I. CPU Terminology Explanation 3DNOW !: (3D NO WAITI) AMD Development SIMD instruction set, can enhance the speed of floating point and multimedia operation, and its instruction is 21. Alu: (Arithmetic Logic Unit, Arithmetic Logic Unit) In the processor is used to calculate, with its own data transmission unit and branch unit. BGA: (Ball Grid Array, Spherical Matrix) A chip package form, Example: 82443bx. BHT: (Branch Prediction Table, Branch Prediction Table) The processor is used to determine a numerical table in the direction of the branch action. BPU: (Branch Processing Unit, Branch Processing Unit) The area used to make branch processing in the CPU. BRACH PEDICTION: (Branch Prediction) A advanced data processing method starting from the P5 era, determines the direction of the program branch, and can operate faster than the CPU. CMOS: (Complementary Metal Oxide Semiconductor, Complementary Metal Oxide Semiconductor) It is a special chip, the most common use is the BIOS (Basic Input / Output System, Basic Input / Output System). CISC: (Complex Instruction Set Computing, Complex Instruction Set Computer) For longer instructions, called complex instructions relative to RISC Such as: x86 instruction length is 87 bits. COB: (Cache On Board, the integrated cache) The integrated cache is integrated on the processor card, usually refers to a secondary cache, an example: Pentium II COD: (Cache on Die, the chip integrated cache) inside the processor chip Integrated cache, usually refers to a second-level cache, an example: PGA Celeron 370 CPGA: (Ceramic Pin Grid Array, Ceramic Needle Raster Array) A chip package form. CPU: (Center Processing Unit, Central Processor) The brain of computer system, is used to control and manage the operation of the entire machine, and perform computing tasks. Data Forwarding: (Data Pre-Administration) CPU Copys the output value of one unit to the input value of the output value of one unit in one clock cycle. Decode: (Instruction Decoding) Due to the inconsistency of the X86 instruction, "translation" must be used in a unit, and the real kernel works by post-translational requirements. EC: (Embedded Controller, Embedded Controller) In a set of specific systems, the control device that has been added to a fixed position, and the control device that completes a certain task is called an embedded controller. Embedded Chips: (Embedded) A special CPU, usually in a non-computer system, such as a household appliance.
EPIC: (Explicitly Parallel Instruction Code, Parallel Instruction Code) Intel's 64-bit chip architecture, itself cannot perform x86 instructions, but can be compatible with the old X86 instructions through the decoder, but the operation speed is more than the real 32-bit chip. decline. Fadd: (Floationg Point Addition, floating point) FCPGA (Flip Chip Pin Grid Array) A chip package form, an example: Pentium III 370. FDIV: (Floationg Point Divide, floating point) FEMMS (Fast Entry / EXIT MULTIMEDIA State, fast entry / exiting multimedia State) In most Pentium, MMX and floating point units cannot run at the same time. The new chip speeds up the switch between the two, which is femms. FFT: (Fast Fourier Transform) A complex algorithm that tests the floating point capability of the CPU. FID: (FID: FREQUENCY IDENTIFY, frequency identification number) Pentium III to check the CPU frequency through the ID number, can effectively prevent Remark. FIFO: (First Input First Output, first-out queue) This is a traditional sequential execution method, first entering the instructions to complete and retract, followed to execute the second instruction. Flop: (Floating Point Operations Per Second, floating point * / second) calculates a unit of CPU floating point capability. Fmul: (Floationg Point Multiplication, floating point) FPU: (Float Point Unit, floating point operation unit) FPU is a processor dedicated to floating-point operations, the previous FPU is a separate chip, after 486, Intel puts FPU With integrated in the CPU. FSUB: (Floationg Point Subtraction, floating point) HL-PBGA: (Surface adhesive, high heat resistant, lightweight plastic spherical matrix package) A chip package form. IA: (Intel Architecture, Intel Architecture) X86 chip structure developed by Intel. ID: (Identify, Identification Number) is used to determine the identification code of different chips. IMM: (Intel Mobile Module, Intel Mobile Module) Intel Develops processor modules for laptops, integrated CPUs and other control devices. Instructions Cache: (instruction cache) Since the speed of the system is slower, when the CPU reads the instruction, the CPU will stop waiting for the memory transmission. The instruction cache adds a fast storage area between the main memory and the CPU, even if the CPU does not require an instruction, the main memory will automatically send the instruction to the instruction cache. When the CPU requires the instruction, you can directly from the instruction cache. Read, no need to access the main memory, reduce the Waiting time of the CPU.
Instruction Coloring: (instruction classification) A technique for manufacturing predicting execution instructions, once predicts that it is determined that the corresponding instruction decision is predicted, the processor will process the same judgment of the same intersection. INSTRUCTION ISSUE: (Instruction Send) It is the first CPU pipe for receiving instructions sent by memory and sent it to the execution unit. IPC (Instructions Per Clock Cycle, Instruction / Clock Cycle) Indicates the number of instructions that can be completed in one clock cycle. KNI: (Katmai New Instructions, Katmai * 碨 SE) Latency (latency) It is more difficult to understand its meaning from the literal, in fact, it indicates that the clock cycle required to complete an instruction, the less the incubation period The better. Strictly speaking, the incubation period includes an instruction from receiving the entire process of sending. Most X86 instructions today require approximately 5 clock cycles, but some of these cycles are overlapping with other instructions (parallel processing), so the incubation period of CPU manufacturers is longer than the actual time. LDT: (Lightning Data Transport, Lightning Data Transfer Bus) K8 Novel Data Bus, Froquency Foundation 200MHz. MMX: (MultiMedia Extensions, Multimedia Extensions) Intel Development The first SIMD instruction set can enhance the speed of floating point and multimedia operations. Mflops: (Million floationg point / second, millions of floating point per second) calculates a unit of CPU floating point capabilities, based on millions of instructions. Ni: (Non-Intel, Non-Intel Architecture) In addition to Intel, there are many other manufacturers who produce X86 systems. Due to patents, their products and Intel are different, but they can still run X86 instructions. Olga: (Organic Land Grid Array, Substrate Grid Array) A chip package form. OOO: (Out of Order) One of the properties of the POST-RISC chip, it is possible to complete the calculation task in order to provide in the order provided, is an architecture that speeds up the processor calculation speed. PGA: (PIN-Grid Array, Pin Grid Array) A chip package form, a disadvantage is that power consumption is large. POST-RISC: A new type of processor architecture, its kernel is RISC, and the periphery is CISC, combined with the advantages of two architectures, with advanced features such as predictive execution, processor reunification, such as: Athlon. PSN: (Processor Serial NumBers, processor serial number) Identifies a set of numbers of processor features, including clocked, production date, production numbers, etc. PIB: (Processor In A Box, Boxing Processor) CPU Manufacturer is officially launched on the market, usually more than the OEM (Original Equipment Manufacturer, original equipment manufacturer) vendors to expensive, but only PIB has The manufacturer's official warranty rights.
PPGA: (Plastic PIN Grid Array, Plastic Needle Matrix Package) A chip package form, the disadvantage is that the power consumption is large. PQFP: (Plastic Quad Flat Package, Plastic Square Planic Pack) A chip package form. RAW: (Read after Write, Write) This is an error in the CPU's chart, that is, before the necessary conditions are not established, the final result is wrong. Register Contention: (Preemption Register) When the last write backward task of the register is not completed, another instruction exists conflicts when this register. Register Pressure: The number of registers required when the software algorithm is executed is restricted. For X86 processors, the register has become the biggest feature of its maximum, so AMD wants to increase the number of registers in the next generation chip K8. Register Renaming: (Register Rename) relocates the output value of an instruction to an arbitrary internal register. In the X86 architecture, this type of situation often occurs, such as: When a FLD or FXCH or MOV command requires the same target register, it is necessary to use it to the register rename. Remark: (Chip Frequency Heavy Identification) Chip Manufacturers set most of the CPUs to freely adjust multiplier and frequency, which is selected in the same batch of CPUs in the same batch of CPUs. Level 1, low performance, is completed in the factory, and is a legal frequency positioning method. But after the factory, the dealer passed the low-end CPU overclock, and the new label was posted, and the illegal frequency positioning as a high-end CPU was called Remark. Because manufacturers have power to change their products, the dealers do this is copyright infringement, don't think that only software has copyright, hardware has copyright. Resource Contention: (Resource Conflict) When an instruction requires a register or pipe, they are used by other instructions, and the processor cannot respond to instantly, which is the resource conflict. Retirement: (Instruction Remove) When the processor performs a command, it automatically removes it from the schedule process. If only the instruction is completed, but still stay in the scheduling process, it is not an instruction to retreat. Risc: (Reduced Instruction Set Computing, Reximel Instruction Set Computer) A computer with a shorter length, which is running fast than CISC. Sec: (Single Edge Connector, Unilateral Connector) A module of a processor, such as Pentium II. SIMD: (Single Instruction Multiple Data) can copy multiple * work and package them in a set of instructions, examples: 3DNow!, SSE. SiO2F: (Fluorided Silicon Oxide, Dioxide) Making the materials for electronic components need to be used.
