Have you ever wondered why that brand new hard drive that was supposedly 80 gigabytes turns out to be a mere 74.5 gigabytes according to your operating system? What happened to that extra space? Did the operating system take it? Was it lost to partitioning? Or, did the Byte Fairy take it? Well, depending on how you count, both the operating system and the manufacturer are right…but, how can that be?

The basic gist to the story is that it all depends on how you count your bytes and how you define prefixes like kilo-, mega-, giga-, and so on.

There really is no discrepancy between the advertised capacity versus the “actual”; they both describe the exact same amount of data. The size of the volume can be denoted as both 80GB (eighty gigabytes), which is what the HDD manufactures typically use, as well as 74.5GiB (seventy-four point five gibibytes), which is what operating systems typically use. The perceived difference stems from the method of notation: SI prefixes versus binary prefixes.

When the first multi-million byte capacity hard drives appeared in the late 1950s, companies started to use the prefix “kilo” to describe 2^10, or 1024, units of storage. Later on, when the SI prefix system was introduced to the industry, 1024 was deemed to be close enough to 1000, as denoted by 10^3, so nothing was changed.

Windows, like many other operating systems, will most often denote the size of a volume using the binary interpretation of the prefix as opposed to the SI interpretation, resulting in a perceived discrepancy. Windows, however, will also display volume size using the decimal system, depending on where you look.

It would seem that the user is losing storage space, which he or she might have been able to use to store more songs, photos or data, but this is incorrect. Sure, this might seem to be the case if the storage capacity of the volume is measured using one definition and the size of the songs were to be measured with the other definition, which, unfortunately, is often times the case. But, if both the file size and storage capacity were to be measured using the same definition, either binary or decimal, then everything would work out fine, mathematically speaking. This is to say that as long as the same definition is used to measure all factors, the exact same number of songs should fit on the volume in question, and there is no “lost” space for an extra 1000, or so, songs.

In short, OEMs will often define a megabyte as 1,000,000 bytes, while operating systems like Windows, Macintosh, and Linux consider a megabyte to be 1,048,576 bytes, resulting in an apparent loss of about 4.5% of the advertised capacity.

So, if this problem is known, why don’t the manufacturers tell the consumers how big their hard drives “really” are. Well, the answer is actually really quite simple: it’s all about marketing. Suppose, that one company were to suddenly decide to be more OS friendly, then their products would appear to be smaller than those of their competitors, which is obviously not the case. Since the average consumer does not know about this counting discrepancy, it makes very little sense for any company to change what they are doing.

Is this false advertisement on the part of the HDD manufacturers? Maybe, maybe not. This all depends on who you talk to, since this whole issue comes down to a matter of opinion. Mathematically speaking, both are right. However, if you really want someone to blame then I would argue that the entire industry is at fault for letting this slide in the first place…