Quantum Clocks have proven to be extremely useful for our everyday life and devices. But what can we use quantum clocks for, what are their fundamental limitations, and finally, are they more efficient than classical clocks? In this talk, after introducing the notion of a finite dimensional quantum clock, we will study two related but different tasks: how well they can autonomously control another quantum system (e.g. perform the gates in a quantum computation) and how well they can tell the time by producing a periodic stream of “ticks”. For the former, we will prove they can do so while only enduring an exponentially small error in clock dimension and clock energy; while for the latter, we will derive the optimal quantum clock as a function of its dimension and compare it to the optimal classical (stochastic) clock of the same dimension; thus proving a separation in space resources between the optimal quantum and classical (stochastic) clocks. We will briefly comment on how this is related to to other fundamental limitations, such as the inescapable entropy gain of a clock, as a function of its accuracy.