Local buses. VESA Local Bus
The bus is an integral part of the motherboard on which connectors (slots) are located for connecting device adapter cards (video cards, sound cards, internal modems, storage devices, input/output devices, etc.) and extensions of the basic configuration (additional empty connectors) . It is not visible externally, but is located between the textolite plates of the motherboard.
As noted earlier, performance computer system In general, the system bus has a big influence. Tires are arteries through which electrical signals are transmitted. Strictly speaking, these are communication channels used to organize interaction between computer devices. And those connectors into which expansion cards are installed are supported by local buses, or interfaces. These connectors are made in the form of slots, and with their help, additional devices (components) are connected via local buses, which, like the system bus, are not visible on motherboards. The structure of the bus interconnection is shown schematically in Fig. 9.
Let us characterize the buses that are present on the motherboard. The main system bus is FSB (Front Side Bus). This bus transfers data between the processor and RAM, as well as between the processor and other devices of the personal computer. This is where there is one pitfall. The fact is that there is a main bus, the processor bus. Some authors claim that the system bus and the processor bus are the same thing, while others do not. Most come to the conclusion: at first the processor was connected to the main system bus through its own processor bus, but in modern systems these buses have become one. We say: “system bus,” but we mean the processor bus; we say: “processor bus,” but we mean the system bus. The phrase: “the motherboard runs at 100 MHz” means that it is the system bus that runs at a clock frequency of 100 MHz. The FSB capacity is equal to the CPU capacity. If you are using a 64-bit processor and the clock speed is system bus 100 MHz, then the data transfer rate will be equal to 800 MB/sec (which is shown in the calculations below).
There are three main indicators of tire performance. These are clock frequency, bit depth and data transfer rate.
Clock frequency. The higher the clock frequency of the system bus, the faster the transfer of information between devices will be carried out and, as a result, the increase in overall performance computer, i.e. the data transfer speed and, consequently, the speed of the computer will increase.
Clock frequency, in relation to personal computers, is measured in MHz, where hertz is one vibration per second, respectively, 1 MHz is a million vibrations per second. Theoretically, if the computer system bus operates at a frequency of 100 MHz, then it can perform up to 100,000,000 operations per second. It is not necessary for every component of the system to necessarily do something with every clock cycle. There are so-called empty clocks (waiting cycles), when the device is in the process of waiting for a response from some other device. Personal computers of the Pentium I class were equipped with motherboards supporting a system bus frequency of 33 MHz, Pentium II - 66 MHz, Pentium III - 133 MHz. Modern motherboards support system bus operation at frequencies of 400, 533, 800, 1066 and even 1600 MHz.
Bit depth. The bus consists of several channels for transmitting electrical signals. If a bus is thirty-two-bit, this means that it is capable of transmitting electrical signals through thirty-two channels simultaneously. A bus of any declared width (8, 16, 32, 64) has, in fact, O More channels. That is, if we take the same thirty-two-bit bus, then 32 channels are allocated for transmitting data itself, and additional channels are intended for transmitting specific information, such as control signals.
Data transfer rate. The name of this parameter speaks for itself. It is calculated by the formula
clock speed * bit depth = data transfer rate.
Let's calculate the data transfer rate for a 64-bit system bus operating at a clock frequency of 100 MHz.
100 * 64 = 6400 Mbps;
6400 / 8 = 800 MB/sec.
But the resulting number is not real. In life, tires are affected by all sorts of factors: ineffective conductivity of materials, interference, design and assembly flaws, and much more. According to some reports, the difference between the theoretical data transfer speed and the practical one can be up to 25%.
In addition to the system bus, the motherboard also has input/output buses, which differ from each other in architecture. They are called local.
On personal computers different generations Bus standards ISA, EISA, VESA, VLB and PCI were used. ISA, EISA, VESA and VLB are now obsolete and not available on modern motherboards. Today all motherboards are based on the PCI bus.
All standards differ both in the number and use of signals, and in the protocols for their maintenance.
ISA (Industrial Standard Architecture). The first 8-bit ISA bus appeared in 1981, and in 1984 its 16-bit version appeared. The first ISA buses were actually the only type, but then differed in clock speeds of 8 MHz and 16 MHz. It should be noted that ISA buses were the only ones on motherboards for almost 10 years and are still found on some of them. Until 1987, IBM refused to publish Full description ISA, many hardware manufacturers decided to develop their own buses. This is how the 32-bit ISA appeared, which was not used, but actually predetermined the appearance of the next generations of MCA and EISA buses. In 1985 Intel company developed a 32-bit 80386 processor, which was released at the end of 1986. There was an urgent need for a 32-bit I/O bus. Instead of continuing to further develop the ISA, IBM created a new MCA (Micro Channel Architecture) bus, which was superior to its predecessor in every way. But this standard did not last long, and soon Compaq developed a new EISA bus.
