最新的PC处理器能提升多少视频处理性能?

Back in 2009, 当时惠普推出了由Nehalem系列cpu驱动的新工作站, 性能提升是如此显著，以至于它们立即呈现了基于以前架构的过时工作站. 现在，英特尔已经推出了基于Ivy bridge的cpu，使早期基于nehalem的工作站的核心数量从4个增加到12个，增加了两倍, can we expect similar performance gains? That’s what I’ll explore in this article, 它比较了12核(24带HTT) HP Z800和24/48核HP Z820在编辑和流编码功能方面的性能.

The Z800, which I reviewed in EventDV in 2010, incorporated two 3.33GHz X5680至强处理器，24GB内存，运行64位版本的Windows 7. The graphics card was an NVIDIA Quadro FX 4800 with 1.5GB of dedicated memory and access to 3.5GB more system memory.

The Z820 (Figure 1, below) includes two 2.7 GHz E5-2697E cpu, 64 GB内存，运行Windows 7. 图形由NVIDIA Quadro K5000提供，具有4 GB的视频RAM. By virtue of its updated architecture, Z820还拥有比Z800更快的系统总线(8gt比6gt).4GT)，更快的内存(1866 MHz相比1333 MHz)和一个额外的内存通道(3 vs . 4GT). 4)，所有这些都有助于提高59的最大内存吞吐量.7gb /秒(GB/秒)，而Z800为32gb /秒.

Figure 1. The Z820 hasn’t changed much on the outside.

尽管盒子里有一些小的硬件差异, 惠普并没有为其最新一代工作站更换机箱. Put the Z800 next to the Z820 and obscure the product name, 只有最敏锐的观察者才能将它们区分开来.

Setting Expectations

All of my tests were rendering tests, 关键问题是Z820的速度比Z800快多少, 我用渲染时间减少的百分比来衡量. If the Z800 took 10 minutes to render a project, and the Z820 took five minutes, the Z820 cut rendering time by 50% (10 min-5 min/10). How much of a performance boost is reasonable to expect?

Let’s start with simple theory. Since the Z820 has twice as many cores as the Z800, performing the same work in half the time sounds reasonable, making the 50% number seem attainable. However, since the CPUs on the Z800 were about 20% faster (3.33 GHz compared to 2.7 GHz)，每个核心的运行速度应该提高20%左右，将50%的速度降低到40%左右. However, on tasks that involve lots of data, like the RED and 4K projects, Z820更快的内存带宽也会带来一些性能红利.

So, 理论上，Z820的运行速度应该快40-60%, depending upon the tasks. And this is actually a pretty decent starting point. However, 请记住，仅仅因为有48个核心并不意味着所有任务都有效地分配到这48个核心上. As an example, Figure 2 (below) 在Z820上显示Windows任务管理器的性能选项卡，同时在Adobe Media Encoder中使用VP6编解码器编码单个文件. 你可以在任何一台Windows电脑上通过三指礼(Ctrl-Alt-Delete)看到这个视图。, choosing Windows Task Manager, and clicking the Performance tab.

Figure 2. This view of Windows Task Manager makes Intel engineers cry.

Why is CPU utilization so low? 因为VP6编解码器是从On2(或者以前是On2)获得许可的，从多处理的角度来看，它总是非常低效, 这意味着它在可用时不能有效地利用额外的CPU. 这在很大程度上是因为VP6是在多核计算机广泛使用之前开发的, 并且在更新代码以利用多核之前就被搁置了.

None of my tests involved outputting to VP6. 高层次的观点是，多核效率因程序而异, and even task to task within a program. 如果一个任务从多核的角度来看效率特别低, the faster clock speed on the Z800’s CPUs (3.3 Ghz)将比CPU速度较慢的Z820上的额外核心(2 Ghz)更大的优势.7 Ghz). In addition, 即使程序在多个核上有效地分割操作, this involves some overhead and management, 哪个从正在进行的渲染或其他操作中窃取百家乐软件.

For all these reasons, it’s not surprising when a particular program, or function within a program, 没有接近于获得理论上的性能优势，额外的核心似乎是可用的. 对于像Premiere Pro这样的应用程序来说尤其如此, which uses a range of third-party codecs to work with DV, HDV, AVCHD, 以及各种输入格式所呈现的其他编解码器的字母汤. 因为程序永远不会比它最慢的操作快, if these codecs are inefficiently written, they can slow the entire operation.

好了，现在我们的期望已经设定好了，接下来看看我们的测试.