We are having quite a funny argument, because it seems we both see the problem. "One of the central tenants of relativity is that what reference frames we choo

We are having quite a funny argument, because it seems we both see the problem. "One of the central tenants of relativity is that what reference frames we choose is irrelevant as long as we account for all the relativistic effects." I think that applies only to inertial frames. At least in my version of SR that is so. "your result is apparently dependent on which twin's reference frame you are using. That is a contradiction." I know it is dependent. I don't see what's the contradiction. The earth twin has a special status - he is non-accelerating. His reference frame is therefore inertial. The other twin is accelerating. His reference frame is non-inertial. Where is the problem here?