EISA (Extended Industry Standard Architecture). Its main difference was 32-bit technology, which led to an increase in data exchange speed. At the same time, compatibility with boards designed to work with ISA was maintained. The data transfer speed was already 33 MB/sec. But still the internal clock frequency remained low - 8.33 MHz. With the increase in clock frequencies and bit depth of processors, an urgent problem arose in increasing the data transfer speed on the buses. In 1992, another extended version of ISA appeared - VLB (VESA Local Bus) - Video Electronic Standard Association. VLB was a local bus that did not change, but complemented existing standards. Simply, several new high-speed local slots were added to the main buses. The popularity of the VLB tire lasted until 1994. The VLB data transfer speed was 128 – 132 MB/sec, and the bit depth was 32. The clock frequency reached 50 MHz, but actually did not exceed 33 MHz due to the frequency limitations of the slots themselves. The main function for which the new bus was intended was data exchange with the video adapter. But the new tire had a number of shortcomings that did not allow it to exist on the market for long.
In 1991, development began on a new local PCI bus. PCI (Peripheral Component Interconnect bus) – bus for connecting peripheral components. And in June 1992, this new standard appeared - PCI (2.0), which was developed by Intel together with other companies Compaq, HP, etc. This was a kind of revolution. The variety of expansion cards using the PCI bus was great. The PCI bus clock speed was 33 MHz and 66 MHz. Bit depth – 32 or 64. Data transfer speed – 132 MB/sec or 264 MB/sec. The PCI bus provides self-configuration of peripheral (additional) equipment - support for the Plug and Play standard, which eliminates manual configuration of hardware parameters of peripheral equipment when it is changed or expanded. An operating system that supports this standard automatically configures equipment connected via the PCI bus without user intervention.
The constant improvement of video cards led to the fact that the physical parameters of the PCI bus became insufficient, which led to the appearance of AGP in 1996. Until 1997 graphics subsystem heavily loaded the PCI bus. The release of the Accelerated Graphics Port (AGP) with the Intel 440LX chipset served two purposes: to increase graphics performance and to remove graphics data from the PCI bus. Because the graphic information began to be transmitted over another “bus”, the overloaded PCI bus was able to be freed up to work with other devices.
This port exists in only one form on the motherboard. Neither physically nor logically it depends on PCI. The first AGP 1.0 standard appeared in 1997 thanks to Intel engineers. This specification corresponded to a clock frequency of 66 MHz. The next version, AGP 2.0, was born in 1998 and the data transfer speed is 533 MB/sec (2x) and 1066 MB/sec (4x). The latest version of AGP was AGPx8 (2004–2005). The main (basic) AGP mode is 1x. In this mode, a single data transfer occurs per cycle. In 2x mode, data transmission occurs twice per cycle, in 4x mode, data transmission occurs four times in each cycle, and so on. AGP 1.0 bandwidth is 32 bits. The great achievement of AGP is that this specification allows fast access to RAM.
However, AGP was only the first step in reducing the load on the PCI bus. PCI Express, formerly known as 3rd Generation I/O (3GIO), is intended to replace the PCI bus and take over the task of interconnecting components within a computer for the next ten years.
As for the cost of implementation, the new bus is designed to meet the PCI level or even be lower than it. Serial bus requires presence smaller number conductors on the PCB, making the board design easier and more efficient since the free space can be used for other components.
The bus maintains PCI compatibility at the software level, meaning existing operating systems will boot without any changes. Additionally, PCI Express device configuration and drivers will be compatible with existing PCI options.
One of the most impressive features of PCI Express is its ability to scale speed using multiple transmission lines. Physical layer Supports bus width X1, X2, X4, X8, X12, X16 and X32 lines. Transmission over multiple lines is transparent to other layers.
Since PCI Express provides transfer speeds of 200 MB/s even at an X1 width, the bus is very effective solution in terms of cost/number of contacts. The PCI Express x16 bus allows you to achieve a throughput of 4 GB/s in each direction (total throughput 8 GB/s) for graphics, more than twice the throughput of AGP 8X.
In other words, the specification describes several types of connections and connectors: PCI Express 1x, 4x, 8x, 16x. The first consists of one so-called Lane. The last one is out of sixteen. Accordingly, the throughput of the first is 500 MB/s in both directions, and the latter is 8 GB/s (4 GB/s in each direction). Moreover, all 20 existing Lane groups can be randomly distributed between 1x, 4x, 8x, and 16x connectors. The slots are compatible from bottom to top, that is, a PCI Express 1x card can be inserted into a PCI Express 4x, 8x, or 16x slot. But not the other way around. It remains to add that desktop PCs mainly use 1x and 16x buses. You should also pay attention to the reduction in size of PCI Express compared to just PCI. In the initial stages, PCI Express was intended to connect video cards, which were quite expensive ($400 or more). Currently, low- and mid-price video cards for the PCI Express bus have become available. And manufacturers of other computer components are beginning to actively develop new devices for this bus. And as indicated in the forecasts, for at least 10 years the PCI Express bus will be the main one for connectivity internal devices The PC will gradually squeeze out the PCI bus.
Chipset
As you can already see from the example of the system and local buses, the motherboard is a rather complex device and includes the next important component - the chipset. All the main characteristics of the motherboard, and therefore the computer system constructed on its basis, directly depend on the chipset.
The chipset is the basis of any motherboard. In fact, the functionality of the motherboard and its performance are 90% determined by the chipset, which determines the supported processor type, memory type, and also functionality for connecting peripheral devices.
