entries, and port-forwarding rules are easier to manage when the target does not move. A device at 192.168.1.20 can be perfectly stable for years and still remain meaningful only inside that network.
A static public address solves a different problem: it gives external systems a stable Internet-facing target. That is useful for business Internet connections, VPN gateways, self-hosted services, mail infrastructure, partner allowlists, site-to-site tunnels, managed database access, or any workflow where another organization expects traffic to come from or return to one known address.
Even then, a server does not automatically need a static public IP. DNS updates, load balancers, reverse proxies, provider-managed public mappings, outbound tunnels, and overlay networks can all provide stable reachability in other ways. The better question is: What does the remote side need to keep pointing at, and what keeps that target stable?
10
CGNAT changes the question
Carrier-grade NAT breaks an assumption many users do not realize they are making: the address on the router’s WAN interface may not be the public IPv4 address seen on the Internet.
In a common CGNAT design, the provider gives the customer router a non-globally routable WAN address and translates many subscribers through a smaller pool of public IPv4 addresses. RFC 6598 reserves 100.64.0.0/10 as Shared Address Space for this purpose. Some deployments use RFC 1918 space internally instead. 100.64.0.0/10 is not RFC 1918 private space; it is a separate allocation with a specific role between subscriber equipment and carrier NAT.
If a router shows 100.64.13.7 while an external lookup reports a different public IPv4 address, that is a strong indication consistent with CGNAT. Provider architecture can vary, so the comparison is evidence, not an absolute proof by itself.
CGNAT address boundary
A fixed WAN address on the subscriber side does not create a fixed public IPv4 address. The provider could keep 100.64.13.7 stable indefinitely and still translate the customer’s traffic through a shared public address upstream. Port forwarding on the home router stops at the subscriber-facing side of the carrier NAT unless the provider has also provisioned the carrier layer for inbound traffic.
A dedicated static public IPv4 therefore requires provider-side provisioning. The provider may assign the subscriber a stable globally routable address directly or use an appropriate one-to-one mapping. A stable public IP-and-port mapping at the carrier NAT can create a persistent inbound endpoint, but that is not the same thing as giving the subscriber a full dedicated static public IPv4.
This is also why a single external IP Lookup cannot tell you whether the address on your service is statically assigned. It can show the public address currently observed and its network context. The assignment policy lives inside the provider’s provisioning systems.
11
Can DDNS replace a static IP?
Dynamic DNS can replace a static address in some remote-access designs, but only for the naming problem.
A DDNS client watches the current public address and updates a DNS record when that address changes. Remote users connect to a hostname instead of memorizing the address. The name stays stable while the address behind it changes.
DDNS update cycle
That solves name-to-address continuity. It does not create a public route, open a firewall, traverse a carrier NAT, or make a blocked port reachable. If a connection is behind CGNAT, DDNS can successfully update a hostname to the observed public address and inbound connections can still fail because that shared carrier address is not mapped back to the subscriber.
There is also a timing window. After an address changes, the update client has to detect the change, the authoritative DNS data must be updated, and resolver caches have to age out according to their TTLs. During that interval, some users may still receive the previous address.
For ordinary home remote access, that may be acceptable. For strict availability targets or source-address allowlists, it can be a design constraint.
12
Do you need a static IP for remote access?
Most remote-access setups do not literally require a static public IP. They require a stable way to find and reach the service.
A static public address makes that simple: DNS records can remain unchanged, remote users always aim at the same target, and address-based allowlists do not need to be updated. That is useful for a NAS, an NVR, a home lab, a self-hosted application, a VPN gateway, or a remote desktop entry point.
Other architectures can meet the same requirement. DDNS works when the connection has a usable public address and inbound traffic is allowed. An outbound tunnel can establish a path to a stable rendezvous server. Overlay networks can connect endpoints without requiring a fixed public address on either side. Reverse proxies and hosted relays can also make an internal service reachable without exposing it directly.
Before choosing any of those approaches, answer a simpler question first: Does this connection have usable public reachability at all? If the router is behind CGNAT, paying attention to static versus dynamic addressing on the customer WAN side will not solve the inbound-path problem.
13
Gaming, speed, and latency
A static address does not lower ping because it is static. Latency depends on the path packets take: access technology, local queueing, congestion, physical distance, peering, transit, routing policy, and the location of the game server all matter. Address-assignment policy does not reduce any of those delays.
If an ISP’s static-IP product uses different routing, aggregation, or service architecture, performance can differ between products. In that case, the difference comes from the service architecture, not from the address being static.
The same applies to throughput. Static addressing does not raise the access rate, remove congestion, or make the endpoints process packets faster. A dynamic address is not slower because it is dynamic.
Where a static public IP can help is hosting. If other players connect directly to a server you run, keeping the public target stable makes configuration easier. That is a reachability and operational convenience, not a latency advantage.
14
Security and privacy
Is a static IP more secure?
Address stability is not a security control. A fixed public address can host a well-protected service behind strict firewall rules, strong authentication, encryption, and careful patching. A changing address can expose an insecure service just as easily.
