Data and File Unit Converter Convert Bits, Bytes, Decimal Units, and Binary Units

What this converter actually does

This tool converts a value from one data unit to another. It accepts bits, bytes, and the two families of prefixed units that sit above them, and it produces the primary answer plus a full breakdown across every supported unit so you can see how the whole system fits together. Nothing about the arithmetic is complicated, but the two competing prefix conventions cause enough confusion that a transparent converter is worth having.

Every conversion runs through bytes as the universal intermediate value. Bits are converted to bytes by dividing by eight. Decimal units are converted to bytes by multiplying by powers of one thousand. Binary units are converted to bytes by multiplying by powers of one thousand and twenty-four. Once a value is expressed in raw bytes, it can be turned into any other unit with the same three rules in reverse. Keeping the intermediate value visible is the main reason the result panel shows raw bytes alongside the answer.

Everything happens in the browser. There is no server request for the conversion itself, no upload, and no history. The tool does not read file metadata, does not inspect any file on your device, and does not know what file you are thinking about. It converts the number you give it and nothing else.

Bits, bytes, and why the ratio matters

A bit is the smallest unit of digital information and can hold one of two values, usually written as zero or one. A byte is a group of eight bits, and it is the standard unit for storage and file size. The ratio is fixed and has been since the earliest computer designs adopted the eight-bit byte as the convention. Eight bits in a byte, always, with no exceptions in modern systems.

That ratio is the source of the single most common misunderstanding in consumer data. Network operators advertise connection speeds in bits per second, because historically that was how telecommunications engineers measured line capacity. File sizes and downloads are measured in bytes, because that is how operating systems and applications report them. A connection advertised as one hundred megabits per second does not deliver one hundred megabytes per second. It delivers at most twelve and a half megabytes per second, because the eight-to-one ratio applies at every step. The capitalisation is the only visual cue: an upper-case B means bytes, a lower-case b means bits. On a specification that has been retyped a few times, the distinction is easy to lose.

Decimal and binary scales, and why both exist

Above the byte, storage units come in two families. The decimal family uses powers of one thousand and follows the standard prefix definitions of the International System of Units. The binary family uses powers of one thousand and twenty-four and was formally defined in 1998 by the International Electrotechnical Commission. Both are in active use, and both describe real things.

The ratio column is the part worth internalising. A kibibyte is about two and a half per cent larger than a kilobyte. A gibibyte is about seven and a half per cent larger than a gigabyte. A tebibyte is about ten per cent larger than a terabyte. Those percentages look small, but they compound every time the unit step goes up, which is why the gap between a one-terabyte label and the operating system's reported figure is roughly nine per cent before the file system has even formatted the disk.

The two families exist because the computer industry adopted the decimal prefixes early on to mean powers of one thousand and twenty-four, at a time when the SI definitions were less of a concern to hardware engineers than convenience. As storage sizes grew and the discrepancy between the two meanings became economically significant, the IEC introduced the binary prefixes to remove the ambiguity. The change has been widely adopted in technical documentation, standards bodies, and developer tools, but consumer packaging still mostly uses the older decimal labels. Both conventions are legitimate. The important thing is to know which one a given figure is using.

The three places the difference bites

The decimal and binary distinction shows up in a small number of high-impact places, and knowing where they sit saves a lot of confusion.

Hard drive and SSD capacity. Manufacturers sell storage in decimal units, because a larger number is easier to advertise and because the raw physical capacity really is a round decimal figure. A drive sold as five hundred gigabytes contains five hundred billion bytes. When a Windows or macOS system reports that drive, it typically uses binary units internally, so the same drive shows up as roughly four hundred and sixty-five gibibytes labelled as GB. The missing space is not lost and it is not being hidden. It is being reported in a different unit system. Formatted capacity and filesystem overhead reduce the visible figure a little further, on top of the unit difference.

Network speeds and download time. Every ISP advertises in bits, every browser reports downloads in bytes, and the two are separated by a factor of eight. A download of five thousand megabytes on a connection rated at one hundred megabits per second takes at least four hundred seconds in theory, or about six minutes and forty seconds, before any real-world slowdowns are counted. Estimating without the divide-by-eight step overstates the connection by a factor of eight, which is the largest single source of error in this whole area.

Cloud storage billing. Most cloud providers bill in decimal gigabytes or terabytes, because their internal accounting and their marketing both follow the SI definitions. Local tools that report the same bucket's size often use binary units. The two figures can differ by seven to ten per cent at typical scales, and reconciling them requires knowing which convention each side is using. The converter on this page shows both so the number can be checked from either direction.

A worked conversion

Suppose someone has a file reported as one point five gigabytes and wants to know how big it is in gibibytes, megabytes, mebibytes, and raw bytes. They enter one point five, select gigabyte as the source, and select byte as the destination first to see the base figure.

A gigabyte is one billion bytes, so one point five gigabytes is one point five billion bytes. In scientific notation that is 1.5 × 10⁹ bytes, or one billion five hundred million bytes. Converting that to bits multiplies by eight, giving twelve billion bits, which is 1.2 × 10¹⁰ bits.

