Hmm, like this. If a gravitational acceleration is a exact equivalence to a uniform motion, also called relative motion, as proven by its 'weightlessness', and if you then look at NIST where clocks start to differ depending on ones elevation relative the other. Then a uniform motion should have a, just as real, time dilation as you define to a real acceleration.
Just a thing I've been pondering now and then, about the question of where a 'time dilation' sets in. in the acceleration (turnaround) or spread through the paths whole ensemble. Like some rocket leaving a origin, then turning around some star to coming back to its origin again.
and it depends on equivalences. The gravitational ('acceleration') as we call it, in fact becoming a geodesic, a uniform 'internally weightless' path taken, to then finally look at NIST experiments proving that moving one of two clocks into a different gravitational potential ( f.ex lifting it up) will make it start to differ temporally from its 'brother'. Showing a time dilation.
If that is true, then you have a proof of time dilation's being spread all over the spectrum.