MagmaNex LogoMagmaNex
Home/Blog/Understanding Data Sizes: Bit vs Byte and KB vs MB vs GB
Understanding Data Sizes: Bit vs Byte and KB vs MB vs GB
GuideJune 14, 2026· 6 min read

Understanding Data Sizes: Bit vs Byte and KB vs MB vs GB

Bit vs byte, the KB (1000) vs KiB (1024) split, why your disk looks smaller, converting Mbps to MB/s for download times, and the basics of number bases.

#bit#byte#data size#kilobyte#megabit

Understanding Data Sizes: Bit vs Byte and KB vs MB vs GB

"I bought a 1 TB drive but my computer shows 931 GB." "My internet is 100 Mbps but files download at 12 MB/s." You hear these complaints constantly, and they almost always come from the same place: not knowing how data-size units actually work. In this article we start from the smallest unit and climb all the way to terabytes and petabytes, explain why decimal and binary units give different numbers, and show you how to estimate download times in your head.

A bit and a byte are not the same thing

Everything starts with the bit (binary digit). A bit can hold only two values: 0 or 1. It is the smallest piece of information a computer understands.

Eight bits make one byte. A byte holds exactly enough to store a single character (the letter "A," for example). The most important rule here: a bit is abbreviated with a lowercase "b," a byte with an uppercase "B." So:

  • 8 bits = 1 byte (8 b = 1 B)
  • Kb = kilobit, KB = kilobyte (an 8x difference!)
  • Mbps = megabits per second, MB/s = megabytes per second

Internet speeds are almost always quoted in bits (Mbps), while file sizes are quoted in bytes (MB). Marketing does this on purpose: 100 Mbps sounds bigger than the 12.5 MB/s you actually get.

The decimal (1000) vs binary (1024) mess

Here is the real reason your disk seems to "shrink." There are two different counting systems:

  • Decimal (SI): 1 KB = 1000 bytes. Used by drive and SSD makers and internet providers.
  • Binary (IEC): 1 KiB = 1024 bytes. Used by operating systems (especially Windows), which usually still print "KB."

The gap widens as the unit grows. A 1 TB drive holds 1,000,000,000,000 bytes. Windows divides by 1024, so it shows roughly 931 GiB on screen, but because it labels it "GB" you get confused. No storage went missing; it was just measured with two different rulers.

Unit Bytes (approx.)
1 bit 1/8 byte
1 byte (B) 1 byte
1 KB (kilobyte) 1,000 bytes
1 KiB (kibibyte) 1,024 bytes
1 MB (megabyte) 1,000,000 bytes
1 MiB (mebibyte) 1,048,576 bytes
1 GB (gigabyte) 1,000,000,000 bytes
1 GiB (gibibyte) 1,073,741,824 bytes
1 TB (terabyte) 1,000,000,000,000 bytes
1 PB (petabyte) 1,000,000,000,000,000 bytes

Rule of thumb: think in powers of 1000 (KB, MB, GB) for marketing and storage, and in powers of 1024 (KiB, MiB, GiB) for memory and OS readouts. The correct terms for the latter are "kibibyte, mebibyte," but in everyday speech everyone keeps saying "KB, MB" - which is exactly where the confusion comes from.

Mbps to MB/s: the download-time math

This is the calculation you'll use most. Your internet speed is in Mbps, but you want to see download speed in MB/s. The rule is simple: divide by 8.

  • 100 Mbps / 8 = 12.5 MB/s (theoretical maximum)
  • 50 Mbps / 8 = 6.25 MB/s

To find how long a file takes: file size (MB) / speed (MB/s).

Example: a 1.5 GB game update on a 100 Mbps connection: 1500 MB / 12.5 MB/s = 120 seconds = 2 minutes. In practice it takes a bit longer due to network congestion and protocol overhead.

If you'd rather not do these by hand, use our bit/byte converter tool to convert between Mbps, MB, GiB and more in a second.

Number bases: binary and hexadecimal basics

Underneath all these units lie number bases. In daily life we use base 10 (decimal: 0-9). Computers use base 2 (binary: only 0 and 1), because a transistor is either on or off.

Because binary numbers get very long, programmers usually prefer base 16 (hexadecimal/hex: 0-9 plus A-F). Hex is shorthand for binary: 4 bits map exactly to one hex digit.

  • Decimal 255 = binary 1111 1111 = hex FF
  • The largest value a single byte can hold: 255 (that is 2^8 - 1)
  • Color codes (#FF5733) and MAC addresses are always hex

This also explains why 1024 is special: 1024 = 2^10. In the binary world, powers of 2 are the "round" numbers, not 1000. When you need to convert between bases, the number base converter flips instantly between decimal, binary, and hex.

Summary

  • bit (b) is not byte (B); 8 bits = 1 byte, mind the upper/lowercase.
  • Manufacturers use 1000 (KB, MB, GB), operating systems use 1024 (KiB, MiB, GiB); that's why a disk seems to "shrink."
  • Internet speed is in Mbps; divide by 8 to get MB/s.
  • Computers run in binary (base 2), and programmers use hexadecimal (hex) as shorthand.

Once you know these four rules, drive labels, internet plans, and the "why isn't my 1 TB a full 1 TB" question will stop surprising you.


🛠 Related Tools

📚 Related posts

← All posts🛠 Explore tools