Speed Answer: Typically, microprocessors (like the ones found in computers), measure their speed in hertz.

A hertz is the measurement of a cycle in a second. So, 10 hertz means 10 cycles per second. In computing, a cycle (or, more specifically, a clock cycle) is the basic unit of measurement that the CPU uses to carry out instructions given to it by software. Therefore, in a CPU running at 900MHz, 900 million clock cycles will occur per second. Software sends commands to the processor called, instructions. These commands are the basis for how all programs run on a computer and are handled by the computer in a very complicated manner. However, a computer running at 3GHz, for example, is not performing 3 billion instructions per second. Some instructions take multiple cycles to complete and some can even have other instructions in the same cycle simultaneously. To complicate matters further, it is not accurate to say that a higher speed processor is better than another one at a lower speed. Certain AMD processors, for example, run at lower speeds than comparable Intel processors of their family but, because they use different architecture, perform at the same (and, sometimes, higher) performance levels than CPU's with high clock speeds. Also, processors with some sort of Hyper-Threading technology or, better yet, multiple cores (like Intel Core 2 Duo processors) will be rated at lowered speeds than other CPU's in their price range but, because of more than one (virtual) processor is running parallel to the others, more instructions are performed per clock cycle.

Read more: http://wiki.answers.com/Q/How_is_a_microprocessors_speed_measured#ixzz1Si0hHOum

There are many microprocessors available to the public. Not knowing the differences can be quite frustrating -- especially when it means saving or spending a couple hundred dollars.

Below is a chart that compares and contrasts important features found on some of the more popular chips in the market today.

Transistors CPU Speed Celeron 7,500,000 1.06 GHz - 2 GHz

L2 Cache 256 KB, full speed 512 KB, half speed 256 KB, full speed

Front-Side Bus Speed 133 MHz and 400 MHz

Pentium II

7,500,000

233 MHz - 450 MHz

100 MHz

Pentium III

9,500,000

450 MHz - 1 GHz

133 MHz

Pentium III Xeon 28,100,000

500 MHz - 1 GHz

256 KB - 2 MB, 100 MHz full speed 256 KB, full speed

Pentium 4

55,000,000

1.4 GHz - 3.4 GHz

800 MHz

K6-II

9,300,000

500 MHz - 550 MHz N/A 256 KB, full speed 256 KB, full speed 384 KB, full speed 64 KB, full speed

100 MHz

K6-III

21,300,000

400 MHz - 450 MHz

100 MHz

Athlon (K7)

22,000,000

850 MHz - 1.2 GHz

200 MHz and 266 MHz

Athlon XP

37,500,000

1.67 GHz

266 MHz

Duron

N/A

700-800 MHz

200 MHz

PowerPC G3

6,500,000

233 MHz - 333 MHz

512 KB, 1 MB, half speed 1 MB, half speed 1 MB, half speed 512 KB

100 MHz

PowerPC G4

10,500,000

400 MHz - 800 MHz

100 MHz

Athlon 64

105,900,000 800 MHz

1.6 GHz

G5

58,000,000

2.5GHz

900MHz - 1.25GHz

What are the applications of microprocessors?

Answer: Microprocessors are mass storage device.They are the advanced form of computers. They are also called as microcomputers.

The impact of microprocessor in different lures of fields is significant. The availability of low cost, low power and small weight, computing capability makes it useful in different applications. Now a days, a microprocessor based systems are used in instructions, automatic testing product, speed control of motors , traffic light control , light control of furnaces etc. Some of the important areas are mentioned below: Instrumentation: it is very useful in the field of instrumentation. Frequency counters, function generators, frequency synthesizers, spectrum analyses and many other instruments are available, when microprocessors are used as controller. It is also used in medical instrumentation. Control: Microprocessor based controllers are available in home appliances, such as microwave oven, washing machine etc., microprocessors are being used in controlling various parameters like speed, pressure, temperature etc. These are used with the help of suitable transduction. Communication: Microprocessors are being used in a wide range of communication equipments. In telephone industry, these are used in digital telephone sets. Telephone exchanges and modem etc. The uses of microprocessor in television, satellite communication have made teleconferencing possible. Railway reservation and air reservation system also uses this technology. LAN and WAN for communication of vertical information through computer network. Office Automation and Publication: Microprocessor based micro computer with software packages has changed the office environment. Microprocessors based systems are being used for word processing, spread sheet operations, storage etc. The microprocessor has revolutionized the publication technology. Consumer: The use of microprocessor in toys, entertainment equipment and home applications is making them more entertaining and full of features. The use of microprocessors is more widespread and popular. Read more: http://wiki.answers.com/Q/What_are_the_applications_of_microprocessors#ixzz1ShVVCtvF

