The speedtest measures the maximum download speed from pixeldrain's servers to your computer. This speed is not affected by the daily download limit for free users.
The speed measured is the maximum speed reached for two continuous seconds during the test. The normal test has a time limit of 12 seconds and the long test 30 seconds. When the maximum speed has not changed for a third of the test duration (4 or 10 seconds) the test will end and the result will be final. The long test will use much more bandwidth, so if you're on a cap it's recommended to use the normal test.
The speed gauge uses a logarithmic scale, which means that each datapoint is 10 times more than the last one.
The first number is the download speed in bytes per second. This is useful for estimating how long a file download will take.
The second number shows the download speed in bits. This is usually the speed your ISP advertises to you. This number is equal to the first number multiplied by eight, because there are eight bits in a byte.
The third number shows the latency to the pixeldrain servers. This is dependent on how far you are removed from the closest pixeldrain server. The lower the latency the faster your downloads will be, generally. The number shows request latency and not ping latency. HTTP requests have some overhead which means this latency number shows multiple round trips instead of one.
The last number shows how much data the speedtest was able to transfer in the duration of the test.
Most speed tests have servers in datacenters which are located very close to your home. They are also connected directly to your ISP which means that the ISP does not have to pay for the bandwidth because it stays within their network.
Pixeldrain does not have this luxury. Because our budget is very small we are only able to afford the cheapest bandwidth available. This means that the data has to travel further and is more likely to be throttled.
VPN services also add overhead to your connections, which lowers effective download speed. VPNs using OpenVPN lower your download speed by approximately 17.2%, and VPNs using the more modern WireGuard protocol lower your download speed by approximately 4.5%. To get most accurate results make sure you have any VPN services turned off.
If the speed is a lot slower than your usual downloads it can mean two things.
Number two is usually not the case, when that happens I will tweet about it. Unfortunately there is nothing I can do about the first scenario. I might be able to get more servers and better bandwidth if the site gets many more subscribers, but for now that's out of reach.