Habari Zetu

Showing posts with label Computer. Show all posts
Showing posts with label Computer. Show all posts

Presentation of FireWire Bus (IEEE 1394)

The IEEE 1394 bus (name of the standard to which it makes reference) was developed at the end of 1995 in order to provide an interconnection system that allows data to circulate at a high speed and in real time. The company Applegave it the commercial name "FireWire", which is how it is most commonly known. Sony also gave it commercial name, i.Link. Texas Instruments preferred to call it Lynx.
FireWire is a port that exists on some computers that allows you to connect peripherals (particularly digital cameras) at a very high bandwidth. There are expansion boards (generally in PCI or PC Card / PCMCIA format) that allow you to equip a computer with FireWire connectors. FireWire connectors and cables can be easily spotted thanks to their shape as well as the following logo:
FireWire Logo

FireWire Standards

There are different FireWire standards that allow you to obtain the following bandwidths:
StandardTheoretical Bandwidth
IEEE 1394a
IEEE 1394a-S100100 Mbit/s
IEEE 1394a-S200200 Mbit/s
IEEE 1394a-S400400 Mbit/s
IEEE 1394b
IEEE 1394b-S800800 Mbit/s
IEEE 1394b-S12001,200 Mbit/s
IEEE 1394b-S16001,600 Mbit/s
IEEE 1394b-S32003,200 Mbit/s
The IEEE 1394b standard is also called FireWire 2 or FireWire Gigabit.

FireWire Connectors

There are different FireWire connectors for each of the IEEE 1394 standards.
  • The IEEE 1394a standard specifies two connectors: 
    • Connectors 1394a-1995:
      connector 1394a-1995
    • Connectors 1394a-2000, called mini-DV because they are used on Digital Video (DV) cameras:
      connector 1394a-2000
  • The IEEE 1394b standard specifies two types of connectors that are designed so that 1394b-Beta cables can be plugged into Beta and Bilingual connectors, but 1394b Bilingual cables can only be plugged into Bilingual connectors: 
    • 1394b Beta connectors:
      1394b Beta connector
    • 1394b Bilingual connectors:
      1394b Bilingual connector

How the FireWire Bus Works

The IEEE 1394 bus has about the same structure as the USB bus except that it is a cable made up of six wires (2 pairs for the data and the clock and 2 wires for the power supply) that allow it to reach a bandwidth of 800 Mb/s (soon it should be able to reach 1.6 Gb/s, or even 3.2 Gb/s down the road). The two wires for the clock is the major difference between the USB bus and the IEEE 1394 bus, i.e. the possibility to operate in two transfer modes:
  • Asynchronous transfer mode: this mode is based on a transmission of packets at variable time intervals. This means that the host sends a data packet and waits to receive a receipt notification from the peripheral. If the host receives a receipt notification, it sends the next data packet. Otherwise, the first packet is resent after a certain period of time.
  • Synchronous mode: this mode allows data packets of specific sizes to be sent in regular intervals. A node called Cycle Master is in charge of sending a synchronization packet (called a Cycle Start packet) every 125 microseconds. This way, no receipt notification is necessary, which guarantees a set bandwidth. Moreover, given that no receipt notification is necessary, the method of addressing a peripheral is simplified and the saved bandwidth allows you to gain throughput.

Another innovation of the IEEE 1394 standard: bridges (systems that allow you to link buses to other buses) can be used. Peripheral addresses are set with a node (i.e. peripheral) identifier encoded on 16 bits. This identifier is divided into two fields: a 10-bit field that identifies the bridge and a 6-bit field that specifies the node. Therefore, it is possible to connect 1,023 bridges (or 210 -1) on which there can be 63 nodes (or 26 -1), which means it is possible to address 65,535 peripherals! The IEEE 1394 standard allows hot swapping. While the USB bus is intended for peripherals that do not require a lot of resources (e.g. a mouse or a keyboard), the IEEE 1394 bandwidth is larger and is intended to be used for new, unknown multimedia (video acquisition, etc.).

Introduction to the USB

USB (Universal Serial Bus) is as its name suggests, based on serial type architecture. However, it is an input-output interface much quicker than standardserial ports. Serial architecture was used for this type of port for two main reasons:
  • Serial architecture gives the user a much higher clock rate than a parallel interface because a parallel interface does not support too high frequencies (in a high speed architecture, bits circulating on each wire arrive with lag, causing errors);
  • serial cables are much cheaper than parallel cables.

USB standards

So, from 1995, the USB standard has been developed for connecting a wide range of devices.
The USB 1.0 standard offers two modes of communication:
  • 12 Mb/s in high speed mode,
  • 1.5 Mb/s in low speed.
The USB 1.1 standard provides several clarifications for USB device manufacturers but does not change anything in the speed. USB 1.1 certified devices carry the following logo:
USB logo
The USB 2.0 standard makes it possible to obtain speeds which can reach 480 Mbit/s/ USB 2.0 certified devices carry the following logo:
USB 2.0 logo
If there is no logo, the best way of determining if something is a low or high speed USB is to consult the product documentation insofar as the connectors are the same.
Compatibility between USB 1.0, 1.1 and 2.0 is assured. However, the use of a USB 2.0 device in a low speed USB port (i.e. 1.0 or 1.1) will limit the speed to 12Mbit/s maximum. Furthermore, the operating system is likely to display a message explaining that the speed will be restricted.