Chipset is a set of chips system logic(abbreviated as NMS or MSL). It is well known that a personal computer consists of a number of devices that are somehow connected to the motherboard and are engaged in receiving, processing and transmitting any information. Chipsets are responsible for the logical organization of all this work. In the first generations of PCs, when NMS did not yet exist, motherboards carried up to a hundred microcircuits that were responsible for the logical organization of the operation of individual devices, which was extremely inconvenient. Here are some of them: interrupt controllers, direct access controller, keyboard controller, clock, system timer, bus controller, and so on and so forth. This situation existed until 1986, when Chip and Technologies proposed a truly revolutionary solution. The chip was called 82C206 and became the main part of the system logic chipset. She performed the following functions:
Bus controller;
Clock generator;
System timer;
Interrupt controller;
Direct Memory Access Controller;
With the advent of the i80486 processor, individual chips began to be combined into one or two large chips, which were called a chipset. Literally translated, chipset means “chip set.” A chipset, also called a system logic set, is one or most often two microcircuits (chips) designed to organize interaction between the processor, memory, I/O ports and other computer components.
With the advent of the PCI bus, individual chipset chips began to be called bridges - this is how the established terms appeared: North Bridge and South Bridge of the chipset, with the north bridge connected directly to the processor, and the south bridge to the north. In some cases, manufacturers combine the north and south bridges into one chip, and this solution is called a single-chip solution, and if there are two chips, then it is a dual-bridge circuit.
The northbridge of the chipset traditionally includes a RAM controller (with the exception of chipsets for processors with AMD64 architecture), a graphics bus controller (AGP or PCI Express x16), an interface for interaction with the southbridge, and an interface for interaction with the processor. In some cases, the northbridge of the chipset may contain additional PCI Express x1 lanes to organize interaction with expansion cards that have the appropriate interface.
The south bridge of the chipset is responsible for organizing interaction with I/O devices. South Bridge contains hard drive controllers (SATA and/or PATA), USB controller, Network Controller, PCI bus and PCI-Express bus controller, interrupt controller and DMA controller. Also, a sound controller is usually built into the southbridge, and in this case a codec chip external to the chipset is also required. In addition, the south bridge connects to two more important chips on the motherboard: the BIOS ROM memory chip and the Super I/O chip, which is responsible for serial and parallel ports and for the floppy drive.
A special dedicated bus is used to connect the north and south bridges to each other, and different manufacturers use different buses (with different bandwidths) for this:
Intel-DMI (Direct Media Interface),
VIA Technologies (main manufacturer for AMD processors)-V-Link;
· SiS (Silicon Integrated System Corporation) - MuTIOL;
· ATI-HyperTransport, PCI Express;
· NVIDIA-HyperTransport.
As a rule, the name of the chipset coincides with the name north bridge, although it is more correct to indicate the combination of north and south bridges, since in many cases the same north bridge of a chipset can be combined with various options southern bridges.
The choice of chipsets today is very large. And if processors are produced by only two companies - Intel and AMD - then chipsets are produced by Intel, VIA, SiS, NVIDIA, ATI, and ULi.
Let's look at some features modern chipsets Intel company. Today Intel company produces a very diverse range of chipsets for Intel Pentium D, Intel Pentium 4 and Intel Celeron D. In 2004–2005 the Intel 915, Intel 925 family was used, and in 2006 - Intel 945. Together with the new Intel Pentium Extreme Edition 8xx and Intel Pentium D processors, Intel introduced and new chipset Intel 955X Express (codenamed Glenwood). All marked chipsets are designed for the LGA775 microprocessor package.
The Intel 955X Express chipset is today the older model and a logical continuation of chipsets Intel series 945, Intel 925X Express. It can support dual core processor Intel Pentium Extreme Edition 8xx with FSB frequency 800 MHz or single-core Intel Pentium 4 Extreme Edition processor with FSB frequency 1066 MHz and regular processors Intel Pentium 4. The Intel Pentium D processor is equipped with the Intel 945X Express chipset. Now let's list the main features of the Intel 955X Express system logic set (Fig. 10) compared to previous series.
The memory controller of this chipset supports DDR2-667 memory in dual-channel mode, and the memory bus has a bandwidth of 8.5 GB/s. In total, up to 8 GB of memory is supported, and support for ECC memory is implemented. In addition, the memory controller implements Performance Memory Optimizations technology.
For compatibility with Intel Pentium 4 Extreme Edition processors, the FSB frequency can be either 800 or 1066 MHz. Northbridge feature Intel chipset The 955X Express also supports dual graphics buses with an external bridge providing two physical PCI Express x16 slots. The south bridge of the ICH7 chipset is a new version of the already well-known ICH6 I/O controller. Functional features include support for a four-channel SATA RAID controller, eight-channel Intel High audio format Definition Audio, PCI bus and six PCI Express x1 bus slots.
Chipsets are developed for specific generations of processors and specific processor models. For example, companies VIA Technologies, NVIDIA, SiS largely develop chipsets for AMD processors. And Intel, of course, works for its own the lineup Pentium 4. The main characteristics of Intel chipsets are reflected in the table. 5. As you can see, the older the model range, the greater the performance and functionality capabilities they contain. Support for high-speed buses (FSB 800/1066 MHz), modern processor socket (LGA 775), fast and large capacity memory (DDR2), increased number of USB ports, high-speed hard drive interfaces (SATA II) and others.
