Why your download is slower than the calculator says
The download time calculator gives you the best-case number. When the real download takes noticeably longer, the gap has a specific, findable cause.
The download time calculator gives you a best-case number: file size divided by your plan's Mbps, converted to MB/s. When a real download takes noticeably longer than that number, the gap has a specific cause, and most of the time you can find it in a few minutes without calling your ISP.
Check what the download client is actually showing you
Steam, browsers, torrent clients and most download managers report speed in megabytes per second (MB/s). Your ISP sells the connection in megabits per second (Mbps). Those are not the same unit, and mixing them up is the single most common reason a download "looks slow" when it is not. See storage units and transfer rates for the full bits versus bytes breakdown. The short version: divide your Mbps plan by 8 to get the MB/s ceiling, then compare that to what the client shows. A 100 Mbps line tops out around 12.5 MB/s, so a download client reading "11.8 MB/s" is not underperforming, it is close to maxed out.
A speed test looks fine, but the download still crawls
A browser speed test measures your connection to one nearby, well-provisioned test server, usually for a few seconds. A single large download instead depends on the path to one specific file host, for the entire length of the transfer. Those are different routes with different congestion, so a clean speed test result does not guarantee the same throughput to a different destination. If a speed test shows 300 Mbps but a specific download sits at a fraction of that, the bottleneck is downstream of your router, not on your line.
Wi-Fi versus wired
Wi-Fi is a shared radio channel, and its real throughput depends on distance, walls, the band in use, and how many other devices and nearby networks are competing for airtime. A router rated for 500 Mbps can still hand a laptop two rooms away only 40 to 80 Mbps over 2.4 GHz. Plugging that same laptop into Ethernet, or moving to 5 GHz close to the router, often closes most of the gap right away. If a wired test on the same machine downloads noticeably faster than a Wi-Fi one, Wi-Fi was the bottleneck, not the ISP.
Server-side and CDN caps
The file host on the other end may cap the speed of an individual connection, throttle free accounts, or simply be busy. This is common with software update servers, less-used mirrors of popular downloads, and services that prioritize paying tiers. A 1 Gbps home connection cannot beat a server that hands out 20 MB/s per connection, no matter what the local math says. A practical check: if a different mirror or host is offered for the same file, try it. If the second source is noticeably faster, the first server was the limit.
Many small files versus one large file
A single large file streams continuously and can use close to all the available bandwidth. A folder of thousands of small files, common with game installs, source code checkouts, or photo libraries, spends a lot of time on per-file overhead: opening a connection, requesting the file, writing it to disk, and repeating. That overhead adds up even when the raw transfer speed is fine, so an installer copying thousands of small assets can report a much lower average speed than a single large file download on the same connection.
Disk write speed, especially on older drives and cards
A fast connection paired with a slow destination drive bottlenecks at the drive. A spinning hard disk typically writes somewhere in the 80 to 160 MB/s range, and a budget SD card or USB flash drive can be far below that, sometimes under 20 MB/s for sustained writes. On a connection capable of delivering more than 100 MB/s, that write speed becomes the limit long before the network does. This shows up most clearly when copying files to a camera memory card, an older external drive, or a console's expansion storage.
Other devices sharing the line
A household streaming video, running a cloud backup, or downloading something else at the same time all draw from the same connection. Home routers generally split bandwidth across active transfers rather than giving one download the full line, so a single download can look slow simply because other things are pulling data at the same moment. Pausing other traffic and rerunning the download is a quick way to confirm this.
A short check order
When a download is slower than expected, working through these in order usually finds the cause fast:
- Confirm the units. Check whether the client is showing MB/s or Mbps, and compare against your plan speed divided by 8.
- Switch to wired, if possible, and see whether speed changes.
- Pause other downloads, streams and backups on the network, then retry.
- Try a different source or mirror for the same file, if one is offered.
- Check whether the download is many small files rather than one large one.
- Confirm the destination drive is not the bottleneck, especially for SD cards, USB sticks and older hard disks.
Whichever step changes the speed is usually the answer. If none of them do, and speed tests to multiple servers are consistently far below your plan speed, that points back to the ISP itself.
Useful tools for this topic
Frequently Asked Questions
My download client shows MB/s and it looks low. Is something wrong?
Probably not. Steam, browsers and most download managers report megabytes per second (MB/s), while your ISP sells the plan in megabits per second (Mbps). There are 8 bits in a byte, so a 100 Mbps plan tops out around 12.5 MB/s and a 500 Mbps plan around 62.5 MB/s. See storage units and transfer rates for the full conversion. If the client's MB/s number is close to your plan divided by 8, the connection is working as expected.
My speed test says 300 Mbps but one download is much slower. Why?
A speed test measures the route to one nearby test server for a few seconds. A download pulls from one specific file host for the whole transfer, over a different route that can be congested or capped even when your line is not. A clean speed test result only confirms your connection to your ISP is fine, not that every server on the internet can fill it.
Why do lots of small files download slower than one big file of the same total size?
Each small file needs its own request, connection handling and disk write, and that overhead repeats thousands of times in a large folder of small files. A single large file streams continuously with far less overhead per byte, so it can use close to the full available bandwidth while a folder of small assets reports a much lower average speed for the same total size.
Can an SD card or old hard drive actually slow down a fast connection?
Yes. A spinning hard disk typically writes at 80 to 160 MB/s, and a budget SD card or USB flash drive can be well under 20 MB/s for sustained writes. On a connection capable of delivering more than that, the drive becomes the bottleneck before the network does, which is common when copying files onto a camera card or an older external drive.