Graphics card and video memory
Secondly, the performance of the monitor and sound card.
Interviewee: satriani 299- Scholar Level 2 4- 14 17:32.
There are two points to consider:
If you buy a computer, you should consider it comprehensively, mainly to ensure the expansibility of the motherboard, followed by CPU+ memory+hard disk (these three influences are relatively large). If you like to play games, you should add a graphics card, and then other accessories.
If you have bought a computer, the easiest way is to test it with software.
Responder: Feutrie-Trainee Magician Level 2 4- 14 17:49.
CPU, memory, graphics card and hard disk, whichever is better, must be compatible.
Responder: wheat breeder-trainee magician level 2 4- 14 17:56
Common performance indicators of computers
First, the main performance indicators of CPU
(1) main frequency, that is, the clock frequency of CPU, is simply the working 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. However, due to the different internal structures of various CPUs, the performance of CPU can not be completely summarized by the main frequency. As for the external frequency, it is the working frequency of the system bus. Frequency doubling refers to the multiple of the difference between CPU external frequency and main frequency. Expressed by the formula: main frequency = external frequency × frequency doubling. We usually say Celeron 433 and PIII 550 refer to the main frequency of CPU.
(2) The memory bus speed or total routing speed of the system is generally equivalent to the external frequency of the CPU.
The speed of memory bus is very important to the performance of the whole system. Because the development of memory speed lags behind that of CPU, in order to alleviate the bottleneck caused by memory, secondary cache appears to coordinate the difference between them. Memory bus speed refers to the working frequency between CPU and L2 cache and memory.
(3) Working voltage. Working voltage refers to the voltage required for the CPU to work normally.
In the early days, due to backward technology, the working voltage of CPU (386,486) was generally 5V. When it developed to Pentium 586, it was already 3.5V/3.3V/2.8V. With the improvement of CPU manufacturing technology and main frequency, the working voltage of CPU gradually decreased. Coppermine, the latest product produced by Intel, uses the working voltage of1.6V. Low voltage can solve the problems of excessive power consumption and excessive heat generation, which is especially important for notebook computers.
(4) coprocessor or mathematical coprocessor. CPU before 486 has no built-in coprocessor.
Because the main function of coprocessor is to be responsible for floating-point operation, the floating-point operation performance of microcomputer CPU such as 386, 286 and 8088 is quite backward. Since 486, CPU generally has a built-in coprocessor, and the function of the coprocessor is no longer limited to enhancing floating-point operations. Now the floating-point unit (coprocessor) of CPU often optimizes multimedia instructions. For example, Intel's MMX technology, MMX is the abbreviation of "Multimedia Extensions Instruction Set". MMX is a new technology adopted by Intel in 1996, which is used to enhance the application of Pentium CPU in audio, video, graphics and communication. The CPU has added 57 MMX instructions, and its ability to handle multimedia has improved by about 60%.
(5) Pipeline technology, excessive. Pipeline was first used by Intel in 486 chips.
An assembly line works just like an assembly line in industrial production. In CPU, an instruction processing pipeline consists of 5~6 circuit units with different functions, and then an X86 instruction is divided into 5~6 steps, which are executed by these circuit units respectively, so that an instruction can be completed within a CPU clock cycle, thus improving the running speed of CPU. Super pipeline means that the pipeline inside a CPU exceeds the usual 5~6 steps. For example, the pipeline of Pentium pro is as long as 14 steps. The more steps (stages) of pipeline design, the faster the completion of an instruction, so it can adapt to CPU with higher working frequency. Overrange means that the CPU can execute multiple instructions in one clock cycle. This is hard to imagine on the CPU before 486, and only the CPU above Pentium level has this superscalar structure; This is because more and more modern CPU adopts RISC technology, so it will exceed the standard CPU.
(6) Out-of-order execution and branch prediction. Out-of-order execution refers to the technology adopted by CPU to allow multiple instructions to be developed and sent to the corresponding circuit units for processing out of the order specified in the program.
A branch refers to a node that needs to be changed when a program is running. There are unconditional branches and conditional branches, in which unconditional branches only need to be executed by CPU in the order of instructions, and conditional branches must decide whether to change the running direction of the program according to the processing results, so it is conditional branches that need "branch prediction" technology to deal with.
