The convert(expr, float) function converts an expression to floating-point to the precision given by the global variable Digits.

If called with an argument n, the expression is evaluated at a value of n digits.

The convert/float command is a call to evalf.