Data Storage Converter
Bits to pebibytes, with the decimal (GB) and binary (GiB) families side by side in every result. Most converters silently pick one and leave you to discover which. This one shows both, because the difference is the whole question.
Convert a storage size
The capacity or file size you have, in the unit chosen below.
Decimal and binary units are listed separately on purpose — GB and GiB are not the same size.
Both families appear in the table below whichever one you pick.
1 TB is 1,000,000,000,000 bytes. Counted in gibibytes — powers of 1024 — those same bytes come to 931.3225746. Neither figure is wrong; they are different units.
- Conversion factor
- 1 TB = 931.3225746 GiB
- Reverse factor
- 1 GiB = 0.001073741824 TB
- In bytes
- 1,000,000,000,000 B
- In bits
- 8,000,000,000,000 bit
- Bytes in one GiB
- 1,073,741,824 B — Binary (IEC): 1 GiB = 2³⁰ = 1,073,741,824 bytes — 7.4% more than a gigabyte. This is the number Windows computes and then labels GB.
| Prefix tier | Decimal (SI) | Value | Binary (IEC) | Value |
|---|---|---|---|---|
| Kilo · 2.4% apart | kB | 1,000,000,000 | KiB | 976,562,500 |
| Mega · 4.9% apart | MB | 1,000,000 | MiB | 953,674.3164 |
| Giga · 7.4% apart | GB | 1,000 | GiB | 931.3225746 |
| Tera · 10% apart | TB | 1 | TiB | 0.9094947018 |
| Peta · 12.6% apart | PB | 0.001 | PiB | 0.0008881784197 |
Two prefix systems, one count of bytes
Every unit here is stored as one number — how many bytes it is — and every conversion is two steps through that base. What makes the subject confusing is not the arithmetic. It is that two different families of prefixes are in daily use and one of them routinely borrows the other’s symbols.
bytes = value × f(from) • result = bytes ÷ f(to)- f(u)
- Bytes in one unit u — 10⁹ for GB, 2³⁰ for GiB
- bytes
- The size in the one unit nobody disputes
- result
- The size in the unit you asked for
Decimal (SI) prefixes are powers of 1000. 1 kB = 10³ bytes, 1 MB = 10⁶, 1 GB = 10⁹, 1 TB = 10¹², 1 PB = 10¹⁵. These are the ordinary SI prefixes — the same kilo and mega used for grams and hertz, standing for exactly the same multipliers. Storage manufacturers, network operators and cloud invoices all use them.
Binary (IEC) prefixes are powers of 1024. 1 KiB = 2¹⁰ bytes, 1 MiB = 2²⁰, 1 GiB = 2³⁰, 1 TiB = 2⁴⁰, 1 PiB = 2⁵⁰. Memory addressing is binary, so these sizes fall out of the hardware naturally. IEC 60027-2 gave them their own names and symbols in 1998 and IEC 80000-13 carries the definitions now.
- The gap compounds and never closes. A kibibyte is 2.4% larger than a kilobyte, a mebibyte 4.9% larger than a megabyte, a gibibyte 7.4%, a tebibyte 10.0% and a pebibyte 12.6%. Each tier multiplies the previous discrepancy by 1.024.
- The symbols are not interchangeable. SI kilo is a lower-case k; an upper-case K has never been an SI prefix at all. KB, KiB and kB are three different strings and only two of them are defined units.
- A byte is 8 bits. That is fixed by IEC 80000-13 too. It matters most where the two are mixed: network throughput is quoted in bits per second and file sizes in bytes, an unmarked factor of eight between them.
- Whole byte counts are printed in full. Results that are not whole numbers carry ten significant figures, but an exact integer count of bytes is reported exactly — one tebibyte is 1,099,511,627,776 bytes here, not 1,099,511,628,000.
Why a “1 TB” drive shows up as 931 GB
You buy a drive labelled 1 TB. The manufacturer means the decimal terabyte, so the drive holds 1,000,000,000,000 bytes. That figure is on the box and it is accurate.
Windows now divides by 1024, three times over. 1,000,000,000,000 ÷ 1024 = 976,562,500. Divide again: 953,674.316406. Divide a third time: 931.3225746. Then it writes “931 GB” next to it.
The number 931.3225746 is correct. The label is not. Those are gibibytes — 931.3225746 × 2³⁰ bytes, which multiplies back out to the same 1,000,000,000,000 bytes you started with. The 68.68 “missing gigabytes” never existed: they are the 6.9% arithmetic difference between dividing by 10⁹ and dividing by 2³⁰, not lost capacity.
The same subtraction happens at every size. A 2 TB drive reports 1.82 TB, a 4 TB drive reports 3.64 TB, and a phone sold with 256 GB of storage shows 238 GB before a single app is installed. Set this converter to 256 GB → GiB and it returns 238.4185791, which is exactly the figure the phone displays and mislabels.