Rice. 10. Block diagram of the Intel 955X Express chipset
BIOS (Basic Input/Output System) is software built into the computer on a chip, which is available to it at the first stage without accessing the disk. It is a set of programs for testing and maintaining computer hardware, in particular those necessary for managing the keyboard, video card, disks, ports and so-called “cold” boot) and reset (“hot” boot) of the system board, tests the board itself and the main units of the computer - the video adapter , keyboard, disk controllers and I/O ports, configures the chipset and transfers control to the operating system bootloader. A sample BIOS chip is shown in Fig. eleven.
Rice. 11. Company BIOS chip American Megatrends Inc (AMI).
Table 5
Main characteristics of chipsets for Intel microprocessors
Essentially, BIOS is a set of drivers (a driver is a device control program) that ensures the system operates when the computer starts or boots in safe mode. When you turn on the computer's power even before loading the operating system, you can control it from the keyboard and see all the actions on the monitor. In addition, if booting occurs in safe mode, then the operating system drivers are abandoned and only the BIOS drivers remain in operation.
When running under the DOS and Windows 9x operating systems, the BIOS also controlled the main devices, that is, it acted as an intermediary between the operating system and the computer hardware. When working under Windows NT/2000/XP, varieties of UNIX, OS/2 and other alternative operating systems, the BIOS is practically not used, performing only initial checking and configuration.
BIOS consists of the following parts:
1. POST (Power On Self Test) is a program responsible for testing the computer hardware when the power is turned on.
2. System Setup - system setup program.
3. A set of programs for controlling the operation of PC equipment.
BIOS, generally speaking, is unique for each computer motherboard model, that is, it is developed taking into account the operating features of that combination of equipment that is typical for this particular model.
BIOS for modern motherboards is most often developed by one of the companies specializing in this - Award Software (which acquired Phoenix Technology, one of the most famous BIOS manufacturers in the past), American Megatrends Inc. ( AMI), Microid Research. Currently most popular Award BIOS. Some motherboard manufacturers - Intel, IBM or Acer - develop BIOS for their boards themselves. They either significantly expand the range of settings or (as in the case of Intel), on the contrary, limit the number of settings to only the minimum necessary.
Originally, the BIOS was located in a ROM (read-only memory) chip located on the computer's motherboard. This technology allows the BIOS to always be accessible despite damage, e.g. disk system. It also allows the computer to boot from other media on its own. Because RAM is accessed much faster than ROM, computer manufacturers have designed systems so that when the computer is turned on, the BIOS is copied from the ROM to the RAM. The memory area involved is called shadow memory.
In all modern boards, the BIOS is stored in electrically reprogrammable ROM (Flash ROM), which allows BIOS flashing using the board itself using a special program. This allows you to correct factory errors in the BIOS, change factory defaults, make other changes, update the BIOS for new motherboards or computer components.
However, in addition to the obvious advantages, this technology also has weaknesses. For example, there is currently a group of viruses that, taking advantage of the ability to change the contents of the BIOS, erase or change it and thus make the computer inoperable. Due to an incorrect or missing BIOS, the computer refuses to boot. This situation can be corrected only in a service center, where in a special device - a programmer - Flash ROM will be written to the chip original version BIOS. For example, famous virus Chernobyl, which occurred on April 26, 1999, destroyed millions of BIOSes around the world. After this epidemic, some manufacturers began to supply their motherboards with two copies of the BIOS. If the primary copy is damaged, the contents of the backup chip are loaded. However, such boards are quite rare.
The BIOS stores its settings in the so-called CMOS RAM. CMOS RAM is so called because it is based on CMOS structures (CMOS - Complementary Metal Oxide Semiconductor), which are characterized by low power consumption. CMOS memory is non-volatile only because it is constantly powered by a battery located on the motherboard. While the computer is turned on, CMOS RAM is powered by the computer's power supply. The power consumption of CMOS RAM is so low that even when the computer is turned off and missing battery its contents can be stored for more than a day only due to residual charges on the capacitors of the power supply.
CMOS RAM stores information about the current clock readings, alarm time, computer configuration: amount of memory, types of drives, etc. If the CMOS RAM chip is damaged (or the battery or battery is low), the BIOS has the ability to use the default settings.
The general principle that should be followed is: if the computer is working stably and no deficiencies in its operation related to the BIOS have been identified, then you should not update the BIOS.
However, there are situations when updating the BIOS is necessary. Usually this is the release of a new processor, support for which was not included in the previous version. Before installing the new version, you need to go to the website technical support motherboard manufacturer, read the specifications new version BIOS and, if necessary, download them, making sure that this version corrects exactly those shortcomings that were identified in your computer.
When you turn on the computer, power is supplied to the processor and it “wakes up”. The first commands read by the processor are instructions from the BIOS chip (the motherboard chips take care of this). The first to run is POST, a self-test program. POST executes next steps:
· initializes system resources and chipset registers, power management system;
· determines the amount of random access memory (RAM) and tests it;
· initializes the video adapter;
· turns on the keyboard;
· tests serial and parallel ports;
· initializes disk drives and hard disk controllers;
· displays summary system information.
All these actions are briefly displayed on the monitor screen (in black and white) and can be monitored and even analyzed by pressing the “Pause” key.
