Subnet calculator

The number after the slash is how many bits at the front of the address are fixed for the network - the rest are free for hosts. So /24 locks the first 24 bits (255.255.255.0) and leaves 8 bits, or 256 addresses, for the block.

For IPv4 it's the total addresses minus two - one for the network address and one for the broadcast - so a /24 has 256 - 2 = 254. The exceptions are /31 (2 usable, a point-to-point link) and /32 (a single host). IPv6 doesn't reserve those two, so every address in the block is usable.

The network address is the first in the block (all host bits 0) and names the subnet itself; the broadcast address is the last (all host bits 1) and reaches every host at once. Neither is assigned to a device, which is why they don't count as usable hosts on IPv4.

A wildcard mask is the subnet mask flipped - subtract each octet from 255. A /26 mask of 255.255.255.192 becomes a wildcard of 0.0.0.63. A subnet mask marks the network bits; a wildcard marks the host bits, which is why Cisco ACLs and OSPF use it to match a range.

Count the 1-bits in the mask written in binary - that count is the CIDR number. 255.255.255.0 is twenty-four 1s, so /24; 255.255.255.240 is /28; 255.255.255.192 is /26. The other way, /24 means the first 24 bits are 1s, which fills the first three octets: 255.255.255.0. The reference table above lists every mask and its prefix.

Pick the smallest prefix whose usable-host count covers what you need, remembering IPv4 loses two addresses per subnet. About 500 hosts needs a /23 (510 usable); 100 hosts fits a /25 (126); 1000 hosts needs a /22 (1022). The rule is usable hosts = 2^(32 - prefix) - 2.

Every extra bit of prefix doubles the number of subnets, so the count is 2 raised to the difference in prefix length. A /16 holds 256 /24s (2^8); a /24 holds 8 /27s (2^3); a /22 holds 4 /24s. The same math runs in reverse when you aggregate routes.

A /64 holds 2^64 addresses - about 18.4 quintillion - and IPv6 reserves no network or broadcast address, so every one is usable. A /64 is the standard size for a single LAN. A /56 is 256 /64s and a /48 is 65,536 /64s, which is why ISPs hand those larger blocks to sites that need many subnets.

A /30 gives 2 usable hosts and was the classic choice for a router-to-router link. A /31 (RFC 3021) squeezes the same point-to-point link into just 2 addresses with no waste, and modern gear supports it. A /32 is a single host - one exact address - used for loopbacks, host routes, and firewall rules that target one machine.

Three IPv4 blocks are reserved for private networks and never routed on the public internet: 10.0.0.0/8, 172.16.0.0/12, and 192.168.0.0/16. Home and office LANs almost always use one of these. The IPv6 equivalent is the unique-local range fc00::/7 (in practice fd00::/8).

Apply the subnet mask to both addresses - keep only the network bits - and compare. If both give the same network address, they are on the same subnet and can talk directly; if not, traffic goes through a router. Enter either address with its prefix above and the tool shows the exact network and host range.