浮点运算

浮点运算是高性能计算研究中的一个重要领域。

The floating point arithmetic is very important for high performance arithmetic .

浮点运算很少是精确的。

Floating point arithmetic is rarely exact .

环境通常以每秒浮点运算次数（FLOPS）加以度量

HPC environments are often measured in terms of FLoating point Operations Per Second ( FLOPS ) . HPC

在定点DSP上实现浮点运算

Floating Point Implementations on Fixed Point DSP Architectures

FPGA中高速浮点运算的设计与实现

Design and Implementation of High-speed Floating-point Operations in FPGA

利用FPGA技术，能方便灵活地设计出浮点运算器。

Using FPGA technology may design a floating point calculator conveniently and quickly .

乔布斯：让我来说这个故事吧，我们得到很多反馈，用户希望Basic操作系统能支持浮点运算，我们都在求沃兹尼亚克求求你！

Jobs : Let me tell the story . We are getting a lot of input that people want this BASIC to be floating point . Please !

而性能较高，尤其适合于浮点运算的DSP处理器正是最佳选择。

Then , DSP processor , higher performance and especially suitable for floating-point calculation , is the most qualified choice .

IEEE浮点运算下三角函数的表驱动方法

Table - driven Implementation of the Trigonometric Functions for IEEE Floating Point s

该记录仪充分发挥了DSP具有浮点运算和处理速度快的优势。

The Electrocardiogram takes fully the advantage of DSP that has the float point computation ability and a rapid process speed .

在具有高速浮点运算能力的DSP芯片上进行高效率编程，生成转子匝间短路在线监测系统。

The integrated efficient diagnosis program at high-speed floating point DSP chip is made to produce a set of on-line inter-turn short circuit monitor system .

定点DSP在做浮点运算时效率不高，影响了定点DSP的应用。

The fixed point DSP is always criticized for its poor performance on float point calculation , which exerts negative influence on the application of fixed point DSP .

相对于CPU，GPU拥有更大的数据传输带宽，更多的计算单元，更强的浮点运算能力。

Compared with the CPU , GPU has greater data transfer bandwidth , more computing units , and more floating-point computing power .

由于计算机中整数运算远快于浮点运算，所以DCT算法采用整数运算，并且通过矩阵变换来减少乘加次数，提高了运算速度。

Through the matrix linear transform the number of operation for multiplications and additions is reduced , so that the speed is greatly raised .

随着硬件技术的发展，图形处理器（GPU）的实际浮点运算性能远远超过同期的CPU。

With the development of hardware technology , the graphical processing units ( GPU ) is far beyond the CPU in floating - point performance .

目前，随着信息应用领域对数据运算精度要求的不断提高和数值运算范围的不断扩大，使得在当今CPU的设计中，浮点运算单元的研究显得越来越重要。

With the increasing demands of digital precision and extending the scope of digital operation , the study of Floating-Point Units in modern CPU design has become more important .

对采用数据采样位置控制方式的三联动数控系统，其实时采样插补程序采用C语言编制，浮点运算。实时数据库平台是实现电厂管控一体化的桥梁。

This paper introduces a method to use C language and floating point calculation to design real time interpolation program in which the CNC system is controlled by data sampling position feedback and three axes can be controlled simultaneously .

这个函数的用途很明显，即处理浮点运算、NaN以及正0和负0的特殊情况。

The reason for this rather obvious function is to handle special cases of floating-point math , NaN , and positive and negative zero .

本论文的另一项工作就是利用GPU强大的浮点运算能力和高度并行的计算特性对本算法进行求解。

It is another work of the paper , using the powerful GPU floating-point computing power and highly parallel computing features to solve this algorithm .

Linpack是目前测试机群浮点运算性能的通用标准。

Linpack is the benchmark of testing the float-point performance of the cluster .

本论文创新地设计了一种基于流水线结构的多个浮点运算器并行计算的IP硬件体系结构，在有限的硬件资源上实现了速度和面积的最优。

This dissertation newly designs a hardware-based architecture for IPs . The architecture based on pipeline structure , has multiple floating point calculators computing at the same time . Speed and area are balanced in the limited hardware resources .

系统前端加了CMOS图像采集芯片OV7120，实现了32位浮点运算和8位数据的采集。

A CMOS image sensor ( OV7120 ) has been added to the system to implement image collecting .

近几年来，GPU计算开启了并行计算的大门，GPU非常适合效率高计算成本低的浮点运算。

In recent years , GPU computing opens the door of parallel computing , GPU is ideal for floating-point operations that require high computational efficiency and low cost .

采用MPU进行浮点运算时的舍入误差现象的分析，提出了一种减小舍入误差的树型分析法。

The author analyses the rounded errors in floating operation with MPU , put forward a tree analytical method for subtracting rounded errors .

浮点运算在MSP430系列微控制器中的应用原因是这些快速算法虽减少了计算复杂性，但在计算过程中仍需要大量浮点运算。

The reason beyond this is that those fast algorithms do reduce the complexity of computing but still one needs to use floating-point arithmetic operations in the computation process .

ICT的压缩性能与DCT相似，但是ICT只有整数运算，而无浮点运算，因此计算复杂度更低，并且精确的整数变换避免了编码器和解码器反变换之间的误匹配问题。

Compared to DCT , ICT has the similar compress performance , only need integer calculation , has low complexity for hardware implementation and no miss matching problem between the encoder and decoder .

数字信号处理器TMS320的浮点运算实现

The Implementation of Floating-point Arithmetic on TMS 320

采用精度敏感的圆周率计算BBP算法验证了CCRG编译器在扩展双精度浮点运算上的正确性。

The precision-sensitive BBP algorithm for the computation of Pi is used to verify the CCRG compiler . 2 .

图形处理器在硬件设计方面具有：浮点运算能力强大，适合大规模、高计算密度并行数据处理等特点。GPU通用计算技术就是利用图形处理器的这些特点去完成那些非图形处理领域的通用计算任务。

GPU general purpose computing is based on the features of graphics processing hard-ware , using its High FLOPS , suitability for large-scale and high-density parallel data pro-cessing to fulfill the general purpose computing task in non-graphical field .

GPU非常适合处理那些能够表示为并行计算的问题，并行计算的算术计算密度非常高，且浮点运算能力大大超过CPU，相同计算量GPU的硬件成本也大大低于CPU硬件成本。

GPU is suited for parallel computing that can be expressed as the problem of parallel computing arithmetic density is very high , and floating-point computing power much higher than the CPU , the same amount of GPU computing hardware cost significantly lower than the CPU cost of hardware .