Propagation Delay in Electron Mode with Tristate Buffers is Inconsistent

[Tynach]

There is almost always a 1-tick delay associated with a tristate buffer's state changing, but not always. Inputs that come from above never have any delay, nor does the uppermost/first of inputs coming from the left. This holds true whether it's one contiguous/solid row of tristate buffers, or separate ones with connected outputs - but multiple sets of tristate buffers that do not share an output behave independently.

Here's a simple demonstration, and here's a more in-depth exploration, along with an example of abusing this.

[cheznewa]

HEYY :::: Why You NOT Donate Your Curcuit Except Your Bug Circuit, Me I Cant Make Circuit At Your Level!?
AND WHY You NOT Starred This Project??

[Tynach]

@cheznewa, what are you talking about?