KB vs KiB: Why Your 1 TB Drive Shows About 931 GB

You buy a 1 TB drive, plug it in, and your computer reports about 931 GB. Nothing has gone wrong, nothing is faulty, and nobody has taken your storage. The drive holds every byte printed on the box — your computer is simply counting those bytes with a different ruler. That ruler is the whole story of KB vs KiB.

The short answer

A kilobyte (KB) counts in thousands: 1 KB is 1,000 bytes. A kibibyte (KiB) counts in powers of two: 1 KiB is 1,024 bytes. The difference looks tiny at the kilobyte level — just 24 bytes — but it compounds at every step up the scale, and by the terabyte level it is nearly 10%. That compounding gap is exactly why a 1 TB drive shows up as about 931 GB. Same bytes, two counting systems, two different numbers.

If you remember one rule, make it this: when a number looks smaller than you expected, check the unit letter before you assume something is missing. The little “i” in KiB, MiB and GiB is doing real work.

KB vs KiB at a glance

There are two families of units, and they look confusingly alike:

The decimal family steps by 1,000 each time, the way the metric system steps everywhere else: a kilometre is 1,000 metres, so a kilobyte is 1,000 bytes. The binary family steps by 1,024 each time, because computers work in powers of two and 1,024 is 2 to the power of 10 — the power of two closest to 1,000.

That is also why the gap grows. At the kilo step the binary unit is 2.4% larger than the decimal one. By the mega step it is 4.86% larger, by the giga step 7.37%, and by the tera step 9.95%. Each step multiplies the small difference again, so the “missing” space looks bigger exactly when the drive is bigger.

Why two systems exist

Early computer engineers needed a “big round number” near 1,000 for memory, and the nearest power of two was 1,024. Close enough, they decided, and “kilobyte” came to mean 1,024 bytes in computing contexts while it kept meaning 1,000 bytes everywhere else. For a floppy disk, a 2.4% difference was invisible. For a modern multi-terabyte drive, it is tens of gigabytes of apparent difference — and a lot of confused buyers.

The fix arrived in December 1998, when the International Electrotechnical Commission (IEC) approved a separate set of names for the binary multiples: kibi, mebi, gibi, tebi, pebi and exbi. Under that standard, KB, MB, GB and TB always mean powers of 1,000, and KiB, MiB, GiB and TiB always mean powers of 1,024. The “i” in the middle — short for “binary” — is the whole point: it tells you which ruler is being used, so the number can no longer be read two ways.

The standard is real and widely documented — the US National Institute of Standards and Technology (NIST) publishes the same definitions — but everyday usage never fully switched. Drive makers label in decimal, many operating systems and tools still report in binary while reusing the decimal labels, and most people have never seen the word “kibibyte” at all.

The 1 TB drive maths

Here is the full calculation, with nothing skipped. A drive sold as 1 TB holds 1,000,000,000,000 bytes — one trillion, counted in thousands. A display that counts in binary divides that byte count by 1,024 three times (once for each step from bytes to gibibytes):

1,000,000,000,000 ÷ 1,073,741,824 = 931.32

So the drive reads as 931.32 gibibytes. When the display labels that figure “GB” instead of “GiB”, you see “931 GB” and conclude that about 69 GB has vanished. It has not. The byte count never changed; only the unit did.

The same maths explains every drive size:

Two honest footnotes. First, an individual drive may show slightly less again after formatting, because the file system itself needs a little space, and pre-built computers often reserve hidden recovery partitions. Second, if a drive shows wildly less than the table above — not 7–10% less, but a large fraction less — that is a different problem, such as an unallocated partition or a counterfeit drive, and it is worth investigating rather than blaming the units.

What your computer actually shows

This is where the label becomes the trap. Some displays use the binary labels honestly: they show KiB, MiB and GiB, and then 931 GiB is exactly what you should expect from a 1 TB drive. Others divide by 1,024 but keep printing KB, MB and GB, so the number looks like a decimal figure and reads as a shortfall.

You cannot tell which ruler a display is using from the number alone. You have to read the unit letter:

• If it says GiB, the figure is binary and a 1 TB drive showing 931 GiB is correct.
• If it says GB but the number matches the binary column in the table above, the display is dividing by 1,024 and labelling it GB.
• If it says GB and the number matches the decimal figure, the display is counting in thousands.

The practical habit is simple: whenever two tools disagree about the same file or drive, compare their unit labels before you compare their numbers. Nine times out of ten, the disagreement is the ruler, not the storage. The same habit helps with photos — a file listed as 4,800,000 bytes in one place is 4.8 MB in decimal and 4.58 MiB in binary, and both descriptions are correct, as our guide to reducing an image’s file size shows from the file-size side.

