Skip to content
Technologydata

Data Transfer Time Calculator

How long a file takes to move over a link, with the bits-to-bytes conversion made explicit and a realistic efficiency rather than the rated speed. A 100 Mbps link does not move 100 MB a second, it moves about 12.5 at best.

Also called: download time calculator, file transfer time.

Speed unit
%

Protocol overhead and congestion mean a link rarely delivers its rated speed.

Transfer time
41.7 minutes

41.7 minutes to move 25,000,000,000 bytes over a link delivering 80 Mbps, which is 80% of its rated speed. The link carries 10 MB a second at this efficiency, not 100. Link speeds are quoted in bits and files are measured in bytes, a factor of eight. At the rated speed it would be 33.3 minutes. The difference is protocol overhead, acknowledgements and contention, which is why the rated figure is a ceiling rather than a forecast.

Seconds
2,500
Seconds at the rated speed
2,000
Effective throughput
80
Bytes to move
25,000,000,000
Megabytes a second
10
Size unit
GB
Speed unit
Mbps
On bits and bytes
The link carries 10 MB a second at this efficiency, not 100. Link speeds are quoted in bits and files are measured in bytes, a factor of eight.
On efficiency
At the rated speed it would be 33.3 minutes. The difference is protocol overhead, acknowledgements and contention, which is why the rated figure is a ceiling rather than a forecast.

The same file on other links

LinkAt this efficiencyAt the rated speed
10 Mbps6.94 hours5.56 hours
50 Mbps1.39 hours1.11 hours
100 Mbps41.7 minutes33.3 minutes
300 Mbps13.9 minutes11.1 minutes
1 Gbps4.2 minutes3.3 minutes
Method and background

How this is calculated

File sizes are quoted in bytes and link speeds in bits, so the size multiplies by eight before it meets the rate. That single factor is why a download looks eight times slower than people expect. Efficiency then accounts for the rest: protocol headers, acknowledgements, congestion control and shared contention mean a link almost never delivers its rated figure, and eighty percent is a reasonable working assumption on a good connection. Parallel streams help on a long fat link where a single stream is limited by the window size rather than by bandwidth, and help not at all when the bottleneck is the link itself.

the file is in bytes and the link is in bits, so the factor of eight is the conversion people forget
S
File size in bytes
R
Link rate in bits per second
e
Efficiency actually achieved

Worked examples

Each of these is asserted on every build. If a change to the engine ever moved one of these answers, the build would fail before the page could print it.

25 GB over 100 Mbps

File size
25
Size unit
GB, base 10
Link speed
100
Speed unit
Mbps
Link efficiency actually achieved
80%
Parallel streams
1

Transfer time41.7 minutes

25e9 x 8 / 80e6

Open this example

base 2 units move more bytes

File size
25
Size unit
GiB, base 2
Link speed
100
Speed unit
Mbps
Link efficiency actually achieved
80%
Parallel streams
1

Transfer time44.7 minutes

boundary: 25 GiB is 7.4% more data than 25 GB

Open this example

Written about this

  • Your 1 TB drive is not missing 69 GB · 1 min · A new 1 TB drive reports 931 GB and nothing is wrong. The manufacturer and the operating system are counting in different bases, and both are telling the truth.

Method and limits

What it assumes

  • A sustained transfer, so ramp-up and connection setup are ignored.

What it deliberately does not model

  • The real bottleneck is often disk, CPU or the far end rather than the link.
  • Parallel streams only help when a single stream cannot fill the pipe.

Formula version 1.0.0 · definition 1.0.0 · United States · Report a problem with this calculator

Frequently asked questions

Why is my download slower than my connection speed?
Partly the factor of eight between bits and bytes, partly overhead. A 100 Mbps link tops out around 12.5 MB per second in theory and closer to 10 in practice.
Do parallel streams make transfers faster?
Only when a single stream cannot fill the link, which happens on high-latency long-distance paths. If the link itself is the bottleneck, splitting the file just splits the same bandwidth.