Internet Download/Upload Time Calculator

Enter the file size and pick its unit; this example uses a 4 GB video file. Enter the connection speed in whichever unit you know it. Most home and mobile plans are advertised in Mbps, so that is the default. If a download manager is already showing you a measured rate in MB/s, switch the speed unit instead of converting by hand.

At 50 Mbps the results panel shows a download time of 10m 40s, 640.0 total seconds, an effective throughput of 6.25 MB/s, and a connection speed of 50.0 Mbps. Entering the same connection as 6.25 MB/s produces exactly the same 640 seconds, because the calculator multiplies an MB/s figure by 8 to get Mbps before dividing.

Use the 6.25 MB/s figure as a check. If your browser consistently reports far less than that mid-download, the bottleneck is somewhere other than your plan's headline speed: the server, Wi-Fi, or other traffic on the line.

File size in bytes = entered size × unit multiplier, where KB = 1,000, MB = 1,000,000, GB = 1,000,000,000 and TB = 1,000,000,000,000 bytes. File size in bits = bytes × 8.

Speed in Mbps = the entered value if the unit is Mbps, or the entered MB/s × 8. Speed in bits per second = Mbps × 1,000,000.

Download time (seconds) = file size in bits ÷ speed in bits per second. Effective throughput (MB/s) = bits per second ÷ 8 ÷ 1,000,000, which is simply Mbps ÷ 8.

The time is then split into whole hours, minutes and seconds, each rounded down, so 2,526.3 seconds displays as 42m 6s while the total-seconds row keeps one decimal place.

Size and direction change the answer far more than people expect. A 150 GB game download on a 500 Mbps connection: 150 × 10⁹ × 8 = 1,200,000,000,000 bits ÷ 500,000,000 bits/sec = 2,400 seconds, exactly 40 minutes, at 62.5 MB/s.

Now upload a 60 GB 4K movie master to a cloud drive from a connection with 20 Mbps upload. The file is less than half the size of the game, yet the calculator returns 24,000 seconds, or 6h 40m, at just 2.5 MB/s. Downloading that same 60 GB file at 500 Mbps would take 16 minutes. The upload is 25 times slower purely because of the asymmetric line.

Both results scale linearly: double the file and the time doubles, double the speed and it halves. So before starting an overnight upload, check the upload figure from a speed test, not the download figure on your bill. The same upload speed decides how long an emailed attachment takes to leave your outbox; for documents, the guide to sending a large PDF sets out the attachment limits of the main email and messaging services and the ways round them. When the attachment is a set of phone photos, Compress Image can shrink the whole batch first, reducing the long edge to 2000 or 1600 pixels if you choose, so there is less to push through a slow upload.

The ×8 conversion is the big one, but units cause a second, smaller error. This calculator uses decimal SI prefixes, where MB means 1,000,000 bytes. Many operating systems display file sizes in binary units, where 1 MiB = 1,048,576 bytes and 1 GiB = 1,073,741,824 bytes, the definitions set by the IEC binary prefixes that NIST documents, while still printing the label "GB".

The gap grows with each prefix: 2.4% at KiB, 4.9% at MiB, 7.4% at GiB and 10.0% at TiB. A file your system shows as "40 GB" that is really 40 GiB holds 42.95 decimal GB. At 100 Mbps, typing 40 GB gives 53m 20s, but the true 42.95 GB takes 57m 15s, 7.4% longer. If your OS uses binary units, convert the displayed size first or expect the extra minutes.

Not every bit on the wire is your file. A full-size Ethernet frame with a 1,500-byte payload occupies 1,538 bytes on the link once the 14-byte header, 4-byte checksum, 8-byte preamble and 12-byte inter-frame gap are counted, and 40 of the 1,500 bytes are IPv4 and TCP headers. That leaves 1,460 bytes of file data in 1,538, or 94.9%. With the common TCP timestamp option it drops to 1,448 bytes (94.1%), and over IPv6 to 1,440 bytes (93.6%). Wi-Fi framing, VPN encapsulation and retransmissions after packet loss push the share lower still.

A practical allowance is to reduce the entered speed by 5-10%. The 150 GB game at 475 Mbps (5% off) takes 42m 6s; at 450 Mbps (10% off) it takes 44m 26s. The 60 GB upload at 19 Mbps stretches to 7h 1m 3s. For why an ISP's "up to" speed and delivered throughput often differ by more than framing alone, see the home internet bandwidth guide.

The result is a theoretical best case from the numbers you enter. The calculator does not measure your connection, and it cannot see Wi-Fi signal quality, server-side rate limits, other devices sharing the line, or the difference between an advertised "up to" speed and what is actually sustained. It applies no overhead on its own, so reduce the speed input yourself if you want a planning figure.

It assumes one uninterrupted transfer at a constant rate. It does not model paused or resumed downloads, multi-connection download accelerators that can beat a single-stream estimate, slow starts on high-latency links, or a storage drive that writes slower than the network delivers.

Before estimating a transfer, the bandwidth calculator sizes how much connection speed a household needs for streams and calls. The data storage calculator turns a photo and video library into the GB total you would actually be moving, and the video file size calculator works out how big a recording will be from its bitrate and length, which is the number to feed in here before an upload. When a transfer runs well below even the overhead-adjusted figure, the TCP throughput and latency calculator shows how round-trip time and packet loss cap a single connection's real speed. A PDF of scanned pages or phone photos is often the file in question; the JPG to PDF converter shows the size before it builds the file, so a 40 MB document can be shrunk before it is sent.