1 Introduction. BIOS and CMOS:
( 1)BIOS:
BIOS is the abbreviation of basic input and output system. It is the basic input and output system of PC, and it is an EPROM or EEPROM integrated circuit loaded with startup and self-test programs, that is, a ROM (Read Only Memory) chip integrated on the motherboard. It stores the most important basic input/output program, system information setting program, power-on self-check program and system boot bootstrap program of PC system.
(2)CMOS:
The English full name of CMOS is complex-educational metal-oxicle-semiconductor, and the Chinese translation is "complementary metal oxide semiconductor".
CMOS is a read-write RAM chip on the motherboard of a microcomputer. It is mainly used to save the hardware configuration of the current system and the operator's setting of some parameters. CMOS RAM chip is powered by the system through the backup battery, so CMOS information will not be lost whether the system is turned off or powered off. Since the CMOS ROM chip itself is only a memory and only has the function of saving data, the setting of various parameters in CMOS has to go through special procedures. Now most manufacturers put CMOS setup programs in BIOS chips, and press? Del? Key to enter the CMOS setup program, which is convenient for setting up the system, so CMOS setup is usually called BIOS setup.
(3) 3) Relationship between BIOS and CMOS:
The system setup program in BIOS is the means to complete CMOS parameter setting; CMOS RAM is not only the storage place for BIOS to set system parameters, but also the result of BIOS setting system parameters. So what's their relationship? Set CMOS parameters through BIOS setup program? .
(4) 4) Differences between BIOS and CMOS: (Thanks to Deng 123 1000 for his guidance)
CMOS is just a piece of memory, and BIOS is PC's? Basic input and output system? Program. Because BIOS and CMOS are closely related to system settings, the statement that BIOS settings and CMOS settings are actually the same thing in actual use, but BIOS and CMOS are two completely different concepts, so don't confuse them.
2. Introduction to 2.PCB:
A PCB is a printed circuit board (PCB). It will appear in almost every kind of electronic equipment. If there are electronic parts in a device, they are all embedded in PCB of different sizes. In addition to fixing various small parts, PCB is mainly used to provide electrical connection between the above parts. With the increasing complexity of electronic equipment, more and more parts are needed, and the lines and parts on PCB are more and more dense.
The motherboard of a computer is a PCB without resistors, chips, capacitors and other components.
3. Motherboard North-South Bridge Chip:
(1) The north bridge is the most important part of the motherboard chipset, also known as the host bridge. Generally speaking, the name of the chipset is named after the name of the North Bridge chip. For example, the northbridge chip of Intel 845E chipset is 82845E, and the northbridge chip of 875P chipset is 82875P. Northbridge chip is responsible for contacting CPU and controlling the transmission of memory, AGP or PCI-E data in Northbridge, providing support for CPU type and main frequency, system front-end bus frequency and memory type (SDRAM, DDR SDRAM and RDRAM, etc.). ) and maximum capacity, AGP or PCI-E slot, ECC error correction, etc. The north bridge chip of the integrated chipset also integrates the display core.
Northbridge chip is the closest chip to CPU on the motherboard, mainly because the communication between Northbridge chip and processor is the closest, and in order to improve communication performance, the transmission distance is shortened. Because the data processing capacity of the Northbridge chip is very large, and the heat generation is also increasing, the current Northbridge chips are covered with heat sinks to enhance the heat dissipation of the Northbridge chip, and some motherboards' Northbridge chips also cooperate with fans to dissipate heat. Because the main function of the Northbridge chip is to control the memory, and the memory standard changes as frequently as the processor, the Northbridge chips of different chipsets are definitely different. Of course, this does not mean that the adopted memory technology is completely different, but there must be some differences between the Northbridge chips of different chipsets.
