Subnetting is the thing people say is hard until suddenly it isn’t. The shift happens when you stop trying to memorize formulas and start building a feel for the math.
Why Do It By Hand
Exam questions give you 90 seconds per question. A calculator isn’t available. More importantly, if you understand what’s happening, you don’t need one.
The Powers of Two
Everything in subnetting comes back to powers of 2. Know these cold:
2^1 = 2
2^2 = 4
2^3 = 8
2^4 = 16
2^5 = 32
2^6 = 64
2^7 = 128
2^8 = 256
A /24 network has 256 addresses. A /25 splits it in half: two networks of 128. A /26 gives four networks of 64. Every bit you borrow from the host portion doubles the number of networks and halves the size of each.
The Block Size Method
Given a subnet mask, find the interesting octet — the one that isn’t 255 or 0. Subtract it from 256. That’s your block size.
Example: 255.255.255.192 → interesting octet is 192 → block size is 64.
Networks start at 0, 64, 128, 192. Given IP 10.0.0.100, it falls in the 10.0.0.64 network. Broadcast is .127. Usable range is .65 to .126.
That’s the whole method. No binary conversion needed for most problems.
What to Drill
The fastest way to get comfortable is repetition. Pick a random IP and prefix length, work out the network address, broadcast, and usable range. Do it until the block sizes feel obvious and the math takes two seconds.
Subnetting stops being a problem and becomes a tool once it’s automatic. At that point you’re not doing subnetting, you’re designing networks — which is the actual skill the CCNA is testing.