CPU is the abbreviation of Central Processing Unit (CPU), which generally consists of logic operation unit, control unit and storage unit. The logic operation and control unit includes some registers, which are used to temporarily store data during the processing of data by the CPU. The main CPU indicators/parameters you need to pay attention to are:

1. Power frequency

The main frequency, that is, the clock frequency of CPU, is simply the working frequency of CPU. For example, we often say P4 (Bense) 1.8GHz, and this 1.8GHz( 1800MHz) is the main frequency of CPU. Generally speaking, the number of instructions completed in a clock cycle is fixed, so the higher the main frequency, the faster the CPU. Main frequency = external frequency x multiple.

In addition, it should be noted that AMD's Athlon XP series processors all have the nominal value of PR (performance level), such as Athlon XP 1700+, Athlon XP 1800+. Athlon XP, for example, actually runs at 1.53GHz and is labeled as 1800+. The system self-check screen also shows the system properties of Windows system, WCPUID and other testing software.

2. External frequency

The external frequency is the external clock frequency of the CPU. The main external frequencies of motherboard and CPU are 66MHz, 100MHz and 133MHz. In addition, the more external frequencies that the motherboard can adjust, the higher the better, especially overclocking.

3. Frequency doubling

Frequency doubling refers to the multiple of the difference between CPU external frequency and main frequency. For example, the CPU of Athlon XP 2000+ has an external frequency of 133MHz, so its frequency doubling is 12.5 times.

Step 4 connect

Interface refers to the interface between CPU and motherboard. There are two main types, one is card interface, called slot. The CPU of the card interface is inserted vertically into the motherboard like all kinds of expansion cards we often use, such as graphics cards and sound cards. Of course, there should also be corresponding slots on the motherboard, and the CPU of this interface has been eliminated at present. The other is the mainstream pin interface, called Socket. The CPU of Socket interface has hundreds of pins, which are called Socket370, Socket478, Socket462, Socket423 and so on.

cache

Cache refers to memory that can exchange data at high speed. It exchanges data with CPU before memory, so it is extremely fast, so it is also called cache. There are two kinds of cache related to the processor-L1cache, also called internal cache; And L2 cache, also known as external cache. For example, the Pentium 4“Willamette "core product adopts a 423-pin architecture, and has a 400MHz front-end bus, a 256KB full-speed second-level cache, an 8KB first-level tracking cache and an SSE2 instruction set.

Internal cache (L 1 cache)

That is, we often say that the first-level cache. The built-in cache of CPU can improve the running efficiency of CPU. The capacity and structure of the built-in L 1 cache have great influence on the performance of CPU. The larger the L 1 cache is, the fewer times the CPU exchanges data with L2 cache and memory with slow access speed, and the operation speed is faster than that of a computer. However, cache memories are all composed of static RAM, and their structures are very complicated. Under the condition that the CPU die area cannot be too large, the capacity of L 1 level cache cannot be too large, and the capacity unit of L 1 cache is generally KB.

External cache (L2 cache)

The external cache of CPU is relatively expensive, so the external cache of Pentium 4 Willamette is 256K, but the Celeron 4 with the same core is only128 K.

6. Multimedia instruction set

In order to improve the computer's application ability in multimedia and 3D graphics, many processor instruction sets came into being, among which the most famous three are Intel's MMX, SSE/SSE2 and AMD's 3D NOW! Instruction Set Theoretically, these instructions play a comprehensive role in strengthening many popular multimedia applications, such as image processing, floating-point operation, 3D operation, video processing and audio processing.

7. Manufacturing process

Early processors were all made with 0.5 micron technology. With the increase of CPU frequency, the original process can no longer meet the requirements of the product, so there are 0.35 micron and 0.25 micron processes. The finer the manufacturing process, the more electronic components are integrated per unit volume. At present, processor products made of 0. 18 micron and 0. 13 micron are the mainstream in the market. For example, Northwood core P4 adopts 0. 13 micron manufacturing process. In 2003, the CPU manufacturing technology of Intel and AMD will reach 0.09 mm. ..

8. Voltage (Vcore)

The working voltage of CPU refers to the voltage required for the normal operation of CPU, which is related to the manufacturing process and the number of integrated transistors. The lower the normal working voltage, the lower the power consumption and the less heat generation. The development direction of CPU is to continuously reduce the voltage required for normal operation on the basis of ensuring performance. For example, the working voltage of the old Athlon XP is 1.75v, while that of the new Athlon XP is1.65v..

9. Packaging form

The so-called CPU packaging is the last process in the CPU production process. Packaging is a protective measure to cure CPU chips or CPU modules with specific materials to prevent damage. In general, CPU can only be delivered to users after packaging. The packaging mode of CPU depends on the installation form of CPU and the integrated design of devices. Generally speaking, the CPU installed in Socket socket is usually packaged in PGA (Grid Array), while the CPU installed in Slot x slot is all packaged in SEC (Single-sided Plug-in Box). Now there are packaging technologies, such as PLGA (Plastic Grid Array) and Olga (Organic Grid Array). Due to the increasingly fierce market competition, the current development direction of CPU packaging technology is mainly to save costs.