Now converting the same one point five billion bytes to binary units. A gibibyte is 1 073 741 824 bytes, so one point five billion bytes divided by that figure gives roughly 1.397 gibibytes. A mebibyte is 1 048 576 bytes, so the same value in mebibytes is roughly 1 430 mebibytes. A megabyte, being one million bytes, gives exactly 1 500 megabytes.

The interesting thing about the example is the shape of the numbers. The decimal figure of 1 500 MB and the binary figure of 1 430 MiB both describe the same physical quantity. Neither is wrong. They are simply using two different rulers, and the ruler difference of about seven and a half per cent at the gigabyte level is the reason a file that is advertised as one and a half gigabytes on a spec sheet looks slightly larger when a binary tool reports it.

Download time, network speed, and the eight-bit rule

The most practical use of a data unit converter is estimating download time. The calculation is one division, but the units on each side have to match before it works.

Take a file of four point seven gigabytes on a connection advertised at two hundred megabits per second. The connection speed in bits needs to be converted to bytes first: two hundred megabits per second divided by eight gives twenty-five megabytes per second. The file size in gigabytes needs to be converted to megabytes: four point seven gigabytes times one thousand is four thousand seven hundred megabytes. Dividing gives one hundred and eighty-eight seconds, or roughly three minutes and eight seconds.

That figure is the theoretical maximum under perfect conditions. Real transfers are slower, usually by a noticeable margin. Protocol overhead, Wi-Fi interference, congestion at the server, and any rate limiting applied by the ISP all reduce the effective throughput. In practice, planning with a cushion of twenty to fifty per cent on top of the theoretical number is the safer approach. A theoretical three minutes often lands closer to four or five minutes in the real world, and on a congested network it can be longer still.

One further subtlety: on a fast local network such as gigabit Ethernet, the bottleneck is often the storage device rather than the connection. A spinning hard drive writes at roughly one hundred and twenty megabytes per second under typical conditions, which is close to the theoretical maximum of a gigabit link. On a solid state drive, the network is almost always the limit. Knowing which side is the bottleneck matters more than the exact transfer estimate when you are diagnosing why a large copy is slow.

Common mistakes with data units

A handful of mistakes show up repeatedly whenever data units come up in a technical conversation.

Treating MB and Mb as the same thing. They are separated by a factor of eight. A file described as one hundred MB is one hundred megabytes, which is eight hundred megabits. A network described as one hundred Mb per second is one hundred megabits per second, which is twelve and a half megabytes per second. The capitalisation is the whole distinction, and copy-and-paste errors propagate it easily.

Assuming 1024 is universal. Many people learn that a kilobyte is one thousand and twenty-four bytes, and they apply that rule everywhere. It is correct for the binary units and incorrect for the decimal ones. On a hard drive spec sheet, a kilobyte is one thousand bytes. On a Windows Explorer file listing, it is one thousand and twenty-four. Both readings are valid in their own context.

Forgetting filesystem overhead. Even after accounting for the decimal-versus-binary difference, a freshly formatted drive shows less usable space than its raw capacity because the file system itself takes up room. On a two-terabyte drive formatted with a typical file system, the visible figure is often closer to one point eight terabytes. That is normal and not a defect.

Mixing bits and bytes inside a single estimate. Converting file size to megabytes but leaving network speed in megabits produces an error of exactly eight times, which is the largest single source of confusion in this topic. Any estimate that mixes the two without converting first will be wrong by a wide margin, and the mistake is easy to make because both units appear in the same product advertisements.

Ignoring reserved space on solid state drives. SSDs reserve a portion of their raw capacity for over-provisioning, which the controller uses for wear levelling and error correction. The reserved space is not visible to the operating system and does not appear in any capacity report, but it is real and it explains a small part of the gap between the labelled figure and the visible one.

Using the wrong unit in a contract. When a service-level agreement or a cloud billing contract quotes a capacity figure, the unit matters. A figure described as one terabyte of storage under a decimal interpretation is a different obligation from a figure described as one tebibyte under a binary interpretation, and the difference at that scale is close to a hundred gigabytes. Reading the unit line in the contract is worth the ten seconds it takes.

Where this tool stops

This converter works on a single value that you enter. It does not read files, does not inspect metadata, and does not know what the file you are thinking about actually contains. That is deliberate: reading a file from a browser page would require uploading it, and the tool is designed so that no such upload ever happens.

It also does not model filesystem-specific overhead, partition schemes, or reserved capacity on solid state drives. Those effects are real, and they explain part of the gap between a drive's labelled capacity and the figure the operating system reports. The unit conversion on this page isolates the decimal-versus-binary difference, which is usually the larger of the two effects on consumer storage, and leaves the rest to the operating system's own reporting.

Finally, the tool treats decimal and binary units as distinct and does not silently guess which one a user meant. If a specification says GB, the tool treats it as decimal GB, and if it says GiB, the tool treats it as binary GiB. That is the correct behaviour under both the SI and IEC standards, and it is the reason the two figures differ in the breakdown panel rather than being averaged into a single number that matches neither convention.