(7)L 1 cache, which is what we often call level 1 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. However, cache memories are all composed of static RAM, and their structures are complex. Under the condition that the die area of CPU can't be too large, the capacity of L 1 level cache can't be too large. Cache with write-back structure. It can provide a cache for read and write operations. However, a cache with a write-through structure is only valid for read operations. Write-back cache is basically used on computers above 486.
(8)L2 cache refers to the cache outside the CPU.
L2 of Pentium Pro processor runs at the same frequency as CPU, but it is expensive, so the running frequency of Pentium II is half that of CPU, and its capacity is 5 12K. In order to reduce the cost, Intel once produced a CPU without L2 called Celeron.
(9) manufacturing process.
The manufacturing technology of Pentium CPU is 0.35 micron, PII and Celeron can reach 0.25 micron, and the latest CPU manufacturing technology can reach 0. 18 micron. Copper wire technology will be adopted, which can greatly improve the integration and working frequency of CPU.
Second, the main performance indicators of memory
Memory has a great influence on the performance of the whole machine. Many indicators are related to memory, and there are many performance indicators of memory itself. Therefore, here are only a few of the most commonly used and important indicators.
(1) speed.
Memory speed is generally used as a performance indicator of the time (in ns) required to access data once. The shorter the time, the faster the speed. Only when the memory matches the speed of the motherboard and CPU can the maximum efficiency of the computer be exerted, otherwise it will affect the full play of the high-speed performance of the CPU. The memory speed of FPM can only reach 70 ~ 80 ns, that of Edo can reach 60ns, and that of SDRAM has reached 7ns at the highest.
The speed index of the memory is usually printed on the chip in some form. Generally, the words -60, -70,-10 and -7 are printed on the back of the chip model, indicating that the access speed is 60ns, 70ns, 10ns and 7ns. The conversion relationship between ns and MHz is as follows:
1 ns = 1000 MHz 6 ns = 166 MHz 7 ns = 143 MHz 10 ns = 100 MHz
(2) capacity.
Memory is the main part of a computer, as opposed to external memory. Windows system, typing software, game software, etc. , generally installed on the hard disk and other external storage, must be adjusted to run in memory to use. For example, entering a paragraph of text or playing a game is actually done in memory. Usually, a large amount of data to be permanently stored is stored in external memory, while some temporary or small amount of data and programs are stored in internal memory. The more memory capacity, the better, but limited by the maximum capacity supported by the motherboard, even for the current mainstream computers, this limitation is still an obstacle. The capacity of a single memory is usually 128MB, 256MB, and the maximum is 5 12MB. In the early days, there were 64MB, 32MB, 16MB and other products.
(3) Parity check of memory.
In order to check whether the memory is accurate during access, every 8-bit capacity is equipped with 1 bit as parity bit, and the parity circuit of the motherboard is used to check the access data correctly, so an extra chip needs to be added to the memory stick. However, in actual use, parity has no effect on system performance, so most memory chips are no longer equipped with parity chips.
(4) Memory voltage.
Both FPM memory and EDO memory use 5V, SDRAM uses 3.3V and DDR uses 2.5V. Pay attention to the jumper on the motherboard when using.
(5) Data width and bandwidth.
The data width of the memory refers to the number of bits of data transmitted by the memory at the same time, and the unit is bits; The bandwidth of memory refers to the data transmission rate of memory.
(6) The number of rows in the memory.
The number of lines in the memory refers to the number of contact points when the memory stick contacts the motherboard. These contacts are golden fingers, including 72 lines, 168 lines and 184 lines. The data widths of 72 lines, 168 lines and 184 lines are 8 bits, 32 bits and 64 bits respectively.
(7) Chemical Abstracts Service
CAS delay refers to the time from the valid read command (issued at the rising edge of the clock) to the time when the output can provide data, which is generally 2 or 3 clock cycles, which determines the performance of the memory. At the same operating frequency, the chip with CAS delay of 2 is faster and better than the chip with CAS delay of 3.