SOI: (Silicon On Insulator, Insulator Wili) SONC (System ON A CHIP, System Integration Chip) Integrates a variety of features in a processor, such as Cyrix Mediagx. SPEC: (System Performance Evaluation Corporation, System Performance Evaluation Test) Benchmark, overall performance of the system. Speculative Execution: (predicting) A area for performing unlimited instruction streams. When the branch instruction is issued, the conventional processor cannot do any work before the correct feedback is not received, and a new processor with predicted execution capability can estimate the upcoming instructions, using a pre-calculated method to speed up. The entire process. SQRT: (Square root calculation, square root calculation) a complex operation, can test the floating point capability of the CPU. SSE: (Streaming SIMD EXTENSIS, single instruction Multi-data stream extension) Intel's second-generation SIMD instruction set, 70 instructions, can enhance the speed of floating point and multimedia operations. SuperScalar: (Overview Architecture) The processor architecture of multiple instruction streams can be performed at the same clock cycle. TCP: (Tape Carrier Package, Film Package) A chip package form is characterized by a small fever. Throughput: (throughput) It includes two meanings: The first type: the minimum number of clock cycles required for executing a command, the better. The faster the speed is executed, the less the next instruction and it seizes the resources. The second: can perform the maximum number of instructions in a certain period of time, and the bigger the better. TLBS: (Translate Look Side Buffers, translation bypass buffer) is used to store instructions and input / output values. VALU: (Vector Arithmetic Logic Unit, vector arithmetic logic unit) is used in the processor for the vector calculation. VLIW: (Very long instruction word) A very long instruction combination, which connects many instructions together to increase the speed of the operation. VPU: (Vector Permutate Unit, vector arrangement unit) The part of the data is used in the processor. Second, memory terminology Bank: Bank refers to the calculation unit of the memory slot (also known as a memory), which is the basic operation unit of the computer system and the memory data exchange. Memory speed: The speed of memory is calculated by processing time consuming with memory between each CPU and the bus cycle (bus cycle) in nanosecond (NS). Memory Module: It is mentioned that the memory module refers to a plurality of memory chip chips on a printed circuit board, and this memory chip is usually a DRAM chip, but recently system design also uses it to hide the chip inlaid. The memory module is a private slot (slot) mounted on the MODULE (CHIPS) to the number of DRAM chip (CHIPs) on the MODULE, which is the main factor in the design of the DRAM chip (CHIPS) on the MODULE. .
SIMM (SINGLE IN-line memory module): The number of memory ICs on the board is not equal to the following 2 types: 72PIN: 72 Foots Single-sided memory module is used to support 32-bit data processing the amount. The single-sided memory module of 30PIN: 30 foot is used to support 8-bit data processing amount. DUAL in-line memory module: (168pin) is used to support 64 bits or a wider bus, and only use 3.3 volts, usually used in a 64-bit table-on computer or server. RIMM: The RIMM module is one of the main specifications of the next generation of memory modules, which is Intel launched the memory module supported by the chipset in 1999 with a bandwidth up to 1.6GBYTE / Sec. SO-DIMM (144PIN): This is a modified DIMM module, which is smaller than the general DIMM module, applied to notebook computers, listing machines, fax machines or various Terminal, etc. PLL: For the lock circuit, used to unify the integrated pulse signal, so that the memory can access the correct access information. Rambus Memory Module (184PIN): The memory module of Direct Rdram is called RIMM module. The module has 184pin feet, the data output is serial, with the current DIMM module 168pin, and there is a large architecture. difference. 6 layers and 4 layers (6 Layers vs 4 Layers): Refers to the circuit printing board PCB Printed Circuit Board is made of 6 or 4 layers of glass fibers, usually SDRAM uses 6-layer plate, although it will increase PCB Cost but exempt noise interference, while 4 layers can reduce the cost of PCB but poor efficiency. Register: It means the cache, its function is to achieve the purpose of synchronization at high speed. SPD: The abbreviation for serial presence detect, it is a code that burned in the EEPROM. When the boot is turned on, the BIOS must detect Memory, but there is no need to detect the action, but directly read the SPD directly by the BIOS. Related information about memory. Comparison of Parity and ECC: Parity Check Codes is widely used on an error detection code, and they add a check bits to each data, and can detect Test a character in a character (even) is wrong, but Parity has a disadvantage that when the computer finds an error, it is not possible to determine which bit of the error is not correct. Buffer and Buffer VS UnBuffer: The buffer DIMM is a method for improving the timing problem. There is no buffer DIMM, although it can be designed for the system, but it can only support four DIMM. If a buffer's DIMM is used for a host board with a speed of 100 MHz, there will be an adverse effect. The DIMM with a buffer can use four memory, but if the buffer speed is not fast enough, it will affect its performance. In other words, there is a buffer's DIMM that has slowed down, but it can support more DIMM usage.
Self-Refresh: DRAM has independent and built-in charging circuitry in a certain period of time, usually used in a computer that is highly saved in a notebook computer or a portable computer. Precharge time (CAS LATENCY): Generally referred to as CL. For example, CL = 3, when the computer system reads the first data autonomous memory, the desired preparation time is 3 external chassis (System Clock). The difference between CL2 and CL3 is only necessary for the first reading data, and there is no significant effect on the performance of the entire system. Clock Signal (CLOCK): The clock signal is provided to the synchronous memory signal synchronization, and the access action of the synchronous memory must be synchronized with the clock signal. Electronic Engineering Design Development Joint Conference (JEDEC): Most of JEDEC is engaged in design, inventive manufacturing, inventive, in accordance with a group consortium composed of computer memory modules, most of the memory products produced by the general industrial, mostly established by JEDEC As an evaluation. Read-only memory ROM (ROM): ROM is only read and cannot be written to the information, because this special is the most common is the BIOS on the motherboard (basic input / output system Basic Input / Output system) Because BISO is the basic hardware setting for computer boot, it is used as a low-order communication interface with the periphery, so the BISO is burned in the ROM to avoid freely clearance. EEPROM (Electrically Erasable Programmable ROM): For a case of writing data, even if the power is turned off, it is necessary to reserve a quite long time, and if the write data does not need to increase the voltage, as long as some handle is written You can write the information into memory. ERASABLE Programmable ROM: After clearing the internal data that can be removed through the ultraviolet rays, the data can be burn into the EPROM, which is the advantage to be repeated burning information. . The read-only memory (PROM): is a memory memory, because it can only be written, so once it is written, if there is an error, it is unable to change and can no longer save other information, so As long as the memory is not available, it cannot be recycled. Mask Rom: It is the manufacturer to produce a ROM or EPROM in advance as a sample, and then produce a large number of ROM samples, this is Mask ROM, The information burning in the Mask ROM will never be modified. Random access memory RAM (Random Access Memory): RAM is a memory that can be read and written. We also read data from RAM when writing data to RAM memory, which is different from ROM memory. However, the RAM must be stabilized by stable and smooth power, so once the power is turned off, the data in the RAM will disappear. Dynamic Random Access Memory: DRAM is the abbreviation of Dynamic Random Access Memory, is usually main memory in the computer. It is a storage action with a capacitor, but due to the problem of electricity itself, The information must be continuously accessible to other information.
FPM DRAM (Fast Page Mode DRAM): is a modified DRAM, most of the 72IPN or 30PIN module, FPM is composed of memory inside the memory Pages, from 512 Bite to several kilobytes, it is characterized by waiting When you re-read, you can read the information within each of the PAGE. EDO DRAM (Extended Data Out DRAM): The access speed of EDO is about 10% faster than traditional DRAM, which is more than 72PIN, 168PIN, which is 12 to 30 times faster than FPM. SDRAM: SYNCHRONOUS DRAM is a new technique for DRAM architecture; it uses Clocks in the wafer to make the input and output can be synchronized. The so-called clock synchronization refers to the time pulse of the memory time and the CPU to synchronize access. SDRAM saves the time of executing instructions and data transmission, so that computer efficiency can be enhanced. DDR: DDR is a higher speed synchronous memory, DDR SDRAM is a DIMM module of 168PIN, which is faster than SDRAM, and DDR is designed to be a higher speed requirement such as servers, workstations, and data transmission. DDRII (Double Data Rate Synchronous Dram): DDRII is the future new standard after DDR's original SLDRAM alliance will be integrated with existing R & D results with DDR after 1199. DDRII's detailed specifications are currently not determined. DRDRAM (Direct Rambus DRAM): is one of the next-generation mainstream memory standards, which is developed by Rambus, which is connected to a common bus, which can not only reduce the volume of the controller. Increase the efficiency of data transfer. RDRAM (Rambus DRAM): It is done independently by Rambus. Its speed is about 10 times more than DRAM, although there is such a strong performance, but the memory controller needs considerable changes, so this Most of memory uses on the game machine or a professional graphics acceleration card. Vram (Video RAM): The biggest difference from DRAM is that it has two sets of output and input ports, so it can be read while reading, while outputting data. WRAM (Window Ram): It belongs to the VRAM's improvement version, which is different from its control lines with an input / output controller of twenty groups, and uses EDO's data access mode. MDRAM (MULTI-BANK RAM): The inner part of MIDRAM is divided into several different small storage libraries, which is made of several college small unit matrices. Each repository is connected to each other than the external data speed, which is applied to a high speed display card or acceleration card. Static Random Handling Memory SRAM (Static Random Access Memory): SRAM is the abbreviation of Static Random Access Memory, usually more stable than the general dynamic random processing memory processing speed. The so-called static meaning means that memory information can be accommodated at any time without Semi-random processing memory is usually used to make caches because of this feature. Async SRAM: For asynchronous SRAM This is a more old SRAM, which is usually used on Level 2 Cache on your computer, which is independent of the computer's systematic time pulse. Sync SRAM: For synchronous SRAM, its working clock and the system are synchronized.
