Computer science: MB vs MiB

Self-tutoring about computer science terms: the tutor mentions one he’s noticed.

The following is according to my understanding.

When I was a kid, computer ideas were measured in Kilobytes, or KB. A computer might have, back then, 5KB of RAM: we called that 5K. By itself, K meant “kilo”, and stood for “thousand.” However, it really stood for 2^10, or 1024. Since 1024 was very close to 1000, being off by just 2.4 percent, people were happy enough to accept the misnomer because of its convenience.

Following that custom, MB, or megabyte, was really 1024×1024=1048576 bytes, which was off by about five percent. Once again, “everyone” knew what it meant. In the context of computers, people accepted slightly different definitions of kilo and mega.

Yet, today I was changing a disk partition, and the application mentioned MiB. I had to look up what that means: “Mebibyte.” A Mebibyte, it turns out, is truly 2^20 = 1048576 bytes. There is a whole system for literal memory measures: KiB means “Kibibyte,” which is 2^10 = 1024 bytes, for instance. GiB means Gibibytes, truly 2^30 bytes.

It seems Kibi, Mebi, Gibi, etc. have been around since 1998. I’ve only started noticing them recently, however.

Interesting, eh?

Source:

techtarget.com