What a subnet tells a device
An IPv4 address identifies a network interface, but the address alone does not tell a device which other addresses are local. It also needs a subnet mask. Together, the address and mask define a subnet: a range of addresses that the device treats as reachable on its local network. Traffic destined for an address outside that range normally goes to a router, often called the default gateway.
Consider a computer with address 192.168.10.25 and mask 255.255.255.0. Its local subnet runs from 192.168.10.0 through 192.168.10.255. A device at 192.168.10.40 is within that range; one at 192.168.11.40 is not. This distinction is about the computer’s network settings, not how close the devices are physically. Two devices plugged into the same switch can still be unable to communicate as expected if their addresses or masks are configured incorrectly.
Reading a CIDR prefix
CIDR notation expresses a mask as a slash followed by the number of leading bits used for the network portion of an address. For example, 192.168.10.25/24 means that the first 24 of the address’s 32 bits identify the subnet. The remaining eight bits identify addresses within it. A /24 corresponds to the mask 255.255.255.0.
A larger prefix leaves fewer bits for addresses within the subnet. For example, a /26 corresponds to 255.255.255.192. It divides a range that would fit within one /24 into four blocks of 64 addresses. Those blocks begin at 192.168.10.0, 192.168.10.64, 192.168.10.128 and 192.168.10.192. Because these boundaries matter, a /26 subnet cannot start at an arbitrary address such as 192.168.10.50.
Finding the usable range
Suppose a device has address 192.168.10.70/26. The address falls in the block from 192.168.10.64 through 192.168.10.127. In a conventional IPv4 subnet, the first address, 192.168.10.64, identifies the network. The last, 192.168.10.127, is the broadcast address. That leaves 192.168.10.65 through 192.168.10.126 for hosts, or 62 usable addresses.
To check another address, find which block it belongs to using the same prefix. Address 192.168.10.100/26 is in the same block as .70. Address 192.168.10.130/26 is in the next block, even though only the final part of the address differs. When doing this check, verify the mask on both devices: a mismatched mask can cause them to disagree about whether a destination is local.
The usual subtraction of two addresses for the network and broadcast addresses is a helpful rule for ordinary IPv4 host subnets, but it is not universal. Specialized /31 point-to-point links and /32 single-address routes have different uses. For a first small-network plan, conventional subnets such as /24 or /26 are easier to reason about.
Choosing a size for a small network
Start by counting the interfaces that will need addresses, including computers, phones, printers, access points and the router interface on that subnet. Allow room for likely additions and for any addresses reserved outside a DHCP pool. A /26 provides 62 conventional host addresses; a /24 provides 254. Choosing a subnet much larger than needed is not automatically a problem, but smaller subnets can help separate groups when paired with appropriate routing and access rules.
Keep the subnet plan separate from the DHCP plan. The subnet defines the full address range and local boundary. A DHCP server assigns settings from a pool you configure within that range. For example, a /26 might reserve several usable addresses for infrastructure and offer the rest through DHCP. Make sure the gateway address and DHCP pool both fall within the intended subnet, and avoid assigning the same address twice.
A quick configuration check
When a device cannot reach a local service, record its IP address, prefix or mask, default gateway and the destination address. Calculate the subnet containing each address rather than relying on a visual similarity between them. If the destination is outside the device’s subnet, check that a suitable gateway and route exist. If it should be inside, look for an incorrect mask, a duplicate address or a device connected to the wrong network segment.
Subnet calculations answer one narrow but important question: which destinations does an interface consider local? They do not guarantee that traffic will pass. Switch configuration, firewall rules, wireless isolation and routing can still affect connectivity. Treat the subnet check as an early step that helps you decide where to investigate next.
