Speed, Throughput and Utilization

When we deal with computations involving time, speed and utilization, we use the following units:
y y

time is measured in seconds speed is measured in "cycles" per second, or Hertz (abbreviated "Hz") In this context, a cycle is one occurence of an event which occurs repeatedly; such events include memory accesses within the PC and bit transfers on a network segment. As a result, Hz also commonly refers to a transfer rate measured in bits per second.

y y

data transfer rate or throughput is a speed, usually measured in bits or bytes per second utilization is dimensionless (it has no units, as we will see below)

Modern computers are very fast; this means that we will be concerned with very small times. For this reason, you will need to be familiar with the numerical abbreviations
y y y y

milli (10 -3), denoted by the prefix m (1 ms = 1 millisecond = 10 -3 seconds); micro (10 -6), denoted by the prefix (the Greek letter "mu"; 1 s = 1 microsecond = 10 -6 seconds); nano (10 -9), denoted by the prefix n (1 ns = 1 nanosecond = 10 -9 seconds) and pico (10 -12), denoted by the prefix p (1 ps = 1 picosecond = 10 -12 seconds).

In addition, when dealing with time and transfer rates the prefixes kilo, mega, giga, etc. all refer to powers of ten:
y y y

1 KHz = 1,000 Hertz 1 MHz = 1,000,000 Hertz 1 GHz = 1,000,000,000 Hertz

This is in contrast to computations of space requirements in which those prefixes refer to powers of two. This is a bit confusing in the beginning, but the correct usage for a given computation should be clear from the context of the problem. Computer speeds and transfer rates are typically measured in terms of frequency (cycles per second or Hertz): a 733 MHz CPU chip has a clock which ticks 733,000,000 times each second, and a 10 MHz Ethernet segment can transfer 10 million bits per second.

In these contexts, a Hertz is one clock tick per second and one bit transferred per second, respectively. Since the clock time (or cycle time) is measured in seconds per clock tick or seconds per cycle, it is just the reciprocal of the frequency. So a 733 MHz CPU chip has a cycle time of 1 / 733,000,000 seconds = 1.36 * 10 9 seconds = 1.36 ns. Similarly, the 10 MHz Ethernet segment transfers one bit in 1 / 10,000,000 seconds = 1 * 10 7 seconds = 0.1 s. A bus in a computer is simply a group of wires over which data or commands are transferred (ie., between the CPU and IDE controller). There are many types of buses found in PCs:
Bus Frequency (MHz) 100 133 33 66 12 480 400 MHz 800 MHz Clock Time* ( s) 0.01 0.0075 0.033 0.016 0.083 0.00208 0.0025 0.00125 .0031 .0016 .0008

PC-100 PC-133 PCI-1 (Peripheral Control Interface) PCI-2 USB-1 (Universal Serial Bus) USB-2 FireWire (IEEE-1394) FireWire-2 Ultra Wide SCSI (Small Computer System 320 Interface) Ultra2 Wide SCSI 640 SCSI-3 1280

* some clock times are approximate There are many factors which determine the speed of a computer: CPU frequency, motherboard frequency, I/O bus frequency, the amount of cache memory, and even the kind of software being run. CPU speeds are often measured in MIPS or millions of instructions per second, or in FLOPS (floating point operations per second). These two will depend both on the CPU architecture and the kind of program being run by the CPU, so there is no simple way to compare the performance of difference CPU chips. Most of the computations involved in computing capacity are products or quotients of numbers with well-defined units. We have seen that we can use these units to construct formulae for various computations. For instance, suppose we wish to compute the lifetime of a packet on a network segment (the length of time it takes to send the packet). We proceed as follows:

1. The results clearly should have units of seconds: ? seconds = 1 packet 2. The computation requires the size of the packet in bytes: assume that there are 150 bytes / packet as well as the frequency in bits per second: let the frequency = 10 MHZ = 10,000,000 bits / second 3. There is a conversion factor of 8 bits / byte. 4. Examining the units, we see that if we multiply the size of the packet in bytes times the conversion factor of 8 bits per byte and divide by 10,000,000 bits per second, we are left with units of seconds: ? seconds = 1 packet * ( 150 bytes / packet ) * ( 8 bits / byte ) / ( 10,000,000 bits / second ) = 1.2 * 10 -4 seconds = .12 ms = 120 s Sometimes we need to perform the same calculation with a number of values for a given parameter (ie., the packet size). For instance, we might need to know the lifetimes of packets of 150 bytes, 350 bytes and 650 bytes. We can use a simple proportional calculation to obtain the remaining answers once we have computed the first:
y y

a 350 byte packet takes ( 350 / 150 ) * 120 s = 280 s a 650 byte packet takes ( 650 / 150 ) * 120 s = 520 s

