F-D(1)

NAME

f-d - convert single-precision floats to double-precision floats

On this page

SYNOPSIS
DESCRIPTION
OPTIONS
EXAMPLES
SEE ALSO
AUTHOR
COPYRIGHT
BUG REPORTS

SYNOPSIS

f-d [-n | scale] < floats > doubles

f-d reads a stream of machine-native single-precision (4-byte) binary floating point values from standard input, converts each one to a double-precision (8-byte) binary float, and writes the result to standard output.

An optional argument controls scaling. If a numeric scale is given, every value is multiplied by it during the conversion. The -n flag selects normalized output, leaving the unit scale of 1.0 in effect. With no argument the values are converted unchanged.

Because the input and output are raw binary, f-d refuses to run when standard input or standard output is a terminal, and it rejects a scale of zero. The inverse conversion, from double-precision floats back to single-precision floats, is performed by d-f(1).

This command is deprecated and is scheduled for removal; it prints a deprecation warning on startup.

OPTIONS

-n: Normalize the data. The unit scale factor of 1.0 is retained so that data already in the range -1.0 to +1.0 is preserved.
scale: A numeric multiplier applied to every value as it is converted. May not be zero.

EXAMPLES

Example 1. Promote single-precision samples to double precision

f-d < signal.f > signal.d

Reads single-precision floats from signal.f and writes the equivalent double-precision values to signal.d, the wider format expected by signal-processing tools such as dfft(1) and dmod(1).

Example 2. Widen and amplify in one pass

f-d 2.0 < quiet.f | dmod -e 2 > loud.d

Converts single-precision floats to doubles while doubling each sample, then pipes the result through dmod(1) for further processing, avoiding an intermediate file.