sales[at]netexpresslabs[dot]com, Silicon Valley, California

Home Page

Contact and Order Information

Network Products


Wireless Networking, Routers, Terminal Servers WAN Hardware and Switches

Please contact us for pricing on Wireless Networking, Routers, Terminal Servers, WAN hardware and switches from Cisco, Extreme Networks, etc. There are too many items for us to list. We carry most brands.

Technical Considerations


An IP addresses is a 32-bit number. That means an IP address has 32 placeholders for zero’s or one’s:


These 32-bits can represent 2 to the 32 or 4,294,967,296 unique numbers. IP addresses can be subdivided by periods into four sets of bytes:


A byte is 8 bits. This means each set of eight numbers can represent 2 to the 8 or 256 unique numbers. All four bytes therefore can yield 256x256x256x256 = 4,294,967,296 addresses. We can convert each binary representation of a byte into its decimal equivalent. This is called dotted quad notation and it’s how we normally write IP addresses:

Therefore IP addresses range from: binary or decimal


11111111.11111111.11111111.11111111 binary
or decimal

IP addresses are grouped into networks. The beginning part of an IP address describes a unique network. The ending portion of an IP address describes a unique host. There are three types of networks, called class A, B and C. In a class A network the first byte designates a network and the remaining three bytes describe unique hosts. In a class B the first two bytes describe a network and the last two bytes describe hosts. In a class C the first three bytes describe a network and the last byte describes host. Class A, B and C networks also begin with specific numbers:

Networks:    First Byte:   Network Bytes:   Host Bytes:
Class A      >128          1                 3
Class B      128-191       2                 2
Class C      192-223       3                 1
Reserved     > 223

For example begins with 204 so it is a class C network. Therefore the first three bytes, 204.182.16, describe a unique network and the trailing byte, 2, describes a unique host. We represent this network address as:

And we specify that the first three bytes belong to the network portion of the address by using  what’s called a subnet mask:

Class A Subnet Mask:
Class B Subnet Mask:
Class C Subnet Mask:

Subnets are normally divided on byte boundaries. For example, the class C network would have a subnet mask of

Subnet Mask:

If we translate each byte back into bits the subnet mask would look like:


However, the subdivision of a subnet on the byte boundary is completely arbitrary. You can subdivide a network at any bit. For a class C there are eight possible bit-wise subnet division points:


From these data we can calculate the number of networks and the number of hosts per network we get with each bit-wise subnet

Bits:  Subnet:    Subnet:	Number of   Number of IP’s 
      (Binary)   (Decimal	Networks:    per Network:
0	00000000	0	1		256
1	10000000	128	2		128
2	11000000	192	4		64
3	11100000	224	8		32
4	11110000	240	16		16
5	11111000	248	32		8
6	11111100	252	64		4
7	11111110	254	128		2
8	11111111	255	256		1

The first column is the number of bits used for the subnet. This is shown by a binary number in the second column. The third column is just the binary number from the previous column converted into decimal. This number is used as the last part of the subnet mask. Note that addresses ending on network boundaries (multiples of the number of bits used for the subnetting) are reserved for subnetwork address and are therefore not available for host addresses. The last number of each subnet is used as the broadcast address. This means there are two less hosts per network then listed above. Therefore the actual numbers are:

Bits:  Subnet:    Subnet:  Number of   Number of Hosts 
      (Binary)   (Decimal) Networks:    per Network:
0	00000000	0		1		254
1	10000000	128		2 		126
2	11000000	192		4 		62
3	11100000	224		8 		30
4	11110000	240		16 		14
5	11111000	248		32 		6
6	11111100	252		64 		2
7	11111110	254		128 		0
8	11111111	255		256 		0

So for example, say we have a class C license,,  and we want to break it into 16 Subnetworks with 14 hosts per
network by using 4-bit subnetting. From the table we can see that we would use a subnet mask of:

This would yield 16 networks:

In this example, the twelfth subnetwork would be It would have the following addresses:

Network Address
Subnet Mask
Broadcast Address:
Host Addresses:

We have listed all 16 subnets with a netmask of for the Class C Network This yields 16 subnets with 224 hosts. To view this example click on this link. To divide your class C address into subnetworks you simply have to choose the proper Subnet Mask and Broadcast address and use these in your configuration files.

There are Four types of network back plates.

Combo cards have all three types. These have various names:

  1. 100Base-T/High-Speed Twisted Pair/RJ-45:
  2. 10Base-T/10Base-10/Twisted Pair/RJ-45/:
  3. BNC/Thin Net/Thin Coax/10Base-5:
  4. AUI/Thick Coax/10Base-2: AUI is the 15-pin interface/adapter extending from the vampire tap.

To create an Office LAN (Local Area Network) you'll need:

  1. One 8-port 10Base-T or 100Base-T HUB
  2. One Twisted Pair or One combo network card for each PC
  3. Some twisted pair cable
  4. LAN software: Win NT, Windows for Workgroups, Novell or UNIX TCP/IP

For Networking two PCs for home you'll need:

  1. Two Twisted Pair or combo network cards
  2. One piece of Cross-over twisted pair cable
  3. LAN software: Win NT, Windows for Workgroups, Novell or UNIX TCP/IP

Hubs restore, boost and retime packets on your network. Hubs have 8, 12, or 24 10Base-T ports to connect 8, 12 or 24 computers respectively. Hubs have one extra BNC or Transceiver ports to link to other hubs. Hubs can be linked to add more computers. Most people use dumb hubs. Smart hubs are only needed to manage very large and complex networks via SNMP. (Simple Network Management Protocol).

Copyright (c) 1989 - 2008 Net Express All Rights Reserved.