When issuing ifort /? to list the options we see:
determine if certain square root optimizations are enabled
Shouldn't this enable (full) precision, or disable (fast) precision of square root?
The compiler uses a faster but less precise implementation of square root.
However, the default is -prec-sqrt or /Qprec-sqrt if any of the following options are specified: /Od, /fltconsistency, or /Qprec on Windows* systems; -O0 , -fltconsistency, or -mp1 on Linux* and OS X* systems.
This option improves precision of square root implementations. It has a slight impact on speed.
This option inhibits any optimizations that can adversely affect the precision of a square root computation. The result is fully precise square root implementations, with some loss of performance.