3.2 Optimizing On- and Off-Chip Caches

Undeniably, on-chip memory can be expensive in terms of die area and cost. Smart designs can be employed to minimize AC power consumption, but cache does contribute to static power. On the positive side though, doubling the L2 cache to 1 MB, for example, can boost performance by 30% to 100%, depending on the application. Even a 30% increase for a device change that requires almost no engineering changes is often worthwhile. Of course, the benefit depends on how often the instruction code fits in L2 cache during a particular application. Performance increases can be even higher if the L2 runs at the core clock rate. The increased static power is on the order of 1 W, which most designs can afford, considering the advantages. Given the costs, additional levels of off-chip memory are hard to justify, especially when one considers that an off-chip cache such as an L3 must be 8 to 16 times larger than the L2 to significantly increase performance. More often than not, the expense of off-chip memory, in terms of die area for the tags, increased pinout packaging, cost, and power, outweighs any performance gain.

3 Integration
As core and memory technologies advance and operation frequencies increase, the surrounding system logic bus limits performance due to board signaling issues at higher speeds. Therefore, external buses are slower than their on-chip equivalents, creating bottlenecks. The difference between internal and external bus speeds, coupled with the intermediate logic delay, further exposes bandwidth deficiencies and latencies of devices operating at higher frequencies. It is a mistake to integrate a high-performance core on a device that does not incorporate enough peripherals and interfaces to make full use of the cores computing power. Pulling the majority of the system logic on chip also reduces board design complexity and offers compatibility protection in the sense that the core, the caches, the peripherals, and the interfaces are truly integrated by the devices vendor.

Integration

Figure 1 compares the performance-to-power ratios for a single core, a dual core, and a single core with doubled frequency.

Single- core F1
Power Performance

Dual- core F1 Power Performance Single- core 2 X F1

Power Performance

Figure 1. Performance vs. Power

CPU Cache, Clock Speed, and Bus Explained

Once upon a time, a processor's clock speed might have been the most reliable indicator of performance, but these days, it's just one of many important factors to consider. It's only meaningful, in fact, when compared to other CPUs in the same series; a quad-core 2.5GHz Core 2 Quad Q9400 from Intel will trounce a 3GHz Core 2 Duo E8400 in many tasks, for example. Those multiple processing cores are more important than the 500MHz difference in clock speed. Another important factor is a processor's L2 cache, which stores memory data and speeds up operation by making recently accessed data immediately available to the processor. Generally, the larger the L2 cache, the bigger the performance increase you can expect, at least with Intel CPUs that utilize one larger shared cache. AMD processors, on the other hand, give each core its own cache. Front-side bus (FSB) speed is the rate at which an Intel processor communicates with the motherboard's memory controller. A high FSB speed boosts the performance of RAM-intensive operations, thus cutting the time it takes for data to move between the CPU and the memory controller. The upper limit of FSB speeds today on Intel's highest-end Core 2 Extreme processors is 1,600MHz, while older processors have FSB speeds of 800MHz, 1,066MHz, or 1,333MHz. Make sure your motherboard supports the FSB speed of the processor you want to buy, however, so you can avoid any potential performance bottlenecks. An AMD CPU doesn't have an FSBinstead, the memory controller is located right on the chip, and the memory transfer takes place over the HyperTransport bus link between the processor and the memory.