Internet-wide scanning is routine, so changing addresses should not be treated as a substitute for filtering or authentication. A stable address does let an attacker return to the same target once it has been identified. The same stability also helps defenders build reliable allowlists, ACLs, monitoring baselines, logging rules, and source-attribution workflows.
What matters is the service and its exposure: what is listening, what the firewall permits, how authentication works, whether software is patched, and whether transport is encrypted.
Privacy
A changing public address can reduce the usefulness of one long-lived IP as a continuity marker. That effect is limited. Providers can retain assignment records, and websites have many other ways to recognize users or sessions, including accounts, cookies and local storage, browser characteristics, and behavior.
Dynamic addressing is not anonymity. A stable address does not reveal a person’s identity by itself either. Public IP data operates at the network level. IP Geolocation explains what location data associated with an address can and cannot support.
If you use a VPN because you want websites to see a different network egress, that is a separate mechanism. VPN Leak Test can help verify whether the paths you expect to be tunneled actually are.
15
Static and dynamic addressing in IPv6
IPv6 makes the words “static” and “dynamic” less useful unless you say what is supposed to stay stable.
A residential ISP commonly delegates a prefix to the customer router. That prefix may remain stable for a long period or may change according to provider policy. DHCPv6 Prefix Delegation is defined in RFC 8415; RFC 7084 covers IPv6 customer-edge requirements, and RFC 9818 updates the DHCPv6 prefix-delegation behavior expected from customer-edge routers. Prefix size is provider-specific. There is no single universal residential size.
Inside that delegated prefix, a host can create or receive addresses in several ways. SLAAC, defined in RFC 4862, lets a host form addresses from an advertised prefix. RFC 7217 describes stable, opaque interface identifiers, and RFC 8064 recommends stable identifiers that do not embed long-lived link-layer information by default. DHCPv6 can also provide repeatable assignments through server policy and DHCPv6 client/Identity Association state. The operational result can resemble an IPv4 reservation, although the identification model is different.
Temporary addresses add another layer. RFC 8981 allows a host to create rotating temporary addresses for outbound connections. That means the same LAN can simultaneously have a delegated prefix that has not changed for months, a server with a deliberately stable address inside that prefix, and laptops whose preferred outbound source addresses rotate over time.
Calling the whole connection “static IPv6” or “dynamic IPv6” hides those distinctions. Ask what is stable: the delegated prefix, a particular interface address, an identifier within the prefix, or the DNS name used to reach the service.
Cloud environments have a similar layering problem. A virtual machine may have a stable private address inside a virtual network while a public address is associated through provider infrastructure. That public association may be ephemeral or persistent, and the guest operating system may never see the public address directly. Provider documentation is the right source for the persistence rules of a specific platform.
Static public address pricing also varies. Some business plans include one or more fixed addresses, some providers sell them as an add-on, and some consumer services do not offer them. On CGNAT-based services, the first useful upgrade may simply be obtaining a dedicated public address, whether static or not.
16
How to tell whether your public IP is static or dynamic
A single lookup cannot prove that an address is static. It tells you what the public address is at that moment.
The strongest evidence comes from the provider. Check the service contract, provisioning portal, product documentation, or ask support whether the address is provisioned as fixed. If the provider explicitly calls it dynamic, months of observed stability do not turn that policy into a static assignment.
Observation over time is still useful. Record the public address periodically and compare the results. If it changes, you have strong evidence of dynamic assignment. If it stays the same, you have evidence of stability, but not proof of a guarantee.
The router’s WAN status adds another piece of the picture. It may show a DHCP lease, a PPP session, or a manually configured WAN address. A DHCP lease proves that DHCP delivered the address; it does not, by itself, prove that the provider treats the assignment as dynamic rather than reserved or sticky.
If the WAN interface shows 100.64.0.0/10 or an RFC 1918 address while an external service reports a different public IPv4 address, there is an upstream NAT layer. With 100.64.0.0/10, that is a strong indication of carrier-grade NAT. At that point, the stability of the router’s WAN address and the stability of the public IPv4 address are separate questions.
| Observation | Reasonable interpretation |
|---|---|
| Same public IPv4 for six months | Could be static or sticky dynamic; not enough evidence to distinguish |
| Public IPv4 changes after reconnect | Strong evidence of dynamic assignment |
| Provider documentation specifies a fixed IPv4 | Static public assignment |
| Router WAN is 100.64.x.x and external IPv4 differs | Strong indication consistent with CGNAT; a fixed WAN address would not create a static public IPv4 |
| Router WAN is RFC 1918 and external IPv4 differs | An upstream NAT layer exists; investigate the topology before deciding whether it is provider CGNAT or local double NAT |
| DDNS hostname keeps working after an address change | The name is stable; the address is dynamic |
| Router WAN shows a DHCP lease | DHCP delivered the address; that alone does not establish whether the assignment policy is dynamic, sticky, or reserved |
| IPv6 prefix stays the same while device addresses rotate | The delegated prefix can be stable while temporary interface addresses change |
What IPRevealed can and cannot tell you
IP Lookup can show the public address observed for the current request and the network context associated with that address. IP Lookup Explained describes those evidence layers in more detail.