SGRAM (Synchronous Graphics RAM): It is improved by SDRAM to block blocks, individually retrieved or modify access, and reduces the number of memory overall read and writes to increase the drawing controller. Cache RAM: A high speed memory is designed to process the operating CPU. The memory of the memory is to use SRAM's particles to do memory. Due to the different connection mode, it can be divided into an external method (External) another is an intersystem. The external method is to place the memory on the host board, also known as Level 1 Cache and the connection method is to refer to the memory in the CPU is called Level 2 Cache. PCMCIA (Personal Computer Memory Card International Association): is a standard card-type expansion interface, mostly used on notebooks or other peripheral products, which can be divided into: Type 1: 3.3mm thickness, often made SRAM , Flash RAM's memory card and the DRAM memory card used by the most recent printer. TYPE 2: 5.5mm thickness, usually designed to have the modem interface (MODEM) used by the note computer. Type 3: 10.5mm thickness, is used to connect the ATA interface to the hard disk. TYPE 4: Small PCMCIA card, mostly used in digital cameras. Flash: Flash memory comparison is a storage device because the information stored in the flash memory is not lost after the power is turned off, and the original information must be cleared before writing the information before it can be written. New information, the shortcomings are too slow to write to the information. Remnant Memory Module: Many merchants in the memory market will sell re-indicated memory modules, the so-called re-indicated memory module is changed to the label on the chip CHIP, making it displayed The message provides more profits to providing merchants. Generally speaking, the industry will indicate faster speeds (-7 to change to -6) or will have no label to change to have a factory. To avoid buying this product, the best way is to purchase a top-level chip manufacturer product to a good reputation. REFRESH: The main memory is a combination of DRAM, and its capacitance needs to be constantly charged to maintain the correct information. Generally there are 2K and 4K refresh classification, and 2K is more fast REFRESH but 2K is 3K to 4K. Third, the main board term interpretation chipset: The chipset is the soul of the motherboard, which determines the functions that the motherboard can support. At present, the common chipches in the market have products such as Intel, VIA, SIS, ALI, AMD and other companies. Among them, Intel's mainstream products are 440bx, i820, i815 / 815e, etc. Via mainly has a chipset from VIA Apollo Pro 133 / 133A, KT 133. SIS is mainly SIS 630 chipset. Ali mainly has an Ali ALADDIN TNT2 chipset, and AMD has AMD 750 chipset. Among them, in addition to Intel's I820, I815 / 815E chipset, all chipsets are composed of two chips: the piece of the CPU is called North Bridge chip, mainly responsible for controlling CPU, memory and display; close to PCI slot The piece called South Bridge, mainly responsible for controlling input and output (such as support for hard disk UDMA / 66/200 mode), soft and sound, etc. And Intel's I820, I815 / 815E chipset uses a new structure, consisting of three chips.
They are MCH (Memory Controller Hub, which is similar to the North Bridge), ICH (I / O Controller Hub, function is similar to South Bridge), FWH (FireWare Hub, function is similar to the BIOS chip). Since the new chipset uses a special bus (generally referred to as accelerator hub structure AHA, Acplerated Hub Architecture) to connect to the main board, rather than using the PCI bus using the PCI bus, so there is a relatively larger Efficiency improvement. CPU interface: Due to most CPUs in the market are Intel and AMD products, only Socket 370 (support Intel new Celeri and Coppermine "copper" processor), Slot 1 (support Intel) Celeron and old PIII processor can also add a connection card to support the Socket 370 processor), Slot A (support AMD Athlon processor), Socket A (supporting AMD New Athlon and Duron processor), etc. Several interfaces. Uncount between different interfaces (only the SLOT 1 interface can be accepted to support the Socket 370 processor). Everyone should recognize it when you purchase. New practical technique: a. Soft jumper technology: The so-called jumper is a set of open-off switches, through the different combinations of communication, break, to achieve the purpose of adjusting the CPU frequency or implement some other functions (such as adjusting the voltage). The previous jumpers are generally consisting of a set of metal pins or dialed switches. Since the classic soft jumper technology of the promotion company SoftMenu, many manufacturers have also added this feature to set the CPU frequency and voltage directly in the BIOS. However, due to the more serious viruses such as CiH, some companies have retained functions such as hard jumpers such as DIP switches. B. New BIOS upgrade technology: Previous BIOS upgrade is considered a "master" patent. Therefore, it has certain risks, so ordinary users don't dare to get involved easily. But some vendors have developed some special BIOS upgrade features, making the BIOS upgrade will never be as dangerous and mysterious as before. For example, MSI new 815 motherboard can be upgraded directly on the Internet, as long as you connect to the network, the system will automatically detect your BIOS version. If you find that the product you use has a new BIOS file, you will download and update it automatically. Reduce the user's * work. Make the BIOS simpler. C. Energy saving function: The current energy-saving function mainly has two types of STD and STR. STD (SUSPEND TO DISK), hangs to the hard disk, refers to the system saved on the hard disk when the system is in depth. When it is turned on again, it can save the restart time. At present, the STD technology has been eliminated, updated Is STR technology. STR (SUSPEND TO RAM), hangs to memory, ie, save the data in memory when the system depth sleeps, and then restart to the original state takes about 3 seconds. This technique is supported by the current newer motherboards (such as 815 motherboards). D. Asynchronous memory adjustment technology: In VIA's chipset VIA Apollo Pro 133 / 133A and KT 133, there is a function of memory and outer frequency asynchronous. That is, under the standard outer frequency (such as 66 MHz or 100MHz, etc.), the frequency can be operated by 33 MHz or higher than the outer frequency. This technology greatly facilitates some old users, so you can use reasonable matching that will compare new memory and comparison CPUs (or more old memory and comparison new CPUs) to give full play to their functions. However, it should be noted that if the frequency of memory operation will not be increased in accordance with this rule, the frequency of memory operation will not be increased in accordance with this rule.
e. Extended slot frequency: Each type of bus has its own rated running frequency, if it exceeds too much, it may cause the device to run abnormal. For example, the rated frequency of the PCI device is 33MHz, and the rated frequency of the AGP device is 66MHz. When the outer frequency is running at 100MHz, the PCI device needs to work in terms of three-thirds to ensure the normal operation of the equipment (such as sound card and other devices), which is usually the three divided division; if the outer frequency is in 1333MHz, PCI equipment requires quadrupler. If the outer frequency rises, even quadre frequency, it is much higher than the standard frequency, and the AGP device usually only supports quadrature, so in high extraction (such as 150MHz), if the PCI device (sound card) Or the AGP device (graphics card) is not good, will strictly affect the overclocking performance of the entire system. At present, the PCI bus only supports quadrestriction, and the AGP bus only supports quadrature. Safety technology: Some safety protection technologies are essential because the current virus is very harmful. For example, in the protection of BIOS, there have been a variety of forms. The simplest is to write a write protection jumper by BIOS to avoid viruses; there is also a double BIOS, even if there is a destroyed, there is another work, such as Gigabyte, this technology; even The manufacturers have developed by the manufacturers, such as Lenovo's "invincible lock", "Zeus", etc., the principle is also to avoid viral infringement BIOS. Motherboard diagnostic technology is also a relatively practical technology. Such as MSI D-LED technology is to repay the fault with four lights. If the graphics card fault is represented by two red lights, the memory fault is indicated by three red lights. This helps some beginners to judge the fault in order to treat the disease. The voice prompt technology developed by Shuettke will cure the voice chip on the motherboard, and the fault can be "said" out (using the chassis little speaker), which is more preferences for some chasing people. New interface: AGP Pro interface: With the power of graphics card processing functions, its energy consumption is getting higher and higher, and traditional AGP slots can no longer meet the needs. The AGP Pro slot is longer than the ordinary AGP slot, which increases some ground lines, so that the signal is more stable, in the interference of the large current, this can improve the accuracy of data transmission, so that the graphics card is more stable. CNR Slots: (Communication and Networking RiSer) is a new slot that appears on the new i815e chipset. It supports Ethernet cards and modems, and features a bit similar to the AMR slot, but more powerful. 440bx chipset: Intel is designed to support high-frequency Pentium II, which has two improvements on 440LX: First, Pentium II of 400MHz; Second, memory can be extended to 1GB. The 440EX chipset: It is an Intel to support the chipset developed by the "Celeron" microprocessor. It is positioned at a low price of personal computers, and the maximum memory it makes up can support 256MB. 440FX chipset: It is a chipset developed by the Pentium Pro microprocessor. It is three-piece structure, 82441FX (system and memory controller), 82442FX (data bus accelerator), 82371SB (PCI, ISA, IDE acceleration controller). 440LX chipset: It is two-piece structure, which introduces QPA four-port acceleration design, making dynamic arbitration speed faster; pipeline diversification is more reasonable; UITRA DMA performance is improved, so that the hard disk transfer is faster. 450nx chipset: It is the super chipset developed by Intel as the high-end server.