In progress BIOS operation Compares the current system configuration data with information stored in CMOS and updates it if necessary. If failures occur during any step, the BIOS informs about this with messages on the monitor screen, and if this is not possible (for example, the video adapter has not yet been initialized), it produces sound signals through system speaker. The number of beeps corresponds to error codes, which can be found in the documentation. Some motherboards are equipped with a liquid crystal indicator that displays the stages of POST tests and error codes that have occurred.
After all POST tasks are completed, the BIOS begins searching for a bootloader program. Modern BIOS version allow you to boot the operating system not only from floppy drives and hard drive, but also with CD-ROM drive, ZIP devices or Flash drives. The bootloader program is usually located in the first sector of the disk (hard drive) on which the operating system is located. The order in which disks are searched when searching for a bootloader is set in the BIOS settings. If the bootloader is found, it is placed in memory and control is transferred to it. After this, it finds and places in memory the actual operating system boot program ( operation system loader), which loads, initializes, and configures the operating system and device drivers. And finally, when the operating system is loaded, all control is transferred to the Windows OS, and then other programs are launched, primarily from the Startup folder.
As mentioned earlier, in systems running DOS or Windows 9x, the BIOS takes on the role of managing the PC hardware and serves as an intermediary between the operating system and the hardware.
The BIOS implements its functions through the software interrupt system. Software interrupts cause the microprocessor to pause the current task and begin executing the interrupt routine.
The BIOS problem is that limited number subroutines cannot optimally cover all the needs of the software and all the operating features of the equipment. Thus, using BIOS routines is not always a good thing. In particular, these routines implement some computer functions very slowly. Another negative point is that the BIOS does not allow you to fully exploit the capabilities of the existing hardware, for example, its capabilities that were implemented after the BIOS was written. Therefore, all modern operating systems, having developed system detecting, configuring and working with computer hardware through drivers, do not use BIOS services.
In the future, a number of motherboard manufacturers intend to abandon using the BIOS. For example, Intel is developing a number of technologies that will redistribute BIOS functions between the chipset and operating system extensions and get rid of the oldest surviving part of the PC.
The full name of the BIOS is ROM BIOS (Read Only Memory Basic Input/Output System). At the initial stages of personal development computer BIOS briefly called ROM (Read Only Memory). ROM is the connecting link between the operating system and the hardware. If there were no ROM BIOS, the operating system would be tied to the hardware (as was the case on almost all microcomputer models) and would be completely dependent on them. Since operating systems have a single interface for working with various equipment, problems with incompatibility between hardware and software, as a rule, do not occur, since the BIOS is between them. Let us recall that in the computer world, according to the accepted terminology, hardware is the hardware part of the computer, and software is the software. All this may look something like this (Fig. 12):
Rice. 12. The role of BIOS in creating a unified hardware and software complex
Each motherboard is equipped with a BIOS chip, of which there are four types:
1. ROM (Read Only Memory) or ROM;
2. PROM (Programmable ROM) or PROM (Programmable ROM);
3. EPROM (Erasable PROM) or EPROM (Erasable PROM);
4. EEPROM (Electrically EPROM) or EEPROM (Electronic - Erasable EPROM), the second name is flash ROM.
ROM. The first ROMs were a matrix on which the program code was burned. The matrix was a silicon crystal. It was not possible to overwrite the data. This technology didn't last very long.
PROM. In the late 70s, Texas Instruments developed the first programmable ROM. The first PROM had a capacity of up to 2 MB. Writing to the PROM chip can be done once. But unlike ROM, PROM could be programmed at home. All you had to do was buy a new IC and have a programming device at home connected to your computer. PROM chips had their own identification numbers by which it was possible to determine the type of PROM and the volume in KB.
EPROM The new microcircuits had a quartz window, which was quite expensive. Through the window, under the influence of ultraviolet rays, a chemical reaction occurred that restored the cells. To erase recorded information, it was used special device. According to physical and functional parameters EPROM chips were not particularly different from PROM.
EEPROM The main advantage of these chips is that reprogramming does not require removing them from the motherboard and does not require any additional hardware. Since 1994, almost all motherboards have been equipped with flash ROM, and at this point in time you will not find another BIOS on a modern motherboard.
Modern computing systems are characterized by:
rapid growth in the speed of microprocessors and some external devices (for example, to display digital full-screen video with high quality, a bandwidth of 22 MB/s is required);
the emergence of programs that require a large number of interface operations (for example, graphics processing programs in Windows, multimedia).
Under these conditions, the throughput of expansion buses serving several devices simultaneously was not enough for comfortable user experience, since computers began to “think” for a long time. Interface developers have taken the path of creating local buses connected directly to the MP bus, operating at the MP clock frequency (but not at its internal operating frequency) and providing communication with some high-speed devices external to the MP: main and external memory, video systems, etc. .d.
There are currently three main universal local bus standards: VLB, PCI and AGP.
TireVLB(VL-bus, VESA Local Bus) introduced in 1992 by the Video Electronics Standards Association (VESA - trademark Video Electronics Standards Association) is often called the VESA bus. The VLB bus is essentially an extension of the internal MP bus for communication with a video adapter and, less often, with a hard drive, multimedia cards, and a network adapter. Outdated (targeting only MP 80386, 80486).
