Why does my 1 TB drive only show about 931 GB?

The manufacturer sold you one trillion bytes, which is a terabyte in the decimal sense. Your operating system divides by 1,024 three times rather than 1,000, so it reports 931.32 of what it calls gigabytes but which are really gibibytes. No capacity is missing; the same bytes are being described with a different sized unit.

What is the difference between GB and GiB?

A gigabyte on the decimal scale is exactly one billion bytes. A gibibyte is 1,073,741,824 bytes, which is 1,024 cubed, and is about 7.4 per cent larger. The IEC defined the gibibyte precisely so the two could be distinguished, but most consumer operating systems still print GB when they mean GiB.

Why is my download slower than my plan speed suggests?

Plans are quoted in megabits per second while files are measured in megabytes, so a 100 Mbps line moves at most 12.5 megabytes per second. On top of that, protocol overhead, encryption, server limits and network contention typically cost another ten to fifteen per cent, and the source server may cap your throughput regardless of your line.

Which standard should I use when planning capacity?

Plan in bytes and convert at the end. Buy against decimal figures, because that is how drives and cloud plans are sold, but size against the binary figures your operating system reports, because that is what your monitoring will show. Leaving fifteen per cent of headroom absorbs both the unit mismatch and filesystem overhead.

Is a megabit the same as a megabyte?

No. A byte is eight bits, so a megabyte is eight megabits. Network speeds use the lowercase b for bits, as in Mbps, while file sizes use the uppercase B for bytes, as in MB. The single-letter difference is the reason a 100 Mbps connection feels eight times slower than people expect when they watch a file download.

Do these units apply to memory as well as storage?

Memory is always binary. A module sold as 16 GB contains 16 gibibytes, because memory is addressed in powers of two and cannot be manufactured in decimal sizes. Storage is the exception, and it is the only place the decimal reading is normal, which is exactly why the two conventions collide there.

Data Storage Converter

Works anywhere

Converts a file or drive size between bytes and every decimal and binary unit at once, and estimates the transfer time at a given connection speed. One gibibyte is 1,073,741,824 bytes, which drive makers would label as 1.0737 GB.

Results

Worked example. Edit the values and press Calculate

Bytes

1,073,741,824

Kilobytes (kB, ×1000)
1,073,741.82
Megabytes (MB, ×1000)
1,073.742
Gigabytes (GB, ×1000)
1.0737
Terabytes (TB, ×1000)
0.001074
Kibibytes (KiB, ×1024)
1,048,576.00
Mebibytes (MiB, ×1024)
1,024.000
Gibibytes (GiB, ×1024)
1.0000
Tebibytes (TiB, ×1024)
0.000977
Transfer time at this speed
85.9 seconds
Assumes the full line rate with no protocol overhead; real transfers run roughly ten per cent slower.

Why use this calculator

Storage is the one area of measurement where the same word means two different numbers, and the gap grows with every step up the scale. Drive manufacturers, cloud plans and internet providers count in powers of a thousand, while operating systems and memory count in powers of 1,024, so a drive sold as one terabyte genuinely contains a trillion bytes but shows up as roughly 931 in the file manager.

Neither party is lying. This converter shows both readings side by side rather than picking one, which is the only way to reconcile a purchase order, an operating system report and a backup plan that disagree. It also converts bytes to bits for the transfer-time estimate, since connection speeds are quoted in megabits and file sizes in megabytes, a factor of eight that catches people out constantly.

How to use it

Enter the number you are looking at in the size box, exactly as it appears on the file properties, the drive label, the storage plan or the invoice.
Choose the magnitude prefix beside it, so a file described as 700 MB uses the mega option.
Then tell the calculator which standard that number is using: pick binary if the figure came from an operating system, a file manager or a memory specification, and decimal if it came from a drive box, a cloud storage plan or a network provider, since those industries have always counted in thousands.
Finally enter your real connection speed in megabits per second, taken from a speed test rather than from the plan headline, to get a realistic transfer estimate.

