Mathematical Explanation for Varying Runtimes and Precision at 256-bit Security Levels in Threshold FHE

The logic is very similar to what is described in In which scenario does the decrease in Q result in reduced ring dimension? It is all driven by the security tables.