Types of connectors

There are two types of USB connectors:
  • Connectors known as type A, where the shape is rectangular and generally used for less bandwidth intensive devices (keyboard, mouse, webcam, etc.);
  • Connectors known as type B, where the shape is square and mainly used for high speed devices (external hard disks, etc.);
Type A and B USB connectors
  1. Power supply +5V (VBUS) 100mA maximum
  2. Data (D-)
  3. Data (D+)
  4. Mass (GND)

Operation of the USB

One characteristic of USB architecture is that it can supply electricity to devices to which it connects, with a limit of 15 W maximum per device. To do so, it uses a cable made up of four wires (the GND mass, the BUS supply and two data wires called D- and D+).
USB cable
The USB standard allows devices to be chained by using a bus or star topology. So, devices can either be connected one to another or branched. 
Branching is done using boxes called "hubs" comprising of a single input and several outputs. Some are active (supplying electric energy), others passive (power supplied by the computer).
bus topology for USB ports
star topology for USB ports
Communication between the host (computer) and devices is carried out according to a protocol (communication language) based on the token ring principle. This means that bandwidth is temporarily shared between all connected devices. The host (computer) issues a signal to begin the sequence every millisecond (ms), the time interval during which it will simultaneously give each device the opportunity to "speak". When the host wants to communicate with a device, it transmits a token (a data packet, containing the address of the device coded over 7 bits) designating a device, so it is the host that decides to "talk" with the devices. If the device recognises its address in the token, it sends a data packet (between 8 and 255 bytes) in response, if not it passes the packet to the other connected devices. Data is exchanged in this way is coded according to NRZIcoding.
Since the address is coded over 7 bits, 128 devices (2^7) can simultaneously be connected to a port of this type. In reality, it is advisable to reduce this number to 127 because the 0 address is a reserved address. (see later).
Due to the maximum length of the cable between two devices of 5 metres and a maximum number of 5 hubs (supplied), it is possible to create a chain 25 meters in length.

USB ports support Hot plug and play. So, devices can be connected without turning off the computer (hot plug). When a device is connected to the host it detects the addition of a new item thanks to a change in the tension between the D+ and D- wires. At this time, the computer sends an initialization signal to the device for 10ms, then it supplies the current using the GND and VBUS wires (up to 100mA). The device is then supplied with electric current and temporarily takes over the default address (0 address). The following stage consists of supplying it with its definitive address (this is the listing procedure). To do so, the computer interrogates devices already connected to know their addresses and allocates a new one, which identifies it by return. The host, having all the necessary characteristics is then able to load the appropriate driver.

Introduction to input-output ports



Input-output ports are material elements on the computer, allowing the system to communicate with exterior elements, in other words to exchange data, hence the name input-output interface (sometimes known as I/O interface).
Serial port

Serial ports (also called RS-232, after the name of the standard they refer to) represent the first interfaces to allow computers to exchange information with the "outside world". The term serial refers to data sent via a single wire: the bits are sent one after the other (refer to section on data transmission for a presentation on transmission modes).



Serial ports were originally able to only send data and not receive it, hence two-way ports were developed (the ports on current computers are two-way); two-way serial ports therefore need two wires for communication.

Serial communication takes place asynchronously, meaning that no synchronization signal (or clock) is required: the data may be sent at random intervals. In return, the peripheral must be able to distinguish the characters (one character is 8 bits in length) among the succession of bits which is sent.
This is why, in this type of transmission, each character is preceded by a STARTbit and followed by a STOP bit. These control bits, which are needed for serial transmission, waste 20% of the bandwidth (for 10 bits sent, 8 are used to code the character and 2 are used for reception).

Serial ports are generally built into the mother board, which is why the connectors behind the casing and connected to the mother board by a wire cable can be used to connect an exterior element. Serial connectors generally have 9 or 25 pins and take the following form (DB9 and DB25 connectors respectively):
DB9 connector
DB25 connector
A personal computer generally has between one and four serial ports.

Parallel port

Parallel data transmission involves sending data simultaneously on several channels (wires). The parallel ports on personal computers can be used to send 8 bits (one octet) simultaneously via 8 wires.
transmission on a parallel port
The first two-way parallel ports allowed for speeds of 2.4Mb/s. Enhanced parallel ports have been developed however to achieve higher speeds:
  • The EPP (Enhanced Parallel Port) achieves speeds of 8 to 16 Mbps
  • The ECP (Enhanced Capabilities Port), developed by Hewlett Packard andMicrosoft. It has the same characteristics as the EPP with in addition a Plug and Play feature, allowing the computer to recognise the connected peripherals.
Parallel ports, like serial ports, are built into the mother board. DB25 connectors allow connection to an exterior element (e.g. a printer).
DB25 connector

Introduction to the AGP bus


The AGP bus (short for Accelerated Graphics Port) was released in May 1997 for Slot One chipsets, then was later released for Super 7 chips in order to manage graphical data flow, which had grown to large to be handled by a PCI bus. The AGP bus is directly linked to the processor's FSB (Front Side Bus) and uses the same frequency, for increased bandwidth.

The AGP interface was developed specifically to connect with the video card, by opening a direct memory access (DMA) channel to the graphics board, bypassing the input-output controller. Cards which employ this graphics bus theoretically require less on-board memory; because they can directly access graphical data (such as textures) stored in central memory, their cost is hypothetically lower.

Version 1.0 of the AGP bus, which used 3.3 V of power, had a 1X mode that could send 8 bytes every two cycles, and a 2x mode for transferring 8 bytes per cycle.

In 1998, AGP version 2.0 added AGP 4X, which could send 16 bytes per cycle. Version 2.0 of AGP was powered by 1.5 V, and AGP 2.0 "universal" connectors which could support either voltage were released.

AGP version 3.0, released in 2002, doubled the speed of AGP 2.0 with a new AGP 8x mode.

Characteristics of AGP

The AGP 1X port operates at 66 MHz, as opposed to 33 MHz for a PCI bus, giving it a top speed of 264 MB/s (vs. 132 MB/s, shared between all the cards, for PCI). This gives AGP better performance, especially when displaying complicated 3D scenes.

When AGP 4X was released, its speed went up to 1 GB/s. This generation of AGP used 25 W of power. The next generation was named AGP Pro and used 50W.

AGP Pro 8x offers speeds of 2 GB/s.

The transfer speeds for the various AGP standards are:

AGP 1X: 66.66 MHz x 1(coef.) x 32 bits /8 = 266.67 MB/s
AGP 2X: 66.66 MHz x 2(coef.) x 32 bits /8 = 533.33 MB/s
AGP 4X: 66.66 MHz x 4(coef.) x 32 bits /8 = 1.06 GB/s
AGP 8X: 66.66 MHz x 8(coef.) x 32 bits /8 = 2.11 GB/s



It should be noted that each of these AGP standards is backwards-compatible, meaning that AGP 4X or AGP 2X cards can be inserted into an AGP 8X slot.

