Computer CPU Common Sense

1.CPU frequency higher performance?





not necessarily. This is in many novice among the misunderstanding. CPU performance depends on a number of comprehensive parameters, not necessarily the root frequency proportional. Of course, in the same series of CPUs, such as Celeron 4, the higher the frequency of high performance.





2. Why Celeron High-frequency Low-energy? And the Fast Dragon series Low-frequency high-energy?





say the question before you talk about what is called caching. Cache (English name cache), a cache memory, is a small but high speed memory located between the CPU and the main memory dram (that is, the hard drive, the burner cache, in fact). CPU Cache Two, one is the internal cache, also known as a cache (L1 cache): Closed in the CPU chip inside the cache, used for the temporary storage of CPU operations in some of the instructions and data, access speed and CPU frequency consistent. The larger the L1 cache, the less time the CPU can exchange data between the L2 cache and the slow access rate, and the faster the computing speed of the computer, the external cache, also known as Level two data cache (L2 cache): CPU external cache, now the processor's L2 The cache is run at the same frequency as the CPU (the previous P2 P3 level two cache runs at half the CPU frequency). 3lian.com




The basic architecture of the
Celeron is similar to its current Pentium, but its FSB is low, the front side bus is low, and the cache is severely shrunk compared to the Pentium series (Northwood Core Celeron 4 's two cache is only 128K, while the Northwood core P4 two-level cache has 512K). Reducing the cache by three-fourths greatly reduces costs, but it also causes a sharp drop in CPU capacity. And the Fast Dragon series of cache up to 128k,ta, TB core of the Fast Dragon two cache for 256k,barton and the core of the Fast Dragon two cache reached 512K, coupled with its more accurate instruction branch prediction and three-way data validation (or triangle data check loop), So the processor does not work very well, but the performance is excellent.





3. Why is my CPU temperature so high?





The first is the radiator problem, in addition, the chassis is not well ventilated, may be the power cord and data line affected the ventilation. Also do not have to the CPU temperature too sensitive, some friends see fifty or sixty degrees is frightened. In fact, not so exaggerated, in general, the CPU under 75 degrees can be safe to work (usually considered safe operating temperature = limit operating temperature of 80%).





4. Why is my CPU FSB only 100 while others are 400, 533 and so on?





ask these kind of questions the friend all has a very big misunderstanding, that is they FSB and the front-end bus concept to confuse. FSB is the base clock frequency provided by the motherboard for the CPU, typically 100, 133, 166, 200. And we say the FSB (Front system bus) refers to the front of the system, it is the processor and the motherboard North Bridge chip or memory control hub between the data channel, the common frequency is 400, 333, 533, 800. As a novice, you don't have to master so many conceptual things, just remember the following few formulas:





frequency = FSB * Frequency multiplier (MHz)





Intel CPU Front-End bus = FSB *4 (MHz)





AMD CPU Front-End bus = FSB *2 (MHz)





CPU Data Bandwidth = Front-End bus *8 (MB/s)





Memory Bandwidth = memory equivalent operating frequency *8 (MB/s)





5. What is a hyper-threading? is hyper-Threading useful to me? Why am I using Hyper-Threading CPU system performance not getting much improvement?





Hyper-Threading Technology (HT) is the use of special hardware instructions, the multithreaded processor inside the two logical cores into two physical chips, so that a single processor can "enjoy" thread-level parallel computing processor technology. Multithreading technology can be used to support multi-threaded operating systems and software, effectively enhance the processor in multitasking, multithreading processing capacity. The simple thing is to simulate two CPUs to work.





CPUs with Hyper-threading technology are significantly more capable of multitasking than non-hyper-threading CPUs, but do not have much of a performance advantage in the task of single tasks, and even have a slight decrease in performance when running software that does not support hyper-Threading technology. In general, hyper-threading CPU is mainly used in high-end machines and servers, ordinary home or office machines, if there is no special requirements, do not need to use HT.





about a lot of friends reflecting the use of Hyper-threading CPU performance is not a big problem to explain, there is a misunderstanding, many friends think that as long as the use of Hyper-threading CPU can use Hyper-Threading technology, in fact, is not the case. It takes five basic conditions to bring the power of the hyper-threading into play





a) CPU to support HT, currently support hyper-Threading Technology CPU has P4C series, P4E series and some models of Xeon





b) motherboard chip to support HT, which is easily overlooked one of the conditions. Currently, the motherboard chipsets that support HT are mainly Intel i925/i915/i875/i865 full range, VIA pt800/pt880, SIS 655fx/655tx/661fx, ATI 9100IGP.





C memory requires a dual-channel DDR400. Due to the hyper-threading of the CPU front-end bus up to 800MHz, data bandwidth up to 6.4gb/s, so the requirements of memory bandwidth must also reach 6.4gb/s, to avoid the emergence of system bottlenecks. Single-channel DDR400 bandwidth is only 3.2g/s, and dual-channel technology can increase its bandwidth by one times to 6.4g/s to meet the needs of hyper-threading CPU. I've seen a lot of friends get 848 with P4C, which is one of the reasons that hyper-threading performance doesn't work. (By the way, i848, PT800 only support Hyper-threading, but do not support dual channels, which some friends call it the reason for the chicken ribs)





d) Support for the operating system. This is the prerequisite that is neglected most by people! Most friends think that as long as the hardware to meet the needs of Hyper-threading is OK, in fact, not so, careful operating system will be a mistake! Currently, the operating system that supports hyper-Threading technology is extremely limited, and only the operating system after WinXP Professional Edition (SP1 patch) supports Hyper-Threading technology, and Windows2000 can also support hyper-threading with the latest SP4 patches.





e) Support for application software. In general, as long as the ability to support multiprocessor software can support hyper-Threading technology, but in fact such software is not many, and biased in favor of graphics, video processing and other professional software, game software very few support. Application software has Office 2000, Office XP, and so on.





in fact, using Hyper-threading is not as simple as people think.





6. What is called overclocking? What's the use of overclocking? Is it harmful to the system? How to overclocking?





the so-called overclocking, is to let the CPU (video card, memory, and then omitted, only CPU for example) work at a higher than its rated frequency, to extract its potential and performance.





overclocking can improve the performance of the CPU, for example, the frequency of P4 2.4 overclocking to 2800MHz, you can let it reach P4 2.8 performance level, so that less money with good things.





Reasonable overclocking on the system will not be too much harm, but only a large number of heat, if the excessive, will seriously shorten the life of accessories, and even burned.





overclocking requires BIOS support for the motherboard, but depending on the motherboard and CPU as well as the cooling system. The most common way: into the BIOS, and then into the "frequency/Voltage control" (the exact name is not necessarily the same, generally with frequency words are, and not necessarily all bios have this), and then the CPU Host Clock control (processor frequency controls) The value of an item is set to Enabled for specific adjustments. Now the CPU is locked frequency multiplier, so we can only tamper from the FSB, FSB a name called CPU Host Frequency, the right side is the corresponding value. Overclocking should be gradual, not once too much, to observe the temperature at any time, adapt to find the most suitable for their own frequency.