-Numeric data (may have a decimal-point part). Requires a third
-\fBint\fR argument controlling the precision, a fourth \fBdouble\fR
-argument constraining minimum value, and a fifth \fBdouble\fR constraining
-maximum value. If your system supports locales, the decimal point character
-to be used must be the one specified by your locale.
-If the maximum value is less than or equal to the minimum value, the range is
-simply ignored. On return the field buffer is formatted according to the
-\fBprintf\fR format specification ".*f", where the '*' is replaced by the
-precision argument.
-For details of the precision handling see \fBprintf's\fR man-page.
+Numeric data (may have a decimal-point part).
+Required parameters:
+.RS
+.bP
+a third \fBint\fP argument controlling the precision,
+.bP
+a fourth \fBdouble\fP argument constraining minimum value,
+.bP
+and a fifth \fBdouble\fP constraining maximum value.
+If your system supports locales,
+the decimal point character must be the one specified by your locale.
+If the maximum value is less than or equal to the minimum value,
+the range is simply ignored.
+.RE
+.IP
+On return, the field buffer is formatted according to the
+\fBprintf\fP format specification \*(``.*f\*('',
+where the \*(``*\*('' is replaced by the precision argument.
+.IP
+For details of the precision handling see \fBprintf\fP(3).