What each result means

Bytes: The exact count with no prefix at all, which is the only figure both standards agree on completely.
Kilobytes (kB, ×1000): The decimal kilobyte of exactly one thousand bytes, as used by drive makers and network providers.
Megabytes (MB, ×1000): One million bytes on the decimal scale, the unit printed on media, camera cards and plan documents.
Gigabytes (GB, ×1000): One billion bytes, the unit used to advertise drive capacity, phone storage and cloud plans.
Terabytes (TB, ×1000): One trillion bytes, the headline capacity on a drive box, which your operating system will report as fewer TiB.
Kibibytes (KiB, ×1024): The binary kilobyte of 1,024 bytes, which is what an operating system usually means when it prints KB.
Mebibytes (MiB, ×1024): 1,048,576 bytes, the unit behind memory sizes, block sizes and most file manager displays.
Gibibytes (GiB, ×1024): 1,073,741,824 bytes, the figure your operating system shows when it reports a drive as having less than advertised.
Tebibytes (TiB, ×1024): 1,099,511,627,776 bytes, about ten per cent more than a decimal terabyte at the same prefix.
Transfer time at this speed: How long the transfer takes at the line rate you entered, before protocol overhead and congestion are allowed for.

The formula

A size is converted to bytes first with bytes = value × base^n

Where the base is 1,000 for the decimal standard or 1,024 for the binary one and n is the prefix position, being 1 for kilo, 2 for mega, 3 for giga, 4 for tera and 5 for peta.

Every output unit is then bytes divided by its own base raised to the same position. Transfer time in seconds is (bytes × 8) ÷ (speed in megabits per second × 1,000,000), where the factor of eight converts bytes to bits because network speeds are quoted in bits.

How this number is used in practice

The split exists for a historical reason. Memory is addressed in binary, so a memory chip naturally comes in a power-of-two size and 1,024 was informally called a kilobyte from the earliest days of computing. Disk capacity has never been binary, so drive manufacturers used the ordinary metric meaning of kilo, mega and giga.

The IEC standardised the binary prefixes kibi, mebi and gibi to resolve the ambiguity, and while Linux tooling and some documentation adopted them, consumer operating systems largely kept printing GB for a value that is actually GiB. The result is a permanent mismatch that no purchase can fix.

The practical consequences show up in capacity planning and in bandwidth billing. A backup target sized in decimal terabytes will hold less than a source measured in binary terabytes, so a nominally like-for-like replacement drive can come up short. Cloud providers bill storage in decimal gigabytes and egress in decimal terabytes, so a workload sized from an operating system report will be under-provisioned by seven to ten per cent.

Network capacity is worse again, because an internet plan quoted at 100 Mbps delivers at most 12.5 megabytes per second even before overhead.

For everyday use the numbers that matter are anchors. A high-resolution photograph is a few megabytes, an hour of streamed high definition video is around three gigabytes, a dual-layer Blu-ray disc is 50 decimal gigabytes, and a typical phone backup runs to tens of gigabytes.

Knowing those figures alongside a connection speed tells you immediately whether a transfer is a coffee break or an overnight job.

Worked examples

1 GiB as the operating system counts it
Exactly 1,073,741,824 bytes, which a drive maker would print as 1.0737 GB.
A drive sold as 1 TB, as the manufacturer measures it
One trillion bytes, which the operating system reports as 931.3226 GiB.
A 25 GB Blu-ray image downloaded on a 100 Mbps line
26.8435 decimal GB of data, taking about 35.8 minutes at full line rate.
A 700 MB file on a slow 8 Mbps connection
734.003 decimal MB, needing roughly 12.2 minutes to move.

When people use it

Reconciling an advertised drive capacity with the figure your operating system reports
Sizing a backup target so it actually holds the source it is replacing
Estimating how long a large download, upload or cloud restore will take
Converting cloud storage and egress quotas between decimal and binary units
Checking whether a file will fit a disc, card or attachment limit before you start

Tips

Bytes are the only unambiguous figure. When two systems disagree about capacity, compare them in bytes.
Network speeds are in bits and file sizes are in bytes. Divide megabits per second by eight to get megabytes per second.
A drive advertised as 1 TB is not faulty when it reports 931 GB; that is 1,000,000,000,000 bytes shown in binary units.
The mismatch grows with scale: about 2.4 per cent at kilo, 7.4 per cent at giga and 12.6 per cent at peta.
Allow ten to fifteen per cent above the theoretical transfer time for protocol overhead, retries and congestion.
Formatting, filesystem metadata and reserved space consume capacity on top of the unit mismatch, typically another one to three per cent.

