boris wrote:
Personal preferences are not bugs ;)
Nah, the bug is not in that he prefers a smaller factor, the bug is this:
Zoom-in factor: 1+x
Zoom-out factor: 1-x
That's not how inverses are made!
You start with a zoom factor of 1. Let's say you zoom in one tic. Zoom factor is now 1.3 * 1, so 1.3. Now you zoom out one tic. Zoom factor is 0.7 * 1.3, so 0.91. Which is not the same as 1...
Zoom-out factor ought to be 1/zoom-in factor, so that the factors remain consistent. So if zoom-in uses 1.3, then zoom-out ought to be 1/1.3.
(I'd suggest a zoom-in factor of 1.25, this way the zoom-out factor is 0.8, and there are no risk of inaccuracies creeping up from loss of precision.)