Bandwidth Converter
Convert megabits per second to megabytes per second and back — the factor of eight that explains why a 100 Mbps connection downloads at 12.5 MB/s.
Bandwidth Converter
Results recalculate instantly on every keystroke. Nothing you type is transmitted.
What this result does not account for
- Line rate is a ceiling; protocol overhead, contention and packet loss are not modelled.
- The transfer time shown is a theoretical minimum illustrating the factor of eight, not a capacity plan.
- Network units here are decimal; some software reports binary rates, which differ by up to about seven per cent at the gigabyte level.
- Upload and download rates differ on most consumer connections and must be converted separately.
In short: Network rates are quoted in bits per second; file sizes are in bytes. Eight bits to a byte, so a 100 Mbit/s link moves 12.5 MB/s at best, and 1 Gbit/s is 125 MB/s. Nothing is being throttled — the units simply differ by eight. Real throughput sits below the line rate because of protocol overhead.
Formula
Each factor is the number of bits per second in one of that unit. Byte-based rows carry a factor of eight. Network prefixes are decimal powers of ten, matching how telecommunications equipment has always been specified.
Worked Example
- Identify the source factor. One megabit per second is 1,000,000 bit/s.
- Convert to bits per second. 100 × 1,000,000 = 100,000,000 bit/s.
- Identify the target factor. One megabyte per second is 8,000,000 bit/s.
- Divide. 100,000,000 ÷ 8,000,000 = 12.5 MB/s.
- Sanity-check the disappointment. A 100 Mbps connection downloading at 12 MB/s is performing exactly as specified.
Why the small b matters. A lowercase b means bits and an uppercase B means bytes, and internet service is always sold in the former while file managers always display the latter. 100 Mbps and 100 MB/s differ by a factor of eight. The convention is not marketing sleight of hand — serial links genuinely move one bit at a time, so bits per second is the natural unit for a wire and bytes is the natural unit for a file.
Strengths & Limits Of This Model
Where this engine is strong
- Makes the bit-versus-byte factor of eight explicit rather than implicit.
- Reports the same rate in bits and bytes simultaneously.
- States plainly that a line rate is a ceiling, not a promise.
- Includes legacy circuit rates at their exact nominal values.
Where it stops
- Cannot predict real-world throughput, which depends on the whole path.
- Not a workload sizing tool — the transfer time is an illustration only.
Practical Use Cases
Interpreting an internet plan
Service is advertised in megabits and downloads are reported in megabytes. Dividing by eight turns the marketing figure into the number you will actually watch tick over.
Network and link capacity planning
Circuit specifications, switch backplanes and interface rates are all in bits per second, while the data volumes they carry are catalogued in bytes. Every capacity calculation crosses that boundary at least once.
Storage and backup windows
Sizing a backup window means combining a byte figure with a bit rate — see the Data Storage Converter for the decimal and binary side of the volume.
Legacy circuit identification
T1 at 1.544 Mbit/s, T3 at 44.736 and OC-3 at 155.52 still appear in contracts and documentation long after the technology has been replaced.
Methodology & Editorial Standards
All conversions route through bits per second. Byte-based units carry a factor of eight exactly, since a byte is eight bits. Network prefixes are decimal powers of ten throughout, matching how telecommunications and networking equipment has always been specified; the mebibyte-per-second row is included because some software reports transfer rates in binary units, and it is labelled distinctly so the two are not confused. Legacy circuit rates are their exact nominal values: T1 at 1.544 Mbit/s, DS3 at 44.736 and OC-3 at 155.52. The transfer-time readout uses decimal gigabytes and is deliberately framed as a theoretical minimum illustrating the eight-times factor rather than as a workload estimate, since real throughput is reduced by protocol overhead, retransmission and contention that no unit conversion can model. All conversion factors are exact by definition under the International System of Units, or exact by international agreement where the unit is defined by treaty. Values are held at full IEEE-754 double precision internally and rounded only for display, so chained conversions do not accumulate drift.
Computation runs in IEEE-754 double precision at full internal precision; rounding to two decimal places occurs strictly at the display layer, so no cumulative drift enters the result. All monetary outputs use accounting presentation — grouped thousands, two decimals, negatives in parentheses — so figures can be transcribed directly into a model or working paper. Division-by-zero and out-of-domain inputs return an em-dash rather than a misleading number.
This engine was reconciled against an independent reference implementation and hand-verified for the worked example above before release. Our full five-stage review process is published on the About Us page.
Disclaimer. This calculator is provided for informational and modelling purposes only and does not constitute financial, tax, legal, medical, or engineering advice. Verify all figures with a qualified professional before acting on them.
Bandwidth Converter — 10 Expert FAQs
10 analyst-written answers to the questions practitioners actually ask — optimised for voice and answer-engine retrieval.
Why does my 100 Mbps connection only download at 12 MB/s?
Because those are the same speed. Megabits per second and megabytes per second differ by a factor of eight, so 100 Mbit/s is 12.5 MB/s at the theoretical maximum. Seeing about 12 MB/s means the connection is delivering essentially its full rated speed, with the small shortfall explained by protocol overhead.
What is the difference between Mbps and MB/s?
The lowercase b means bits and the uppercase B means bytes. One byte is eight bits, so MB/s is eight times larger than Mbps for the same number. Internet plans are sold in bits per second; file transfers are displayed in bytes per second. The capitalisation carries the entire distinction, which is why it is so easily missed.
How long does it take to download a 25 GB file?
At 100 Mbit/s the arithmetic minimum is 2,000 seconds, or about 33 minutes: 25 GB is 200 gigabits, divided by 100 megabits per second. In practice it takes longer because of protocol overhead, server limits and contention. Treat the calculated figure as a floor, not a forecast.
Why are network speeds measured in bits rather than bytes?
Because a serial link genuinely transmits one bit at a time, so bits per second describes what the wire does. The convention also predates the eight-bit byte becoming universal, when byte width varied between machines and only the bit was unambiguous.
Do network prefixes use 1,000 or 1,024?
1,000. Networking has always used decimal prefixes, so a megabit per second is exactly 1,000,000 bits per second. This differs from memory, where binary prefixes are the norm, and it is one reason storage and bandwidth figures should not be mixed without converting.
What is protocol overhead?
The portion of a link's capacity consumed by headers and control traffic rather than payload. Ethernet framing, IP and TCP headers, acknowledgements and any encryption all take a share, typically a few per cent in total. Retransmissions on a lossy path can cost far more.
What is a T1 line?
A legacy digital circuit carrying 1.544 Mbit/s, made up of 24 channels of 64 kbit/s plus framing. It was the standard business connection in North America for decades. T3, at 44.736 Mbit/s, bundles 28 T1s. Both still appear in older contracts and network documentation.
Is upload speed the same as download speed?
Rarely on consumer connections. Most residential broadband is asymmetric, with upload capacity a small fraction of download. The unit conversion is identical in each direction, but the two rates must be converted separately because they are different numbers.
What does Gbps mean in practical terms?
One gigabit per second is 125 megabytes per second, so a gigabit link can in principle move a 25 GB file in about 200 seconds. Reaching that in practice requires the storage at both ends to sustain 125 MB/s, which is often the real bottleneck rather than the network.
Why is my measured speed slightly below the rated line rate?
Because the rated figure is the raw signalling rate on the wire and your measurement counts application payload. The gap is protocol overhead, and a few per cent shortfall is normal and healthy. A much larger gap points to congestion, packet loss, Wi-Fi conditions or a slow server.