It can be almost as annoying as one of those math problems from high school -- you know the ones, where two nimrods with nothing better to do each get on a train, and you have to calculate where and when they meet, or lose their luggage, or something:

This file is 12 MB, and my download speed is 15 Mbps, so it should download in... eight tenths of a second? Wait, that can't be right... Oh, okay, one's a B and one's a b, so that means... uhh... Lemme just download it and see... Wait, Kbps? Shouldn't that K be an M? Can't these people just pick a winner here?

Well, the good news is that with a 15 Mbps FiOS connection (not even the fastest one available, by the way), your file probably would've downloaded in less time than it took you to read that last paragraph. The bad news is that, yes, all these megas and kilos and bits and bytes can get pretty confusing. So let's start with the humble little bit.

A bit, abbreviated with a lowercase b, is the smallest unit of binary information, usually a 1 or 0. A byte is made up of eight bits, and is abbreviated with a capital B.

Store Bytes, Move Bits

What can make things confusing is the fact that bytes usually are used as units for data storage (the size of your files), while bits usually are used as units for data transfer (the amount of info moving through a network). So while your file may be 12 megabytes in size, your download speed is 15 megabits per second. If you want to know how many megabytes of data you're downloading per second, tragically, some arithmetic may be inevitable. (Don't get too scared -- you'll actually be dividing by 10, not 8, which is much easier to do in your head. More on that later.)

Kilos and Megas and Gigas, Oh My!

But wait, your computer's transfer rate -- signified by the little meter you see when you download something -- is probably in kilos, not megas. Oh brother, more math.

Making matters even more confusing is a decades-old argument over what a "kilo" is. Back in the day, when computers were the size of armored tanks and had the memory capacity of a goldfish, computer engineers working in base 2 (binary) noticed that 210 (or 1024) was pretty darn close to 1,000, so they adopted the prefix "kilo" for short. Thus, if you said the word "kilobyte" to some computer engineers, they'd all know you actually were talking about 210, or 1024, bytes.

As computers began moving into the homes of people who weren't computer geeks, though, and who were used to the decimal and metric systems, it was hard to shake the old notion that "kilo = 1000." Which isn't too far off if you're just talking about a few thousand bytes. Even the term "mega," which originally meant 2gn: super;">20, or 1,048,576, was fairly close to one million, so people thought of a megabyte as about a million bytes. However, as the speed and memory capacities of computers skyrocketed over the years into prefixes like "giga," "tera" and "peta," the discrepancies between binary and decimal units grew exponentially bigger.

To try to put an end to the confusion, standards boards such as the IEEE decided in the late 1990s to go with the decimal system for units of data:

kilo = 103 = 1000
mega = 106 = 1,000,000
giga = 109 = 1,000,000,000

...and so on. (Although the file managers on many computers still use the old binary units (kilobyte = 1024 bytes) which is why your 100 GB hard drive might appear to be "missing" a couple of gigabytes when you check your system profile.)

Okay, good deal, IEEE. Confusion vanquished, thanks. Understanding transfer rates, and figuring out if I'm downloading as fast as I should be, ought to be a snap now, right? Well, yes and no.

Throughput: Info, Minus Two Bits

To really understand how it all works, you should know a bit (no pun intended) about s y_mce/themes/advanced/langs/en.js" type="text/javascript"> // --> omething ca lled throughput as well. Throughput is a measure of the actual data delivered through a network, and will always be less than your maximum download rate. This is because data is delivered over networks in small chunks, called packets, usually consisting o f 8 bits. So for every 8 bits of data that move, there are 2 additional bits that go with them, to signify where the packet starts and stops.

Those 2 extra bits do make it easier to eyeball how much data you're actually moving, though, since you're dividing by 10 now instead of 8.

Let's say you've got a 15 Mbps (megabit per second) FiOS connection -- that speed is counting all those "meaningless" (to you, anyway) start/stop bits as well as the "useful" data you're downloading. So...

15 Mbps ÷ 10 = 1.5 MBps actual throughput (approx.)

In other words, if your download speed is 15 megabits per second, you can expect to download roughly 1.5 megabytes per second of that MP3 or movie trailer.

Other variables, such as packet loss and how busy the server on the other end is at a given time, can affect your throughput rates as well, so throughput is not always an exact science. But dividing your download speed (in Mbps) by 10 to determine your ideal throughput in MBps is the quickest and easiest way to give you a rough idea.

So, with a throughput rate of about 1.5 megabytes per second, and the 12 megabyte file mentioned above, we're looking at....

12 MB ÷ 1.5 MB/s = 8 s

...or about 8 seconds, give or take, to download that 12 MB file.

A more accurate way to measure your real throughput speed, though, is to download several large files off the Internet (ideally files from your ISP), divide the size of each file by the time it took to download it, and average them out.

There are also several websites that measure throughput speed, such as speedtest.verizon.net.

When you have FiOS, though, the best thing to do is just relax and let your file download -- it'll probably be done before it occurs to you to bust out the math formulas, anyway.

Posted by Jim 2.0 | tagged: bandwidth, speed, Mbps, internet, upload, download, FiOS