AGP Connectors

Recent motherboards are built with a general AGP connector which can be identified by its brown color. There are three types of connectors:
  • AGP 1.5 volt connector:

AGP 1.5 volt connector
  • AGP 3.3 volt connector:

AGP 3 volt connector
  • Universal AGP connector:

Universal AGP connector:

Summary

Here is a table summarising the technical specifications for each version and mode of AGP: 
AGPVoltageMode
AGP 1.03.3 V1x, 2x
AGP 2.01.5 V1x, 2x, 4x
AGP 2.0 universal1.5 V, 3.3 V1x, 2x, 4x
AGP 3.01.5 V4x, 8x

8 Navigation Tricks Every iPad User Needs to Know

ipad-air
Like any piece of technology, iPads aren’t completely obvious when you first pick them up. They have their own language of gestures, swipes, and button presses you should learn to become more comfortable using them.

Here’s 6 Great Tricks for Windows 8 that You Probably Don’t Know


image
We’ve covered a lot of tips, tricks, and tweaks for Windows 8, but there are still a few more. From bypassing the lock screen to instantly taking and saving screenshots, here are a few more hidden options and keyboard shortcuts.
Whether you love Windows 8, hate it, or just wish Metro would go away, these options will help you make Windows 8 work the way you want it to.

PCI Express Bus (PCI-E)

The PCI Express bus (Peripheral Component Interconnect Express, written PCI-E or 3GIO for "Third Generation I/O"), is an interconnect bus that allows you to add expansion boards to a computer. The PCI Express bus was developed in July 2002. Contrary to the PCI bus, which runs in parallel interface, the PCI Express bus runs in serial interface, which allows it to reach a bandwidth that is much higher than that PCI bus.
PCI Express Logo

Characteristics of the PCI Express Bus

The PCI Express bus comes in several versions (1X, 2X, 4X, 8X, 12X, 16X and 32X), which provide throughputs of between 250 Mb/s and 8 Gb/s, or close to 4 times the peak throughput of AGP 8X ports. Because its manufacturing cost is that similar to that of the AGP port, the PCI Express bus will progressively replace the former.

PCI Express Connectors

PCI Express connectors are not compatible with older PCI connectors. They vary in size and require less electricity. One of the interesting characteristics of the PCI Express bus is that it is hot pluggable, i.e. it can be plugged in or unplugged with out turning off or restarting the machine. PCI Express connectors can be recognised thanks to their small size and dark grey color.
  • The PCI Express 1X connector has 36 pins and is intended for high-bandwidth I/O use


PCI Express 1x connector
  • The PCI Express 4X connector has 64 pins and is intended to be used on servers:


PCI Express 4x connector
  • The PCI Express 8X connector has 98 pins and is intended to be used on servers:


PCI Express 8x connector
  • The PCI Express 16X connector has 164 pins, is 89 mm long and is intended to be used on the graphics port:


PCI Express 16x connector



The PCI Express standard is also intended to replace PC Card technology with "PCI Express Mini Card" connectors. What is more, contrary to PCI connectors which can only be used for to make internal connections, the PCI Express standard can be used to connect external peripherals by using cables. Despite that fact, it is not in competition with USB or FireWire ports.

Introduction to the AGP bus

The AGP bus (short for Accelerated Graphics Port) was released in May 1997 for Slot One chipsets, then was later released for Super 7 chips in order to manage graphical data flow, which had grown to large to be handled by a PCI bus. The AGP bus is directly linked to the processor's FSB (Front Side Bus) and uses the same frequency, for increased bandwidth.
The AGP interface was developed specifically to connect with the video card, by opening a direct memory access (DMA) channel to the graphics board, bypassing the input-output controller. Cards which employ this graphics bus theoretically require less on-board memory; because they can directly access graphical data (such as textures) stored in central memory, their cost is hypothetically lower.
Version 1.0 of the AGP bus, which used 3.3 V of power, had a 1X mode that could send 8 bytes every two cycles, and a 2x mode for transferring 8 bytes per cycle.
In 1998, AGP version 2.0 added AGP 4X, which could send 16 bytes per cycle. Version 2.0 of AGP was powered by 1.5 V, and AGP 2.0 "universal" connectors which could support either voltage were released.
AGP version 3.0, released in 2002, doubled the speed of AGP 2.0 with a new AGP 8x mode.

Characteristics of AGP

The AGP 1X port operates at 66 MHz, as opposed to 33 MHz for a PCI bus, giving it a top speed of 264 MB/s (vs. 132 MB/s, shared between all the cards, for PCI). This gives AGP better performance, especially when displaying complicated 3D scenes.
When AGP 4X was released, its speed went up to 1 GB/s. This generation of AGP used 25 W of power. The next generation was named AGP Pro and used 50W.
AGP Pro 8x offers speeds of 2 GB/s.
The transfer speeds for the various AGP standards are:
  • AGP 1X: 66.66 MHz x 1(coef.) x 32 bits /8 = 266.67 MB/s
  • AGP 2X: 66.66 MHz x 2(coef.) x 32 bits /8 = 533.33 MB/s
  • AGP 4X: 66.66 MHz x 4(coef.) x 32 bits /8 = 1.06 GB/s
  • AGP 8X: 66.66 MHz x 8(coef.) x 32 bits /8 = 2.11 GB/s



It should be noted that each of these AGP standards is backwards-compatible, meaning that AGP 4X or AGP 2X cards can be inserted into an AGP 8X slot.

AGP Connectors

Recent motherboards are built with a general AGP connector which can be identified by its brown color. There are three types of connectors:
  • AGP 1.5 volt connector:

AGP 1.5 volt connector
  • AGP 3.3 volt connector:

AGP 3 volt connector
  • Universal AGP connector:

Universal AGP connector:

Summary

Here is a table summarising the technical specifications for each version and mode of AGP:
AGP Voltage Mode
AGP 1.03.3 V1x, 2x
AGP 2.01.5 V1x, 2x, 4x
AGP 2.0 universal1.5 V, 3.3 V1x, 2x, 4x
AGP 3.01.5 V4x, 8x

PCI Bus

The PCI bus (Peripheral Component Interconnect) was developed by Intel on 22 June 1992. Contrary to the VLB bus, it is not so much a traditional local bus but rather an intermediate bus located between the processor bus (NorthBridge) and the I/O bus (SouthBridge).

