Why are we 10 years into Kindle's evolution and the renderer still does not draw pages exactly the same way every time you load the same page (paginating forwar

Why are we 10 years into Kindle's evolution and the renderer still does not draw pages exactly the same way every time you load the same page (paginating forward vs. paginating backward, at the same font size)? I have to expect that this will reduce spatial-visual memory where people remember what position a figure or text had on a page versus a better renderer (PDF, though impractical on small screens) or a physical book and I'd like to see a study which distinguishes this effect.