== What now? And... ==

"When first released, the Dhrystone benchmark fulfilled a useful function – at least it gave an alternative indicator to vendors’ literal MIPS ratings. However, almost more than twenty years later, there are undoubtedly better benchmarks available for measuring processor performance. Code profiling demonstrates that Dhrystone is not a particularly good predictor of real-world performance, especially for complex embedded systems. The benchmark’ssusceptibility to being optimized, yielding large variations in results, is a considerable weakness if the intent is to use the numbers for comparative purposes.  In general, more representative data will be obtained by using a suite of benchmark data.First, this enables benchmarks to reflect the requirements of specific application domains.Second, the effects of any one optimization will be mitigated by the benchmark having awide spread of algorithms and coding styles. Results based on such an approach are morelikely to be representative of processor performance than compiler or tool chainperformance."