PCI Connectors

At least 3 or 4 PCI connectors are generally present on motherboards and can generally be recognised by their standardized white color.
The PCI interface exists in 32 bits with a 124-pin connector, or in 64 bits with a 188-pin connector. There are also two signalling voltage levels:
  • 3.3V, for laptop computers
  • 5V, for desktop computers
The signalling voltage does not equal the voltage of the motherboard power supply but rather the voltage threshold for the digital encryption of data.
There are 2 types of 32-bit connectors:
  • 32-bit PCI connector, 5V:
    PCI connector, 5V
  • 32-bit PCI connector, 3.3V:
    PCI connector, 3.3V
The 64-bit PCI connectors offer additional pins and can accommodate 32-bit PCI cards. There are 2 types of 64-bit connectors:
  • 64-bit PCI connector, 5V:
    64-bit PCI connector, 5V
  • 64-bit PCI connector, 3.3V:
    64-bit PCI connector, 3.3V

Interoperability

Generally, it is not possible to make a mistake when plugging a PCI card into a PCI slot. If the card plugs in correctly, it is compatible. Otherwise, there are foolproof devices to keep you from installing it.
PCI - Interoperability
There are expansion boards that have what are called "universal" connectors, i.e. that have two types of foolproof devices (two notches). These expansion cards can detect signalling voltage and adapt to it, and can therefore can be inserted independantly in 3.3V or 5V slots.

Bus Updates

The original version of the PCI bus is 32-bits wide and has a clock speed of 33 MHz, which allows it to theoretically provide a throughput of 132 Mb/s on 32 bits. On 64-bit architectures, the bus operates on 64 bits and offers a theoretical throughput of 264 Mb/s.
An interest group made up of a large number of manufacturers, dubbed PCI-SIG (PCI Special Interests Group), was created to upgrade the PCI standard. Bus updates were published. Version 2.0 from 30 April 1993 defined the shape of the connectors and additional cards and gave it a clock speed of 66 MHz versus 33 MHz for version 1.0, therefore doubling its theoretical throughput to reach 266 Mb/s on 32 bits.
On 1 June 1995, revision 2.1 of the PCI bus improved its use to 66 MHz. At the time, engineers anticipated a progressive move from 5V signalling voltage toward 3.3V.
Version 2.2 of the PCI bus, which appeared on 18 December 1998, allowed peripherals to be plugged in when hot (hot plug).
Revision 2.3, edited on 29 March 2002, did away with the possibility of using additional 5V cards but permitted the use of cards that support both voltages in order to ensure downward compatibility. Revision 3.0 of the PCI standard completely did away with the use of 5V compatible cards.
In September 1999, a major change to the PCI bus was made, dubbed PCI-X. The PCI-X 1.0 bus supports 66, 100 and 133 MHz frequencies. The PCI-X bus is fully compatible with the PCI format. PCI-X slots support PCI format cards and vice versa.
Revision 2.0 of the PCI-X bus supports 66, 100, 133, 266 and 533 MHz frequencies and allows throughputs of 4.27 Gb/s on 64 bits.
The table below summarises the different PCI bus revisions:
Revision Release Date Frequency Voltage Width
PCI 1.0 1992 33 MHz Nil 32 bits 133 Mb/s
64 bits 266 Mb/s
PCI 2.0 1993 33 MHz 3.3V / 5V 32 bits 132 Mb/s
64 bits 264 Mb/s
PCI 2.1 1995 33 MHz 3.3V / 5V 32 bits 132 Mb/s
64 bits 264 Mb/s
66 MHz 3.3V 32 bits 264 Mb/s
64 bits 528 Mb/s
PCI 2.2 1998 33 MHz 3.3V / 5V 32 bits 132 Mb/s
64 bits 264 Mb/s
66 MHz 3.3V 32 bits 264 Mb/s
64 bits 528 Mb/s
PCI 2.3 2002 33 MHz 3.3V / 5V 32 bits 132 Mb/s
64 bits 264 Mb/s
66 MHz 3.3V 32 bits 264 Mb/s
64 bits 528 Mb/s
PCI-X 1.0 1999 66 MHz 3.3V 32 bits 264 Mb/s
64 bits 528 Mb/s
100 MHz 3.3V 32 bits 400 Mb/s
64 bits 800 Mb/s
133 MHz 3.3V 32 bits 532 Mb/s
64 bits 1,064 Mb/s
PCI-X 2.0 2002 66 MHz 3.3V 32 bits 264 Mb/s
64 bits 528 Mb/s
100 MHz 3.3V 32 bits 400 Mb/s
64 bits 800 Mb/s
133 MHz 3.3V 32 bits 532 Mb/s
64 bits 1,064 Mb/s
266 MHz 3.3V / 1.5V 32 bits 1,064 Mb/s
64 bits 2,128 Mb/s
533 MHz 3.3V / 1.5V 32 bits 2,128 Mb/s
64 bits 4,256 Mb/s

ISA, MCA and VLB Buses

Expansion Bus

Expansion buses (sometimes called peripheral buses) are buses that have connectors that allow you to add expansion cards (peripherals) to a computer. There are different types of standard internal buses that are characterised by:
  • their shape
  • the number of connector pins
  • the type of signals (frequency, data, etc.)

ISA Bus

The original version of the ISA bus (Industry Standard Architecture) that appeared in 1981 with PC XT was an 8-bit bus with a clock speed of 4.77 MHz.
In 1984, with the appearance of PC AT (the Intel 286 processor), the bit was expanded into a 16-bit bus and the clock speed went from 6 to 8 MHz and finally to 8.33 MHz, offering a maximum transfer rate of 16 Mb/s (in practice only 8 Mb/s because one cycle out of every two was used for addressing).
The ISA bus permitted bus mastering, i.e. it enabled controllers connected directly to the bus to communicate directly with the other peripherals without going through the processor. One of the consequences of bus mastering is direct memory access (DMA). However, the ISA bus only allows hardware to address the first 16 megabytes of RAM.
Up until the end of the 1990s, almost all PC computers were equipped with the ISA bus, but it was progressively replaced by the PCI bus, which offered a better performance.
  • 8-bit ISA Connector:
    8-bit ISA Connector
  • 16-bit ISA Connector:
    16-bit ISA Connector

MCA Bus

The MCA bus (Micro Channel Architecture) is an improved proprietary bus designed by IBM in 1987 to be used in their PS/2 line of computer. This 16 to 32-bit bus was incompatible with the ISA bus and could reach a throughput of 20 Mb/s.