Converting between the systems

Converting is just division or multiplication by 1,024 at each step, and it works in both directions.

Decimal to binary. Take the byte count and divide by 1,024 for each step up. A 5,000,000-byte file is 5 MB in decimal (5,000,000 ÷ 1,000,000), and 4.77 MiB in binary (5,000,000 ÷ 1,048,576 = 4.768…). The same file, two honest labels.

Binary to decimal. Take the binary figure and multiply by 1,024 for each step, or convert through bytes. A file of 100 MiB is 100 × 1,048,576 = 104,857,600 bytes, which is 104.86 MB in decimal. So a limit written as 100 MiB is roomier than a limit written as 100 MB — by almost 5 MB.

A shortcut for drive shopping. To estimate what a decimal-labelled drive will show in binary units, multiply the advertised figure by about 0.93 at the terabyte level (and about 0.91 if you want the tebibyte figure). A 2 TB drive will show about 1,862 GiB; a 500 GB drive about 466 GiB. If the number you see after plugging in matches that estimate, your drive is fine.

Which unit to use when

The right unit depends on whose ruler the other side is using:

• Buying storage. The box uses decimal. Compare drives by their byte counts, not by what different displays will later report, and use the table above to set your expectation before you plug the drive in.
• Upload limits and hosting plans. These are usually written in decimal, matching the way drive makers and download pages count. Convert your file’s byte count into the same unit the limit uses, or a file that “fits” in one system will look too large in the other.
• Memory and system tools. Memory is addressed in powers of two, so RAM and many system displays naturally speak binary — this is where KiB and MiB labels, when they appear, are most at home.
• Comparing two figures. Always match the units first. A 931 GB figure and a 1 TB figure can describe the identical drive; a 4.8 MB figure and a 4.58 MiB figure can describe the identical photo.

Try it: decimal conversions with the File Size Converter

For the decimal side of the job, the File Size Converter converts a size between bytes, KB, MB, GB and TB instantly in your browser. Enter a value, choose its current unit and the target unit, and read the converted size. Its own sample shows the decimal maths at work: 2,048 KB converts to 2.048 MB, because each step multiplies by 1,000.

One limit, stated plainly because it matters for this article: the File Size Converter is decimal-only. It converts bytes, KB, MB, GB and TB at 1,000 steps, and it does not convert KiB, MiB, GiB or TiB. If a figure must match a binary-based display exactly, use the division shown above rather than the tool — the tool’s own page says the same. For everything in decimal, and for checking upload limits and drive labels written in decimal, it is the quick check. You can browse the rest of the File Tools category or all free tools for related checks.

Frequently asked questions

Is a kilobyte 1,000 bytes or 1,024 bytes?
Both, depending on the ruler — and that is the problem the IEC names were invented to solve. Strictly, a kilobyte (KB) is 1,000 bytes and a kibibyte (KiB) is 1,024 bytes. In older computing usage, “kilobyte” was also used for 1,024 bytes, which is why you still see both. When it matters, look for the “i”.

Is my drive faulty if it shows less than the box says?
Almost certainly not, if the shortfall matches the table above — about 7% at the gigabyte level, about 10% at the terabyte level. That is the unit gap. Formatting and hidden partitions can account for a little more. A much larger shortfall is worth checking in your system’s disk manager.

Does the File Size Converter handle KiB and MiB?
No. It is decimal-only: bytes, KB, MB, GB and TB at 1,000 steps. For binary figures, use the maths in this article — divide bytes by 1,024 per step to reach KiB, MiB and GiB.

What are MiB and GiB, in one sentence?
They are the unambiguous binary versions of MB and GB: 1 MiB is 1,048,576 bytes and 1 GiB is 1,073,741,824 bytes, always, whatever the display calls them.

Why hasn’t everyone just switched to KiB and MiB?
Because the old labels are everywhere — on boxes, in software, in habits formed over decades — and the new words sound unfamiliar. Until displays label their ruler consistently, the reader has to do the checking: read the unit letter, not just the number.

Check the unit, then trust the number

KB vs KiB is not a trick and not a fault. It is two honest ways of counting the same bytes, separated by a factor of 1,024 against 1,000, and the gap between them grows with every step up the scale. Once you know which ruler a figure uses, the “missing” gigabytes stop being a mystery: a 1 TB drive showing 931 GB is a full drive, correctly described.

Next time a size looks wrong, pause before you blame the drive. Check the unit letter, run the division, and if the figure is decimal, convert it with the File Size Converter — then compare like with like.