(8) Rated usable frequency (GUF)
Reduce the maximum frequency given by the manufacturer, and the value obtained in this way is called the rated available frequency GUF. For example, if the maximum available frequency of an 8ns memory chip is 125MHz, the rated available frequency (GUF) should be 1 12MHz. The maximum available frequency keeps a certain margin with the rated available frequency (front-end system bus working frequency), which can ensure the stable operation of the system to the maximum extent.
Third, the main performance indicators of the graphics card
The main performance indicators of the graphics card include the following aspects:
(1) Refresh frequency: refers to the speed at which the image is updated on the screen, that is, the number of times the whole picture is displayed on the screen every second, in Hz. The flicker caused by the refresh frequency above 75Hz is generally not easy to be detected by human eyes, therefore, in order to protect eyes, it is best to adjust the refresh frequency of the display to above 75Hz. However, not all graphics cards can achieve a refresh rate of more than 75Hz at the maximum resolution green (this performance depends on the speed of RAM-DAC on the graphics card), and the display may not meet the requirements because of insufficient bandwidth. Some low-end graphics cards can only set the refresh frequency to 60Hz at high resolution.
(2) Color number of bits (color depth): The color of each pixel in the graph is described by a set of binary trees, and the length (number of bits) of this set of binary numbers describing color information is called color number of bits. The more color bits, the richer the color of the displayed graphics. Usually the standard VGA display mode is 8-bit display mode, that is, 256 colors can be displayed in this mode; The enhanced color (16 bit) can display 65,536 colors, also called 64K color; 24-bit true color can display16.77 million colors, also called 16M color. In this mode, you can see that there is no difference between the color of a true color image and a high-definition photo. In addition, there are 32-bit, 36-bit and 42-bit color bit trees.
(3) Resa resolution: refers to the product of horizontal pixels and vertical pixels that make up an image (the image is displayed on the display screen). The higher the display resolution, the more image pixels displayed on the screen, and the clearer the image display. The display resolution is closely related to the monitor and graphics card.
The display resolution is usually expressed as "horizontal points× vertical points", such as 1024×768. Maximum resolution refers to the highest resolution that a graphics card or monitor can display. At the highest resolution, the luminous point of the display corresponds to one pixel. If the set resolution of the display is lower than the highest resolution of the display, the pixel can be composed of multiple light-emitting points.
(4) Memory capacity: The higher the resolution supported by the graphics card, the more memory installed, the stronger the function of the graphics card, but the higher the price.
Four, the main performance indicators of CRT display
(1) The size of the picture tube is what we usually call 14, 15 and 17 inches. Note that the length mentioned here refers to the diagonal length of the display screen in inches (1 inch =25.4 mm). Although monitors usually use indicators such as 15 inch and 17 inch to measure the screen size, in fact, their display sizes are different. The maximum visible image size (VIS) depends on the available display size of the CRT and the opening size of the front panel of the display. Generally, the VIS of 15 inch CRT is about 13.8 ~ 14 inch, and that of 17 inch CRT is about 15.5 ~ 16 inch. Therefore, when choosing the size of the display, we should also pay attention to its nominal maximum display area.
(2) Point spacing: Point spacing (or stripe spacing) is a very important hardware indicator of the display. The so-called point distance refers to the distance between the luminous point of a given color and the nearest adjacent luminous point of the same color. This distance cannot be changed by software, which is different from the resolution. At any same resolution, the smaller the point distance, the clearer and finer the displayed image, and the higher the resolution and image quality. At present, most household monitors adopt a pitch of 0.28mm, such as 0.25mm from Sony's Teletron and Mitsubishi's Diamondtron, 0.26mm from BenQ and some Philips, and 0.27mm, such as Samsung 750S, which can fully meet the needs of various industries. For ordinary users, you can consider a display with a point distance below 0.28 mm.
(3) Resolution: Resolution refers to the sum of pixels that make up an image, that is, how many pixels the screen contains. Generally expressed as the product of horizontal resolution (the number of pixels in a scanning line) and vertical resolution (the number of scanning lines). For example, 1024×768 means that the horizontal direction can contain at most 1024 pixels, and the vertical direction is 768 pixels. The total number of pixels on the screen is their product. The higher the resolution, the more pixels the picture contains, and the finer and clearer the image is. The resolution of the display is affected by the display size, the distance between picture tubes and the circuit characteristics.