 Wysłany: Pią 13:30, 26 Lis 2010 Temat postu: ghd hair straightener price CPU Parameters |
|
| 1, the frequency
in the electronics, the pulse signal is a voltage according to a certain extent, offern a certain time interval of successive pulses. The time interval between pulses, known as the cycle; and in unit time (eg 1 second) the number of pulses induced within a known frequency. Is to describe the periodic signal frequency cycle (including pulses) in unit time the number of pulses arevolution from the measurement of the number of names; frequency proclamation unit of measurement is Hz (Hertz). Computer system clock frequency is a typical very accurate and stable pulse generator. Frequencies used in the mathematical expression of \Hz) where: 1GHz = 1000MHz 1MHz = 1000kHz 1KHz = 1000Hz pulse cycle time calculation unit and the corresponding conversion relationship is: s (seconds) ms (milliseconds) μs (microseconds) ns (nanoseconds) where: 1s = 1000ms 1ms = 1000μs 1μs = 1000ns CPU's frequency, that is, the work of CPU core clock frequency (CPU Clock Speed). Commonly referred to as certain how much MHz CPU, while the number of MHz, that is, \Many people believe that the frequency of CPU is its speed, it is not. CPU's frequency, said digital pulse signal within the CPU speed of shock, and the actual computing strength of CPU and not directly related. The actual speed of operation frequency and a certain relationship, but there is no one formula to determine the quantitative relationship between the two values, because the CPU's operating speed depends on the CPU pipeline performance in all aspects of (cache, instruction set , CPU number of bits, etc.). Frequency does not directly represent the computing speed, so in certain circumstances, is likely to occur with higher frequency of the actual speed of operation with lower CPU phenomenon. For example, AMD's AthlonXP series CPU, most of the lower frequency can reach Intel's Pentium 4 series CPU higher frequency of the CPU's performance. Therefore, Athlon XP series CPU only way to name the PR value. Therefore, the frequency is only one aspect of CPU performance, and do not represent the overall CPU performance. CPU's frequency does not mean that CPU speed, but increase the frequency of operation for improving the CPU speed is importante. For example, suppose a CPU operation in one clock cycle to execute a command, then when the CPU running at 100MHz frequency, the frequency than when it is running in the 50MHz speed doubled. Because the 100MHz clock cycle takes more than 50MHz clock cycle time is abated by half, that is, work in the 100MHz frequency of the CPU to execute an manner instruction, the time is only 10ns, than the work in the 50MHz frequency cut in half the time of 20ns, Natural computing speed will quickly doubled. But the overall speed of the computer not only depends on the CPU operation speed, but also with the operation of other related subsystems, and only in improving the frequency at the same time, the subsystems and the operating speed of data transmission speed between subsystems can improved, the overall speed of the computer, can really be improved. improve CPU work frequency, limited mainly by the production process. Since CPU is manufactured in the semiconductor silicon, silicon components in the wire for connection between the need to, because at high frequencies requires finer wires as short as possible, so as to reduce stray capacitance and other stray wires to ensure that interference CPU operation is correct. Therefore, the manufacturing process constraints, CPU frequency is one of biggest obstacles to evolution. 2, front side bus bus is the source of information to one or more sections of the purpose of transmission to one or more components of a transmission length. Popular to say, that is, public connections between multiple components, used to transfer information between the various components. It is often expressed in MHz bus speed to describe the frequency. Many deviateent types of bus, front side bus of the English name is Front Side Bus, FSB is usually said, is the North Bridge chip connected to the CPU bus. The computer's front side bus frequency is shared by the CPU and north bridge chip decision. North Bridge chip (the motherboard will feature in the future to do explain) responsible for the contact memory, graphics and other data throughput of the largest components and bonds and Southbridge chips. CPU is through the front side bus (FSB) to connect to the Northbridge chip, and then through the north bridge chip and memory, graphics exchange data. FSB is the CPU and the outside world the most