What it assumes, and where it stops being accurate

The binary standard means powers of 1,024 and the decimal standard means powers of 1,000, as defined in IEC 80000-13.
Transfer time assumes the full line rate is available for the whole transfer, which real networks rarely deliver.
Protocol overhead, packet loss, disk speed limits and contention are excluded from the transfer estimate.
Filesystem overhead, formatting loss and vendor over-provisioning are not modelled, so usable capacity will be lower again.
Compression and deduplication are ignored; a compressed transfer can move far more logical data than the figure suggests.
Values are capped at one billion of the chosen unit so a shared link cannot request an unbounded conversion.

Terms used on this page

Byte: Eight bits, the smallest addressable unit of storage and the only figure both standards agree on.
Gibibyte (GiB): 1,073,741,824 bytes, the binary gigabyte defined by IEC 80000-13.
Gigabyte (GB): One billion bytes on the decimal scale used by drive makers and cloud providers.
Megabit per second: One million bits per second, the unit internet plans are sold in; eight of them make one megabyte per second.
Over-provisioning: Capacity a drive reserves for wear levelling and spare blocks, reducing usable space beyond the unit mismatch.

Questions people ask

Why does my 1 TB drive only show about 931 GB?: The manufacturer sold you one trillion bytes, which is a terabyte in the decimal sense. Your operating system divides by 1,024 three times rather than 1,000, so it reports 931.32 of what it calls gigabytes but which are really gibibytes. No capacity is missing; the same bytes are being described with a different sized unit.
What is the difference between GB and GiB?: A gigabyte on the decimal scale is exactly one billion bytes. A gibibyte is 1,073,741,824 bytes, which is 1,024 cubed, and is about 7.4 per cent larger. The IEC defined the gibibyte precisely so the two could be distinguished, but most consumer operating systems still print GB when they mean GiB.
Why is my download slower than my plan speed suggests?: Plans are quoted in megabits per second while files are measured in megabytes, so a 100 Mbps line moves at most 12.5 megabytes per second. On top of that, protocol overhead, encryption, server limits and network contention typically cost another ten to fifteen per cent, and the source server may cap your throughput regardless of your line.
Which standard should I use when planning capacity?: Plan in bytes and convert at the end. Buy against decimal figures, because that is how drives and cloud plans are sold, but size against the binary figures your operating system reports, because that is what your monitoring will show. Leaving fifteen per cent of headroom absorbs both the unit mismatch and filesystem overhead.
Is a megabit the same as a megabyte?: No. A byte is eight bits, so a megabyte is eight megabits. Network speeds use the lowercase b for bits, as in Mbps, while file sizes use the uppercase B for bytes, as in MB. The single-letter difference is the reason a 100 Mbps connection feels eight times slower than people expect when they watch a file download.
Do these units apply to memory as well as storage?: Memory is always binary. A module sold as 16 GB contains 16 gibibytes, because memory is addressed in powers of two and cannot be manufactured in decimal sizes. Storage is the exception, and it is the only place the decimal reading is normal, which is exactly why the two conventions collide there.

Use this with

Data Storage Converter answers one part of the question. These pick up where it stops: IPv4 Subnet Calculator, Cloud Cost Comparison Calculator, Screen PPI Calculator and Number Base Converter.

Where these figures come from

Works anywhereLast reviewed 2026-09·How we build and check calculators

This calculation is jurisdiction-neutral arithmetic. No country-specific rules are applied.

NIST: Prefixes for binary multiples. Why kibi, mebi and gibi exist and how they differ from kilo, mega and giga
International Electrotechnical Commission. Publisher of IEC 80000-13, the standard that defines the binary prefixes
NIST Special Publication 811: Guide for the Use of the International System of Units. Definitive table of exact conversion factors between SI and customary units

Data Storage Converter produces an estimate for general information. It is not financial, tax, medical, legal or engineering advice. read the full disclaimer. Spotted a wrong number? Report it and it gets fixed. The form picks up this calculator and the inputs you used.