Or perhaps we are interested in the lifetime if we speed up the network segment to 100 MHz: 120 s / ( 100 MHz / 10 MHz ) = 12 s Here we have divided by the ratio of the speeds because the speed was in the denominator of the original calculation: the lifetime is inversely proportional to the speed. We define utilization as

the ratio of the amount of a resource actually used to the maximum amount that could possibly be used. Utilization therefore has no units, and should be between zero and one, or between 0 and 100% (since it is not possible to use less than nothing or more than is possible). If we suppose that five hundred of our 150 byte packets are sent each second on a network segment, we can compute the utilization as follows:
1. ? = actual throughput / maximum possible throughput

2. Actual throughput is usually measured in bits per second for network segments (which is the same units as Hz): 500 packets / second * 150 bytes / packet * 8 bits / byte = 600,000 bits / second 3. Possible throughput is 10 MHz, so the utilization is 600,000 bits / second / ( 10,000,000 bits / second ) = .06 = 6 % Note here that the "per cent" ("%") is literally " / 100 ", and is not a unit or dimension at all. We can use proportionalities to compute further utilizations as before:
y y

if the packets are 1200 bytes long, the utilization is .06 * ( 1200 / 150 ) = .48 = 48 % if the speed of the network segment is 100 MHz, the utilization is .06 / ( 100 MHz / 10 MHz ) = .006 = .6 %

Storage Requirements
We can also use units to guide our computations of space requirements for multimedia applications. Consider the question of how much video RAM (Random Access Memory) is required for a given monitor (or display) setting. The display setting is specified by
y y

the width and height (or resolution) in pixels (picture elements), and the number of colors (depth) per pixel.

A typical resolution might be

800 pixels wide by 600 pixels in height, with a depth of 65 thousand colors per pixel. Vertical resolution is sometimes given in lines, so that the above would be denoted as 600 lines, each of which is 800 pixels wide. In this usage, a line is not a real unit; the resolution is computed by simplying multiplying the horizontal pixel resolution by the vertical line resolution, and the result is given in pixels. Thus a two dimensional video image is represented as a simple list of pixels; in this example, the computer would start a new line after every 800 pixels. Since 65,536 is 2 16, we will need 16 bits for each pixel. We then compute the amount of video memory required:
1. ? bytes = number of pixels * bytes per pixel

2. number of pixels = 800 * 600 = 480,000 pixels 3. bytes per pixel = 16 bits / pixel / ( 8 bits / byte ) = 2 bytes / pixel 4. bytes of video RAM = 480,000 pixels * 2 bytes / pixel = 960,000 bytes We can again use proportionalities to do additional problems:
y

For 1024 by 768 16 bit resolution, we need 960,000 bytes * ( 1024 * 768 pixels / ( 800 * 600 pixels)) = 1,572,864 bytes / ( 2 20 bytes / MB ) = 1.5 MB

a Truecolor (24 bits per pixel) digital camera image which is 2048 by 1536 pixels requires 960,000 bytes * 24 bits / pixel / ( 16 bits / pixel) * 2048 * 1536 pixels / ( 800 * 600 pixels) = 9,437,184 bytes / ( 2 20 bytes / MB ) = 9 MB

Here we have used powers of 2 for conversions to MB, since we are dealing with storage requirements. The space computations for audio files can be done in much the same way. A microphone converts sound into changing voltages:

In order to digitize the sound, we need to convert the changing voltages into a series of numbers. To do this, we sample the voltages periodically:

For CD quality sound, the sample rate is 44,100 samples per second. Of course, for stereo recordings we need twice that many samples per second, one for each channel. We also need to decide how many bits are necessary to hold the voltage for each sample:

For CD quality sound 16 bits are used. This corresponds to 2 16 voltage divisions in the graph above. The amount of space required for one minute of CD quality sound is then ? bytes = 1 minute * ( 60 seconds / minute) * ( 44,100 samples / second / channel) * 2 channels * ( 16 bits / sample) / ( 8 bits / byte) = 10,584,000 bytes / ( 2 20 bytes / MB) = 10.09 MB

Here, the 44,100 samples per second per channel was represented with units of samples / second / channel. For a 3:21 (3 minutes, 21 seconds or 201 seconds) mono (one channel) audio file with a sample rate of 22050 Hz and a sample depth of 8 bits, we require 10.09 MB * ( 201 seconds / 60 seconds) * (1 channel / 2 channels) * ( 8 bits / sample / ( 16 bits / sample)) = 8.45 MB We can perform a similar computation for full motion video. Using the NTSC standard and VHS quality, full motion video involves a resolution of 352 pixels in width by 240 lines in height for each field; two fields are shown for each frame, and there are 29.97 frames per second. If we use 24 bits for each pixel (TrueColor), we can compute the space requirements for 1 minute of VHS video: ? bytes = ( 352 pixels * 240 lines / field) * ( 2 fields / frame) * ( 29.97 frames / second) * ( 60 seconds / minute ) * ( 24 bits / pixel) / ( 8 bits / byte * 2 20 bytes / MB) = 869.25 MB Note that the vertical "lines" units were omitted in the computation, as per the discussion above. For a 3:21 full-motion video clip, we would need 869.25 MB * ( 201 seconds / 60 seconds ) = 2912 MB / ( 2 10 MB / GB) = 2.844 GB It is clear that capturing full motion video without some sort of compression is a daunting task. It is not even possible to do so on many computers; for one second of full motion video, we need 869.25 MB / minute / ( 60 seconds / minute) = 14.5 MB /second throughput; many PCs cannot achieve the disk throughput necessary to store full motion video. A series of compression standards have been established by the Motion Picture Experts Group (MPEG) which makes the use of digital audio and video practical for today's computers. The compression techniques are "lossy" in that information is lost during compression; they are variable in that you can choose the degree of loss. In each case, a target bit rate is established and the space requirement is then the product of the bit rate and the length of time of the recording. For instance, an audio mp3 file

