...sorta. More like the semi-definitive/your results may vary/it worked for me/not exactly scientific, old guy's observations on fuel consumption under the following conditions:
the ride: 06, v6, auto, K&N drop in, Gibson cat back.
the driver: old but not slow, not usually in a hurry but drive with a purpose, don't really care how fast you can pass me, do nail it thru a couple of gears on a regular basis.
the route: mostly rural roads and state hiways, in and out of town every day, not many stop lights. No city driving and maybe 50 miles a tank on the hiway.
the conditions: 5000'+, not many flat roads around, midrange octane = 87, hi test octane = 91, AC on when needed (most afternoons), I generally let the trans shift between 2,000 & 2500 rpm, no cruise control, regular fuel..no oxygenated (oxy-junk) fuel.
the quest: salesman said 'put reglar in her grandpa', book said midrange (with hi test for 'added' performance), Service manager said computer not that smart...mid range. So..does Hi test make a difference?
the test: 3 tanks of mid range; 3 tanks of hi test (and a tank of 93 octane at the end...scarce and pricey.....just for shitsNgrins). Same station/brand for each.
the results: mid range averaged 19.5 (19.2, 19.9, 19.5) give or take a tenth. hi test averaged 20.0 (19.9, 20.0, 20.2) give or take a tenth, 93 octane? save ur $$...19.5 (service manager partially right..computer not that smart).
observations: hi test did do better (if you get better mileage with a different octane, under similiar conditions...the motor's happier with that octane) The $$ are close to a wash...5 tenths better = 10 miles more per tank @ 20 gallon fill. At $3 a gallon that a buck 50; hi test here is 10 cents a gallon more or 2 bucks more a 20 gallon fill...your call. But I can think of a lot dumber ways to blow 50 cents. But then of course your results may vary.....