I tried the pre-release but I really would not limit any of the adjustments.
I wonder if there is some difference in how we implemented the tracking that prevents you from allowing full access to all of the settings. With mine, after I collect a sweep, I calculate the transition's center switch point. Prior to the next sweep, I change the delay to recenter. I do this on every sweep.
The transition point is recalculated at the end of each sweep, then the timebase position is adjusted accordingly.
The concern is that the user may be in the middle of adjusting the timebase when the automatic adjustment is being applied. To solve this, we can temporarily pause tracking as long as the user is interacting with the timebase settings. The pre-release was a quick workaround in case you wanted to run additional experiments.
There are also many edge cases to consider due to interactions with other features. (For example, what to do if the user switches on/off power-user mode while the tracking is active? What if de-embedding is enabled?) For the official release, we will implement the necessary logic to handle this gracefully (or disable features for which we can't find a good resolution).