EISA Bus

The EISA bus (Extended Industry Standard Architecture) was developed in 1988 by a consortium of companies (AST, Compaq, Epson, Hewlett-Packard, NEC, Olivetti, Tandy, Wyse and Zenith) in order to compete with the MCA proprietary bus that was launched by IBM the previous year. The EISA bus used connectors that were the same size as the ISA connector but with 4 rows of contacts instead of 2, for 32-bit addressing.
The EISA connectors were deeper and the additional rows of contacts were placed below the rows of ISA contacts. Thus, it was possible to plug an ISA expansion board into an EISA connector. However, they did not plug as deep into the connector (because of the bezels) and thus only used the top rows (ISA) of contacts.

Local Bus

Traditional I/O buses, such as ISA, MCA our EISA buses, are directly connected to the main bus and there are forced to work at the same frequency. However, some I/O peripherals need a very low bandwidth while other need higher bandwidths. Therefore there are bottlenecks on the bus. In order to solve this problem, the "local bus" architecture offers to take advantage of the system bus, or front side bus (FSB), by interfacing directly with it.

VLB Bus

In 1992, the VESA local bus (VLB) was developed by the VESA (Video Electronics Standard Association under the aegis of the company NEC) in order to offer a local bus dedicated to graphics systems. The VLB is a 16-bit ISA connector with an added 16-bit connector:
VESA Local Bus (VLB) Connector
The VLB bus is a 32-bit bus initially intended to work a bandwidth of 33 MHz (the bandwidth of the first PC 486s at that time). The VESA local bus was used on the following 486 models (40 and 50 MHz, respectively) as well as on the very first Pentium processors, but it was quickly replaced by the PCI bus.

What is a computer bus?


Introduction to the concept of a bus

A bus, in computing, is a set of physical connections (cables, printed circuits, etc.) which can be shared by multiple hardware components in order to communicate with one another.
The purpose of buses is to reduce the number of "pathways" needed for communication between the components, by carrying out all communications over a single data channel. This is why the metaphor of a "data highway" is sometimes used.
schematic diagram of a bus



If only two hardware components communicate over the line, it is called a hardware port (such as a serial port or parallel port).

Characteristics of a bus

A bus is characterised by the amount of information that can be transmitted at once. This amount, expressed in bits, corresponds to the number of physical lines over which data is sent simultaneously. A 32-wire ribbon cable can transmit 32 bits in parallel. The term "width" is used to refer to the number of bits that a bus can transmit at once.
Additionally, the bus speed is also defined by its frequency (expressed in Hertz), the number of data packets sent or received per second. Each time that data is sent or received is called a cycle.
This way, it is possible to find the maximum transfer speed of the bus, the amount of data which it can transport per unit of time, by multiplying its width by its frequency. A bus with a width of 16 bits and a frequency of 133 MHz, therefore, has a transfer speed equal to:
16 * 133.106 = 2128*106 bit/s, 
or 2128*106/8 = 266*106 bytes/s 
or 266*106 /1000 = 266*103 KB/s 
or 259.7*103 /1000 = 266 MB/s

Bus subassembly

In reality, each bus is generally constituted of 50 to 100 distinct physical lines, divided into three subassemblies:
  • The address bus (sometimes called the memory bus) transports memory addresses which the processor wants to access in order to read or write data. It is a unidirectional bus.
  • The data bus transfers instructions coming from or going to the processor. It is a bidirectional bus.
  • The control bus (or command bus) transports orders and synchonisation signals coming from the control unit and travelling to all other hardware components. It is a bidirectional bus, as it also transmits response signals from the hardware.

The primary buses

There are generally two buses within a computer:
  • the internal bus (sometimes called the front-side bus, or FSB for short). The internal bus allows the processor to communicate with the system's central memory (the RAM).
  • the expansion bus (sometimes called the input/output bus) allows various motherboard components (USB, serial, and parallel ports, cards inserted in PCI connectors, hard drives, CD-ROM and CD-RW drives, etc.) to communicate with one another. However, it is mainly used to add new devices using what are called expansion slots connected to the input/outpur bus.

The chipset

A chipset is the component which routes data between the computer's buses, so that all the components which make up the computer can communicate with each other. The chipset originally was made up of a large number of electronic chips, hence the name. It generally has two components:
  • The NorthBridge (also called the memory controller) is in charge of controlling transfers between the processor and the RAM, which is way it is located physically near the processor. It is sometimes called the GMCH, forr Graphic and Memory Controller Hub.
  • The SouthBridge (also called the input/output controller or expansion controller) handles communications between peripheral devices. It is also called the ICH (I/O Controller Hub). The tem bridge is generally used to designate a component which connects two buses.

system architecture of a PC



It is interesting to note that, in order to communicate, two buses must have the same width. The explains why RAM modules sometimes have to be installed in pairs (for example, early Pentium chips, whose processor buses were 64-bit, required two memory modules each 32 bits wide).
Here is a table which gives the specifications for the most commonly used buses:
StandardBus width (bits)Bus speed (MHz)Bandwidth (MB/sec)
ISA 8-bit88.37.9
ISA 16-bit168.315.9
EISA328.331.8
VLB3233127.2
PCI 32-bit3233127.2
PCI 64-bit 2.16466508.6
AGP3266254.3
AGP (x2 Mode)3266x2528
AGP (x4 Mode)3266x41056
AGP (x8 Mode)3266x82112
ATA33163333
ATA1001650100
ATA1331666133
Serial ATA (S-ATA)1
180
Serial ATA II (S-ATA2)2
380
USB1
1.5
USB 2.01
60
FireWire1
100
FireWire 21
200
SCSI-184.775
SCSI-2 - Fast81010
SCSI-2 - Wide161020
SCSI-2 - Fast Wide 32 bits321040
SCSI-3 - Ultra82020
SCSI-3 - Ultra Wide162040
SCSI-3 - Ultra 284040
SCSI-3 - Ultra 2 Wide164080
SCSI-3 - Ultra 160 (Ultra 3)1680160
SCSI-3 - Ultra 320 (Ultra 4)1680 DDR320
SCSI-3 - Ultra 640 (Ultra 5)1680 QDR640