(2) The South Bridge chip is an important part of the motherboard chipset, which is generally located under the CPU slot and near the PCI slot on the motherboard. This layout is based on the fact that there are many I/O buses connected to it, and being far away from the processor is beneficial to wiring. Compared with the north bridge chip, its data processing ability is not great, so the south bridge chip generally does not cover the heat sink. Southbridge chip is not directly connected with the processor, but through a certain way (different chipsets are different from different manufacturers, such as Intel Hub architecture of Intel and multi-threading of SIS? Wonderful canal? ) Connect the North Bridge chip.
The South Bridge chip is responsible for the communication among I/O buses such as PCI bus, USB, LAN, ATA, SATA, audio controller, keyboard controller, real-time clock controller and advanced power management. These technologies are generally stable, so the South Bridge chip may be the same in different chipsets, but the difference is only the North Bridge chip. Therefore, the number of north bridge chips in the motherboard chipset is far more than that of south bridge chips. The development direction of Southbridge chip is mainly to integrate more functions, such as network card, RAID, IEEE 1394 and even WI-FI wireless network.
4. Expansion slot on the motherboard:
Expansion slots are slots on the motherboard for fixing expansion cards and connecting them to the system bus, also known as expansion slots and expansion slots. Expansion slots are a way to add or enhance computer features and functions. For example, if you are not satisfied with the performance of the motherboard integrated graphics card, you can increase the independent graphics card to enhance the display performance; If you are not satisfied with the sound quality of the onboard sound card, you can add a separate sound card to enhance the sound effect; Motherboards that do not support USB2.0 or IEEE 1394 can obtain this function by adding corresponding USB2.0 expansion cards or IEEE 1394 expansion cards.
At present, the main types of expansion slots are ISA, PCI, AGP, CNR, AMR, ACR, relatively rare WI-FI, VXB, PCMCIA for notebook computers. There are MCA slots, EISA slots and VESA slots, which have appeared in history and have long been eliminated. At present, the mainstream expansion slot is PCI Express slot.
(1)AGP slot (accelerated graphics port) is developed on the basis of PCI bus, which is mainly optimized for graphics display and specially used for graphics display cards. The AGP standard has also developed for several years, from the initial AGP 1.0 and AGP2.0 to the present AGP 3.0. If divided by multiple speeds, it has mainly experienced AGP 1X, AGP 2X, AGP 4X and AGP PRO, and the latest version is AGP 3.0, namely AGP 8X. The transmission rate of AGP 8X can reach 2. 1GB/s, which is twice that of AGP 4X. AGP slots are usually brown (the purpose of distinguishing the above three interfaces with different colors is to facilitate user identification). It should be noted that they are not on the same level as PCI and ISA slots, but are built-in, so that PCI and ISA cards cannot be inserted.
(2)PCI-Express is the latest bus and interface standard. What was its original name? 3GIO？ , proposed by Intel, obviously Intel means that it represents the next generation I/O interface standard. It was renamed after being certified by PCI-SIG(PCI Special Interest Organization). PCI-Express？ . This new standard will completely replace the existing PCI and AGP, and finally realize the unification of bus standards. Its main advantage is its high data transmission rate, which can reach above 10GB/s at present, and its development potential is considerable. PCI Express also has many specifications, from PCI Express 1X to PCI Express 16X, which can meet the needs of low-speed equipment and high-speed equipment at present and in the future.
The difference between PCI-E and AGP:
First of all, PCI-E x 16 bus channel is wider than AGP. Maximum speed limit? Higher;
Second, what is the PCI-E channel? Two lanes? , that is? Duplex transmission? Is it allowed at the same time? Become? And then what? Out? Two digital signals pass through at the same time, while AGP has only one channel, that is, only one direction of data flow is allowed at a time. Due to these improvements, the transmission bandwidth of PCI-E x 16 can reach 2? 4Gb/s=8Gb/s, while the maximum size of AGP 8x is only 2Gb/s. The advantages of PCI-E can be seen.
