There's no such thing as a reliable performance source - magazines/websites will either use manufacturer's data, their own testing, or a third party (and sometimes chop and change between all 3). And amongst that, unless all cars were tested in the exact same method at the same track at the same day, then results aren't exactly comparable (unlike power figures which are constrained by standards and tested on isolated controlled dynos). For instance, one magazine might test cars on the main straight of Eastern Creek (which is not quite flat), with 2 people in the car and a full tank of fuel. Another one might test at Avalon airstrip, which is flatter but has a much different surface (which will affect grip and thus launches), and with only 1 person in the car. For the most part, Wheels/Motor/Top Gear etc will use their own method relatively consistently (only using other figures if they are unable to test the car properly, or if weather was not conducive, and they'll usually say that in the review too), so the closest you'll get to a reliable source for comparisons is to stick with one particular media source.