Mainly developed by DESCHUTES (enhanced Pentium II) chips. Target is positioned in the server, high-end workstation. Its Cache can be extended to 2MB. 5591 5595 Slices: It is a chipset developed by SIS company specially to support the high frequency Pentium level CPU of the Socket-7 structure. It can support the AGP graphics accelerator card. Some can also support 100MHz bus frequencies, the CPU main frequency can support 266MHz; SDRAM memory can be extended to 768MB. ACPI power interface: It is a new feature that Pentium is unique to the motherboard. The role is to save energy when managing various components within the management of the computer. SMP symmetrical multi-processing mode: It is characterized by inserting two CPUs, support alternating operations to improve the work efficiency of the CPU. But the characteristics of the two CPU must be exactly the same. AGP Slots: (Accelerated-Graphics-Port: Accelerated graphics port) It is a bus structure that is built to alleviate video bandwidth. It connects the display card directly to the chipset of the motherboard to perform point-to-point transmission. But it is not a regular bus, because it can only be connected to the AGP graphics card, there is no general purpose and scalability. The frequency of its work is 66MHz, which is doubled by the PCI bus, and can provide 528MB / s data transfer rate for video devices. Therefore, it is actually a super-collection of PCI. AMD-640 chipset: This chipset is AMD's product. Its features are: support all Pentium level CPUs, specifically optimize AMD-K6-CPU; can truly play SDRAM high-speed performance over 66MHz; also has remote wake-up function; and there is a USB interface controller, but it does not Support AGP. ASUS slot: It is a design that ASUS is doing new sink on the motherboard it produced. Its structure is a short slot after the PCI slot is used in conjunction with ASUS's own supporting sound card. ATX Plate: Its layout is "horizontal" board design, just put the Baby-AT board type, this increases the space of the motherboard to lead the port, so that the motherboard can integrate more extension. ATX Power: The ATX power is a power supply for the ATX motherboard, which adds some new roles to it; one is to increase the power supply in the shutdown state to provide a set of micro-current (5V / 100mA). The second is to increase the 3.3V low voltage output. Baby-AT board type: The "vertical" plate design, that is, the short side is located behind the chassis, so that the space of the various extraction ports on the motherboard is small, which is not conducive to plugging various leads and peripherals. BIOS: BiOS (Basic-Input - & - Output-System: Basic Input / Output System) is a special manager in a dedicated EPROM chip in advance in advance. The motherboard is the control and coordination between the various components through this manager. CMOS: CMOS is a readable write-writable RAM chip on the computer motherboard, which protects the hardware configuration of the current system and the setting of the user to certain parameters. The current manufacturers make CMOS programs to the BIOS chip, and can set the system when booting, and enter the CMOS setup program to set the system. So being called BIOS settings again. COM port: A motherboard typically has two COM serial ports. Usually used to connect the mouse and communication devices (such as connected external Modems), etc. Concurrent PCI: Concurrent PCI bus technology, it is actually a reinforced structure of PCI. Used to improve between CPUs and PCI, CPUs and memory, which is Intel first in use in 440FX.
DIMM: (DUAL-inline-menory-modules) is a new 168-wire memory slot. It is to be longer than the SIMM slot, you can plug a single SDRAM that does not exceed 64MB. And it also supports a new 168-wire EDO-DRAM memory. EIDE: EIDE (Enhanced IDE: Enhanced IDE) is a standard interface for Pentium above the motherboard. Two EIDE interfaces are usually available on the motherboard. In the Pentium above the motherboard, EDIE is integrated into the motherboard. EISA Bus: EISA (Extended Industy Standard Architecture: Extended Industrial Standard Structure) is a bus extension criterion designed with 32-bit CPUs. It absorbs the essence of IBM microchannel bus and is compatible with ISA bus. But today has been eliminated. Flash: Flash (flash-memory: Quick modified memory) It is Pentium or more main boards to store BIOS programs. I / O chip: There are I / O control circuits on the board, on the board, on the board. It is responsible for providing a serial, parallel interface, and floppy drive control interface. IDE: IDE (Integrated Device Electronics): An interface type of a disk drive, also known as an ATA interface. Up to two IDE interface devices can be connected, allowing the maximum hard disk capacity 528 megabytes, control lines and data cables to connect with a 40-core flat cable to the hard disk interface card. The data transfer rate is 3.3 Mbps-8.33Mbps. ISA Bus: Industry Standard Architecture: Industrial Standard Architecture: The bus standard made by IBM's PC / AT computer, which is 16-bit architecture, only supporting 16-bit I / O equipment, data transmission rate is approximately 8MB / S. Also known as the AT standard. MVP3 chipset: It is the latest product launched after VIA after VP3. It supports a 100MHz bus frequency. Motherboard memory can be extended to 1GB, support ECC function, and Cache can support 2MB. PCI bus: PCI (Peripheral Component Interconnect: External Device Interconnection) is a bus structure that is launched by the SIG Group. It has 132 Mb / s data transfer rate and strong belt load capacity, which can be applied to a variety of hardware platforms, compatible with ISA, EISA bus. POST: POST (POWER-ON-Self-Test) is a major part of the BIOS function. It is responsible for completing detection of CPU, motherboard, memory, soft and hard disk subsystem, display subsystem, including display cache, keyboard, CD-ROM optical drive, etc. PS / 2 mouse interface: Today's popular Pentium motherboards are multi-use PS / 2 Mouse interface, and the common serial interface is given to mouse interface. The advantage of this is that it can save a regular serial interface, but also make the mouse faster response speed. SCSI: SCSI (Small Computer System Interface: Small Computer System Interface) It can drive at least 6 (SCSI-3 standard expansion up to 32) external devices; and its data transfer rate can reach 40Mbps, SCSI-3 is more up to up to 80Mbps. SIMM: (Single-in-line-menory-modules) is a memory slot we often use, which is a 72-wire structure. Most of the memory modules today is to integrate several memory chips on a small board.
VL Local Bus: (Local Bus: Local Bus) is an open bus structure designed by VESA organization. Its width is 32 bits, the operating frequency is 33MHz, and the data transfer rate is 132Mb / s. However, its definition standard is not strict, the compatibility is not good, and the load capacity is relatively low, so it has been replaced by the PCI. VP3 chipset: It is the latest product launched in the fourth quarter of 1997 in 1997. It is a motherboard for the Socket 7 structure. Its main performance indicators are: support all Pentium level CPUs, the highest frequency of the CPU can be 300MHz, support second-generation SDRAM memory; maximum scalable to 1GB. Battery: The Pentium Level Motherboard is mostly lithium batteries, only a few fully enclosed structural batteries. It is used to keep the motherboard CMOS data. Slipboard motherboard: It refers to the frequency of the CPU, the operating voltage, and the main board bus operating frequency settings do not use the regular jumpers, but "soft" settings via Setup (System BIOS). Memory: Memory is essentially one or more sets of integrated circuits, with data input and output and data storage. Since its storage information is different, it is divided into read-only, rewritten, read-only and random memory. Chipset: (chipset) is the core of constituting the motherboard circuit. In a certain sense, it determines the level and grade of the motherboard. It is the "Nanqiao" and "North Bridge", is the chipset that maximizes previously complex circuits and components into several chips. Fourth, the hard disk term explains the rotational speed of the hard disk: the speed of the hard disk motor spindle, the speed is one of the key factors that determine the internal transfer rate of the hard disk, and its quick slowness has largely affected the speed of the hard disk, while speeding speed The quick slow is also one of the important signs that distinguish the hard disk. The spindle motor of the hard disk drives the disc at high speed to generate buoyancy to float the head over the disc. To bring the sector you want to access to the head below the head, the faster the speed, the shorter waiting time. Therefore, the speed determines the speed of the hard disk to a large extent. At present, there are usually 5400 rpm, 7200 rpm, or even 10,000 rpm on the market. In theory, the faster the speed is, the better. Because higher speeds can shorten the average search time and actual read and write time of the hard disk. However, the faster the speed, the larger the heat, which is not conducive to heat dissipation. The mainstream hard drive speed is typically 7200 rpm. As the hard disk capacity is increasing, the speed of the hard disk is also increasing. However, the improvement of the speed also brings a series of negative effects such as wear, temperature rise, and increased noise. Thus, the Fluid Dynamic Bearing Motors in the precision machinery industry is introduced into the hard disk technology. The liquid bearing motor uses a mucosa liquid oil bearing that is replaced by an oil film. This can avoid direct friction of the metal surface, minimize the noise and temperature; the oil film can effectively absorb the vibration, enabling the seismic ability; more reduced wear, improve the life. Average seek time: Refers to the time that the hard disk moves the read writing on the disk to the specified track to find the corresponding target data, which describes the ability of the hard disk to read data, and units in milliseconds. When the single disc capacity is increased, the seek action and moving distance of the magnetic head are reduced, thereby reducing the average search time and speeds up the hard disk speed. At present, the average roof of the mainstream hard drive on the market is generally 9ms or less, and the hard drive greater than 10ms is a earlier product, which is generally not worth buying.