(technically MPEG audio layer 3) containing one minute of audio recorded at 128K bits per second requires 1 minute * ( 60 seconds / minute) * ( 128,000 bits / second) / ( 8 bits / byte) = 960,000 bytes / ( 2 10 bytes / KB) = 937.5 KB Note that we used 128,000 for 128 K, since it is a frequency and not a storage requirement. The MPEG 1 standard is used for video CDs and the MPEG 2 standard is used for DVDs; the bit rates range between 1.15 Mbits per second (fixed for MPEG 1) and between 5 and 9.8 Mbits per second (variable) for MPEG 2. So, for example:
y

a 3:21 mp3 file with a bit rate of 64 KHz requires 201 seconds * ( 64,000 bits / second) / ( 8 bits / byte) = 1,608,000 bytes / ( 2 20 bytes / MB) = 1.534 MB

a 3:21 mpeg1 video file requires 201 seconds * ( 1,150,000 bits / second) / ( 8 bits / byte) = 28,894,750 bytes / ( 2 20 bytes / MB) = 27.56 MB

and a 3:21 mpeg2 video file recorded at 5 MHz requires 201 seconds * ( 5,000,000 bits / second) / ( 8 bits / byte) = 125,625,000 bytes / ( 2 20 bytes / MB) = 119.8 MB

It is interesting to compare the storage requirements for a similar amount of multimedia data; it is clear that computer multimedia is not practical without compression technologies, and that multimedia fuels the trend toward larger disk drives for computers.

KB, MB, GHz, and All of That Stuff

In case you haven't noticed (fat chance), the computer biz is filled with words that start with "kilo" and "mega", and abbreviations like "K" and "M" and "G". These words and abbreviations don't represent "things". Rather, they represent numbers. Kind of like the old gangster movies where a bad guy would say fifty g's rather than fifty thousand dollars. Computer nerds use a different slang for numbers. But the idea is the same.

K. M, and G are Numbers

Let's start with just the numbers -- the K and M and G part. We'll talk about the "B" and "Hz" and stuff in a moment. Table 1 shows what the abbreviations mean, how they're often spoken, the approximate number each represents, and the way-too-trivial-a-difference-to-worry-about actual number each represents.
Abbreviation K M G Stands for Spoken as Approximate # Actual #

Kilo Mega Giga

kay or killa meg gig or giga

1,000 (a thousand) 1,000,000 (a million)

1,024 1,048,576

1,000,000,000 (a 1,073,741,824 billion)

Table 1 If you ignore the boring "actual" numbers, you'll see there's a simple pattern to it. Each time you go to from K to M to G, you stick another ,000 onto the end of

the preceding number (also known as multiplying the previous number by a thousand), as you can see below:
y y y

1,000 1,000,000 1,000,000,000

K (kilo) M (mega) G (giga)

So "K" means "thousand" or ",000", and "M" means "million" or ",000,000" and G means "billion" or ",000,000,000". Though maybe I should just shut up now before I make it sound more complicated than it is. Suffice it to say if you're gonna buy a used car, and it has 80K miles on it, then that means the car has 80,000 miles on it. If the car has 20M or 20G miles on it, don't buy it.
Tip: Just in case you're some kinda math brain who's wondering where the actual numbers come from, K=210, M=220, G=230. (Yawn)

B is for Byte
Information in your head doesn't have any particular "size" to it. Just because Albert Einstein was a genius doesn't mean his head was the size of hot-air balloon or the Good Year blimp. His head was probably about the same size as anyone else's, give or take a couple inches. That's because the human brain stores information in some really weird abstract way that nobody understands. Computers have no brains, and really don't store "information" the way a human brain does. In fact, computers don't really store "information" per se. Except in the sense that a book stores information -- as letters, numbers, pictures, and words. The information in a book has no meaning to the book. Likewise, the information in a computer has no "meaning" to the computer. Books and computers are a lot alike in that way -- they both can be used to store text, numbers and pictures. And they're also alike in that the text, pictures, and numbers inside have no "meaning" to either the book or the computer. Anyway, the point is it takes a certain amount of "space" to store information outside of our brains. That's because the information needs to be stored as words, numbers, pictures, or something that takes up space. In a computer, the basic "unit" of measure is a byte, which is the amount of space it takes to store one character, like the letter "A" or an exclamation point (!). So it takes exactly three bytes to store the word "cat". It takes about 2,000 bytes to store one double-spaced page of typed text. When you see an uppercase letter "B", that stands for "byte". So instead of saying it takes "three bytes" to to store the word "cat", I could have said it takes

about 3B to store the word "cat". Likewise, I could have said it takes about 2,000B to store the a typed page of text. So now, given all you know about K and M and G, I bet you can figure out what KB. MB, and GB mean before you even peek at Table 2.
Abbreviation KB MB GB Stands for Approximate # (or) Actual #

