Bits vs Bytes: The Most Confusing Concept in Networking, Explained

There is one convention in computing that trips up more people than any other, and it comes down to whether a letter is capitalised. Mb and MB look almost identical. They differ by a factor of eight.

The short answer

A bit is a single 1 or 0. A byte is a group of eight bits. Lowercase b means bits, uppercase B means bytes. Network speeds use bits; file sizes use bytes.

What a bit actually is

A bit — short for binary digit — is the smallest possible piece of information: a single yes or no, on or off, 1 or 0. It is the atom of computing. Everything else is built from stacks of them.

One bit on its own carries almost nothing. Two bits can represent four states, three bits eight, and so on, doubling each time. To be useful you need a group.

Why a byte is exactly eight bits

Eight bits gives 2⁸ = 256 possible combinations, which was enough to represent every character an early computer needed: the alphabet in both cases, digits, punctuation and control codes. Eight is also a power of two, which makes address arithmetic clean in hardware.

Early machines did experiment with other sizes — 6, 7, 9 and even 12-bit bytes existed. The 8-bit byte won decisively with the IBM System/360 in the 1960s and has been the universal standard ever since.

The number 8 does not come from the decimal system, which is why you divide by 8 and never by 10.

Reading any abbreviation

Once you know the rule, every abbreviation decodes the same way. Read it right to left:

Why networks chose bits and storage chose bytes

It is not arbitrary. The two fields measure genuinely different things.

A network transmits a stream. Bits go down the wire one after another, and they do not necessarily align to byte boundaries — a packet has headers, error-correction bits and framing that are not part of any file. Measuring in bits describes what the wire is really doing.

Storage holds files. A file is a definite number of bytes, and every byte is addressable. Measuring in bytes describes what you actually have.

So when your browser reports 12 MB/s, it is telling you how fast the file is growing on disk. When a speed test reports 96 Mbps, it is telling you how fast the wire is carrying bits. Same connection, two honest answers.

The second trap: 1,000 or 1,024?

Having settled bits versus bytes, there is a smaller sequel. The prefix 'mega' also has two meanings in practice:

• Decimal (SI): 1 MB = 1,000,000 bytes. Used by network speeds, hard drive manufacturers and the official standard.
• Binary (IEC): 1 MiB = 1,048,576 bytes. Used by Windows, Steam and most operating systems — but usually still labelled 'MB'.

The gap is 2.4% at the megabyte level and 7.4% at the gigabyte level. It is why a '1 TB' drive shows up as 931 GB in Windows: the manufacturer counted in decimal, Windows displays in binary, and neither is lying.

The download time calculator lets you pick which convention your file size uses, because mixing them is the most common source of a wrong estimate.

Putting it to work

Three conversions cover essentially every situation you will meet:

1. 1Connection speed to download speed: divide Mbps by 8. A 300 Mbps plan downloads at 37.5 MB/s.
2. 2Download speed to connection speed: multiply MB/s by 8. A transfer at 9 MB/s is using 72 Mbps.
3. 3Gigabits to megabytes: multiply by 1,000, then divide by 8. 2 Gbps = 2,000 Mbps = 250 MB/s.

The takeaway

Lowercase b is bits and describes the wire. Uppercase B is bytes and describes the file. Eight of the first make one of the second, always.

That single sentence resolves the mismatch between your speed test and your download manager, the mismatch between your ISP's marketing and Steam's progress bar, and most arguments about whether an internet plan is delivering what it promised.