OS X 10.10 vs. Ubuntu 15.04 vs. Fedora 21 Tests: Linux Sweeps The Board

With having a new Apple Mac Mini in our testing labs this week, I ran some basic benchmarks comparing Mac OS X 10.10.2 to Ubuntu 15.04 to Fedora 21 in a few different configurations.

The new Mac Mini up for testing has a 1TB Apple hard drive, 8GB of DDR3-1600MHz memory, and has an Intel Core i5 4278U processor (dual-core with Hyper Threading, 3.1GHz Turbo) with Intel Iris Graphics. While this is a Haswell CPU, the Iris Graphics make this petite computer still an interesting test target.

I'll have more concentrated OS X vs. Linux benchmarks out soon, but in this article I just ran benchmarks stressing various subsystems under the following conditions:

- OS X 10.10.2 as loaded on this Late 2014 Apple Mac Mini with Xcode 6.2 providing LLVM Clang 3.5 SVN. The Journaled HFS+ file-system remains the OS X default.

- Ubuntu 15.04 in its current development state with the Linux 3.19 kernel, GCC 4.9.2, and Mesa 10.5.0.

- Ubuntu 15.04 when upgraded to the Linux 4.0 Git kernel as well as the Oibaf PPA to gain Mesa 10.6-devel and other updated graphics components.

- A clean install of Fedora 21 with all stable upgrades that led it to the Linux 3.19.1 kernel, GCC 4.9.2, and Mesa 10.4.3.

- Fedora 21 when switching from the default GCC 4.9 compiler to instead using LLVM Clang 3.5 as packaged for F21.

This is just an assortment of test configurations I found interesting for the initial tests of this Apple Mac Mini.

All of the benchmarks under both OS X and Linux were facilitated using the open-source Phoronix Test Suite benchmarking software. All of the hardware was the same throughout testing: the reported differences on the automated table above just come down to differences in what the OS reports, such as the difference between the CPU base frequency and turbo frequency, etc. On the following pages are the initial results with more interesting data points to come shortly.