Data Size Units Explained: KB, MB, GB and Bytes vs bps

Your phone has "128 GB," a photo is "4 MB," your connection is "100 Mbps"—you see these numbers daily, yet "how many MB is a GB?" or "which is the Byte, B or b?" can be surprisingly hard to answer on the spot. Data units become far clearer once you grasp a few basic rules. This page walks through the relationship between bits and Bytes, the KB–MB–GB–TB ladder, the confusing 1000-versus-1024 issue, how the bps used for connection speed differs, and rough file-size benchmarks—focusing on the points where people most often stumble.

The difference between a bit and a Byte

Everything starts with the bit, the smallest unit of information a computer handles. A bit is just a 0 or a 1—think of it as a single switch that is either on or off. Because 0 and 1 alone are too limited, bits are bundled in groups of eight: 1 Byte = 8 bits, and the Byte serves as the basic unit of data size.

One Byte (8 bits) allows 2 to the 8th power = 256 combinations—just enough to represent a single character such as a letter, digit, or symbol, which is why "1 Byte ≈ one character" is a historical rule of thumb. Notation matters too: by convention, an uppercase B means Byte and a lowercase b means bit, so "KB" and "Kb," or "Mbps" and "MB/s," denote completely different quantities. Miss this capitalization and your sense of size or speed can be off by a factor of eight, so it is the first thing worth keeping in mind.

A way to remember it: data amounts (file sizes, storage capacity) are given in Bytes (uppercase B), while connection speeds are given in bits (lowercase b). "Storage is bytes, the line is bits" keeps the two straight.

The KB–MB–GB–TB ladder

Building on the Byte, larger amounts add a prefix (K = kilo, M = mega, G = giga, T = tera). Each step up is roughly 1000 times the one below it, forming a ladder. Remember this order and you can handle almost any everyday situation.

A "4 MB photo" is about 4,000 KB, and a "2 GB video" is about 2,000 MB = roughly 2,000,000 KB. Knowing each step multiplies by about 1,000, you can estimate "how many MB is 128 GB?" as 128 × 1,000 ≈ 128,000 MB in your head. There are even larger units like PB (petabyte) and EB (exabyte), but individuals rarely go beyond TB. To check conversions between units at once, the Unit Converter gives accurate values from a single input.

1000 versus 1024—KB versus KiB

This is the trickiest part of data units. The section above said "about 1,000 times," but computers run on binary (0s and 1s), so the natural breakpoint is not the round 1000 but 2 to the 10th power = 1024. Strictly, then, one school counts 1 KB = 1024 Bytes, 1 MB = 1024 KB, and so on. Another school uses "kilo = 1000," just like the metric system. The coexistence of these two conventions is the source of the confusion.

To resolve the ambiguity, international standards gave the 1024-based units their own names: 1024 Bytes = 1 KiB (kibibyte), 1024 KiB = 1 MiB (mebibyte), 1024 MiB = 1 GiB (gibibyte), inserting an "i." In short:

• Decimal (1000-based): KB, MB, GB, TB. Common on storage-maker spec sheets.
• Binary (1024-based): KiB, MiB, GiB, TiB. Often used for OS file-size displays and memory capacity.

This gap creates the "I bought 1 TB but my PC shows only about 931 GB" effect. The maker counts 1 TB = 1,000,000,000,000 Bytes (decimal), while the OS divides by the binary base (actually TiB), so 1,000,000,000,000 ÷ 1024 ÷ 1024 ÷ 1024 ≈ 931—the figure appears to shrink. No capacity has vanished and nothing is broken; it is merely an apparent difference caused by how the units are counted. The gap widens with size: about 7% between GB and GiB, and about 9% between TB and TiB.

How connection speed (bps) differs

File sizes are in Bytes, but internet speed is given in bps (bits per second), a bit-based unit. "100 Mbps" means 100 megabits can be sent each second, and here the 8-times gap between Bytes and bits comes into play.

Consider "how long to download an 800 MB file over a 100 Mbps line." The file is in Bytes and the speed is in bits, so first match the units: 800 MB = 800 × 8 = 6400 Mbit. Divide by 100 Mbps and you get 6400 ÷ 100 = 64 seconds (in theory). If you glance only at the speed and assume "100, so about 8 seconds," you underestimate by eight times. Real throughput is usually below the theoretical value due to congestion and hardware, so remember the rule of thumb: divide Mbps by 8 to get MB/s (download per second), and your time estimates will be far closer.

Rough file-size benchmarks

The fastest way to build a feel for units is to know the typical size of files you handle. Sizes vary greatly with content and quality, but here are rough guides.

• Text and documents: a memo-sized document is a few KB to tens of KB; a PDF with images runs from hundreds of KB to several MB.
• Photos: a JPEG from a phone is about 2–5 MB each; high-resolution or RAW files are larger.
• Music: a typical compressed track is about 3–10 MB.
• Video: depending on quality, full-HD video is roughly tens of MB to over 100 MB per minute.
• Apps and games: small apps are tens of MB; large games can reach several GB to tens of GB.

With these in mind, you can quickly judge things like "200 photos is about 1 GB" or "is 5 GB of free space enough?" The same instinct helps when choosing a cloud storage plan or a data contract.

Summary

Four points keep data units from tripping you up. First, the smallest unit is the bit, 8 bits = 1 Byte, with uppercase B for Byte and lowercase b for bit. Second, KB → MB → GB → TB is a ladder of roughly 1000 times per step. Third, because of how computers work, the 1024-based units (KiB, MiB, GiB) are also in use, which is why "1 TB looks like 931 GB." Fourth, since bps is bit-based, divide by 8 to convert it into a download amount. To check these relationships all at once, enter your numbers into the Unit Converter for accurate conversions without mental math.