Well, I "splurged" and bought it. I've thought many times of getting a GTech Pro RR for autocross and trackday feedback, but never felt the need to spend the $300 for it. The SS version is cheaper at $200, but for my purposes would just be a toy without much practical use. So only
thirteen bucks for the toy seemed like an easy decision.
By the way, I have a first-gen iPhone. So far, playing with it in my living room, I don't see any functionality that requires GPS.
I've never been to a drag strip and don't plan on it, so I have no real way of determining if it's accurate or not. I'll leave that up to the professional auto sites/mags.
In the meantime, I'll answer some of the questions and confusions about the product and the principles behind it.
First, you initiate a test run by hitting the start button, but the recording doesn't actually begin until it passes a certain g-force threshold. The default is 0.1 G's. Meaning you can secure the iPhone, hit the start button, put your gloves on, tighten belts, get everything ready and whatnot, then drop the clutch.
Second, GPS is not required to calculate your speed or distance. The software knows the time and acceleration, and using highschool physics concepts can figure out the distance and speed bits.
Third, the acceleration runs end at 1/4 mi. No top speed runs with this.
Fourth, the skidpad function only shows the max for each of acceleration, braking and cornering. There's no g force trace, so you can't see how often you were at certain accelerations, say during an autocross run.
Fifth, using it in a pocket won't work - it would move around too much, and it must be in-line with the car (vertical or flat, doesn't matter.)
Sixth, a Lamborghini would do about 0.95 g's cornering depending on tires and road surface IIRC. No idea what it's acceleration g-force would be, I'd guess somewhere around 0.6 or so?
Seventh, when you enter a car into your garage, you input its weight (it's TOTAL weight including driver and all cargo.) Based on that, and the measured acceleration, the software can do some math to figure out how much force was required, and how much power was needed to generate that force. If your weight is off, you won't get a "true" horsepower figure, and if the weight fluctuates any (a full tank of gas can weigh nearly 100 lbs depending on capacity, for example) you'll get different readings. So it's unlikely to be significant enough to tell you if that intake
really gave you 10 hp or not, but it's fun to play around with.
Eighth, the reporting is rudimentary so far. It'll give you its 1/4 mi timeslip, but so far I don't see any way to get a hold of the g-force/hp/mph graph to view.
Finally, I'd love to see this software developed further with roadrace/autocross elements.