Average Latency Time: Refers to the time to continue to rotate to the magnetic head after the head moves to the data where the data is moved, and then waits for the time to continue to rotate under the magnetic head, generally between 2 ms-6ms. Average Access Time: The average time to find the specified data is usually the sum of the average roof time and the average latency. The average access time can represent the time used by the hard disk to find a certain amount of data, the better the average access time, generally between 11 ms-18ms. Note: Most of the average access hours in many hard drive advertisements are now replaced by average finding time. BURST DATA TRANSFER RATE: Refers to the highest rate of the computer read data from the hard disk internal cache area through the data bus. Also called external data transfer rates. The external transmission rate of the hard disk currently adopted UDMA / 66 technology has reached 66.6Mb / s. The maximum internal data transfer rate: refers to the maximum data transfer rate between the head to the hard disk cache, generally depending on the disk speed of the hard disk and the disk data line density (refers to the data spacer on the same track). Also called sustaind transport rates. The internal transmission rate of the hard disk of the UDMA / 66 technology is generally 25-30Mb / s, only a few products exceeds 30Mb / s, because the internal data transfer rate is the real bottleneck of the system, so everyone should distinguish these two times when buying Concept. However, in general, when the speed of the hard disk is the same, the single-disc capacity is high; the internal transmission rate of the hard disk having a high speed is high when the single disc capacity is the same. Automatic detection analysis and reporting technology (S.M.A.R.T): The factory hard drive is now basically supports S.M.A.R.t technology. This technique can monitor the head unit, disc motor drive system, hard disk internal circuit of the hard disk, and disk surface media materials, etc. loss. S.m.a.r.T technology must only work on the premise of motherboard support, and S.m.a.r.T technology does not guarantee the possible hard disk failure. The magnetoresistive head technology MR (Magneto-Resistive Head): MR (Magneto-ResitiveHead) is a referred to as a magnetoresistive head. MR technology can record density, record data, increasing the hard disk capacity and increase data throughput. The current MR technology has several generations of products. Maxtor's diamond three-generation / four generations have adopted the latest MR technology. The working principle of the magnetoresistive head is based on the magnetoresistance effect. The core is a small sheet metal material, which varies with the change in the magnetic field. Although its change rate is less than 2%, but because the magnetoresistive element is connected to a very sensitive The amplifier, so that the tiny resistance change can be measured. MR technology can increase the hard disk capacity by more than 40%.
GMR (GIANTMAGNETORESIVE) GMR head is the same as the MR magnetic head, and the resistance value of the special material is used to read the data on the disk with the principle of magnetic field changes, but the GMR head uses a magnetoresistive effect, better materials and multiple layers. The thin film structure is more sensitive than the MR head, the same magnetic field change can cause a larger resistance value to change, so that higher storage density can achieve a higher storage density can be achieved by 3Gbit-5Gbit / IN2 ( Gigabit per square inch), while the GMR head can reach 10Gbit-40Gbit / IN2 or more. At present, the GMR magnetic head is already in a mature promotion period. In the next few years, it will gradually replace the MR head and become the most popular magnetic head technology. Cache: The cache is a place where the hard disk is exchanged with external bus. The process of reading data of the hard disk is to convert the magnetic signal into an electrical signal, fill the empty, fill, then empty, step by step by cached, then fill, then emptying, according to the cycle of the PCI bus, can be seen, the role of the cache is quite important. When interface technology has evolved to a relatively mature stage, the size and speed of the cache are important to directly related to the transmission speed of the hard disk. At present, there are several mainstream hard drives, including several 512KB and 2MB. Its type is generally EDO DRAM or SDRAM, which is generally mainly SDRAM. Based on the way the write mode, there are two written forms and backcrops. When reading hard disk data, the system checks the request instruction first to see if the required data is in the cache, if the response is sent out by the cache, this process is called hit. Such the system does not have to access data in the hard disk. Since the speed of SDRAM is much faster than the magnetic medium, the speed of data transmission is accelerated. The reply is that it is also looking for a cache when writing to the hard disk data. If it is found to be written by the cache, most hard drives currently use the reply hard disk, which greatly improves performance. Continuous failure time (MTBF): Refers to the longest time of the hard disk from starting to the failure. The MTBF of the typical hard drive is at least 300,000 or 40000 hours. Partial response fully matched technology PRML (Partial Response Maximum Likelihood): The disc can store more information while effectively increase the reading and data transfer rate of the data. It is one of the advanced technologies currently applied to the hard disk data reading channel. The PRML technology is to divide the hard disk data reading circuit into two "* on-line water lines". The first segment of the pipeline transmits the signal reads the signal and then only selects the part "standard" signal to transfer the second segment to continue processing, the second paragraph will The received signal is compared to the PRML chip preset signal model, and then selects the smallest difference with the smallest, and outputs the combined output to complete the data read. PRML technology can reduce the error rate of the hard disk read data, so the disk data intensity can be further improved. Single track time (Single Track Seek Time): When the head is transferred from a track to another track. Super Digital Signal Processor (ULTRA DSP) Technology: Mathematical operation with Ultra DSP, with a speed of 10 to 50 times higher than the general CPU. Ultra DSP technology, a single DSP chip can also provide a dual function of the processor and drive interface to reduce the use of other electronic components, which can greatly improve the speed and reliability of the hard disk. Interface technology can greatly improve the maximum external transmission rate of the hard disk, the biggest benefit is that data can be directly transmitted from the hard disk to the main memory without occupying more CPU resources, and improving system performance. Hard disk surface temperature: Refers to the temperature generated during hard disk to increase the temperature of the hard disk sealing.
The temperature generated during the hard disk will affect the data read sensitivity of the thin film magnetic head (including MR head), so the hard disk having a lower temperature of the hard disk has a better data read, write stability. Max Full Seek Time: Refers to the full time to move the magnetic head until finally find the required data blocks. Interface technology: Oral technology can greatly improve the maximum external data transfer rate of the hard disk, and now the Ultraata / 66 that is generally used has greatly improved the performance of the E-IDE interface. The so-called Ultradma66 refers to a synchronization of Intel and Quantum design. DMA protocol. The hard disk of this technology is used in conjunction with the corresponding chipset, and the maximum transmission speed can be increased from 16Mb / s to 66 ms / s. Its maximum advantage is to liberate the CPU from a large amount of data transmission, you can directly transfer data from HDD to the main memory without occupying more CPU resources to a certain extent, to a certain extent, to a certain extent. Performance of the entire system. Since the overall performance of the Ultraata technology is more than 20% to 60% compared to the ordinary hard drive, it has become the current E-IDE hard disk in fact standard. The interface technology of the SCSI hard drive is also growing rapidly. Ultra160 / MSCSI is introduced to the hard disk world, which has the ability to extend the performance extension of the high computing amount application field, and dealing with the server, graphics workstation, redundant disk array (RAID) such as a critical task, will therefore be improved. From the perspective of technology development, Ultra160 / MSCSI is only a ring on the road of hard disk interface. The 200MB of fiber optic technology has not reached the end, and the future interface technology will make today's users. Fiber Channel technology has the advantages of high data transmission rate, far data transmission distance, and simplified large storage system design. At present, the Fiber Channel supports 200MB of data transmission rates per second, and can accommodate up to 127 drivers on a loop, and the local cable can be run within 25 meters, and the remote cable can be run within 10 km. Some specialized storage applications, such as small storage area networks (SAN), and digital video applications often require up to 200MB of data transfer rates and strong networking capabilities per second, and the launch of Fiber Channel technology is adapted to this requirement. At the same time, its long data transmission distance is greatly facilitating the technical implementation of remote communication. Due to the superiority of Fiber Channel technology, hard disk products that support fiber optic interface began to appear in the market. These products are generally a large-capacity hard disk, and the average finding time is short, adapted to the application requirements of high speed, high data volume, will provide good guarantees for medium high-end storage applications. IEEE1394: IEEE1394 is also known as FireWire, which is a high-speed serial bus. Existing IEEE1394 standards support 100Mbps, 200Mbps, and 400Mbps transmission rate, will reach 800Mbps, 1600Mbps, 3200Mbps, and even higher, so The high rate makes it possible to use a large-capacity storage device such as a hard disk, a DVD, a CD-ROM. IEEE1394 is expected to replace existing SCSI buss and IDE interfaces, but due to high cost and technical lack of mature, there is still only a small number of products that use IEEE1394 interfaces, and the hard disk is less. Hard disk: English "hard-disk" referred to as HD. It is a huge storage of storage, and the role is to store data required for computer runtime. The computer's hard drive is mainly composed of discs, magnetic heads, magnetic head arms, magnetic head arm service positioning systems, and underlying circuit boards, data protection systems, and interfaces. The technical indicators of computer hard drives are mainly surrounded on disk size, disk, single-disc capacity, disk speed, head technology, service positioning system, interface, secondary cache, noise, and S.m.a.r.t. Disc: All data of the hard disk is stored on the disc, and the disc is made of a hard alloy, and a glass disc is also present.
