Not sure I understand the question. Is it off at zero and then the same at 96 inches? I dunno. Off means out of square. Never made checks away from the center of travel.
However, since the post we mounted a camera and introduced a variable by hand pushing the gantry to observe the end "twist" or "racking." What seems to have happened is that the error was reduced substantially. I suppose that it is under some mechanical tension as it always ends up off in the same direction.
So the aforementioned asks another question. How does one set dual homing for a dual drive gantry running in slave mode? Perhaps there is a combination of error. That is, or could be, a manufacturing defect, setup defect, lost steps, and so on. I am thinking that there is some mechanical tension that could be eliminated by loosing the various attachment points and re-torquing them. That's a project, but combined with dual homing could take the boat anchor aspect out of this machine.
We made index marks with using the camera and the diagonal at 48x48 was, as best a tape will tell, off by about 1/32.