Types of Memory Card - xD picture card

xD Picture memory (for eXtreme Digital) is a type of memory card created by Fuji and Olympus in August 2002.
The architecture of xD cards is based on NAND type flash memory circuits (EEPROM)
xD picture card memory is smaller in size than a postal stamp (20.0mm x 25.0mm x 1.7mm) and weighs barely 2 grams.
xD picture card



Access to the data is carried out via a lateral connector with 18 pins, allowing a transfer rate of 1.3 Mb/s to be reached and potentially up to 3Mb/s for writing and around 5 Mb/s for reading.
In time it is expected that xD picture cards will reach a capacity of 8Gb.

Type of Memory Card-SmartMedia cards

SmartMedia memory is a type of memory card created by Toshiba and Samsung.
Its architecture is based on NAND type flash memory circuits (EEPROM)
SmartMedia memory is equivalent in size to a postal stamp (45.0mm x 37.0mm x 0.76mm) and weighs barely 2 grams.
There are two types of SmartMedia card with different voltages:
  • 3.3V SmartMedia cards have a notch on the right
    3.3V SmartMedia card
  • 5V SmartMedia cards have a notch on the left
    5V SmartMedia card
Access to the data is carried out via a chip with 22 pins. Whatever the capacity of the SmartMedia card, the dimensions and location of the chip are the same.
Access time for the memory is approximately 25µs for the first access and cycles of 50 ns for the following ones.

Compatibility

There are two adapters making it possible to insert a SmartMedia card in a PCMCIA location, so as to enable the transfer of data directly from a SmartMedia card to a laptop.

Type of Memory Card - SD Card (Secure Digital)

Secure Digital memory (known as SD or SD Card) is a type of memory card created by Matsushita Electronic, SanDisk and Toshiba in January 2000. Secure Digital memory is a memory specifically developed to meet new safety requirements in the field of electronic audio and video devices. It therefore includes a copyright protection system that satisfies the SDMI (Secure Digital Music Initiative) standard.
The architecture of the SD cards is based on NAND-type flash memory circuits (EEPROM).
The Secure Digital memory has small dimensions (24.0mm x 32.0mm x 2.1mm), equivalent to those of a postage stamp, and weighs barely 2 grammes.
SD Card - Secure Digital


Data are accessed using a 9-pin lateral connector achieving a transfer speed of 2 Mb/s with the potential to go up to 10 MB/s.
SD memory access time is around 25µs for first access and cycles of 50 ns for subsequent cycles.

Type of Memory Card - Multimedia Cards(MMC)

Multimedia card memory (abbreviated as MMC) is a type of memory card created jointly by SanDisk and Siemens in November 1997.
Its architecture is based on a combination of read-only memory (ROM) for read-only applications and flash memory for read/write purposes.
Multimedia cards are very small (24.0 mm x 32.0 mm x 1.4 mm), which is equivalent to the size of a postage stamp, and weigh only 2.2 grams.
MMC Card - Multimedia Card



There are two types of MMC cards that have different voltages:
  • MMC 3.3V, with a notch on the upper left-hand corner
  • MMC 5V, with a notch on the upper right-hand corner



Data can be accessed by way of an edge connector with 7 pins, for a throughput of up to 2 Mb/s (perhaps even 2.5 Mb/s).

Type of Memory Card - Memory Stick

The Memory Stick (written as MS or MS Card) is a type of memory card created jointly by Sony and SanDisk in January 2000.
The architecture of Memory Stick cards is based on NAND flash memory circuits (EEPROM).
Memory stick memories are very small (21.5 mm x 50.0 mm x 2.8 mm), which is equivalent to the size of a small box of matches, and weigh only 4 grams.
MS Card - Memory Stick



Data can be accessed by way of an edge connector with 10 pins, for a throughput of up to 14.4 Mb/s (up to a maximum of 19.6 Mb/s).
There are two types of Memory Sticks: the "normal" Memory Stick and the "Magic Gate", which protects documents that are copyright protected.

Type of Memory Card - Compact Flash

Compact Flash memory (sometimes called CF) is a kind of memory card created in 1994 by the company SanDisk. Compact Flash is made up of a memory controller and a flash memory chip contained within a miniature casing (42.8mm wide and 36.4mm high), which is smaller than a matchbox and weighs only 11.4 grams.
There are two types of Compact Flash cards, with different dimensions:
  • Type I Compact Flash cards, which are 3.3mm thick;

3.3mmCompact Flash card
  • Type II Compact Flash cards, which are 5mm thick.

5mm Compact Flash card



CompactFlash cards comply with the PCMCIA/ATA standard, although the connector has 50 pins instead of 68, as PCMCIA do. For this reason, a CompactFlash card can be inserted into a passive Type II PCMCIA slot

Introduction to Flash memory

Flash memory is a kind of semiconductor-based, non-volatile, rewritable computer memory; that is, it has many of the same characteristics as RAM, except that the data is not wiped out when the machine is turned off. Flash memory stores bits of data in memory cells, but the data remains saved even when electrical power is cut.
Due to its higher speed, durability, and low energy consumption, flash memory is ideal for many applications, such as digital cameras, mobile phones, printers, PDAs, laptop computers, and device that can record and play back sound, such as mp3 players. What's more, this kind of memory has no moving parts, which makes it very shock-resistant.

