Are the accelerometers in older iPhone/iPad/iPod models comparable to current models in terms of sensitivity/accuracy (such that no perceivable difference in readings)? For example, if someone takes accelerometer readings on their iPhone 3GS, and I take readings with an iPhone 4S, and someone else does it with an iPad, assuming all else equal, will we get similar results? What kind of deviation would one see? For example, let's say you put all the devices on a table that is vibrating enough to show up on an accelerometer. Would different device versions show results within the same level of accuracy? What are the limitations? I know that the older iPhones have an older version of the accelerometer, and although I found their datasheets, I could not tell whether the differences were significant. Of course I can do an experiment to determine if their are perceivable differences in my application, but I was wondering if anyone had prior experience with this.