1. 15
  1.  

  2. 4

    Why would you ever benchmark something by changing two variables at once?

    1. 7

      Despite the (very misleading) title, it doesn’t look like it was even really an attempt at careful benchmarking, just two tests run a year apart, differing by whatever happened to change on the author’s laptop in that year (one test was run now, and one is just pulling numbers from an old blog post). In that time, they appear to have upgraded OSX, changed filesystems, and applied Meltdown patches, among who knows what other changes.