Kilobyte Megabyte Gigabyte

1,000 bytes 1,000,000 bytes 1,000,000,000 bytes Table 2

A thousand bytes A million bytes A billion bytes

1,024 bytes 1,048,576 bytes 1,073,741,824 bytes

So before, when I was talking about a typed, double-spaced page of text taking up about "two thousand bytes" or "2,000B", I could have said it take about 2 KB to store that page of text. Often, the "B" is assumed, so it would be just as accurate for me to say it takes about 2 K to store that page. If you already have files stored on your computer, and know how to get around in folders, you can see that every file has a size. You'll need to use the Details view (choose View > Details from the menu bar above the icons). Figure 1 shows an example where you can see the sizes of some pictures in a folder on my computer.

Figure 1 The first file in in Figure 1 has a size of 735 KB, (or roughly 735,000 bytes). The biggest file in that folder is 1,732 KB. That could actually be expressed as 1.7MB (because a megabyte is about 1,000 kilobytes). But Windows always shows the file sizes in kilobytes (KB) just to keep all the numbers on the same scale. Of course, not all information in a computer is typed text. A computer can also store pictures, sound, and video. We generally refer to pictures, sound, and video asmultimedia or media files. But like typed text, those files have sizes too. Table 3 provides some examples of multimedia files.
Example Size Or about...

The picture shown in Figure 1 This entire Web page you're reading A 3-minute CD-quality song 1 minute of DVD-quality video with sound Table 3

61 KB 125 KB 3.5 MB 11 MB

61,000 bytes 125,000 bytes 3,500 KB or 3,500,000 bytes 11,000 KB or 11,000,000 bytes

Exactly how much "stuff" you can get on a disk depends on the capacity of the disk. This is no different from the capacity of a container for water (e.g. cup, bucket, bathtub). But of course, we don't use "pint", "quart", "gallon" and such for describing disk capacities. We use (what else?), KB, MB, and GB. There are lots of different "types" of disks out there. You've probably already heard of most of them -- floppy disks, CDs, DVDs, and of course the hard disk that lives inside every computer. Hard disks come in many different sizes., usually in the range of 20GB to 120GB. Table 4 lists the storage capacities of common types of disks. The column on the right describes the capacity in relation to the capacity of a floppy disk, just to give you some perspective on how greatly these capacities vary.
Disk type Capacity Equals this many floppy disks

Floppy disk CD DVD Hard disk

1.4 MB 700 MB 4.7 GB 20 GB to 120 GB Table 4

1 700 4,700 20,000 to 120,000

Tip: To see the storage capacity of your own computer's hard drive, and how much of that space is still available for storing more files, click here.

bps is for "bits per second"

The letters bps are short for bits per second. A byte (described earlier) is actually 8 bits. But that's not important. What's important is that unlike a byte, which is a measure of size, bps is a measure of speed. Simply stated, the higher the bps, the less time it takes. It's a lot like Miles Per Hour (MPH) in that sense. Think how long it would take to get from New York to California in a car going 55MPH. Now think how long that same trip would take in a jet going 700MPH. Simply stated, the higher the MPH (or bps), the less time it takes. As a rule, we use bps as a measure of the speed of getting data from one computer to another across a network (including the Internet). And we use "K" for a thousand, "M" for a million, and "G" for a billion, as usual. Table 5 sums it all up. The fourth column shows some alternative abbreviations. The rule-ofthumb is that an uppercase B always stands for "bytes" while a lowercase "b" always stands for "bits per second".
Abbreviation Spoken Bits per Second (English) Alternative Abbreviations

Kbps Mbps Gbps

Kilobits Megabits Gigabits

1,000 Thousand Kb or Kbits 1,000,000 1,000,000,000 Table 5 Million Billion Mb or Mbits Gb or Gbits

The most common use for the bps measurements is in types of Internet accounts. To give you a sense of how the speeds relate to one another in terms of "wait time", Table 6 compares the amount of time it would take to download (copy) a 1MB file from a computer on the Internet to your own computer. (Since a byte is equal to 8 bits, a 1 MB file is roughly 8,000,000 bits.) Also, I'm using "Broadband" as a general term for Cable and DSL accounts, which are actually available in speeds ranging from about 256 Kbps to 1,000 Kbps. Note, too, that 1,000 Kbps is the same as 1 Mbps.

