So now that we’ve built and programmed our very own CPU, we’re going to take a step back and look at how CPU speeds have rapidly increased from just a few cycles per second to gigahertz! Some of that improvement, of course, has come from faster and more efficient transistors, but a number hardware designs have been implemented to boost performance. And you’ve probably heard or read about a lot of these - they’re the buzz words attached to just about every new CPU release - terms like instruction pipelining, cache, FLOPS, superscalar, branch prediction, multi-core processors, and even super computers! These designs are pretty complicated, but the fundamental concepts behind them are not. So bear with us as we introduce a lot of new terminology including what might just be the best computer science term of all time: the dirty bit. Let us explain.

现在，我们已经构建并编程了自己的CPU，我们将后退一步，看看CPU速度是如何从每秒几个周期快速提高到千兆赫兹的！当然，其中一些改进来自更快、更高效的晶体管，但为了提高性能，已经实施了一些硬件设计。你可能听过或读过很多这样的东西——它们是几乎每一个新的CPU版本都附带的流行语——比如指令流水线、缓存、触发器、超标量、分支预测、多核处理器，甚至超级计算机！这些设计相当复杂，但背后的基本概念并不复杂。