It's clear that the ballistics of all the tanks are consistant with each gunsight as I tested them all at ranges from .target 400 to .target 3200 and the shells hit exactly where they should. So if the gunsights have not changed since the previous version (as one person clearly showed by overlaying them) then the ballistics have not changed.
What's perhaps changed is the default field of view and the apparent visual size of targets in the viewfinder. Some people are apparently highly skilled at judging range via the apparent size of the target. If the FOV is different, or the image quality different, this judgement may be thrown off. For example, some are saying things look "bigger" and others are saying their shots are falling "short". Well if the target looks bigger and you judge range by target size then you're going to estimate a shorter range and the shell is going to "lob short".
I am not highly skilled at judging range just by looking at the target. Instead I use the gunsight hash marks (for example if a Panzer is presenting its side view and it is exactly one hash mark wide, it is at 1200 yds range). This gives an internal calibration so if the FOV changes I still aim the proper distance.
Perhaps this is the source of the discrepancy?