Types of memory cards

There are many competing, incompatible memory card formats, almost one for every manufacturer. Among these formats of memory cards, the most common are
  • Compact Flash
  • Secure Digital cards (called SD Card)
  • Memory Stick
  • SmartMedia
  • MMC (MultimediaCard)
  • xD picture card

Comparison


Dimensions (mm)Volume (mm3)Weight (g)# of connectorsTransfer rateTheoretical capacityTheoretical size
Compact Flash type I43 x 36 x 3,35 1083,35020 MB/s137 GB128 GB
Compact Flash type II43 x 36 x 57 74045020 MB/s137 GB12 GB
SmartMedia37 x 45 x 0,81 2652222 MB/s128 MB128 MB
MMC24 x 32 x 1,41 0751,3720 MB/s128 GB8 GB
MMC Plus24 x 32 x 1,41 0751,3752 MB/s128 GB4 GB
RS-MMC MMC Mobile24 x 16 x 1,45381,3138 MB/s128 GB2 GB
MMC Micro14 x 12 x 1,1185< 113
128 GB2 GB
Memory Stick Standard, Pro21,5 x 50 x 2,83 0104102 MB/s128 MB128 MB
Memory Stick Duo, Pro Duo20 x 31 x 1,699221020 MB/s32 GB16 GB
Memory Stick Pro-HG20 x 31 x 1,699221060 MB/s32 GB32 GB
Memory Stick Micro M212,5 x 15 x 1,222521020 MB/s32 GB8 GB
SD24 x 32 x 2,11 6132920 MB/s32 GB32 GB
mini SD20 x 21,5 x 1,460211112 MB/s32 GB4 GB
micro SD15 x 11 x 11650,3810 MB/s32 GB12 GB
xD25 x 20 x 1,88902,8189 MB/s8 GB2 GB

Memory card readers

It should be noted that there are multi-format memory card readers, most of which can be plugged into a USB port.

All About Read-only memory (ROM)

There is a type of memory that stores data without electrical current; it is the ROM (Read Only Memory) or is sometimes called non-volatile memory as it is not erased when the system is switched off.
This type of memory lets you stored the data needed to start up the computer. Indeed, this information cannot be stored on the hard disk since the disk parameters (vital for its initialization) are part of these data which are essential for booting.
Different ROM-type memories contain these essential start-up data, i.e.:
  • The BIOS is a programme for controlling the system's main input-output interfaces, hence the name BIOS ROM which is sometimes given to the read-only memory chip of the mother board which hosts it.
  • The bootstrap loader: a programme for loading (random access) memory into the operating system and launching it. This generally seeks the operating system on the floppy drive then on the hard disk, which allows the operating system to be launched from a system floppy disk in the event of malfunction of the system installed on the hard disk.
  • The CMOS Setup is the screen displayed when the computer starts up and which is used to amend the system parameters (often wrongly referred to as BIOS).
  • The Power-On Self Test (POST), a programme that runs automatically when the system is booted, thus allowing the system to be tested (this is why the system "counts" the RAM at start-up).

Given that ROM are much slower than RAM memories (access time for a ROM is around 150 ns whereas for SDRAM it is around 10 ns), the instructions given in the ROM are sometimes copied to the RAM at start-up; this is known as shadowing, though is usually referred to as shadow memory).

Types of ROM

ROM memories have gradually evolved from fixed read-only memories to memories than can be programmed and then re-programmed.

ROM

The first ROMs were made using a procedure that directly writes the binary data in a silicon plate using a mask. This procedure is now obsolete.

PROM

PROM (Programmable Read Only Memory) memories were developed at the end of the 70s by a company called Texas Instruments. These memories are chips comprising thousands of fuses (or diodes) that can be "burnt" using a device called a " ROM programmer", applying high voltage (12V) to the memory boxes to be marked. The fuses thus burnt correspond to 0 and the others to 1.

EPROM

EPROM (Erasable Programmable Read Only Memory) memories are PROMs that can be deleted. These chips have a glass panel that lets ultra-violet rays through. When the chip is subjected to ultra-violet rays with a certain wavelength, the fuses are reconstituted, meaning that all the memory bits return to 1. This is why this type of PROM is called erasable.

EEPROM

EEPROM (Electrically Erasable Read Only Memory memories are also erasable PROMs, but unlike EPROMs, they can be erased by a simple electric current, meaning that they can be erased even when they are in position in the computer.
There is a variant of these memories known as flash memories (also Flash ROM or Flash EPROM). Unlike the classic EEPROMs that use 2 to 3 transistors for each bit to be memorised, the EPROM Flash uses only one transistor. Moreover, the EEPROM may be written and read word by word, while the Flash can be erased only in pages (the size of the pages decreases constantly).
Lastly, the Flash memory is denser, meaning that chips containing several hundred mega octets can be produced. EEPROMs are thus used preferably to memorise configuration data and the Flash memory is used for programmable code (IT programmes).
The action involving reprogramming of an EEPROM is known as flashing.

All About Random access memory(RAM) Part 2

DRAM PM

The DRAM (Dynamic RAM) is the most common type of memory at the start of this millennium. This is a memory whose transistors are arranged in a matrix in rows and columns. A transistor, coupled with a capacitor, gives information on a bit. Since 1 octet contains 8 bits, a DRAM memory module of 256 Mo will thus contain 256 * 2^10 * 2^10 = 256 * 1024 * 1024 = 268,435,456 octets = 268,435,456 * 8 = 2,147,483,648 bits = 2,147,483,648 transistors. A module of 256 Mo thus has a capacity of 268,435,456 octets, or 268 Mo! These memories have access times of 60 ns.
Furthermore, access to memory generally concerns data stored consecutively in the memory. Thus burst mode allows access to the three pieces of data following the first piece with no additional latency time. In this burst mode, time required to access the first piece of data is equal to cycle time plus latency time, and the time required to access the other three pieces of data is equal to just the cycle time; the four access times are thus written in the form X-Y-Y-Y, for example 5-3-3-3 indicates a memory for which 5 clock cycles are needed to access the first piece of data and 3 for the subsequent ones.

DRAM FPM

To speed up access to the DRAM, there is a technique, known as paging, which involves accessing data located in the same column by changing only the address of the row, thus avoiding repetition of the column number between reading of each row. This is known as DRAM FPM (Fast Page Mode). FPM achieves access times of around 70 to 80 nanoseconds for operating frequency between 25 and 33 Mhz.