(3)PCI slot is an expansion slot based on PCI local bus (Pedpherd Component Interconnect), generally ivory in color, located below AGP slot and above ISA slot on motherboard. Its bit width is 32 bits or 64 bits, its working frequency is 33MHz, and its maximum data transmission rate is 133 MB/ s (32 bits) and 266 MB/ s (64 bits). Card card, sound card, network card, built-in modem, built-in ADSL modem, USB2.0 card, IEEE 1394 card, IDE interface card, RAID card, TV card, video capture card and other expansion cards. PCI slot is the main expansion slot of motherboard. By plugging in different expansion cards, almost all the external functions that can be realized by computers at present can be obtained.
(4)PCI-X is an extended architecture of PCI bus. Unlike PCI bus, PCI bus must frequently exchange data between the target device and the bus, while PCI-X allows the target device to see that only one PCI-X device has exchanged data. At the same time, if the PCI-X device has no data transmission, the bus will automatically remove the PCI-X device to reduce the waiting time between PCI devices. Therefore, at the same frequency, PCI-X will provide 14-35% higher performance than PCI.
Another advantage of PCI-X is its extensible frequency, which means that the frequency of PCI-X will not be fixed like PCI, but will change with the change of equipment. For example, if a device works at 66MHz, it will work at 66MHz, and if the device supports 100MHz, it will work at 100MHz. PCI-X can support 66, 100 and 133 MHz, and may provide more frequency support in the future.
5. Storage controller
Memory controller is an important part of computer system, which controls memory and exchanges data between memory and CPU. The storage controller determines the maximum storage capacity, the number of memory banks, the type and speed of memory, the depth and width of data in the storage area group and other important parameters, that is to say, it determines the storage performance of the computer system and has a great influence on the overall performance of the computer system.
In the traditional computer system, the memory controller is located in the north bridge chip of the motherboard chipset, and the CPU needs to go through? CPU-north bridge-memory-north bridge-CPU? Five steps, in this mode, the data is transmitted in multiple stages, and the data delay is obviously large, thus affecting the overall performance of the computer system; AMD's K8 series CPU (including various processors with sockets 754/939/940 and other interfaces) integrates a memory controller, so the data exchange process between CPU and memory is simplified to? CPU-memory-CPU Three steps, omitting two steps, obviously has lower data delay than the traditional memory controller scheme, which is helpful to improve the overall performance of the computer system.
The advantage of integrating the memory controller into the CPU is that it can effectively control the memory controller and the CPU core to work at the same frequency, and because the data exchange between the memory and the CPU does not need to go through the north bridge, the transmission delay can be effectively reduced. For example, this is like moving the goods warehouse directly to the processing workshop, which greatly reduces the time required for the round-trip transportation of raw materials and finished products between the goods warehouse and the processing workshop and greatly improves the production efficiency. In this way, the overall performance of the system has also been improved.
The biggest disadvantage of the memory controller integrated in CPU is its poor adaptability and flexibility to memory, and it can only use specific types of memory, with limited capacity and speed. To support new types of memory, the memory controller integrated in CPU needs to be updated, that is, a new CPU needs to be replaced; The traditional memory controller is located in the north bridge chip of the motherboard chipset, so there is no such problem. It can use different types of memory only by replacing the motherboard, such as Intel Pentium 4 series CPU. If the original motherboard does not support DDR2, it can only use DDR2 by replacing a motherboard that supports DDR2. If you are equipped with a motherboard that supports both DDR and DDR2, you can use DDR2 directly without replacing the motherboard.
6. Frequency division effect of memory controller
When the system is working, the running frequency of memory changes with the running frequency of CPU. The element that controls this change is the memory controller. This way of adjusting the running frequency of memory according to the actual frequency of CPU is called the frequency division effect of memory controller. The specific frequency division method varies from platform to platform.
(1)AMD platform
At present, the mainstream AMD CPU integrates a memory controller, so no matter what motherboard it matches, its memory frequency division mechanism is certain. Every AMD platform with hardware configuration has its fixed memory frequency division coefficient, which affects the actual working frequency of memory.