TirePCI(Peripheral Component Interconnect, connection of external components) is the most common and universal interface for connecting various devices. Developed in 1993 by Intel. The PCI bus is much more versatile than VLB; allows connection of up to 10 devices; has its own adapter, allowing it to be configured to work with any MP from 80486 to modern Pentium. Using this interface, video cards, sound cards, modems, SCSI controllers and other devices are connected to the motherboard.
The PCI bus, although local, also performs many of the functions of an expansion bus. Expansion buses ISA, EISA, MCA (and it is compatible with them), if there is a PCI bus, are connected not directly to the MP, but to the PCI bus itself (via an expansion interface). Thanks to this solution, the bus is independent of the processor and can work in parallel with the processor bus without accessing it for requests. Thus, the processor bus load is significantly reduced. For example, the processor works with system memory or cache memory, and at this time over the network HDD information is written.
TireAGP(Accelerated Graphics Port) - an interface for connecting the video adapter to a separate AGP bus, which has access directly to the main memory.
Comparative characteristics of tires
|
Characteristic | |||||
|
Digit capacity of KShD/KShA, bit | |||||
|
Frequency, MHz | |||||
|
Bandwidth, MB/sec. | |||||
|
Number of connected devices, pcs. |
ISA, MCA and EISA buses have one common drawback - relatively low performance. The four types of buses described in the following sections are local. The main types of local buses used in PCs include the following.
- VL-Bus (VESA Local Bus)
This limitation existed back in the days of the first PCs, in which the I/O bus operated at the same speed as the processor bus. The speed of the processor bus increased, and the characteristics of I/O buses improved mainly due to an increase in their capacity. It was necessary to limit the speed of the buses because most of the adapter boards produced could not operate at increased data exchange rates.
Some users are haunted by the thought that their computer is running slower than it can. However, the speed of the I/O bus does not matter in most cases. For example, when working with a keyboard or mouse, high performance is not required, since in this situation the computer's performance is determined by the user. It is really only needed in subsystems where high data transfer rates are important, for example in graphics and disk controllers.
The problem associated with bus speed has become relevant due to the spread of graphical user interfaces(eg Windows). They process such large amounts of data that the I/O bus becomes the system's bottleneck. Ultimately, the high performance of a processor with a clock frequency of 66 or even 450 MHz turns out to be completely useless, since data is transferred on the I/O bus several times slower (the clock frequency is about 8 MHz).
An obvious solution to this problem is to have some of the data exchange carried out not through the I/O bus connectors, but through additional high-speed connectors. The best approach to solving this problem is to locate the additional I/O connectors on the fastest bus, i.e. on the processor bus (this is similar to connecting external cache memory). The corresponding block diagram is shown in the figure below. This design is called a local bus, since external devices (adapter boards) now have access to the processor bus (i.e., the bus closest to it). Of course, the local bus connectors must be different from the I/O bus slots so that “slow” adapter cards cannot be inserted into them.
It is interesting to note that the first 8- and 16-bit ISA buses had a local bus architecture. These systems used the processor bus as the primary bus, and all devices operated at processor speed. When the clock speed in ISA systems exceeded 8 MHz, the main computer bus was separated from the processor bus, which could no longer perform these functions. Introduced in 1992, an enhanced version of the ISA bus, called VESA Local Bus (or VL-Bus), marked a return to local bus architecture. Subsequently, the local VESA bus was replaced by the PCI bus, and its addition was the AGP bus.
Note!
To organize a local bus in a computer, it is not at all necessary to install expansion slots: a device that uses a local bus can be mounted directly on the motherboard. The first local bus computers used exactly this approach.
The local bus does not replace previous standards, but complements them. The main computer buses, as before, remain ISA and EISA, but one or more local bus slots are added to them. At the same time, compatibility with older expansion cards is maintained, and high-speed adapters are installed in local bus slots, while all their capabilities are realized. Thus, until now the most common connectors are AGP, PCI and ISA. Older boards sometimes turn out to be compatible with new connectors, but all the capabilities of the local AGP and PCI buses allow you to use only new adapter models. As the popularity of the ISA bus decreases and emphasis shifts to the LPC interface, the role of the ISA bus is gradually reduced, and other buses are used instead.
GUI performance Windows user or Linux (such as KDE or GNOME) has increased significantly after ISA video adapters were replaced by PCI and AGP adapters.
Local bus
All the previously described buses have a common drawback - relatively low throughput. This is due to the fact that the buses were designed with slow processors in mind. Subsequently, the processor speed increased, and the bus characteristics were improved mainly “extensively”, due to the addition of new lines. The obstacle to increasing the bus frequency was great amount released boards that could not operate at high transfer speeds (this applies to MCA to a lesser extent, but for the reasons stated above, this architecture did not play a noticeable role in the market). At the same time, in the early 90s in the world personal computers changes have occurred that require a sharp increase in the speed of exchange with devices:
- creation of a new generation of processors Intel type 80486, operating at frequencies up to 66 MHz;
- increasing the capacity of hard drives and creating faster controllers;
- development and active promotion to the market graphical interfaces user ( Windows type or OS/2) have led to the creation of new graphics adapters that support more a high resolution and more colors (VGA and SVGA).