important channel to exchange data. Therefore, the front-side bus data transmission capacity, plays a character in the overall performance of the computer. If there is no front-side bus fast enough, then strong computer CPU can not be vitally improved overall speed. Maximum data transfer bandwidth, all the while depending on the width of the data transmission and transmission frequency, ie data bandwidth = (bus frequency × Data Width) ÷ 8. PC, now the front-side bus frequency can be carry outd, there is 266MHz, 333MHz, 400MHz, 533MHz,ghd hair straightener price, 800MHz several. The greater the front-side bus frequency, describeing the CPU and north bridge chip data transmission capacity between the larger, more CPU into full play the function. Current CPU technology developed rapidly, fast computing speed, while the front-side bus big enough, you can pledge there is enough data supplied to the CPU, the lower front side bus, will not supply enough data to the CPU, which limits CPU performance may play into the system bottleneck. FSB and the FSB frequency difference: front-side bus speed, referring to the CPU and north bridge chip between the bus speed, more substantive, said the CPU and external data transmission speed. The concept of the FSB, is based on the digital pulse signal based on the shock speed, that is, 100MHz FSB, especially a digital pulse signal of a hundred million times per second shock, it was more of the PCI and other bus frequency. The reason why the FSB Front Side Bus and confusing the two concepts, the main reason is that in the past for a long period of time (mostly before the advent of the Pentium 4 and Pentium 4 first appeared),ghd australia, the front-side bus frequency and the FSB is the same The. Therefore, often directly for the FSB said the front-side bus, eventually leading to such a misunderstanding. With the development of computer technology, it was found higher than the FSB needs front side bus frequency, so using a QDR (Quad Date Rate) technology, or other similar technology to achieve this goal. The principle of these technologies, similar to the 2X or AGP 4X, which makes the front-side bus frequency as FSB 2 times 4 times or more. Since then, the difference between front-side bus and the FSB only deviation to be seriously. In addition, the FSB, the more special is the AMD 64's HyperTransport. HyperTransport AMD was first proposed in 1999, a bus, AMD 64-bit platform with the release and promotion, HyperTransport more and more widely, it is increasingly well known. HyperTransport is a board-based integrated circuit interconnect technology designed to end the bus, it can be in memory controller, disk controllers, and PCI bus controller, providing higher data transmission bandwidth. DDR HyperTransport similar work, work in the 400MHz frequency, the equivalent of the transmission frequency of 800MHz. Also in the same HyperTransport bus to simulate two separate two-way data link for point to point data transmission, so in theory can be regarded as double the maximum transfer rate, with 4,8,16, and 32-bit high-speed serial link bandwidth function. At 400MHz, the two-step bus 4bit mode bandwidth 0.8GB/sec, 8bit mode bidirectional bus bandwidth 1.6GB/sec; 800MHz, the bi-directional bus bandwidth 8bit mode 3.2GB/sec, two-way 16bit bus bandwidth mode 6.4GB/sec, 32bit mode bidirectional bus bandwidth 12.8GB/sec. To 400MHz, the bi-directional 4bit mode, for example, the method of calculating bandwidth of 400MHz × 2 × 2 × 4bit ÷ 8 = 0.8GB/sec. HyperTransport there is a major feature, that is, when the data bit width is not 32bit, you can transfer data in batches to achieve the same impact with the 32bit. For example 16bit data can be transferred in two batches, 8bit data can be transmitted in four batches. Sub-transmission method such data, to the HyperTransport greater flexibility in the application space. 2004 年 2 月, HyperTransport Technology Alliance (Hyper Transport Technology Consortium) has officially released the HyperTransport 2.0 specification, the use of the Dual-data technology, the frequency of successful uprank to the 1.0GHz, 1.2GHz and 1.4GHz, data transfer bandwidth per channel 1.6Gb/sec by the elevated to 2.0GB/sec, 2.4Gb/sec and 2.8GB/sec, maximum bandwidth from the 12.8Gb/sec elevated to 22.4GB/sec. When applied to the memory controller, HyperTransport, in fact, also similar to the traditional front-side bus (FSB, Front Side Bus), so for the memory controller HyperTransport technology for CPU, its HyperTransport, FSB frequency can be equal to the frequency. 