Account type Speed

Bits per Second Time to download 1MB

Dial-up ISDN Broadband T1

56 Kbps 128 Kbps 1 Mbps 1.5 Mbps

56,000 128,000 1,000,000 1,500,000 Table 6

143 seconds 63 seconds 8 seconds 5 seconds

Hz is for Hertz
The Hz abbreviation stands for Hertz, but has nothing to do with pain or car rentals. In this context, "Hertz" pays homage to a guy named Heinrich Rudolf Hertz who figured out that radio and electricity have frequencies that you can measure in cycles per seconds. But that's getting more technical (and boring) than we need to be. The workhorse of your computer is its microprocessor - a little chip about the size of a toenail that can work at a pretty fast speed. Exactly how fast it can work is measured in terms of "instructions per second", which roughly corresponds to Hertz's idea of frequency. Of course, microprocessors can do thousands, millions, or billions of instructions per second. So that brings good old K, M, and G into the picture once again, as summarized in Table 7.
Abbreviation Stands for Cycles per Second That is...

KHz MHz

KiloHertz MegaHertz

1,000 Thousand 1,000,000 Million

GHz

GigaHertz Table 7

1,000,000,000

Billion

As is always the case with speed, "faster" means "less time waiting" for a human. The speed of the processor only affects "local tasks". It has nothing to do with how long it takes to get your e-mail or download stuff from the Internet. The speed of your Internet connection is all that matters there. The speed of your microprocessor has more to do with how long it takes to render changes to huge graphic images, or produce movies, or similar complex tasks that require a lot of internal calculations on the computer's part. So there you have it. To summarize, K, M, and G are all abbreviations for numbers:
y y y

K = Thousand (1,000) M = Million (1,000,000) G = Billion (1,000,000,000)

The letters that come after are either a measure of size or capacity, or a measure of speed, as follows:
y y y

B = Bytes or "how much" information. bps (or b) = "bits per second" or "how fast across some wire" Hz = Hertz of "how many calculations can be done in one second"

You're becoming dangerously close to be a full-fledged computer nerd here...

Hertz abbreviated Hz, is a unit of frequency. One Hz means one cycle per second. Hz is named after the German physicist Heinrich Hertz who made considerable contributions to electromagnetism. In computers, the CPU (central processing unit) and other components such as RAM are measured in Hz or MHz (Megahertz) and now GHz (Gigahertz).

One KHz equals 1,000 (one thousand) Hz, while 1,000,000 (one million) Hz equals one MHz, and 1,000,000,000 (one billion) Hz equals one GHz. Hopefully the table below will make it easier to understand.

Hertz Kilohertz Megahertz Gigahertz

1 Hz 1,000 Hz 1,000,000 Hz 1,000,000,000 Hz

0.001 kHz 1 kHz 1,000 kHz 1,000,000 kHz

0.000001 MHz 0.001 MHz 1 MHz 1,000 MHz

For a CPU the MHz refers to the clock speed of the processor. This means that a number of bits (either 8, 16, 32, or 64) are manipulated at one million cycles per second. Sound fast? Well that was a 1MHz CPU, (very,very slow). Now a days a 3GHz CPU is common and that means that a certain number of bits are being manipulated at 3 billion cycles per second. Now thats fast! Common examples of megahertz are: A Pentium II CPU (300 MHz), running at 300,000,000 Hz.Or a 256MB RAM stick running at 166,000,000 Hz.

Bit rate
From Wikipedia, the free encyclopedia

Bit rates

Decimal prefixes (SI)

Name

Symbol Multiple

kilobit per second

kbit/s

103

megabit per second Mbit/s

10

gigabit per second

Gbit/s

10

terabit per second

Tbit/s

10

12

Binary prefixes (IEC 60027-2)

kibibit per second

Kibit/s

10

mebibit per second Mibit/s 2

20

gibibit per second

Gibit/s

30

tebibit per second

Tibit/s

40

In telecommunications and computing, bitrate (sometimes written bit rate, data rate or as a variable R[1]) is the number of bits that are conveyed or processed per unit of time. The bit rate is quantified using the bits per second (bit/s or bps) unit, often in conjunction with an SI prefix such as kilo- (kbit/s or kbps), mega- (Mbit/s or Mbps),giga- (Gbit/s or Gbps) or tera- (Tbit/s or Tbps). Note that, unlike many other computer-related units, 1 kbit/s is traditionally defined as 1,000 bit/s, not 1,024 bit/s, etc., also before 1999 when SI prefixes were introduced for units of information in the standard IEC 60027-2. The formal abbreviation for "bits per second" is "bit/s" (not "bits/s", see writing style for SI units). In less formal contexts the abbreviations "b/s" or "bps" are often used, though this risks confusion with "bytes per second" ("B/s", "Bps"). 1 Byte/s (Bps or B/s) corresponds to 8 bit/s (bps or b/s).
Contents
[hide]

1 Bit rates at various protocol layers

o o o o o

1.1 Gross bit rate 1.2 Information rate 1.3 Network throughput 1.4 Goodput (data transfer rate) 1.5 Multimedia encoding bit rate

