The control IC is running at 250Mhz. Does that mean the fault timings are limited to 4 nanosecond granularity? Is that enough to make a reliable attack? Is

The control IC is running at 250Mhz. Does that mean the fault timings are limited to 4 nanosecond granularity? Is that enough to make a reliable attack? Is anything done to sync with the startup and locking of the host clock PLL, since that would presumably vary quite a bit based on temperature?