What is the Difference Between Mbps and MBps?
MBps is an abbreviation for Megabytes per second, whereas Mbps is an abbreviation for megabits per second. Eight megabits equal one Megabyte. These abbreviations are commonly used to specify how much data can be transferred per second between two points. In some cases, mbps is used as an abbreviation for megabits per second; however, the lowercase m technically means "milli" not "mega," so it doesn't really mean the same thing.
To put megabits and Megabytes in perspective, let's back up for just a moment. One bit of data is a single "on" or "off" digit, a one or zero. It takes eight bits to represent a single character, or one byte of data.
- 8 bits = 1 byte
- 1000 bytes = 8 kilobits (kb) = 1 Kilobyte (KB)
- 1000 Kilobytes (KB) = 8 megabits (Mb) = 1 Megabyte (MB)
As a point of possible confusion it should be mentioned that there are two different systems for calculating multiples of data: the decimal system as noted above, and the binary system.
According to the binary system, used in relation to computer storage and memory, it takes not 1000 bytes to equal a KB, but 1024 bytes. This is because the binary system is base 2, and 210 = 1024. Technically, the designations in this case are Kibibyte (KiB) and Mebibyte (MiB), but these haven't caught on in the public sector, leading many uses of "MB" to mean 1024 kilobytes, and others to mean 1000 kilobytes. When considering MBps, however, the decimal system applies, as the reference is to data transfer rates and not data storage.
Data transfer rates are quite handy for gauging performance levels of various hardware devices. Everything from USB and Firewire® ports to memory card readers and mobile devices are associated with corresponding transfer rates, often measured in megabits or Megabytes per second.
We must also translate speed to value when considering Internet service plans, advertised by download and upload speeds expressed in kilobits per second (kbps) or megabits per second. For example, a typical Digital Subscriber Line (DSL) plan might have an upper transfer limit of 1,500 kbps, which can also be expressed as 1.5 Mbps. A cable plan might be advertised with speeds up to 5,000 kbps or 5 Mbps; and fiber optic Internet can reach speeds of 50 Mbps or more.
The wireless G network (802.11g) has a maximum transfer rate of 54 Mbps, making it much faster than all but the fastest fiber optic Internet plans. Thankfully, going wireless won’t slow your surfing. The more current wireless N standard (802.11n) can’t speed up your Internet connection, but will allow faster data transfer rates between local networked computers of up to 100 Mbps, or about twice the data transfer rate of G networks.
As if the abbreviations aren't close enough to cause confusion, it doesn’t help that they are often expressed in the wrong case. When in doubt look for translations such as the kilobit or Kilobyte equivalent, or simply ask someone if the specification is indeed megabits per second or Megabytes per second.
If you have read the fine print, a Megabyte has always been 1024 bytes as well as a megabit being 1024 bits. It was never rounded using base 10, it was used as a shortcut rather than specify the actual size of discs and memory. The REASON it IS 1024 and not 1000 is because computers "think' in base 2 - on or off. When you get close enough to being a thousand, it is 1024: 2,4,8,16,32,64,128,256,512,1024, etc. Which is created by: 2^1, 2^2, etc. Since you can derive all the letters of the alphabet in 5 bits, but you want to add the special characters and letters for foreign languages (ñ, ü, ø, ã, etc), you need more than 5 and since it's a computer in base 2, you go up and use 8 bits.
When you start to count in base 2, you get: 0, 1, 10, 11, 100, 101, 110, 111, (in base 10 this is 1,2,3,4,5,6,7) etc. If you put these bits together you can group them in 8 bits so using the 8 bits you can get 00000000 to 11111111. Reading this can be difficult, so programmers have split the 4 bits into 2 groups of 4: 0000 0000 up to 1111 1111. This can be shortened further if you use base 16 because 1111 in base 2 is 15 in base 16. and to count in base 16, you count this way: 0,1,2,3,4,5,6,7,8,9,A,B,C,D,E,F, 10, 11, 12, 13, 14 -> which is how you count to 20(base 10) in base 16.
You can shorten the reading of the bits to base 16: 0000 0000 is zero, 1111 1111 is FF. (0 to 128 base 10). SO the the max number in 1 word in 8 bit word computers is 1111 1111 in base 2, or FF in base 16 which is 255 in base 10.
The first microprocessors had 8 bit words. Each word was a computer instruction (sometimes it took more than 1 word, 2, 3 or more). In modern computers the microprocessors use 64 bit words! So the max number per word is 1111 1111 1111 1111 1111 1111 1111 1111 1111 1111 1111 1111 1111 1111 1111 1111, which is FFFFFFFFFFFFFFFF, which is 2( ^ 64) -1 which is 1.8 x 10^ 19 base 10 (a large number!).
So, for example, if you buy a 1MB disk (which is small in today's market) you get not 1 million bytes, but 1,048,576 (base 10) bytes of storage (in 8 bit word size. That size: 8 bits per word for storage, has not changed since the early storage devices despite the instruction word size in the processors, now up to 65 and some I've seen at 128).
Sajith: In Kerala ISP's are providing internet in downloading speed. But what they mean I don't understand. They tell me the line is 2 Mbps speed, but we will get downloading speed below 256kbps only. So, 2 Mbps divided by 8. Therefore, why they are telling us the downloading speed is 2 Mbps?
I've been in the Information Technology field for over eight years and I have noticed an increase in techs old and new using mb and MB interchangeably. I always correct these said techs. Truth is, I know what they mean and when they type it, they type it correctly. It's when it's spoken that it turns into megabits.
