Bytes, bits - What is my speed?
Users are often confused as to the different definitions of computer data sizes. In turn this leads to misunderstanding of service speeds.
1 bit = a 1 or 0 (b)
8 bits = 1 byte (B)
1024 bytes = 1 Kilobyte (KB)
1024 Kilobytes = 1 Megabyte (MB)
1024 Megabytes = 1 Gigabyte (GB)
1024 Gigabytes = 1 Terabyte (TB)
Computer data is effectively held in binary code either a "0" or a "1". These individual units of binary are the bits.
The odd multipliers relate to how the early computers could move a small number of bits at a time, in groups of 8. Although the metric kilo is in fact 1000, for computing purposes it is in fact 1024. This 1024 principle then applies to the higher Mega and Giga type numbers.
Most confusion usually stems from writing something as bits or Bytes. Think of it this way: Bytes are larger so have a Capital B!
How does this relate to Internet Speeds?
Internet speeds are often written in terms of Mbps, (mega bits per second). The internet provider companies (ISPs) use the bits standard as it makes your connection sound faster, by allowing the use of bigger numbers!
Remember too that the ISPs usually promote the download speed (how fast stuff gets to you), rather than the upload speed (how quick you can send stuff). The upload is usually only a fraction of the download.
Confusion is caused because software often report's speed using different units to your ISP.
Quick Speed Comparision
1 Mbps = 0.125 MBps = 128 KBps = 1,024 Kbps = 131,072 Bps = 1,048,576 bps
4 Mbps = 0.5 MBps = 512 KBps = 4,096 Kbps = 524,288 Bps = 4,193,304 bps
10 Mbps = 1.25 MBps = 1,280 KBps = 10,240 Kbps = 1,310,720 Bps = 10,485,760 bps
20Mbps = 2.5 MBps = 2,560 KBps = 20,480 Kbps = 2,621,440 Bps = 20,971,520 bps
The above figures are based on the traditional methodology. However it is becoming more common for the SI systen to be used, whereby kilo, mega etc all use simple multipliers of 1000. Indeed for purposes of quick calculation of speeds and general capacities, even if you aren't sure of the true units being used, the 1000 multiplier will normally suffice. Strictly speaking the traditional binary systems should be written as KiB, MiB to denote the difference between binary and SI, but this is rarely seen.
Computer purists might argue for slightly a slightly different or more precise basis of calculation, but as a general working principle the above figures are sufficient.