Introduces the next generation of processorsYiming XiaoAs the need for microarchitecture performance modeling power level will continue to work with future workload and increasing performance requirements, the designer must make the right choices in defining the next generation of low-power microprocessors. In this article I will mainly talk about the next generation of processors and the processors used in smartphones and how they are designed in different situations. next generation processorFirst, the design team at the beginning of the design cycle, consider limiting power consumption and heat dissipation, it is possible to use an inherently low-power microarchitecture line. This more than the teams will benefit from some advantages. Additionally, power consumption has become a major limiting factor in microprocessor design. At the lower end of the performance spectrum, the main design concern has been one of power rather than performance (execution time). Battery and design team forced by system costs to consider power performance in this case. However, increased power is a critical issue in the design of the desktop and server markets. Here we present the use of energy performance simulation in the early stages of the design method with saving functionality. We looked at some of the new paradigms in processor design and their intrinsic motivation, comments on performance characteristics. The most effective way to reduce processor power consumption is a method designed to operate at a high frequency: a reduced voltage (and therefore frequency). This is the main control mechanism of the Transmeta Crusoe chip for . There is a limit, however, that can reduce the number of VDDs (for a given technology), namely because the circuit can be......half of paper......replace the desktop processor as the development of innovative microprocessor technology. These processors essentially have the greater power, command execution, and support of multi-core processors. Many visual computing tasks have two common characteristics: 1) window used repeatedly for scrolling and 2 image data) involving a large number of data-dependent branch operations. For example, most, if not all, algorithm point detection methods use a sliding window to apply a filter cartridge to each pixel in the image. Many decision tree algorithms involve many judgments to decide based on the current results of the next steps. Both types of calculations will not cause problems with most desktop CPUs, they may cost a mobile processor Assets: No sliding window data-driven access pattern and cache size encountered among compact mobile CPUs .