The specific calculation method of AMD platform memory frequency division coefficient is as follows:
Frequency division coefficient n = n = default main frequency of CPU? 2? Memory nominal frequency
The number you got is reused? Become law? Take an integer. Attention? Become law? The number after the decimal point is discarded, and the integer part in front is added with 1 instead of rounding.
At this time, the actual running frequency of memory = the actual running frequency of =CPU? Frequency division coefficient n.
For example, when Athlon64 3000+ of AM2 interface is matched with DDR2 667 memory, we set the memory frequency to DDR2 667 in BIOS, but the memory actually works under DDR2 600, which is caused by the memory frequency division coefficient. Because the setting value of BIOS at this time is not the actual working frequency of memory, we call the setting value in BIOS the nominal frequency of memory.
Take am2aathlon 64 3000+with DDR2 667 memory as an example:
N= 1800？ 2? 667? 5.397, integer =6,
At this time, the actual running frequency of the memory = 1800MHz? 6=300MHz, which means DDR2 600.
If the memory is set to DDR2 533 in BIOS, then the frequency division coefficient N=7 is calculated by the above formula, and the memory actually works under DDR2 5 17.
When the memory with different frequency matches the CPU with different frequency, the memory frequency division coefficient is different.
If the CPU is changed to 3200+, the default frequency is 2GHz.
And then at DDR2 667: N=2000? 2? 667, taking the integer as 6,
In DDR2 533, N=2000? 2? 533, take the integer as 8,
The coefficient n varies with different hardware configurations of the platform.
For AMD platform, three decisive factors directly related to overclocking amplitude are: CPU, memory and HT bus, and any one of them delays the overclocking amplitude of the whole platform. We can artificially reduce CPU frequency doubling and HT bus frequency doubling to reduce the influence of CPU and HT bus on overclocking results. At this time, overclocking can determine the overclocking limit of memory.
(2) Intel platform
The memory controller of Intel platform is generally integrated on the motherboard chip, and its frequency division mechanism is also determined by different motherboard chips.
Memory frequency division coefficient of Intel platform =CPU external frequency: memory running frequency.
Take the mainstream Intel 965/975 chipset as an example, its frequency division mechanism is very clear, and several fixed frequency division coefficients are directly provided in the BIOS. For example, 1: 1, 1: 1.33, 1: 1.66 and so on.
The default external frequency of E6300 is 266MHz. If the frequency division coefficient is set to 1: 1.33,
So the actual running frequency of the memory =266MHz? 1.33=353.78MHz, which is DDR2 707.
The three decisive factors directly related to overclocking amplitude on Intel platform are CPU, memory and FSB bus, in which the FSB bus value is fixed at four times of the CPU external frequency. The frequency division coefficient of Intel 965/975 chipset is less than 1. The smaller the frequency division coefficient, the greater the multiple of the memory operating frequency relative to the external frequency of CPU. We can reduce the influence of CPU composition on the overall overclocking result of the platform by choosing a smaller frequency division coefficient, so as to test the limit overclocking frequency of the memory. NVIDIA's nForce680i chipset also provides a frequency division factor greater than 1, which can make the memory run below the external frequency of CPU.
7. Explain the names and locations of various components on the ATX motherboard.
(Take ASUS P5B-E PLUS motherboard as an example)
/tips/show_bbs_pic.php？ Picid=72859 ASUS P5B-E PLUS motherboard
(1) motherboard power supply design:
Mainboard power supply design
(2)CPU slot: (red box in the figure below)
CPU slot (socket 775)
(3) North-South Bridge Chip:
Main board North Bridge and South Bridge chips (covered with radiator)
(4) Memory slot: (red box in the figure below)
DDR2 DIMM memory slot
(5) Hard disk interface: (red box below)
Summary of the most comprehensive basic knowledge of computer DIY in history (2)
2008-1 2-311:52: 37 Source: Author: large, medium and small viewpoints: 37206 Comments:1Collection articles.
