I am still a beginner to subnetting, so please forgive the basic nature of the doubt.

When a network is subdivided into a subnet, won't the network broadcast address be the same as the last subnet's broadcast address.

For example, in the subnet, the broadcast address is This is the same as the broadcast address of the whole network with network ID Please help me understand this.

  • Broadcast address will always be in relation with the network address, then, if you have an network address, you can not use because the first is contained on the second one. Once you subnet a network, you need to use the small networks that results from the subnetting operation. I guess you need to read a little more about subnetting Oct 5, 2016 at 18:11
  • Suppose I subnet the network into 16 subnets. The network broadcast address will enable me to send a broadcast message to all the 16 subnets. Meanwhile the subnet broadcast will enable me to send a broadcast to just the last subnet. Is this understanding correct? Oct 5, 2016 at 18:16
  • If you send a broadcast message, you will sended to all the equipments on the subnet where you are located. To place a pc on a subnet you need to configure an IP address and mask. Routers will not let you configure subnets that overlaps each others, then, you need to keep on mind that router divide broadcast domain, thats it, each of the 16 subnets will be a broadcast domain. Oct 5, 2016 at 18:31
  • If you divide a network into smaller networks, each of the smaller networks has its own network broadcast address. Broadcasts cannot, under normal circumstances, cross networks. Routers route between networks, and broadcasts stop at routers. Many years ago, it was the default for routers to allow you to send a broadcast from one network to another, but that was changed. The Internet could not survive under those circumstances, and you do not want just anyone to perform a DoS on your network by flooding it with broadcasts from another network. IPv6 has eliminated broadcasts altogether.
    – Ron Maupin
    Oct 5, 2016 at 19:55

2 Answers 2


When you divide a network into smaller subnets, you no longer have the original network. Cut a piece of cake in half and you have two pieces of cake. The whole cake no longer exists. So it makes no sense to talk about the original network broadcast address.


If it's a class C /28 then you get 16 networks. Each network has a network address, a range of useable hosts, and one broadcast. It usually has one gateway too, usually the first usable address. The first network would be .0 and the range .1-.14 with the broadcast .15. Then net two is .16, range .17-.30, and the broadcast (.31) is always one less than the next network, which is .32 because /28 is nets divided by 16s.

Network Masks

A network mask helps you know which portion of the address identifies the network and which portion of the address identifies the node. Class A, B, and C networks have default masks, also known as natural masks, as shown here:

Class A: Class B: Class C: http://www.cisco.com/c/en/us/support/docs/ip/routing-information-protocol-rip/13788-3.html

  • 1
    Please let network classes rest in peace. Classful networking was killed last century by VLSM and CIDR. See RFCs 1518 and 1519 from 1993.
    – Ron Maupin
    Oct 6, 2016 at 21:22
  • Your classroom is 20 years behind the times. Modern networking has been classless for a couple of decades. The Internet would not function with classful routing.
    – Ron Maupin
    Oct 6, 2016 at 22:59
  • Your definition of the classes is incorrect. Class A addresses start with the first bit a 0, regardless of the mask, Class B addresses have the first two bits as 10, regardless of the mask. Class C addresses have the first thee bits as 110, regardless of the mask. Class D addresses have the first four bits as 1110, regardless of the mask. Class E addresses have the first four bits as 1111, regardless of the mask. Yes, there are default masks for each class, but the masks don't make the class.
    – Ron Maupin
    Oct 6, 2016 at 23:03
  • In fact, the link you provided describes what you have for the classes is incorrect. RIPv1 is a classful routing protocol, but it was obsoleted by the classless RIPv2, many years ago (1998) in RFC 2454.
    – Ron Maupin
    Oct 6, 2016 at 23:07
  • In any case, Network Engineering provides real-world answers to business networking problems, and the real world is classless.
    – Ron Maupin
    Oct 6, 2016 at 23:14

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.