The obvious way out of this situation is the following: carry out some data exchange operations that require high speeds not through the I/O bus, but through the processor bus, in much the same way as connecting an external cache. This design is called a local bus. The figures clearly demonstrate the difference between conventional architecture and local bus architecture.
The local bus did not replace previous standards, but complemented them. The main buses in the computer were still ISA or EISA, but one or more local bus slots were added to them. Initially, these slots were used almost exclusively for installing video adapters, and by 1992, several incompatible local bus options had been developed, the exclusive rights to which belonged to the manufacturers. Naturally, such confusion held back the spread of local buses, so VESA (Video Electronic Standard Association) - an association representing more than 100 companies - proposed its local bus specification in August 1992.
VESA local bus (VL-bus)
The main characteristics of VL-bus are as follows.
- Support for 80386 and 80486 series processors. The bus is designed for use in single-processor systems, while the specification provides the ability to support x86-incompatible processors using a bridge chip.
- The maximum number of bus masters is 3 (not including the bus controller). If necessary, it is possible to install several subsystems to support more masters.
- Although the bus was originally designed to support video controllers, it can also support other devices (for example, hard disk controllers).
- The standard allows the bus to operate at frequencies up to 66 MHz, but the electrical characteristics of the VL-bus connector limit it to 50 MHz (this limitation, of course, does not apply to devices integrated into the motherboard).
- The bi-directional 32-bit data bus also supports 16-bit communication. The specification includes the possibility of 64-bit exchange.
- DMA support is provided only for bus masters. The bus does not support special DMA "initiators".
- The maximum theoretical bus bandwidth is 160 MV/sec (at a bus frequency of 50 MHz), the standard is 107 MV/sec at a frequency of 33 MHz.
- Batch exchange mode is supported (for 80486 motherboards that support this mode). 5 lines are used to identify the type and speed of the processor, the Burst Last (BLAST#) signal is used to activate this mode. For systems that do not support this mode, the line is set to 0.
- The bus uses a 58-pin MCA connector. A maximum of 3 slots are supported (on some 50 MHz buses, only 1 slot can be installed).
- The VL-bus slot is installed in a line behind the ISA/EISA/MCA slots, so all lines of these buses are available to VL-boards.
- Both the integrated processor cache and the motherboard cache are supported.
- Supply voltage is 5 V. Devices with 3.3 V output are supported provided they can handle 5 V input.
The VL-bus was a huge improvement over the ISA in both performance and design. One of the advantages of the bus was that it made it possible to create cards that work with existing chipsets and do not contain large quantities expensive control logic circuits. As a result, VL cards were cheaper than similar EISA cards. However, this tire was not without its drawbacks, the main ones being the following.
- Orientation towards the 486th processor. VL-bus is hardwired to the 80486 processor bus, which is different from the Pentium and Pentium Pro/Pentium II buses.
- Limited performance. As already mentioned, the real VL-bus frequency is no more than 50 MHz. Moreover, when using processors with a frequency multiplier, the bus uses the main frequency (for example, for the 486DX2-66 the bus frequency will be 33 MHz).
- Circuit restrictions. The quality of signals transmitted over the processor bus is subject to very stringent requirements, which can only be met with certain parameters load of each bus line. According to Intel, installing insufficiently carefully designed VL boards can lead not only to data loss and synchronization problems, but also to system damage.
- Limitation on the number of boards. This limitation also arises from the need to comply with the load restrictions on each line.
Despite the existing shortcomings, VL-bus was the undoubted leader in the market, as it made it possible to eliminate the bottleneck in two subsystems at once - the video subsystem and the hard disk exchange subsystem. However, the leadership was short-lived, as Intel developed its new product - the PCI bus. According to the company, VL-bus was based on 11-year-old technologies and was just a “patch”, a compromise between manufacturers. True, VESA stated that both buses can “coexist” together in one system. Intel agreed that such a neighborhood was possible, but asked a counter-deadly question: “Why?” In fairness, it must be said that PCI was indeed freed from most of the disadvantages inherent in VL-bus.
With the increase in clock frequencies and bit depth of processors, an urgent problem arose in increasing the data transfer speed on the buses (what is the point of using a stone with a clock frequency of, say, 66 MHz, if the bus operates at a frequency of only 8.33 MHz). In some cases, such as a keyboard or mouse, high speed is useless. But engineers from expansion board manufacturers were ready to produce devices at speeds that buses could not provide.
IN
The following way out of this situation was found: some of the data exchange operations that require high speeds should be carried out not through standard I/O bus connectors, but through additional high-speed interfaces - the processor bus, in much the same way as an external cache is connected.
The fact is that these very high-speed interfaces are connected to the processor bus. It follows from this that the connected boards will have access directly to the processor through its bus. This design was called local bus (LB, Local Bus). The local bus did not replace previous standards, but complemented them. The figure shows the difference between a conventional architecture and a local bus architecture. By the way, the first ISA buses were local, but when their clock frequency exceeded 8 MHz, separation occurred.
The main buses in the computer were still ISA or EISA, but one or more local bus slots were added to them. Initially, these slots were used almost exclusively for installing video adapters, and by 1992, several incompatible local bus options had been developed, the exclusive rights to which belonged to the manufacturers.