The current disk size of the current hard disk products is: 5.25, 3.5, 2.5 and 1.8 inches (after two commonly used in notebooks and partial pocket precision instruments, there are often 3.5-inch discs in the desktop. The head of the head: The head of the hard disk is made by coil on the magnetic core, and the initial magnetic head is read and written, and the amplitude of the induction signal is transmitted by current. For most computers, in the process of exchanged data with the hard disk, read * is far from writing *, and read / write is two different characteristics, so that hard disk manufacturers have developed a read / Write the separation head. In 1991, IBM proposed a magnetoresistive (MR) technology-based reading technology - various anisomagnetic, magnetic heads and rotated disc phase contacts, read data by changes in magnetic field on the induction disc . In the hard disk, the disc's single-disc capacity and head technology are mutually restrictions and promote each other. AMR (Anisotropic Magneto Resistive, AMR): A magnetic head technology, AMR technology can support 3.3Gb / square inch record density, in 1997 AMR is the mainstream technology in the market at the time. GMR (GIANT Magneto Resistive, GMR): 2 times higher than the AMR technology, GMR head is made of four-layer conductive material and magnetic material film: a sensor layer, a non-conductive interposer, a magnetic property Trouble layer and a switching layer. The first three layers control the resistance of the magnetic head. In the ingress layer, the magnetic field strength is fixed, and the exchange layer of the magnetic field is maintained. Moreover, the magnetic field strength and direction of the free layer is changed as the slight magnetization zone of the disk surface under the magnetic head is changed, and the change in the magnetic field strength and direction results in a significant change in the magnetic head resistance, and below a fixed signal voltage. You can pick the signal for the hard disk circuit processing. OAW (Optical Auxiliary Voltage Technology): The OAW, which is developed by Seagate is the direction of future magnetic head technology. OAW technology can write more than 10,5000 tracks in 1 inch wide, and single-disc capacity is expected to break through 36GB. The increase in single-disc capacity can not only improve the total capacity of the hard disk, which reduces the average seek time, but also reduces costs, improve performance. PRML (the largest propaganda of local response, Partial Response maximum likelihood): In addition to the chance of magnetic head technology, magnetic recording technology is also a major factor affecting the performance of hard disk performance. When the magnetic recording density reaches a certain degree, the phenomenon of mutual interference between the two signals will be very serious. In order to solve this problem, people have added PRML technology in the design of the hard disk. The PRML read channel method can be simply divided into two parts. The first is to digitize the signal read from the disc and abandon the standard signal without reaching the standard, and no signal output. This part is called a local response. The maximum planning part is compared to the signal model library with the PRML chip itself, identify the signal model of the PRML chip itself, and then re-combines these signals directly to directly output data. Using the PRML mode, no signal strength as the pulse detection mode is required, or the mutual interference due to the signal recording is too intensive. Advances in magnetic head technology, plus the development of current recording materials and processing technology, will increase the storage density of the hard disk to 10GB per square inch, which means that 40GB or more hard disk capacity can be implemented. Space Factor: The number of physical sectors between the two logical sectors of the hard disk track. Because the information on the hard disk is organized in the form of a sector, each sector has a number, access * For this sector number, so use a specific spacing to give the sector number to help Get the best data transfer rate.
Landing area (LZ): To make the hard disk has a starting position, generally specify an inner cylinder as a landing area, which stops the hard disk head until the power is turned back before the power is turned back. The landing area is not used to store data, which can avoid the loss of data during the fire, and the power supply is avoided. Currently, a general hard disk automatically stops the head in the landing area when the power is turned off, and the old hard disk needs to perform a park command to place the head. Reaction time: refers to the working conditions of the turn * in the hard disk. The reaction time is a most direct reaction index of the hard disk speed. The 5400 rpm hard drive has a reaction time of 5.55 ms, while 7200 rpm can reach 4.17 ms. The reaction time is that the hard disk will take advantage of how long to complete the first turn * rotation. If we determine that a hard drive reaches a speed of 120 weeks, the rotation of a week will be 1/120 is 0.008333 seconds. If our hard drive is 0.0041665 seconds per week, we can also call the reaction time of this hard disk is 4.17 ms (1 ms = 1/1000 per second). Average Latency: Refers to the time when the head moves to the data where the data is located, then waits for the required data block to continue to rotate (half a ring or less) to the magnetic head at milliseconds (MS). The average incubation period is, the better, the latency is small in the hard disk read data, which is equal to having a higher hard disk data transfer rate. SINGLE TRACK SEEK: When the head is transferred from a track to another, in milliseconds (MS). Max Full Seek: Refers to the full time that the magnetic head starts until finally find the required data block, unit is in milliseconds (ms). External data transfer rate: Burst Data Transfer Rate: Refers to the rate of reading data from the hard disk buffer, often replaced by the data interface rate, and the unit is MB / s. At present, the mainstream hard drive is Ultra ATA / 66. Its largest external data rate is 66.7Mb / s, the Ultra ATA / 100 launched in 2000. Theoretically, the largest external data rate is 100Mb / s, but due to internal data transmission The restriction of the rate is often not so high. Spindle speed: refers to the rotational speed of the motor spindle in the hard disk. At present, the spindle speed of the ATA (IDE) hard disk is generally 5400-7200 rpm. The rotational speed of the mainstream hard disk is 7200 rpm. As for the SCSI hard disk, the spindle speed of the SCSI hard disk can be reached from 7200 to 10,000 rpm. And the highest speed of the SCSI hard drive speed is up to 15,000 rpm. Data caching: Refers to the high speed memory inside the hard disk, which is temporarily stored in the computer, which is temporarily saved to read and read it. At present, the hard disk is typically 512KB-2MB. At present, the mainstream ATA hard disk is 2MB, and the highest data cache in the SCSI hard drive has now reached 16MB. Has a great advantage for the hard disk of big data cache is accessed. Hard disk surface temperature: It refers to the temperature generated when the hard disk is working, the hard disk sealing case temperature rise. The temperature generated during the hard disk will affect the data read sensitivity of the head of the head, so the hard disk having a lower temperature surface temperature is better data read, write stability. MTBF (continuous failure time): It refers to the maximum time that the hard disk is running from the beginning to the failure, and the unit is hour. The MTBF of the typical hard drive is at least 300,000 or 40000 hours.
SMART (self-monitoring, analysis, reporting technology): This is now the universal data security technology used by the hard disk. When the hard disk works, the system analyzes the state of the motor, circuit, disk, and heads. When there is an abnormality, it will WARNING, some will automatically reduce and back up data. DPS (Data Protection System): Quantum built DPS in the first time in the eight-generation hard disk of the fireball, the first 300MB memory of the hard disk * Make important information such as the system, DPS automatically detects recovery system data within 90 seconds after the system. If it is not possible to start with the DPS floppy disk, it will automatically analyze the fault, try to ensure that the data is not lost. Data Guard: It is a unique hard disk data security technology that western data (WD). This technology can automatically automatically scan, detect, detect, and repair the sectors of the disk every 8 hours in the spare time of the hard disk. MaxSafe: It is the technology applied by Mai Tuo in the gold diamond second-generation. Its core is to save additional ECC check digits on the hard disk, so that the read and write procedures are verified to ensure data integrity. DST: Drive self-testing technology, is the data security technology used in Sejour's hard drive, which ensures the security of the data stored in the hard disk. DFT: Drive Health Test Technology, is the data security technology used in IBM's hard drive, which can greatly improve data security as many technologies. Noise and Shock Seismic Technology: The hard disk spindle is inevitable to generate noise during high speed rotation, and wear and heat resistance due to metal friction, "Liquid bearing motor" can solve this problem. It uses the mucosa liquid oil bearings, which can effectively reduce the above problems in an oil film. At the same time, the liquid oil bearing can also effectively absorb the vibration, so that the earthquake resistance of the hard disk is increased by one or two hundred g, so the life and reliability of the hard disk can also be improved. Quantum has applied SPS vibration protection systems after the hard disk after the Fireball Seven Generation (EX) series; Mai Tuo applied ShockBlock anti-seismic protection system in the gold diamond second generation, and their purpose is to disperse the impact energy, try to avoid the head and disk. The impact of the film; the Seashield system in the gold medal of Seagate is achieved by the protective soft cover of the shock absorbing material and the shock-shock design between the disc. ST-506/412 Interface: This is a hard disk interface developed by Seagate, first using the hard disk of this interface for Seagate's ST-506 and ST-412. The ST-506 interface is quite simple, it does not require any special cables and connections, but it supports the transmission speed is very low, so this interface is basically eliminated in 1987, using the old hard disk capacity of the interface. Most are less than 200MB. The hard drive used in the early IBM PC / XT and PC / AT machines is the ST-506/412 hard drive or called the MFM hard disk-MFM (Modified Frequency Moduction) refers to a coding scheme. ESDI interface: Enhanced Small Drive Interface interface, it was developed in 1983. It is characterized by placing the codec in the hard disk itself, rather than on the control card, the theoretical transmission speed is 2 ... 4 times the ST-506 described above, usually 10 Mbps. However, its cost is high, and there is no advantage compared to the Ide interface generated by the Ide, so it is eliminated after the 1990s.