10, FSB FSB CPU and the whole computer system is the base frequency, in units of MHz (megahertz). In the early computer, the synchronization between the memory and the motherboard is equal to the speed of running FSB. In this way, can be understood as memory, CPU FSB directly connected to pass, to achieve synchronous operation between the two states. For the current computer systems, two doly different. However, the significance of the FSB remains, most of the frequency of computer systems are based on the FSB, multiplied by the multiplier to achieve a certain, this factor can be greater than 1, it can be less than 1. said processor FSB, it is necessary that the two closedownly related concepts: frequency and frequency, frequency is the CPU clock frequency; multiplier and FSB frequency ratio is multiples. Frequency, FSB, multiplier, and its relationshipship: frequency = FSB × multiplier. before the 486, CPU's frequency is still at a lower stage, CPU's frequency is generally equal to the FSB. Appear in the 486, befactor of the increasing frequency of CPU, and PC, some of the other devices (such as cards, rough forces, etc.) are subject to process constraints, can not afford the higher frequency, hence limiting the further improvement of CPU frequency. Resulting in a frequency doubling technology,ghd hair, which enables the internal operating frequency changes to the external CPU frequency multiplier, thereby enhance the frequency multiplier to achieve the purpose of upgrading. Frequency technology, is to make the external device can work in a lower FSB, while the CPU FSB frequency is a multiple. in the Pentium era, CPU FSB is generally 60/66MHz, starting from the Pentium Ⅱ 350, CPU FSB up to 100MHz, the current FSB CPU has reached 200MHz. Because under normal circumstances, the same FSB and memory bus frequency, so when CPU FSB was raised, and the exchange rate between the memory has been increased as a result, to increase overall computer speed greater impact. FSB and the front-side bus (FSB) frequency, it is easy to be confused. Front-side bus speed, referring to the CPU and north bridge chip between the bus speed, more substantive, said the CPU and external data transmission speed. The concept of the FSB, is based on the digital pulse signal based on the shock speed, that is, 100MHz FSB especially digital pulse signal of a hundred million times per second shock, it was more of the PCI and other bus frequency. The reason why the FSB Front Side Bus and confusing the two concepts, the main factor is that in the past for a long period of time (mostly before the advent of the Pentium 4 and Pentium 4 first appeared), the front-side bus frequency and the FSB is the same, so often called direct FSB FSB, eventually leading to such a misunderstanding. With the development of computer technology, it was found higher than the FSB needs front side bus frequency, so using a QDR (Quad Date Rate) technology, or other similar technology to achieve this goal. The principle of these technologies like AGP to 2X or 4X, which makes the front-side bus frequency as FSB 2 times 4 times or even higher, since then, the difference between front-side bus and the FSB only starting to be seriously. 3, CPU frequency multiplier, the full name is the multiplication factor. The core CPU FSB frequency and between, there is a relationship between the ratio, the ratio is the multiplication factor, referred to as frequency. In theory, the frequency is from 1.5 up to infinite. Note, however, the multiplier is 0.5 for a unit interval. Multiplying the FSB and multiplier, that is frequency. Therefore, any one to improve, can increase the CPU's frequency. the birthal concept is not frequency, CPU clock speed and system bus speed is the same. But with faster CPU speed, frequency doubling technology also came into being. It allows system bus working at relatively low frequencies, while the CPU speed can increase frequency to infinity. So the calculation of CPU frequency, becomes: frequency = FSB x multiplier. That is, the multiplier is the difference between the CPU and system bus in multiples of the same when the FSB,ghd price sydney, increasing the frequency, CPU frequency, the higher. 