That's a good point, but I think as a source of error it's pretty small. I've never had a rollout figure more than 5mm different from the ones you'd get form the chart. As a source of error, that's only +/-0.25%.

BTW, the bigger issue for the iBike with changing tire pressure is that it can change rolling resistance, and if there's a difference front to back, it can also cause the bike to tilt slightly under your weight. So long as I kept my tires in the 90-120psi range, where rolling resistance curves are fairly flat, I didn't see much problem from rolling resistance changes. But if for some reason I had substantially different pressures in the tires, that could produce a noticeable change in power. It only happened if one tire had a slow leak, or if I fixed a flat out on the road and was too lazy to pump the tire up to full pressure.

BTW2, as a speedometer and odometer, the iBike is the most accurate of any device I've used. I suspect they got that part right because they had too -- the iBike needs that data to be right more than other devices. I also suspect that's why the speed sensor isn't (and maybe never will be?) wireless.