In the accelerating frame, though, measurement of a circle would get pi. If you were in a frame stationary to an accelerating circle, the circle would look oval (if it were oriented correctly).
I don’t think pi changes with gravity, but that’s an interesting concept.
I was thinking of an area of space where if you measured a circle you’d get 3.00000. Lok at all the physical calculations that involve pi! They’d all be different, including the behavior of inductors.
A Flatlander would be surprised to find that - and it would be possible if his circle is curved in our 3rd dimension, normal to his flat world, like when you hit a tambourine. This is because the Flatlander can measure the diameter only by crawling over the stretched area, without noticing that it is stretched.