10. Integer units and floating-point units

Alu-arithmetic logic unit, which is what we call "integer" unit. Mathematical operations such as addition, subtraction, multiplication and division, and logical operations such as "OR, AND, ASL and ROL" are all performed in the logical operation unit. In most software programs, these operations account for most of the program code.

FPU (floating point unit) is mainly responsible for floating point operation and high precision integer operation. Some fpu also have the function of vector operation, while others have special vector processing units.

Integer processing ability is the most important embodiment of CPU operation speed, but floating-point operation ability is an important index related to CPU multimedia and 3D graphics processing, so for modern CPU, the strength of floating-point unit operation ability can better reflect CPU performance.

Second, look at the parameters to understand the motherboard.

Motherboard is a common platform for all computer accessories, so when you buy or use a motherboard, you must first know what the core functions of your motherboard are, what types of CPU, memory and graphics card are, and how many PCI devices you can support.

1. plate type

If the circuit board wants to be the main board on the computer, it needs to be made into different board types. Here is a brief introduction to the common motherboard types. AT board is a basic board with simple structure and low price. The standard size is 33.2cmX30.48cm. The AT motherboard needs to be used together with the AT chassis power supply. Baby AT is an improved motherboard with more reasonable structure and layout, which can support AT/ATX power supply, but it has declined because of the popularity of ATX architecture.

The atX board is like a large horizontal AT board, which is convenient for the fan of the ATX chassis to dissipate heat to the CPU, and many external ports on the board are integrated on the motherboard, unlike many COM ports and printing ports on the AT board, which are output by wiring. In addition, ATX also has a miniature ATX small board, which can support up to four expansion slots, reducing the size, power consumption and cost.

NLX board, which is favored by brand manufacturers, looks like a motherboard with a graphics card inserted. It consists of two parts: one part is an ISA substrate with logic control chips and basic input and output ports, and the other part is an additional board with slots such as AGP, PCI and ISA. It can be inserted into the special port of the substrate like a graphics card, which can increase space and facilitate disassembly.

2. Core

Motherboard chipset is the core of computer motherboards and represents the main technical characteristics of motherboards. With the adoption of different motherboard chipsets, various computer motherboards support different functions. For example, the motherboard uses Intel's i845D motherboard chipset. The main change of i845D motherboard chipset compared with its predecessor i845 is that it provides support for mainstream DDR memory. Its main features are introduced in the motherboard manual. "The I845D chipset is composed of i845D chip and ICH2 chip, which supports Pentium 4 processor of Socket478, 400MHz FSB (Front Side Bus), AGP4X, integrated AC97 sound effect and ATA 100 hard disk transmission specification.

3. Socket type

The CPU socket is where the processor is installed on the motherboard. The mainstream CPU sockets mainly include Socket370, Socket 478, Socket 423 and Socket A. Socket370 supports processors such as PIII, New Celeron and CYRIXIII. The early Pentium 4 processor used Socket 423, while the current mainstream Pentium 4 processor used Socket 478. Socket A(Socket462) supports AMD's processors such as Duron and Athlon. In addition, the CPU socket type is Socket7 socket, which supports Pentium/Pentium MMX and K6/K6-2 processors. SLOT 1 socket supporting PII or PIII and SLOTA socket used by AMD ATHLON.

4. Supported memory types

At present, there are two main types of memory used by people: 168 SDRAM and 184 DDR SDRAM. SDRAM memory, 168 line, bandwidth of 64 bits, working voltage of 3.3v, supporting different specifications such as PC66/100/150; The main feature of DDR memory is that it can use the rising edge and falling edge of clock pulse to transmit data, so it can double the speed of DRAM without increasing the working frequency.

At present, DDR memory has several specifications, including PC1600/pc2100/pc2700/pc3200. For example, a motherboard specification states that "two DDR memory slots with 184 pins are supported, which can support 2GB of memory capacity." This sentence shows that it does not support SDRAM with 168 line, and it has two DDR memory slots, which can insert two DDR memories. In addition, as can be seen from other DDR texts, this motherboard can only support DDR memory of PC 1600/PC2 100 specifications.

5. Supported AGP slot types

Graphics card slots such as AGP 1X(266Mbps), AGP2X(533Mbps), AGP4X( 1066Mbps), AGP Pro, AGP universal slot (1066Mbps) and AGP8X(2 133Mbps). For example, the AGPX graphics card slot can use the AGPX graphics card, but the AGPX graphics card on the AGPX graphics card slot cannot be used normally (note: there are also general AGP2X/4X slots).

So what kind of graphics card your motherboard supports is the key to your correct choice of graphics card. For example, a motherboard uses an AGP4X slot, so you can buy an AGP 1X/2X/4X graphics card for normal use.

Third, look at the parameters to understand the hard disk.

As we all know, hard disks on the market are mainly divided into IDE and SCSI. SCSI hard disk has the characteristics of high speed, large capacity and stable use, and is the vanguard of hard disk technology, but its price is too expensive, and it is mainly used in more professional occasions.

