LinuxInsider: Macs Are Faster

September 30th, 2004 at 3:00 PM - News by Bryan Chaffin

LinuxInsider has published an interesting editorial from Paul Murphy who asks which is faster, Macs or PCs? In answer to that question, he says: "The short answer can be based entirely on raw hardware capabilities, and that answer is pretty simple: the Mac wins hands down." He goes on to say:

For example, the NCSA "Tungsten" cluster computer built last year was recently upgraded to include 2,500 dual Xeon Dell Poweredge 1750 servers at 3.2 GHz. According to NCSA public affairs, this thing has a theoretical peak capacity of about 32 Tereflops and yields about 15.36 teraflops in operation -- meaning that each CPU contributes about 3.1 Gigaflops to actual throughput.

In contrast, the cluster built last year at Virginia Tech using 1,100 Mac desktops has a theoretical peak of about 18.2 Teraflops and initially benchmarked at 8.1 Teraflops to deliver a contribution of 3.7 Gigaflops per CPU.

Although that was 19 percent better than the most recent Dell Xeonis, later machines built with Appleis X-Serves do much better because they have fewer I/O bottlenecks. Thus the Mach5 cluster built by Colsa Corp. and the U.S. Army, uses 1,566 dual CPU X-serves to deliver an expected 15 Teraflops in sustained throughput. Thatis 4.8 Gigaflops per CPU -- more than 50 percent faster than the Xeon -- and thatis with last yearis 2.0 GHz G5.

It can be argued, of course, that this comparison isnit fair because the Xeon is an older 32 bit processor and not generationally comparable to the G5. It might be better, therefore, to compare the X-serve cluster to machines built using a more modern 64 bit processor like AMDis (NYSE: AMD) Opteronis. "Lightning," at Los Alamos National Laboratory , uses 2,816 Opterons to produce a peak of 11,264 or 4.0 Gigaflops per CPU -- 30 percent better than the Xeons, but still 20 percent less than the G5s.

The article in its entirety goes on to address many of the issues on what makes such comparisons difficult, as well as other related issues. Included in those issues is that despite his quantitative answers, many questions regarding which platform is better simply can not be unresolved. We recommend the full article as a good read.