I found this article great because after hearing all this megabit of a mess, I started saying to myself, "Have I been doing this long enough to not even realize I have been using the terminology incorrectly? Or has the whole world gone mad?" Needless to say, it was neither. I wasn't aware of there being two "systems." I've always equated bits and bytes to binary and the almighty 1024. I think the ISP's use the mbps in combination with the 1000 system to mislead uninformed consumers. It's all a sales gimmick, in my opinion.
Furthermore, for those who still don't quite grasp it, I subscribe for a 10(mbit) line with Comcast, which equates to 10,240(kBytes) for binary value or 10000(kbits) on the 1000 system depending on ISP. Now if I go download a file with a 10meg (snicker) line, I theoretically should pull that file at a transfer rate of 1.25MBytes per second. Now, of course, again, that's the binary value. If Comcast uses the 1000 system, I will pull that file at a theoretical 1.2MBytes per second.
And ladies and gents, as for me and my brain, the binary is much easier to process. And even if I'm wrong in guessing an ISP is using actual binary values, I can calculate it much faster than the other system.
Humans think in base 10
1 kilo is normally 1000, as per the metric system's prefixes.
One would expect a kilo of bytes to be 1000 bytes ... but it's 1024 usually.
This is because 1024 is 2^10 (2 to the 10th power), conveniently close to 10^3 (1000). In computers, base-2 shows up over and over again.
So, 1 kilobyte is 2^10=1024
1 megabyte is not 1000x1000, but instead 1024x1024 or 2^20
and 1 gigabyte is 1024 megabytes or 1024 * 1024
and so on (terabytes, petabytes ...)
There are some computer companies however who have very confusingly broken tradition for example you buy a 2 terabyte hard drive but its actually 2tb based on the metric system of counting not base 2 so its actually 1.7 - 1.8tb and then you loose some size to formatting it.
Personally I feel that 2tb hard drives should be based around computer methods of counting using base 2 because I feel anything else is actually ripping people off.
@post 11: You forget to take into account the bandwidth between the computer and your router is limited and the fact that it takes time for the signal to travel from your computer to your isp and from your isp maybe halfway around the world. The server that has the song takes time to send it to his isp and from his isp to your isp and from your isp to your router and to you and your computer to read the info it just got and figure out how to use it and then finally start downloading and then it's done.
You cannot expect it to do this in one second so it's not that the speed isn't 8MBps -- it probably isn't. Anyway, because the ips only says up to, so it depends on your line, but that's not the point. The point is that there isn't enough bandwidth between your computer and your router to get 1 MB on a 8MB connection in under one second. To do that, you would have to have a fiber optics connection.
@anon48950: I need some help to understand the comment. My question is: Why are there 1024 bytes? (instead of bits). Why is there a change from bits to bites? For example, 32, 64, 512 bits, but 1024 bytes.
2.73 mbps for download and 0.90 up is good speed or I have to get a better one.
What really needs to be clarified is the following: Mbps is indeed used by ISP to their advantage since the number is eight (8) times higher than the actual transfer rate you, the end user, are interested in (you are transferring data calculated in bytes not the bites ;) ).
Example: You want to download an 8MB MP3 file and you have a 8Mbps bandwidth. You may figure it out that it will take 1 sec, when in fact it will take you 8 seconds since the real speed is 1MBps.
In other words, when you go to an ISP and get a subscription, divide the advertise speed by 8 to see what will actually get ;)
Also, don't bother too much with the 1024 vs. 1000, it is a 2.4% difference.
Thanks for the lesson. My problem is finally solved.
I need to know in details about that Mbps and kbps. Whatever information is given is there in detail, but i want to know elaborately.
I agree with Barnsk, although 2^10 is 1K, i.e 1024 (as opposed to 1k which is 1000)
Barnsk brings up the one obvious error in the article. Without going into a detailed history, computer hardware manages memory best in powers of 2 with 1 bit (short for binary digit) being the smallest possible unit of memory. 2 bits is a dibit or di-bit, 4 bits is a nybble. 8 bits is a byte. 16 bits is a word. 32 bits is a double word or full word. Jumping up some past 64, 128, 256, and 512, we arrive at 1024 which, for convenience purposes, became known as a kilobyte even though it wasn't exactly 1000 bytes. I guess you could consider it rounding. I suspect what the wiseGeek was referring to was that in data communications every packet of data transmitted consists of a packet header, some data, and a packet trailer. The header and trailer perform a function similar to that which an envelop serves for the post office. It keeps that data separate from all the other data floating around in a communication network as well as identifying who sent it and who is supposed to receive it. There are many different standards defined on how to build a packet and at least one of them uses 24 bytes of header and trailer to transmit the 1000 bytes of raw data. - jsiegman
Very nice article! Must read, as many people (like me) are confused between MBps and Mbps. And that can be harmful when buying a product or service. Thank you wiseGEEK! :D
This is an article that needed to be written 'because it confuses people, but sadly...' This is because 24 extra bytes are used to store indexing or mapping information about the 1000 bytes of data.'. What is this about? 1024=2^10, which is near to 1000 and is therefore called 1k. Nothing to do with any 'extra' for 'storing' or 'mapping'. What computer accesses data in powers-of-ten chunks? Computers, though digital, don't have fingers on which to count, unlike their ape-like creators.
Post your comments