IDE and EIDE interface: Ide (Integrated Drive Electronics) actually refers to a hard drive that integrates a controller with a tray, and we often say the IDE interface, also called the ATA (Advanced Technology Attachment) interface, now PC Most of the hard drives used are IDE compatible, just use one cable to connect them with the motherboard or interface card. The practice of integrating the disk with the controller reduces the number of hard drive interfaces and length, the reliability of data transmission is enhanced, and the hard disk is made easier, because the manufacturer does not need to worry about whether the hard disk is worried about whether the hard disk is The controller produced by the manufacturer is compatible, and it is more convenient to install the hard disk. ATA-1 (IDE): ATA is the official name of the earliest IDE standard, the IDE actually refers to the hard disk itself connected to the hard disk interface. ATA has a socket on the motherboard, supports a master and a slave device, the maximum capacity of each device is 504MB, and the earliest supported PIO-0 mode (Programmed I / O-0) in ATA is only 3.3MB / s, and ATA -1 specifies three types of PIO mode and 4 DMA modes (without actual application), to upgrade to ATA-2, you need to install an EIDE adapter card. ATA-2 (EIDE ENHANCED IDE / FAST ATA): This is an extension of ATA-1, which increases 2 PIOs and 2 DMA modes, raising the highest transfer rate to 16.7Mb / s, and introduce LBA address conversion The way, breaks through the old BIOS inherent 504MB limit, supports up to 8.1GB hard drives. If your computer supports ATA-2, you can find (LBA, LogicalBlock Address) or settings in CMOS settings. The two sockets can be connected to a master device and a slave setting, so that four devices can be supported, and the two sockets are also divided into the main socket and from the socket. The fastest hard disk and CD-ROM can usually be placed on the main socket, while the secondary device is placed on the jack, which is necessary for the 486 and the early Pentium computer, which makes the main socket Connect to the fast PCI bus and connect from the socket on a slower ISA bus. 5. Graphics interpretation of EDO DRAM (Extended Data Out DRAM): Extended Data Output DRAM. Some improvements to the DRAM access mode, shorten the time of memory efficient access. Vram (Video DRAM): Video RAM. This is specifically for dual port memory optimized for graphics (data exchange with RAMDAC and CPU), which can effectively prevent conflicts that occur when accessing other types of memory. WRAM (Windows Ram): Enhanced VRRAM, performance than VRAM increases by 20%, accelerating video functions such as transmission and mode padding. SDRAM (Synchronous DRAM): Synchronous DRAM. It works synchronously with the system bus to avoid additional waiting time required for asynchronous DRAM on the system bus to synchronize the data. SGRAM (Synchronous Graphics DRAM): Synchronous Graphics RAM, enhanced SDROM. It supports writing masks and blocks. Write mask can reduce or eliminate read - modifications - write * written; block write facilitates the foreground or background fill.
SGRAM greatly accelerates data exchange between the memory and the bus. (, Such as: Ritai S680, Banshee) MDRAM (Multibank DRAM): Multi-segment DRAM. MDRAM can be divided into multiple independent valid segments, reducing each process at the time loss of display refresh, video output, or graphic acceleration. RDRAM (Rambus DRAM): Mainly used for special high-speed bursts *, the access frequency is up to 500MHz, and the traditional memory can only be accessed at 50mHz or 75MHz. RDRAM's 16 Bit bandwidth can reach 1.6 Gbps (EDO's limit bandwidth is 533Mbps), and 32bit bandwidth is up to 4 Gbps. Second, 3D graphics card basic 3D function: 1. Alpha Blending: Alpha mix. Alpha is a special passage in the color characteristics of the 3D texture element, using it to color mixing with the texture image, producing a transparent effect. 2. Billinear Filternig: Double line filtering. A texture mapping technique can reduce block diagrams due to unevenness of color distribution during texture. 3. DITHERING: Jitter. This is the alignment of the change color pixel (Pixel) to get a new color process. 4. Flad Shading: A basic drawing technology that uses the same color inside each triangle drawn. 5. Fogging: Atomization. Mix some color with background to hide background to achieve haze effect. 6. Gouraud Shading: Use the color of the triangle vertex to get the interposition of the triangle to each point in the triangle. 7. MIPMAP: MIP mapping. It can save different resolutions and sizes of texture graphics in memory, which allows texture smooth and change when 3D objects are moved. 8. Perspective Correction: Perspective correction. Different angles and distances can be more realistic in 3D scenes in 3D scenes. 9. Point Sampled: Point sampling. A simple texture mapping technology, with the nearest texture element to determine the color of the current point. 10. TEXTURE MAPPING: Texture Mapping. A bitmap or image is attached to the 3D object, making the object. 11. Transparency: Transparent. 12. Z-Buffer: It is a memory area for storing the field of field pixels. 13. Gamma Correction: Gamma correction. In order to compensate for graphical distortion due to display deviation, the gamma correction is corrective correctively. Third, 3D graphics card three API API (Application PROGAM Interface Application): It is a software interface for 3D applications and 3D graphics cards. 1.Direct 3D: It is the middle interface interface in Microsoft's Direct X. When some 3D functions cannot be implemented by hardware, Direct 3D can use software simulation most 3D functions, improve 3D graphic display speed, and its animation feature is quite high, which is very suitable for game development.
2.Heidi (also called Quick Draw 3D): It is a purely scheduled window, mainly suitable for application development, HEIDI flexible, can handle very complex geometric graphics, strong extension, support interactive rendering, most important It has been supported by Autodesk (Autodesk is a famous AutoCAD and 3D SUTDIO, 3DMAX Manufacturer) 3.opengl (Open 3D graphics library) is developed by SGI (SGI production non-PC graphics workstation) Companies, including its software Waterfull Alias Maya, its popularity is equivalent to Intel of the PC. OpenGL is a stand-alone platform with portability. It can quickly draw 2D and 3D objects, working together in distributed environments, is a standard application interface for large science and engineering for high complex 3D graphic design. 16-, 24- and 32-bit color: 16-bit color can display 65,536 different colors in the display, 24-bit color can display 16.7 million colors, and the 32-bit color is different, it is just technology One concept, it is true that the number of colors is just the same as 24-bit color, only 16.7 million colors. For the processor, processing 32-bit color graphic images is higher than the processing 24-bit color load, more workload, and the user needs to be more within 32-bit color mode. 2D card: There is no 3D acceleration engine ordinary display card. 3D card: show card with 3D graphic chip. Its hardware feature can complete the processing of three-dimensional images, to reduce the work burden for the CPU. Usually a 3D accelerator card also includes a 2D acceleration function, but there is a separate display card only having 3D image acceleration capabilities, such as VOODOO2. Accelerated Graphics Port (AGP) High-speed Graphics Acceleration Interface: AGP is a PC bus system, and its appearance is to make up for some shortcomings of PCI. AGP has a higher operating frequency than PCI, which means it has a higher transmission speed. AGP can use the system's memory as a material cache, and in the 3D graphics card of the PCI, the material can only be stored in the display card. Alpha Blending: It is used to make the object to generate transparent sense, such as blurred transparent sights that are blurred through water, glass, and the like. Previous software transparent processing is to give all transparent objects to give the same transparent parameters, which is obviously very unreal; today's hardware transparent mixing treatment adds a value to the pixel in red green blue, which has added a value to specifically store the transparency of the object. Advanced 3D chips should support at least 256 transparency, all objects (whether water or metal) are all values of transparency, only high and low. Anisotropic Filtering: (Please refer to the second-line filtration and tri-linear filtration) Anisotropic filtration is the latest filtering method, it needs to sample 8 or more pixels around the map point, obtain The average is mapped to the pixel point. For many 3D acceleration cards, it is almost impossible to filter an anisotropy filtration using more than 8 pixels because it requires more pixel filling rates than tri-linear filtration. But for 3D games, anisotropic filtering is a very important feature because it can make the screen more realistic, naturally processed more slower than tri-lineary filtration. Anti-aliasing (edge softening or anti-aliasing): Since the object edge of the object in the 3D image exhibits a triangular sawtooth, the anti-aliasing is smooth and natural, improve the image quality to make it soft. method.
The latest full-screen anti-aliasing can effectively eliminate the displacement phenomenon of polygon binding (especially smaller polygons), and reduce the distortion of the image. When the panoramic anti-aliasing is treated, the pixels near the image shall be sampled 2-4 times to achieve different levels of anti-aliasing effects. 3DFX adds the choice of 2x2 or 4x4 anti-aliasing effect in the drive. According to the series chip, the two-chip VOODOO5 will provide 2x2 anti-aliasing effect, and the card of the four-chip can provide a higher 4x4 anti-aliasing level. In short, it is to mix the image edge and the pixel colors on both sides, and then replace the point in the original position with the newly generated point to achieve the shape of the softened object, eliminate the effect of the zigzag. API (Application Programming Interface) application interface: API is an interface exists between 3D programs and 3D display cards, which makes the software over hardware. To use a 3D acceleration function, you must use the API supported by the display card to write programs, such as Glide, Direct3D, or OpenGL. Bi-linear filtering: It is the most basic 3D technology, now almost all 3D acceleration cards and games support this filtering effect. When a texture is made small, it will inevitably appear "mosaic" phenomenon, and filter can effectively solve this problem, which is a smooth image of flexible processing using the difference algorithm between different pixels in the raw material. of. Its work is the center of the target texture, and the four pixel color values near this point are averaged, and then this average color value is passed to the position of the target image. By using double linear filtration, although the transition between different pixels is more sleek, the image after bilinear processing will appear blurred. Environment Mapped Bump Mapping: The surface of the object in the real world is not smooth, so it is necessary to reflect the uneven and pleats of the real object by concave convection simulation techniques. The traditional 3D graphics card is used to be embossed (EMBOS) effect approximately a convex and concave mapping. This relief effect is limited, it is difficult to show the reflective effect at a subtle angular angle and the effect in a complex multi-aerative light source, and it is more unable to show water waves. Effect of special fluids such as airflow. The environment mapping convex and concave map is to map a layer of texture on the standard surface texture, the texture is the same but the location is changed, and the displacement depth is determined by the depth information and the light source position, and then the lower texture is further processed as an upper layer according to the performance object. The textured shadow or bottom surface, which realizes the convex pleat effect of the surface of the real object. Gouraud Shading: This is currently a more popular coloring method, which provides continuous color discs for each point on the polygon, which is rendered, and each polygon can use unlimited colors. It renders have extremely rich colors and smooth discoloration effects. MIP-MAPPING: MIP-mapping core feature is based on changes in the direction of the object's depth of field, and the MIP mapping is based on different sizes of material map according to different far, such as the large material of 512x512, and A smaller map is attached to the distal object. This can not only produce better visual effects, but also saves system resources. Phong Shading: This is the best, most complicated coloring method, and the effect is better than Gouraud Shading. Its advantage is to process the "mirror reflective", which gives each point on the model to the total strength value of the projected light, so the extremely high surface brightness can achieve the effect of "mirror reflectance".
