Convert Mbps ↔ MB/s ↔ Gbps and calculate download time for any file size at any internet speed, with real-world overhead and file presets (4K movie, OS install, game patch).
Built and maintained by Paul Clark, Redmoon Software · Checked against what this tool computes on
| Conversion | Value |
|---|
1 byte = 8 bits. ISPs advertise in megabits per second (Mbps); files are in megabytes (MB). MB/s = Mbps ÷ 8. Time = file size ÷ throughput. We use decimal (1 GB = 1 000 MB) since storage and ISP marketing follow SI prefixes.
Real-world overhead (~10%) accounts for TCP / IP headers, retransmits and bursty servers. Theoretical time × 1.111 ≈ practical time.
Download time = file size in bytes divided by (bits per second / 8). Every unit here is decimal SI rather than binary: 1 Mbps is 1,000,000 bits per second and 1 GB is 1,000,000,000 bytes, matching how ISPs advertise speed and how drives are labelled. Tick Apply ~10% real-world overhead and the time is multiplied by 1.111, the reciprocal of 90% goodput, covering TCP and IP headers, retransmits and bursty servers.
The speed Unit field is the one people get wrong. Kbps, Mbps and Gbps are bits; KB/s, MB/s and GB/s are bytes, and the eightfold gap between them is the whole reason this tool exists. Defaults are 100 Mbps against a 5 GB file, which comes out at 6m 40s. Six presets fill the file size in: MP3 song (5 MB), 1080p movie (5 GB), 4K movie (25 GB), OS install (5 GB), Game patch (50 GB) and AAA game (100 GB).
This is a straight division, so it assumes you hold your full advertised line rate for the entire transfer. It models nothing about the source server's own upload cap, Wi-Fi loss, evening congestion, disk write speed, or TCP slow start at the beginning of a connection. The overhead option is a single flat 10% allowance rather than a protocol simulation, and the conversion table always lists both the theoretical and the padded time regardless of the checkbox.
12.5 MB/s, because there are eight bits in a byte and the tool uses decimal megabits. The conversion table prints that directly, alongside the same speed expressed in Mbps and Gbps, whichever unit you originally typed in.
At 100 Mbps, about 2h 13m theoretically, or roughly 2h 28m once the 10% overhead allowance is applied. The AAA game (100 GB) preset fills the file size in for you, so you only need to set your line speed.
1,000. Kilo, mega, giga and tera are all powers of ten here, for both speeds and file sizes, because ISPs and drive manufacturers use SI prefixes. Your operating system may report the very same file as a smaller number of gigabytes.
The maths assumes your full line rate end to end. In practice the source server may throttle you, Wi-Fi and peak-time congestion cut throughput, and connections need time to ramp up. The overhead checkbox only covers protocol overhead, around 10%.
It is a fixed reference point: how long a single gigabyte takes at whatever speed you entered, independent of the file size field. Useful for mental arithmetic, since you can multiply it by however many gigabytes you actually need to move.
Two conversions stack here and both catch people. Network speeds are quoted in bits per second and file sizes in bytes, a factor of eight — so a 100 Mb/s connection moves at best 12.5 MB/s. Then decimal and binary prefixes diverge: a decimal megabyte is 10⁶ bytes and a mebibyte 2²⁰, about 4.9% larger, and the gap widens with each prefix step to about 7.4% at giga and 10% at tera.
That second gap is why a 1 TB drive shows as roughly 931 GB — the manufacturer counted decimal terabytes and the operating system reported binary ones. Nothing is missing.
Protocol overhead consumes part of every link: TCP/IP headers, acknowledgements and retransmissions typically cost several percent before anything else. Latency matters as much as bandwidth for many transfers, because a high-latency link with a small window size cannot keep enough data in flight to saturate its capacity regardless of how wide it is.
Storage on either end is frequently the real bottleneck, and on shared or asymmetric consumer connections the upload figure is often a small fraction of the download one. Plan transfers on measured throughput rather than on the line rate.
The standard or formula behind this tool, so you can check it rather than take it on trust.
IEC 80000-13 binary prefixes (KiB, MiB) versus SI decimal prefixes (kB, MB).
A "100 Mb" line moves at most 12.5 MB per second — bits versus bytes is a factor of eight, and decimal versus binary prefixes is a further ~7% at the gigabyte scale.
Cron Expression Builder & Translator
Build cron expressions visually and translate any cron string to plain English with the next 5 fire times.
Unix Timestamp Converter (with Timezone)
Bidirectional Unix timestamp ↔ human date converter with timezones, milliseconds and ISO 8601.
Regex Tester with Plain English Explainer
Test regular expressions live with match highlighting and a plain English breakdown of every part of the pattern.
