When physical motion controllers are attached there seems to be an opportunity to get close to zero latency.
Presuming that the user's action is reasonably repeatable it seems possible to anticipate the strike. If this could be translated into several network packets as the bell is coming up or down, each containing an updated prediction of the actual strike time, that seems like it should allow for [apparently] zero latency and good robustness to packet loss / delay.
When physical motion controllers are attached there seems to be an opportunity to get close to zero latency.
Presuming that the user's action is reasonably repeatable it seems possible to anticipate the strike. If this could be translated into several network packets as the bell is coming up or down, each containing an updated prediction of the actual strike time, that seems like it should allow for [apparently] zero latency and good robustness to packet loss / delay.