Although there is a technical gap between IDE hard disk and SCSI hard disk in Shang Tong, there is no doubt that the gap is getting smaller and smaller. The current IDE hard disk also has the characteristics of high speed, large capacity and low price, making it the first choice for home occasions.

IDE hard disks can be divided into 5.25, 3.5, 2.5, 1.8 inch hard disks according to their inner disk diameters. 2.3 and 1.8 inch hard disks are mainly used for notebook computers and other devices; Hard disks with diameters of 5.25 and 3.5 are mainly used on desktops, and now the most commonly used hard disks with diameters of 3.5 inches are on desktops.

1. Hard disk capacity

When we buy a hard disk, the first thing we ask is, how big is this hard disk? Answer: 40GB and 80GB refer to the capacity of the hard disk. Generally refers to the formatted capacity of the hard disk. The larger the hard disk capacity, the better.

Secondly, when selecting capacity, you can also give priority to the large products stored in each disk. The larger the storage per disk, the more advanced the technology, and the easier it is to control the cost. For example, for the same 40GB hard disk, if the storage of each disk is 10GB, it needs 4 disks with 8 heads, and if the storage of each disk rises to 20GB, it needs 2 disks with 4 heads. For the storage of each 40GB hard disk, only 1 disk and 2 magnetic heads are needed, which can save a lot of costs and improve the working stability of the hard disk.

2. The speed of the hard disk

This is also a matter of concern to everyone. It refers to the rotational speed of the spindle in the hard disk. At present, IDE hard disks on the market are mainly divided into 5400 rpm and 7200 rpm. In the case of similar capacity and price, you can give priority to 7200 rpm fast hard disk products.

3. Hard disk transfer rate

The transmission rate of hard disk is also one of the important parameters of hard disk. Mainly refers to the transmission rate of external and internal data of hard disk, and their units are Mb/s (megabits per second) or MB/s( 1MB=8Mb). The burst data transmission rate of hard disk is the burst data transmission rate of hard disk, which generally refers to the data interface speed of hard disk. At present, the hard disk transmission rate of ATA/66/ 100/ 133 interface can reach 66-133 MB/s.

The internal data transfer rate of hard disk refers to the maximum data transfer rate between the magnetic head and the hard disk cache. In this respect, the maximum internal data transfer rate of mainstream hard disks on the market can generally reach more than 350Mb/S, and the maximum internal data transfer rate of excellent hard disks can reach 500 MB/s.

4. Hard disk cache

The cache size of hard disk is also one of the important indicators of hard disk. The cache of hard disk refers to the high-speed memory inside the hard disk. At present, most cache types used in hard disks are SDRAM, but there are exceptions, such as using EDO DRAM. The larger the cache capacity, the better, which is directly related to the reading speed of the hard disk. At present, the cache capacity of hard disks is mostly 2M, and it is moving towards 8M. However, there are a few products with only 5 12K cache, so you should pay attention.

5. Hard disk head

There are two main types of magnetic heads used in hard disks: MR and GMR. GMR giant magnetoresistive head has begun to replace MR head as the mainstream of hard disk head.

MR magnetoresistive head adopts a head structure with separate reading and writing heads. It senses the signal amplitude through the change of resistance value, which is quite sensitive to the change of signal and improves the accuracy of reading data accordingly. Moreover, because the signal amplitude read by Mr magnetoresistive head has nothing to do with the track width, the track can be made very narrow, thus improving the density of the disk and making the capacity of the hard disk very large.

Compared with MR head, GMR head adopts materials with better magnetoresistance effect and multi-layer thin film structure, which is more sensitive than MR head, so it can achieve higher storage density. At present, the disk storage density of MR head can reach 3Gbit-5Gbit/in2 (per square inch per gigabit), and that of GMR head can reach more than 10 gbit-40 gbit/in2.

6. Hard disk seek time

Hard disk seek time is also one of the important parameters to understand hard disk. Mainly refers to average seek time, single track seek time, maximum seek time, average hard disk delay, etc. Are all in ms (milliseconds).

Average seek time of hard disk refers to the time required for the hard disk head to move to the track where the data is located. The smaller the number, the better. At present, the average seek time of IDE hard disk is mostly below 9 ms .. The seek time between tracks of hard disk refers to the time when the head moves from one track to another, and the shorter the time, the better.

The maximum seek time of hard disk refers to the total time it takes for the hard disk head to move from the beginning to finally find the required data block, and its value is as small as possible. The maximum seek time of mainstream IDE hard disks in the market is mostly within 20 ms As for the average waiting time of hard disks, it refers to the time when the head moves to the track where the data is located, and then waits for the required data block to continue rotating under the head. The smaller the number, the better. Four: About the graphics card: 1, chip 2, video memory frequency 3, video memory bit width 4, rendering pipeline 5, special edition N card has GTX (GT eXtreme) enhanced version, which becomes GT after frequency reduction or pipeline, and then continues to shrink into GS version five: About memory: performance parameters of memory: capacity, frequency, delay value.