It cannot read the provider’s subscriber-assignment policy from the address itself. That policy is not carried in the packet and is not published in public registries at subscriber granularity. Repeated observations can show practical stability. Provider information is what establishes whether that stability is actually guaranteed.
Check what is currently observed
See the public address, ASN and routing context an external service currently sees for your connection, then compare it against your router’s WAN address.
Open IP Lookup17
When a static IP is actually useful
A static public address is useful when another system needs a stable network identity. Common examples include site-to-site VPNs, partner allowlists, mail infrastructure, remote-access gateways, managed database access rules, monitoring systems, and services whose DNS records should not depend on a DDNS update client. Providers also commonly pair configurable PTR/rDNS service with dedicated or static address products, although that is a provider policy rather than a protocol requirement.
A static private address or DHCP reservation is useful when local configuration points directly at a host: printers, NAS devices, cameras, firewall rules, port-forwarding rules, backup jobs, monitoring systems, or scripts. If the device must boot with the correct address even while DHCP is unavailable, manual configuration is the safer choice.
You probably do not need a static public address for ordinary outbound browsing, streaming, calls, or gaming as a client. You may also not need one when an outbound tunnel, overlay network, or hosted relay already provides remote access. And if the connection is behind CGNAT, the immediate requirement may be a usable public address rather than a static one.
18
Common misconceptions
“Static IP means public IP.”
A static address can be private or public. Many printers, servers, cameras, and gateways use fixed private addresses.
“Dynamic IP means private IP.”
Residential public IPv4 service is often dynamically assigned and still fully public when the provider gives the subscriber a routable address.
“A dynamic IP changes every day.”
Some services change often; others can keep the same assignment for months. There is no universal schedule.
“Rebooting the router always changes the public IP.”
A reboot may return the same lease or session. It is not a reliable test by itself.
“If my IP hasn’t changed for months, it must be static.”
Long-term stability fits both static assignment and sticky dynamic behavior. Provider policy is what distinguishes them.
“Static IP is faster.”
Address stability does not increase throughput.
“Static IP improves ping.”
Latency is determined by the network path and conditions along it, not by whether the address is static.
“Static IP is more secure.”
Security depends on exposure, filtering, authentication, patching, and encryption. Address stability is not a security mechanism.
“Dynamic IP makes me anonymous.”
It does not. Providers retain assignment records, and websites can identify users through many non-IP signals.
“DDNS gives me a static IP.”
DDNS keeps a hostname pointed at a changing address. It does not stop the address from changing and does not create reachability through CGNAT.
“A DHCP reservation is the same as manual static configuration.”
The result can look similar, but the configuration lives in a different place. A reservation depends on DHCP; manual configuration does not.
“A fixed WAN address behind CGNAT gives me a static public IP.”
It does not. The carrier still controls the public side of the translation boundary.
19
FAQ
What is the difference between a static and dynamic IP address?
A static address is deliberately kept stable through configuration or assignment policy. A dynamic address is allocated automatically without a guarantee that it will remain fixed. The distinction is about expected stability, not how often the address happens to change.
Is a static IP always public?
No. Static private addresses are common on local networks. A printer or NAS can keep the same private address for years without being directly reachable from the Internet.
Can a private IP be static?
Yes. You can configure it manually on the device or create a DHCP reservation so the network keeps assigning the same address.
Does a dynamic IP change every day?
Not necessarily. Some dynamic addresses remain unchanged for weeks or months because leases and sessions keep renewing and the provider keeps returning the same assignment.
Why has my dynamic IP stayed the same for months?
The provider may be exhibiting sticky dynamic behavior. Your lease or session keeps renewing, the subscriber identity has not changed, and the old address remains available. That shows stability in practice, not a guaranteed static assignment.
Does rebooting my router change my public IP?
Sometimes, but not reliably. A brief reboot often returns the same address. Longer outages or provider-side changes may increase the chance of a new assignment.
Is a static IP faster?
No. Throughput depends on the access service, network conditions, and endpoints. Address stability does not make data move faster.
Is a static IP better for gaming?
Not for ordinary play. It can make hosting easier because other players have a stable address to connect to, but it does not reduce latency.
Is a static IP more secure than a dynamic one?
Neither is inherently more secure. Security depends on firewall policy, service exposure, authentication, software maintenance, and encryption.
Can DDNS replace a static public IP?
In many remote-access setups, yes, as long as the connection already has usable public reachability. DDNS keeps a hostname updated when the address changes. It cannot bypass CGNAT or provider filtering.
How can I know whether my public IP is static?
Check authoritative provider information: the service contract, provisioning portal, product documentation, or direct provider confirmation. Monitoring the address over time can show that it is stable, but only the provider can tell you whether that stability is guaranteed.
Does IPv6 use static or dynamic addresses?
Both can exist on the same connection. A provider may delegate a stable prefix while individual devices use temporary addresses that rotate for outbound traffic. Servers inside the same prefix can also use deliberately stable addresses. With IPv6, always ask what is stable: the prefix, a specific interface address, or the DNS name.
20