The reverse trip is worth knowing too. Eight gibibytes of RAM — which is what “8 GB of RAM” always means, because memory really is addressed in powers of two — is 8.589934592 GB in decimal terms. Memory and storage genuinely use different conventions, and both are usually printed with the same two letters.
What this converter assumes
- Decimal prefixes mean exactly what SI says: kB is 10³ bytes and never 1,024. If you have a figure from a tool that labels its 1024-based numbers “KB” or “GB”, enter it as KiB or GiB — that is what the tool actually computed.
- A byte is 8 bits. Historical machines with 6-, 7- or 9-bit bytes are outside this table, as are nibbles, machine words, disk sectors and filesystem blocks: those depend on an architecture or a filesystem, not on a definition.
- This converts sizes, not rates. A megabit per second is not a megabyte per second, and dividing by time is your job — mixing a rate into a size table is how the factor of eight goes missing.
- Advertised capacity is not usable capacity, and no unit conversion accounts for the difference. Filesystem metadata, spare blocks, over-provisioning on SSDs and a recovery partition all consume real bytes. Those are product facts; this page only settles the unit arithmetic.
- Sizes are magnitudes, so negative values are rejected. If you are converting the difference between two capacities, convert its size and keep the sign yourself.
Data storage conversion FAQ
Why does my 1 TB drive show up as 931 GB?
Because two different units are both being called GB. The drive holds 1,000,000,000,000 bytes — a terabyte in the decimal sense the manufacturer used, where tera means 10¹². Windows divides that byte count by 1024 three times, which gives 931.3225746, and then prints the label 'GB' next to it. The honest label for that number is GiB — gibibytes, powers of 1024. Nothing is missing from the drive and nothing was taken by formatting. One number is 10¹² bytes and the other is 931.32 × 2³⁰ bytes, and those are the same bytes.
Is a kilobyte 1,000 bytes or 1,024 bytes?
1,000. The SI prefix kilo has meant 10³ since 1795 and means it for grams, metres and hertz without argument. Early computing borrowed it for 1,024 because 2¹⁰ was conveniently close, and the 2.4% error was harmless when files were small. It stopped being harmless as the prefixes grew: the gap is 2.4% at kilo, 4.9% at mega, 7.4% at giga, 10.0% at tera and 12.6% at peta. IEC 60027-2 fixed this in 1998 by giving the binary sizes their own names — kibibyte (KiB), mebibyte (MiB), gibibyte (GiB) — and IEC 80000-13 carries that today.
What is the difference between GB and GiB?
1 GB is 10⁹ = 1,000,000,000 bytes. 1 GiB is 2³⁰ = 1,073,741,824 bytes, which is 7.4% larger. Read the other way round, a gigabyte is 6.9% smaller than a gibibyte, which is why converting 1,000 GB into GiB gives 931.32 rather than 1,000. Both units are legitimate; the only error is using the symbol of one for the size of the other.
Why does macOS report a bigger size than Windows for the same drive?
Because they use different prefixes and only one of them says so. Since Mac OS X 10.6 Snow Leopard in 2009, macOS reports storage in decimal units, so a drive sold as 1 TB reads as roughly 1 TB in Finder. Windows still divides by 1024 and labels the result GB or TB, so the same drive reads as 931 GB. The disk did not change between the two machines; the divisor did.
Is 100 Mbps the same as 100 MB/s?
No — it is eight times slower. Mbps is megabits per second and MB/s is megabytes per second, and a byte is 8 bits. A 100 Mbps connection delivers at most 12.5 MB/s, so a 1 GB file takes about 80 seconds at the theoretical maximum. The lower-case b for bit and upper-case B for byte is the only thing distinguishing them, which is why the abbreviation is worth spelling out whenever the number matters.
How many megabytes are in a gigabyte?
1,000, if both are decimal — that is the definition of the prefixes. If you mean how many mebibytes are in a gibibyte, the answer is 1,024. If you mean how many megabytes are in a gibibyte, it is 1,073.741824. Three sensible questions hide behind one sentence, which is the entire reason this page shows both families at once instead of picking one for you.
Sources and review notes
- NIST — Prefixes for binary multiples, the reference table for kibi, mebi, gibi, tebi and pebi
- NIST Special Publication 811 — Guide for the Use of the International System of Units (SI), for the decimal prefixes
- BIPM — The International System of Units (SI) Brochure, which defines the SI prefixes and states that they are never used for powers of two
- IEC 80000-13, Quantities and units — Information science and technology, which defines the byte and the binary prefixes. It supersedes IEC 60027-2, where kibi, mebi and gibi were introduced in 1998.
These are defined values, not measured ones, so they do not go stale: 1 GB has been 10⁹ bytes and 1 GiB has been 2³⁰ bytes since the binary prefixes were standardised in 1998. What does change is which convention a given piece of software uses to report a size — macOS moved to decimal in 2009, Windows has not — and that is a labelling choice, never a change in the units themselves.