S3TL ("Transform and Lighting): This technology is similar to NVIDIA's latest T & L technology, which greatly reduces the geometric operation process of the CPU's 3D pipeline. The "Deformation and Light Source" engine can be used in future OpenGL and DirectX 7 graphics interfaces to increase the polygonal generation rate in the game to 4 to 10 times. This greatly alleviates the complexity of the software, and also maximizes the CPU's calculation, so for systems with slower CPU floating point, there can be high speed graphics Treatment ability. S3TC (S3 TEXTURE COMPRESSION) / DXTC / FXT1: S3TC is a texture compressed format proposed by S3 companies, with the aim of achieving a system bandwidth and improving efficacy. The S3TC is compressed, and the limited texture cache space is used to store more textures because it supports the 6: 1 compression ratio, so the 6m texture can be compressed to 1M in the material cache, so that the cache is saved. It also improves the display performance. DXTC and FXT1 are similar to S3TC, which are Microsoft and 3DFX development texture compression standards, although supported in Direct 6, but has not yet been supported by games, and FXT1 can provide more than S3TC High compression ratio, reaching 8: 1, and it will also be supported in the 3DFX version of Glide. T & L and Light Processing: This is a new type of technique studied by NVIDIA to improve the image quality. In the past graphics technology, in order to make the object image true, it has to increase the polygonal design, so It will result in a decrease in speed, and less polygons are used, and the picture is very rough. This T & L technology used in GeForce256 is characterized by further improving the degree of edge smoothness of the surface of the object without increasing the object polygon, making the image more realistic and accurate. In addition, the role of the light source has also been paid to: the traditional light source processing is more single, free, and the geforce 256 has powerful light source processing power, which supports 8 independent light sources on the hardware, plus GPU support, instant processing The light source will make the picture more vividly, which can generate a light source effect with reflective properties. Trilinear Filtering: Three-line filtration is a combined overlap in mitigating or eliminating different combined grade texture transitions. It must be used in conjunction with double linear filtration and combined processing mapping. Three-linear filtration is obtained from two most similar LOD grade textures to obtain new pixel values by sampling in two most similar LOD grade textures, making the two different depth grades can be more smoothing. Because of this, triplet filtering must use two double linear filtration, which must calculate the value of 2x4 = 8 pixels. For many 3D acceleration, this requires the calculation time of their two clock cycles. W-Buffer: The role of W-buffer is similar to Z-buffer, but its range is smaller, and the accuracy is higher. It can process more detailed processing between different objects and the location of the same object part. Z-Buffer: This is a technique for processing 3D object depth information. It records the current Z coordinates of different objects and the same object. When coloring, the structure behind other objects is blamed, so that They are not displayed. The higher the number of digits used in z Bufer, the more accurate the depth value it can provide.
Now the graphic chip supports 24bit Z-Buffer and adds 8Bit template buffer, called 32bit Z-Buffer. Display memory: Like the memory function on the motherboard, the memory is also used to store data, but it is stored in the display chip processing. 3D Display Card Save Differences in General Display Card Different in: 3D Display Card There is also a memory of texture data or z-buffer data, such as a 6M memory VOODOO I display card, where 2M memory is Used in the above use. Since the 3D application is increasing, and the requirements of high-resolution, high-colored deep graphics treatment, the requirements for memory is getting faster, from early DRAM, transition to EDO-DRAM, until now often see SDRAM and SGRAM, speed is getting faster and higher in performance. The memory of Figure 4 is SGRAM, pay attention to its four sides, very good difference; Figure 5 The memory is EDO-DRAM, and two-side leads are used as SDRAM. Distinguishment of EDO-DRAM and SDRAM to see the number above the memory, generally marked with "08", "10", "12" and other words are SDRAM, labeled "80", "70", "60", " Most of the words such as -6 "," - 7 "is EDO-DRAM. In addition to the above three common memory, there are more professional memory such as VRAM (Dual Port Video Memory), WRAM (Window Memory), RDRAM, Cacheram, etc., mostly used on graphics processing workstations. The size of the memory is not fixed, from a single 256k, 512K, 1M to a single strip 2m, so you don't only look at the number of memory chips to guess how much memory capacity on the display card. There are also some empty sockets on the old display card to expand the memory (such as the right map, there are already inserted in the outlet), we should pay attention to the memory speed set with the display card when expanding, such as the original memory is 80ns, newly expanded memory is also 80ns, so that you can fail after expansion. BIOS: Also known as "VGA BIOS", mainly used to store control programs between chip and drivers, and store information such as display card model, specification, manufacturer, factory delivery time. When you open your computer, you feed back these information onto the screen by displays a control program within the BIOS. Figure 6 is showing BIOS on 3 different display cards, which can be seen that the shape is not the same. Early display BIOS is cured in the ROM, and cannot be modified, and the current most display card uses a large capacity EPROM, which is called "flash -bios", which can be rewritten through a dedicated program. Don't underestimate this feature, a lot of display cards are to modify the errors in the original program by continuously introducing the upgrade driver, adapt to new specifications, and enhance the performance of the display card. For users, the practice of software improvement performance is deep. Bus interface: The display card is inserted on the motherboard to exchange data with the motherboard. Main ISA, EISA, VESA, PCI, AGP, etc. connected to the motherboard. The ISA and EISA bus are wide and narrow, slow speed, and VESA bus extension ability, these three bus have been eliminated by the market. It is now common to PCI and AGP interfaces. The PCI interface is a bus interface, operating in a 1/2 or 1/3 system bus frequency (usually 33MHz), and if other data is to be processed while processing image data, all data flowing through the PCI bus must be respectively During the treatment, it is necessary to exist in data retention. When the amount of data is large, the PCI bus is very nervous.
The AGP interface is designed to solve this problem. It is a dedicated display interface (that is, can insert a sound card, display card, video capture card, etc. in the motherboard, but not on the motherboard Insert any board in addition to the AGP display card in the AGP slot), with the characteristics of exclusive bus, only image data can pass through the AGP port. In addition, AGP uses a higher bus frequency (66MHz), which greatly increases the data transfer rate. The current development trend of the current display card interface is the AGP interface. It should be noted that the AGP technology is divided by AGP1 × and AGP2 ×, the maximum theoretical data transmission rate of the latter is twice the former, and the AGP4 × display card (such as Savage4) will appear this year, its maximum theoretical data transmission rate Will reach 1056MB / s. Distinguishing the AGP interface and the PCI interface, the front width width of the former is very popular, commonly known as "gold fingers", the latter leads up and down. VGA outlet: It is a socket with 15 jacks, a bit like a larger written "D" (prevent interpretation). Different from the MIDI connector on the sound card, the jack of the VGA socket is set in 3 rows, 5 holes per row, the MIDI connector has 9 holes, 2 rows set, a little longer than the former, flat. The VGA socket is the output interface of the display card, which is connected to the D-shaped plug of the display for the output of the analog signal. Feature connector: It is a channel that displays the data exchange data in the video device, usually a 34-pin, and 26-pin. It doesn't matter, and early use to connect MPEG hard solution card as a channel transfer. Other parts: crystal oscillator: stainless steel housing, more conspicuous. Its action is to generate a fixed oscillation frequency that has a reference based on the operation of the display card. S terminal: Partial Display Card The function of 5 jacks is a semi-circular distribution through it to the TV (or monitor), which is identical to the S terminal on the TV. Patch resistance: Medium, high-grade display card is high in operating frequencies, using uncommon patch resistors to reduce interference. They are part of the display card electrical line. 6. Display terms explained the scanning method: The display of the display is divided into "progressive scan" and "interlaced". If the scanning system is scanned only (even), only a scanning method of scanning the odd (odd), and the display is used in the horizontal retraction, and the display is used as an interlaced display, this display Although the price is low, the human eye will clearly flicker, and the user uses the eyes for a long time to be tired, and it has been eliminated. The progressive display overcomes the above disadvantages, and the progressive scan is a horizontal scan, and the vertical scan is performed, and there is no parity. The progressive scan minimizes the visual flicker, and the screen does not feel fatigue. It is also a point that the interlaced display is actually displayed under low resolution, and only the resolution is increased to a certain extent to be changed to the interlaced display. Refresh frequency: From the display principle, any characters you see on the screen, the images, etc., the images are all arranged in the vertical direction and the horizontal direction. Since the imaging tube phosphor is tapped by the hit of the electron beam, the scanning display point array must be continuously refreshed. The refresh frequency is the speed of the screen refresh. The lower the refresh frequency, the worse the image flashes and jitter, the faster your eyes are fatigue. Sometimes it will cause eye pain, dizziness, and other symptoms. Excessive refresh frequencies will produce uncomfortable stroboscopic effects. When the refresh frequency is used in 75 Hz, it can be basically eliminated. Therefore, the refresh frequency of 75 Hz should be the minimum requirement of the display stabilization.
