As anecdotal evidence, their results seem to be indicative of the real-world browser performance I'm seeing with my (largish) application. There is a perceptible difference between Safari, Chrome and Firefox, in that order.<p>I really hope that Firefox catches up with JavaScript performance, BTW.