In my experience the timing jitter on a general-purpose PC - the typical difference between runs of the same code on the same data - is in the ballpark of a couple of percent (and no, back-to-back runs aren't fully independent of each other so you can't average it out over a couple of dozen runs).

The claimed timing difference was less than one percent so while I certainly can't say it wasn't real, I'm not seeing any evidence that it was.