DRAM EDO

DRAM EDO (Extended Data Out, sometimes also called hyper-page") was introduced in 1995. The technique used with this type of memory involves addressing the next column while reading the data in a column. This creates an overlap of access thus saving time on each cycle. EDO memory access time is thus around 50 to 60 nanoseconds for operating frequency between 33 and 66 Mhz.
Thus the RAM EDO, when used in burst mode, achieves 5-2-2-2 cycles, representing a gain of 4 cycles on access to 4 pieces of data. Since the EDO memory did not work with frequencies higher than 66 Mhz, it was abandoned in favor of the SDRAM.

SDRAM

The SDRAM (Synchronous DRAM), introduced in 1997, allows synchronized reading of data with the mother-board bus, unlike the EDO and FPM memories (known as asynchronous) which have their own clock. The SDRAM thus eliminates waiting times due to synchronization with the mother-board. This achieves a 5-1-1-1 burst mode cycle, with a gain of 3 cycles in comparison with the RAM EDO. The SDRAM is thus able to operate with frequency up to 150 Mhz, allowing it to achieve access times of around 10 ns.

DR-SDRAM (Rambus DRAM)

The DR-SDRAM (Direct Rambus DRAM) is a type of memory that lets you transfer data to a 16-bit bus at frequency of 800Mhz, giving it a bandwidth of 1.6 GB/s. As with the SDRAM, this type of memory is synchronized with the bus clock to enhance data exchange. However, the RAMBUS memory is a proprietary technology, meaning that any company wishing to produce RAM modules using this technology must pay royalties to both RAMBUS and Intel.

DDR-SDRAM

The DDR-SDRAM (Double Data Rate SDRAM) is a memory, based on the SDRAM technology, which doubles the transfer rate of the SDRAM using the same frequency.
Data are read or written into memory based on a clock. Standard DRAM memories use a method known as SDR (Single Data Rate) involving reading or writing a piece of data at each leading edge.
SDR - Single Data Rate



The DDR doubles the frequency of reading/writing, with a clock at the same frequency, by sending data to each leading edge and to each trailing edge.
DDR - Double Data Rate



DDR memories generally have a product name such as PCXXXX where "XXXX" represents the speed in Mo/s.

DDR2-SDRAM

DDR2 (or DDR-II) memory achieves speeds that are twice as high as those of the DDR with the same external frequency.
QDR (Quadruple Data Rate or quad-pumped) designates the reading and writing method used. DDR2 memory in fact uses two separate channels for reading and writing, so that it is able to send or receive twice as much data as the DDR.
QDR - Quad Data Rate



DDR2 also has more connectors than the classic DDR (240 for DDR2 compared with 184 for DDR).

summary table

The table below gives the equivalence between the mother-board frequency (FSB), the memory (RAM) frequency and its speed:
MemoryName Frequency (RAM)
[!Frequency (FSB)
Speed
DDR200 PC1600200 MHz 100 MHz1.6 GB/s
DDR266 PC2100266 MHz 133 MHz2.1 GB/s
DDR333 PC2700333 MHz 166 MHz2.7 GB/s
DDR400 PC3200400 MHz 200 MHz3.2 GB/s
DDR433 PC3500433 MHz217 MHz3.5 GB/s
DDR466 PC3700466 MHz 233 MHz3.7 GB/s
DDR500 PC4000500 MHz 250 MHz4 GB/s
DDR533 PC4200533 MHz 266 MHz4.2 GB/s
DDR538 PC4300538 MHz 269 MHz4.3 GB/s
DDR550 PC4400550 MHz 275 MHz4.4 GB/s
DDR2-400 PC2-3200400 MHz 100 MHz3.2 GB/s
DDR2-533 PC2-4300533 MHz 133 MHz4.3 GB/s
DDR2-667 PC2-5300667 MHz167 MHz5.3 GB/s
DDR2-675 PC2-5400675 MHz 172.5 MHz5.4 GB/s
DDR2-800 PC2-6400800 MHz 200 MHz6.4 GB/s

Synchronization (timings)

It is not unusual to see scores such as 3-2-2-2 or 2-3-3-2 to describe the parameterization of the random access memory. This succession of four figures describes the synchronization of the memory (timing), i.e. the succession of clock cycles needed to access a piece of data stored in the RAM. These four figures generally correspond, in order, to the following values:
  • CAS delay or CAS latency (CAS meaning Column Address Strobe): this is the number of clock cycles that elapse between the reading command being sent and the piece of data actually arriving. In other words, it is the time needed to access a column.
  • RAS Precharge Time (known as tRP, RAS meaning Row Address Strobe): this is the number of clock cycles between two RAS instructions, i.e. between two accesses to a row. operation.
  • RAS to CAS delay (sometimes called tRCD): this is the number of clock cycles corresponding to access time from a row to a column.
  • RAS active time (sometimes called tRAS): this is the number of clock cycles corresponding to the time needed to access a row.



The memory cards are equipped with a device called SPD (Serial Presence Detect), allowing the BIOS to find out the nominal setting values defined by the manufacturer. It is an EEPROM whose data will be loaded by the BIOS if the user chooses "auto" setting.

Error correction

Some memories have mechanisms for correcting errors to ensure the integrity of the data they contain. This type of memory is generally used on systems working on critical data, which is why this type of memory is found in servers.

Parity bit

Modules with parity bit ensure that the data contained in the memory are the ones required. To achieve this, one of the bits from each octet stored in the memory is used to store the sum of the data bits. The parity bit is 1 when the sum of the data bits is an odd number and 0 in the opposite case.
Thus the modules with parity bit allow the integrity of data to be checked but do not provide for error correction. Moreover, for 9 Mo of memory, only 8 will be used to store data since the last mega octet is used to store the parity bits.

ECC modules

ECC (Error Correction Coding) memory modules are memories with several bits dedicated to error correction (they are known as control bits). These modules, used mainly in servers, allow detection and correction of errors.

Dual Channel

Some memory controllers offer a dual channel for the memory. The memory modules are used in pairs to achieve higher bandwidth and thus make the best use of the system's capacity. When using the Dual Channel, it is vital to use identical modules in a pair (same frequency and capacity and preferably the same brand).

 
Copyright © March 2012. MARIDADI AGENCY
-Haki zote zimehifadhiwa
- All Rights Reserved
Template Modified by Hassan Moh Toziri
Published by Maridadi Agency