2 Prefixes 3 Progress trends 4 Bitrates in multimedia

4.1 Audio

  o o

4.1.1 MP3 4.1.2 Other audio

4.2 Video 4.3 Notes

5 See also

6 References 7 External links

[edit]Bit

rates at various protocol layers

bit rate

[edit]Gross

In digital communication systems, the physical layer gross bitrate,[2] raw bitrate,[3] data signaling rate[4] gross data transfer rate[5] or uncoded transmission rate[3] (sometimes written as a variable Rb[2][3] or fb[6]) is the total number of physically transferred bits per second over a communication link, including useful data as well as protocol overhead. In case of serial communications, the gross bit rate is related to the bit transmission time Tb as:

The gross bit rate is related to, but should not be confused with, the symbol rate or modulation rate in baud, symbols/s or pulses/s. Gross bit rate can be used interchangeably with "baud" only when there are two levels or symbols, representing 0 and 1 respectively, meaning that each symbol of a data transmission system carries exactly one bit of data; something not true for modern modemmodulation systems and modern LANs, for example. For most line codes and modulation methods: Symbol rate Gross bit rate More specifically, a line code (or baseband transmission scheme) representing the data using pulse-amplitude modulation with 2N different voltage levels, can transfer N bit/pulse. A digital modulationmethod (or passband transmission scheme) using 2N different symbols, for example 2N amplitudes, phases or frequencies, can transfer N bit/symbol. This results in: Gross bit rate = Symbol rate N An exception from the above is some self-synchronizing line codes, for example Manchester coding and return-to-zero (RTZ) coding, where each bit is represented by two pulses (signal states), resulting in: Gross bit rate = Symbol rate/2 A theoretical upper bound for the symbol rate in baud, symbols/s or pulses/s for a certain spectral bandwidth in hertz is given by the Nyquist law: Symbol rate Nyquist rate = 2 bandwidth

In practice this upper bound can only be approached for line coding schemes and for so-called vestigal sideband digital modulation. Most other digital carriermodulated schemes, for example ASK,PSK, QAM and OFDM, can be characterized as double sideband modulation, resulting in the following relation: Symbol rate Bandwidth In case of parallel communication, the gross bit rate is given by

where n is the number of parallel channels, Mi is the number of symbols or levels of the modulation in the ith channel, and Ti is the symbol duration time, expressed in seconds, for the i-th channel.

[edit]Information

rate

The physical layer net bitrate,[7] information rate,[2] useful bit rate,[8] payload rate,[9] net data transfer rate,[5] coded transmission rate,[3] effective data rate[3] or wire speed (informal language) of a digital communication channel is the capacity excluding the physical layer protocol overhead, for example time division multiplex (TDM) framing bits, redundant forward error correction(FEC) codes, equalizer training symbols and other channel coding. Error-correcting codes are common especially in wireless communication systems, broadband modem standards and modern cupper-based high-speed LANs. The physical layer net bitrate is the datarate measured at a reference point in the interface between the datalink layer and physical layer, and may consecuently include data link and higher layer overhead. In modems and wireless systems, link adaptation (automatic adaption of the data rate and the modulation and/or error coding scheme to the signal quality) is often applied. In that context, the termpeak bitrate denotes the net bitrate of the fastest and least robust transmission mode, used for example when the distance is very short between sender and transmitter.[10] Some operating systems and network equipment may detect the "connection speed"[11] (informal language) of a network access technology or communication device, implying the current net bit rate. Note that the term line rate in some textbooks is defined as gross bit rate,[9] in others as net bit rate. The relationship between the gross bit rate and net bit rate is affected by the FEC code rate according to the following. Net bit rate Gross bit rate code rate The connection speed of a technology that involves forward error correction typically refers to the physical layer net bit rate in accordance with the above definition.

For example, the net bitrate (and thus the "connection speed") of a IEEE 802.11a wireless network is the net bit rate of between 6 and 54 Mbit/s, while the gross bit rate is between 12 and 72 Mbit/s inclusive of errorcorrecting codes. The net bit rate of ISDN Basic Rate Interface (2 B-channels + 1 D-channel) of 64+64+16 = 144 kbit/s also refers to the payload data rates, while the signalling rate is 160 kbit/s. The net bit rate of the Ethernet 100Base-TX physical layer standard is 100 Mbit/s, while the gross bitrate is 125 Mbit/second, due to the 4B5B (four bit over five bit) encoding. In this case, the gross bit rate is equal to the symbol rate or pulse rate of 125 Mbaud, due to the NRZI line code. In communications technologies without forward error correction and other physical layer protocol overhead, there is no distinction between gross bit rate and physical layer net bit rate. For example, the net as well as gross bit rate of Ethernet 10Base-T is 10 Mbit/s. Due to the Manchester line code, each bit is represented by two pulses, resulting in a pulse rate of 20 Mbaud. The "connection speed" of a V.92 voiceband modem typically refers to the gross bit rate, since there is no additional error-correction code. It can be up to 56,000 bit/s downstreams and 48,000 bit/supstreams. A lower bit rate may be chosen during the connection establishment phase due to adaptive modulation - slower but more robust modulation schemes are chosen in case of poor signal-to-noise ratio. Due to data compression, the actual data transmission rate or throughput (see below) may be higher. The channel capacity, also known as the Shannon capacity, is a theoretical upper bound for the maximum net bitrate, exclusive of forward error correction coding, that is possible without bit errors for a certain physical analog node-to-node communication link. Net bit rate Channel capacity The channel capacity is proportional to the analog bandwidth in hertz. This proportionality is called Hartley's law. Consequently the net bit rate is sometimes called digital bandwidth capacity in bit/s.