This diversity was holding back the spread of local buses, so the VESA (Video Electronic Standard Association), representing more than 100 companies, proposed its VESA Local Bus (VL-bus or VLB) specification in August 1992, which did not change , but complemented existing standards. The VLB bus was designed to increase the bandwidth between the main processor and the video card by simply adding several new high-speed local slots to the main buses. The main function for which the new bus was intended was data exchange with the video adapter.
It was a 32-bit bus that used the third and fourth connectors as an extension of the regular ISA slot. The bus operated at a nominal frequency of 33 MHz and provided a significant increase in performance compared to ISA. Subsequently, the VLB bus began to be used by manufacturers of hard drive controllers and other devices requiring high-speed data transfer. Even 100-megabit ones were produced Ethernet controllers with VLB bus. The widespread use of the VESA bus is due to its relative low cost and top-down compatibility with its predecessor, the ISA bus. The VLB connector is an ISA connector with a “continuation”.
The main characteristics of VL-bus are:
support for processors of the 80386 and 80486 series. The bus is designed for use in single-processor systems, while the specification provides for the ability to support x86-incompatible processors using a bridge chip;
The maximum number of bus masters is 3 (not including the bus controller). If necessary, it is possible to install several subsystems to support a larger number of masters. Despite the fact that the bus was originally designed to support video controllers, it can also support other devices (for example, hard disk controllers);
the bus can operate at frequencies up to 66 MHz, but the electrical characteristics of the VL-bus connector limit it to 50 MHz (this limitation, of course, does not apply to devices integrated into the motherboard);
The bi-directional 32-bit data bus also supports 16-bit communication. The specification includes the possibility of 64-bit exchange;
DMA support is provided only for bus masters. The bus does not support special DMA "initiators";
maximum theoretical bus bandwidth 160 Mb/s (at a bus frequency of 50 MHz), standard - 107 Mb/s at a frequency of 33 MHz;
support for batch exchange mode (for 80486 motherboards that support this mode). Five lines are used to identify the processor type and speed, the Burst Last (BLAST#) signal is used to activate this mode. For systems that do not support this mode, the line is set to 0;
using a 58-pin MCA connector. A maximum of 3 slots are supported (on some 50 MHz buses, only 1 slot can be installed). The VL-bus slot is installed in a line behind the ISA/EISA/MCA slots, so all lines of these buses are available to VL-boards;
support for both integrated cache processor and cache on the motherboard. Supply voltage is 5 V. Devices with 3.3 V output are supported provided they can handle 5 V input.
This 32/32-bit bus was designed for 386, 486 and Pentium processors. The VLB bus is most widely used on 486 motherboards. On them, VESA is the address, data and control lines of the processor, output to the connector. This circumstance imposes significant restrictions on VLB expansion cards - time and load parameters must be strictly adhered to. As indicated in the instructions for many motherboards, the number of VLB cards at a clock frequency of 25 MHz should not exceed three, at 33 MHz - two, at 40 and 50 MHz - one. If these requirements are violated, the system will be unstable because the load capacity of the processor has been exceeded.
To estimate the bus speed, you can make the following calculation: if the expansion card operates at a frequency of 50 MHz, then the bus bandwidth will be equal to 32 * 50 * 10 6 = 1.6 * 10 9 Mbit/s = 200 MB/s, which is quite a lot. However, we should not forget that such a speed can almost never be used, since data from video memory cannot be read at such a speed. In addition, while accessing the VLB card, the processor cannot do anything else, no matter how slow the device on this card is (for example, a serial port).
The VL-bus was a huge improvement over the ISA in both performance and design. One of the advantages of the bus was that it allowed the creation of cards that worked with existing chipsets and did not contain a large number of expensive control logic circuits. As a result, VL cards were cheaper than similar EISA cards. However, this tire was not without its drawbacks, the main ones being the following:
targeting 486 processor. VL-bus is hardwired to the 80486 processor bus, which is different from the Pentium and Pentium Pro/Pentium II buses.
limited performance. As already mentioned, the real VL-bus frequency is no more than 50 MHz. Moreover, when using processors with a frequency multiplier, the bus uses the main frequency (for example, for the 486DX2-66 the bus frequency will be 33 MHz);
circuit limitations. The quality of signals transmitted over the processor bus is subject to very stringent requirements, which can only be met with certain load parameters for each bus line. According to Intel, installing insufficiently carefully designed VL boards can lead not only to data loss and synchronization problems, but also to system damage;
limiting the number of boards. This limitation also arises from the need to comply with the load restrictions on each line.
The popularity of the VLB tire lasted until 1994. main feature tires that made it possible to achieve high performance, was also the reason for VLB leaving the market. The bus was a direct extension of the 486 processor/memory bus, running at the same speed as the processor (hence the name local bus). Direct connection means that the connection is too large number devices ran the risk of interfering with the processor itself, especially if signals passed through the slot. VESA recommended using no more than two slots per clock speeds 33 MHz or three slots if they used special buffer. At higher clock speeds, no more than two devices should be connected, and at 50 MHz, both VLB devices should be built into the motherboard.
Since the VLB runs synchronously with the processor, increasing the processor frequency resulted in problems with the VLB peripherals. The faster the peripherals were required to work, the more expensive they were due to the difficulties associated with the production of high-speed components. Few VLB devices supported speeds above 40 MHz.