. cn/BBS/73/a 72865 . jpg " onload = " if(this . width/>530){ this . resize = true； this.width = 530} " & gt
Hard disk interface
Includes six SATA 3.0 Gb/s interfaces, 1 ultradma 1 33/100/66 interfaces,1internal SATA 3.0 Gb/s interface and1external SATA 3.0 Gb/s interface.
(6) JMB363 chip supports hard disk interface: (below)
(7) On-board sound card chip: (below)
(8) On-board network card chip: (below)
(9) expansion slot:
Expansion slot on motherboard
In the above figure, the green boxes are the graphics card slots PCI-E X 16 (longer blue slot) and PCI-E X4 (shorter black slot).
The red box in the above picture is an ordinary PCI expansion slot.
(10) Input/output device interface:
Input output device interface
8. Intel chipset naming rules
(1) From 845 series to before 9 15 series.
PE is the mainstream version, without integrated graphics card, supporting the mainstream FSB and memory at that time, and supporting AGP slots.
E is not a simplified version, but an evolutionary version. In particular, 845E only has the suffix e, which increases the support for 533MHz FSB compared with 845D and ECC memory compared with 845G, so 845E is often used in entry-level servers.
G is the mainstream chipset of integrated graphics card, which supports AGP slot, and other parameters are similar to PE.
GV and GL are simplified chipsets for integrated graphics cards, and AGP slots are not supported. Other parameters GV are the same as G, while GL decreases.
Compared with G, GE is an evolution chipset of integrated graphics card and also supports AGP slot.
There are two kinds of P, one is the enhanced version, such as 875P, and the other is the simplified version, such as 865P.
(2)9 15 series and higher versions
P is the mainstream version, without integrated graphics card, supporting the mainstream FSB and memory at that time, and supporting PCI-E X 16 slots.
Compared with P, PL is a simplified version, which has shrunk in FSB and memory, and has no integrated graphics card, but it also supports PCI-E X 16.
G is the mainstream integrated graphics chipset, supporting PCI-E X 16 slots, and other parameters are similar to P.
GV and GL are simplified chipsets of integrated graphics cards and do not support PCI-E X 16 slots. Other parameters GV are the same as G, while GL decreases.
Compared with XE, P is an enhanced version, which has no integrated graphics card and supports PCI-E X 16 slots.
(3) After 965 series
Starting from 965 series chipsets, Intel changed the naming method of chipsets, changed the letters representing chipset functions from suffixes to prefixes, and subdivided them according to different user groups, such as P965, G965, Q965 and Q963.
P is the mainstream chipset version for individual users, without integrated graphics card, supporting the mainstream FSB and memory at that time, and supporting PCI-E X 16 slots.
G is the mainstream integrated graphics chip set for individual users, supporting PCI-E X 16 slots, and other parameters are similar to P.
Q is an enterprise-class desktop chipset for business users, which has an integrated graphics card similar to G. Besides all the functions of G, it also has special functions for business users, such as active management technology.
In addition, in the case of the same function prefix, the performance is distinguished by the following numbers, and the smaller the number, the simpler the supported memory or FSB. For example, compared with Q965, Q963 only supports DDR2 667.
9. Interface between mouse and keyboard: PS/2 interface
PS/2 interface is the most common mouse and keyboard interface at present, which was originally patented by IBM. Small mouth? . This is a 6-pin circular interface. But the mouse only uses four pins to transmit data and power, and the other two pins are empty. PS/2 interface is slightly faster than COM interface, which is the standard interface of ATX motherboard, but it still can't make the high-end mouse perform well and doesn't support hot plug. In the BTX motherboard specification, this is also the interface that will be eliminated soon.
It should be noted that when connecting a mouse with a PS/2 interface, you can't insert the PS/2 interface of the keyboard by mistake (of course, you can't insert the PS/2 keyboard into the PS/2 interface of the mouse by mistake). Generally speaking, the mouse interface is green and the keyboard interface is purple for motherboards that meet the PC99 specification. In addition, judging from the relative position of PS/2 interface, the keyboard interface is close to the motherboard PCB, and the mouse interface is above it. (pictured)