13, secondary cache capacity CPU cache (Cache Memoney) is located between the CPU and memory, temporary memory. Smaller than the capacity of its memory, but the exchange faster. Data in the cache only memory, a small share of the data, but this small character of the CPU will visit within a short time, when the CPU calls a lot of data, you can avoid the call directly from the cache memory to speed read speed. Thus, adding in the CPU cache is an efficient solution. Thus, the internal memory (cache + memory) becomes both the high-speed cache, there are large-capacity memory stofury system. CPU cache performance of great impact. Mainly because the CPU order and exchange of data between the CPU and cache bandwidth caused. cache works is when a CPU to read the data, the first from the cache to find, if found, immediately read and sent to CPU processing; if not found, use a relatively slow speed read from memory and sent to CPU processing, while the data block where the data transferred to the cache, you can make the future of the entire data is read from the cache for all, without having to call the memory. read mechanism is such that CPU reads the cache hit rate is very high (most of the CPU up to 90% or so), that is,ghd australia sale, CPU to read the data the next time 90% in the cache, only about 10% of the need to read from memory. This greatly saves time CPU memory read directly, but also the CPU reads the data base without having to wait. In general, CPU read the data in the order, is the first cache, then memory. the CPU cache before the first is a whole, and the capacity is very low, Intel Pentium era from the start, the cache are classified. Was integrated in the CPU core in the CPU cache has been insufficient to meet the requisition of the manufacturing process constraints, can not indispensablely improve the capacity of the cache. Resulting in an integrated circuit with the CPU with a cache on the board or motherboard, this time, put the CPU core integrated cache, called a cache. The outside is called the secondary cache. A cache, data cache is also divided (Data Cache, D-Cache) and instruction cache (Instruction Cache, I-Cache). Both are used to store data and execute instructions of these data, and both can be CPU access, reduce the contention caused by the conflict Cache to improve the performance of the processor. Intel Pentium 4 processor at the launch, also added a level trace cache with a capacity of 12KB. CPU manufactionuring mode with the advance of the secondary cache can be easily integrated in the CPU core, the volume also increased year by year. Now and then integrated in the CPU or not to define an internal, secondary cache, is imprecise. Moreover, with the secondary cache is integsized into the CPU core in the past, large gap between the secondary cache and the CPU frequency the situation is deviated, this time its the same speed in the frequency of the toil, can supply higher transmission speed CPU . CPU performance of the secondary cache is one of the crucials. In the case of the same CPU core to increase the secondary cache capacity, can greatly improve performance. The same high-low-end CPU core, divided, and often there are differences in the secondary cache. Thus, the importance of the secondary cache for the CPU. CPU found in the cache of helpful data is known as the \In theory, a secondary cache in a CPU, a cache read hit rate of 80%. That is, CPU cache to find useful data, accounting for 80% of the total data, the remaining 20% from the secondary cache read. Since we can not accurately predict the data to be performed, secondary cache read hit rate was 80% (read from the secondary cache of useful data of the total data, 16%). Well, there's data would have to call from memory, but this is a very small proportion. The current high end CPU, but also with a level three cache, it is to read the secondary data cache misses the kind of cache design, CPU cache in a three-level, only about 5% of the data needs call from memory, which further improves the CPU efficiency. CPU access in order to ensure a higher hit rate, cache contents should be replaced by a certain algorithm. A more common method is the \Therefore, the need to set up a counter for each line, LRU algorithm is to hit the line counter is obvioused, the other row counter plus 1. When the need to be replaced, the maximum out of the line counter counts the rows of data out. This is an efficient and scientific algorithms. The process counter is cleared, you can often call and then do not need some data out of the cache to improve cache utilization. CPU products, a cache of 4KB to 64KB in size between the basic, secondary cache capacity is divided into 128KB, 256KB, 512KB, 1MB, 2MB and so on. Cache capacity, little difference among products, while the secondary cache capacity, it is substantial to improve CPU performance. Upgrading the secondary cache capacity is determined by the CPU manufacturing process, the capacity increases will inevitably bring about the increase in the number of transistors within the CPU, the CPU in a limited area on the integration of larger caches, requirements for the higher manufacturing process . |
|