[edit]Network

throughput

Main article: Throughput The term throughput, essentially the same thing as digital bandwidth consumption, denotes the achieved average useful bit rate in a computer network over a logical or physical communication link or through a network node, typically measured at a reference point above the datalink layer. This implies that the throughput often excludes data link layer protocol overhead. The throughput is affected by the traffic load from the data source in question, as well as from other sources sharing the same network resources. See also Measuring network throughput.

[edit]Goodput

(data transfer rate)

Main article: Goodput

Goodput or data transfer rate refers to the achieved average net bit rate that is delivered to the application layer, exclusive of all protocol overhead, data packets retransmissions, etc. For example, in the case of file transfer, the goodput corresponds to the achieved file transfer rate. The file transfer rate in bit/s can be calculated as the file size (in bytes), divided by the file transfer time (in seconds), and multiplied by eight. As an example, the goodput or data transfer rate of a V.92 voiceband modem is affected by the modem physical layer and data link layer protocols. It is sometimes higher than the physical layer data rate due to V.44 data compression, and sometimes lower due to bit-errors and automatic repeat request retransmissions. If no data compression is provided by the network equipment or protocols, we have the following relation: Goodput Throughput Maximum throughput Net bit rate for a certain communication path.

[edit]Multimedia

encoding bit rate

In digital multimedia, bit rate often refers to the number of bits used per unit of playback time to represent a continuous medium such as audio or video after source coding (data compression). The encoding bit rate of a multimedia file is the size of a multimedia file in bytes divided by the playback time of the recording (in seconds), multiplied by eight. For realtime streaming multimedia, the encoding bit rate is the goodput that is required to avoid interrupt: Encoding bit rate = Required goodput The term average bitrate is used in case of variable bitrate multimedia source coding schemes. In this context, the peak bit rate is the maximum number of bits required for any short-term block of compressed data.[12] A theoretical lower bound for the encoding bit rate for lossless data compression is the source information rate, also known as the entropy rate. Entropy rate Multimedia bit rate

[edit]Prefixes
When quantifying large bit rates, SI prefixes (also known as Metric prefixes or Decimal prefixes) are used, thus:

1,000 bit/s

rate = 1 kbit/s (one kilobit or one thousand bits per second) rate = 1 Mbit/s (one megabit or one million bits per second)

1,000,000 bit/s

rate = 1,000,000,000 1 Gbit/s (one gigabit or bit/s one billion bits per second)

Binary prefixes have almost never been used for bitrates, although they may occasionally be seen when data rates are expressed in bytes per second (e.g. 1 kByte/s or kBps is sometimes interpreted as 1000 Byte/s, sometimes as 1024 Byte/s). A 1999 IEC standard (IEC 60027-2) specifies different abbreviations for Binary and Decimal (SI) prefixes (e.g. 1 kiB/s = 1024 Byte/s = 8192 bit/s, and 1MiB/s = 1024 kiB/s), but these are still not very common in the literature, and therefore sometimes it is necessary to seek clarification of the units used in a particular context.

[edit]Progress

trends

These are examples of physical layer net bit rates in proposed communication standard interfaces and devices:

WAN mod Ethern WiFi WL Mobile data ems et LAN AN

1972: Aco 1972: I WiFi W 1G: 1981: NMT 1 ustic EEE LANs coupler 30 802.3 0 baud 1977: 1200 200 bit/s Ethern 1997: 80 2G: et 2.94 2.11 2 Mbit/s Mbit/s 1991: GSM C SD and D-

baud Vadi 1985: c and Bell 10b2 212A 10

1999: 80 AMPS 14.4 2.11b 11 kbit/s Mbit/s 2003: GSM

1986: ISD Mbit/s 1999: 80 EDGE 296 N introduc coax ed with two 64 kbit/s channels (160 kbit/s gross bit rate) 2.11a 54 kbit/s down, 118.4 kbit/s up 3G:

thinwir Mbit/s e 1990: (72 Mbit/s

10bT 1 gross bit 2001: UMTS0 rate) FDD Mbit/s 2003: 80 (WCDMA) 1995: 2.11g 54 384 kbit/s 2007:

100bT Mbit/s

1990: v.32 100 bis mode

(72

UMTS HSDPA 14.4 Mbit/s

Mbit/s Mbit/s

ms: 2400 / (125 4800 / 9600 / 19200 bit/s

gross bit 2008: UMTS HSPA 1

Mbit/s rate) gross bit rate)

2005: 80 4.4 Mbit/s 2.11g down, 5.76

(proprie Mbit/s up tary)108 2009: HSPA+ (Without

1994: v.34 1999: modems

1000b Mbit/s

with 28.8 T kbit/s

2007: 80 MIMO) 28

(Gigabi 2.11n 60 Mbit/s 0 Mbit/s downstreams (56 Mbit/s with 2x2 MIMO), 22 Mbit/s upstreams 2010: CDMA2000 E V-DO Rev. B 14.7 Mbit/s downstreams Pre-4G: 2007: Mobile WiMAX (IEEE 802.16e) 144 Mbit/s down, 35 Mbit/s up. 2009: LTE 10 0 Mbit/s downstreams (360 Mbit/s with MIMO 2x2), 50 Mbit/s

1995: v.90 t) 1 modems with 56 kbit/s Gbit/s (1.25 Gbit/s

downstrea gross ms, 33.6 kbit/s bit rate)

upstreams 2003: 1999: v.92 10GBA modems with 56 kbit/s downstrea ms, 48 kbit/s upstreams 1998: ADS L up to 8 Mbit/s, 2003: ADS L2 up to 12 Mbit/s 2005: ADS L2+ up to SE 10 Gbit/s

24 Mbit/s

upstreams See also Compari son of mobile phone standards

For more examples, see List of device bit rates, Spectral efficiency comparison table and OFDM system comparison table.

[edit]Bitrates

in multimedia

In digital multimedia, bitrate represents the amount of information, or detail, that is stored per unit of time of a recording. The bitrate depends on several factors: The original material may be sampled at different frequencies

The samples may use different numbers of bits

The data may be encoded by different schemes

The information may be digitally compressed by different algorithms or to different degrees

Generally, choices are made about the above factors in order to achieve the desired trade-off between minimizing the bitrate and maximizing the quality of the material when it is played. If lossy data compression is used on audio or visual data, differences from the original signal will be introduced; if the

compression is substantial, or lossy data is decompressed and recompressed, this may become noticeable in the form of compression artifacts. Whether these affect the perceived quality, and if so how much, depends on the compression scheme, encoder power, the characteristics of the input data, the listeners perceptions, the listener's familiarity with artifacts, and the listening or viewing environment. The bitrates in this section are approximately the minimum that the average listener in a typical listening or viewing environment, when using the best available compression, would perceive as not significantly worse than the reference standard:

[edit]Audio [edit]MP3  
32 kbit/s MW (AM) quality 96 kbit/s FM quality - This is questionable since FM broadcast is transmitted in analog 30 Hz-15khz. Similarly one cannot compare directly an LP record to CD using kbit/s.

100160 kbit/s Standard Bitrate quality; difference can sometimes be obvious (e.g. lack of low frequency quality and high frequency "swashy" effects)[citation needed]

192 kbits/s is the highest level supported by most MP3 encoders when ripping from a Compact Disc.

224320 kbit/s VBR to highest MP3 quality.

[edit]Other audio 
800 bit/s minimum necessary for recognizable speech (using specialpurpose FS-1015 speech codecs)

2.15 kbit/s - minimum bitrate available through the open-source Speex codec

8 kbit/s telephone quality (using speech codecs)

32-500 kbit/s -- lossy audio as used in Ogg Vorbis

256 kbit/s - Digital Audio Broadcasting (DAB) MP2 bit rate required to achieve a high quality signal [13]

400 kbit/s1,411kbit/s lossless audio as used in formats such as Free Lossless Audio Codec, WavPack or Monkey's Audio to compress CD audio

1,411.2 kbit/s Linear PCM sound format of Compact Disc Digital Audio

5,644.8 kbit/s DSD (A trademarked implementation of PDM) sound format of Super Audio CD[14]

[edit]Video
16 kbit/s videophone quality (minimum necessary for a consumeracceptable "talking head" picture using various video compression schemes)

128 384 kbit/s businessoriented videoconferencing quality using video compression

1.15 Mbit/s max VCD quality (using MPEG1 compression)[15]

3.5 Mbit/s typ - Standard-definition television quality (with bit-rate reduction from MPEG-2 compression)

9.8 Mbit/s max DVD (using MPEG2 compression)[16]

8 to 15 Mbit/s typ HDTV quality (with bit-rate reduction from MPEG-4 AVC compression)

  24 Mbit/s max - AVCHD (using MPEG4 AVC compression)[18]

19 Mbit/s approximate - HDV 720p (using MPEG2 compression)[17]

 25 Mbit/s approximate - HDV 1080i (using MPEG2 compression)[17]  29.4 Mbit/s max HD DVD  40 Mbit/s max Blu-ray Disc (using MPEG2, AVC or VC-1 compression)[19