The type(x, complex) function returns true if x is an expression of the form $a+\mathrm{I}b$, where a (if present) and b (if present) are finite and of type realcons.

The type(x, complex(d)) function returns true if $\pm \Re \left(x\right)$ or (if present) and $\pm \Im \left(x\right)$ (if present) are both of type d.

The type(x, complex[8]) function returns true if the real and imaginary parts of x are Maple hardware floats.

The "8" in complex[8] refers to the number of bytes allocated for the underlying hardware floating-point numbers.  As a complex[8] number has two parts, real and imaginary, each complex[8] requires 16 bytes.

You can build complex[8] expressions using the HFloat command.

$\mathrm{type}\left(5\mathrm{I}\,\mathrm{complex}\right)$

$\mathrm{true}$

$\mathrm{type}\left(x\,\mathrm{complex}\right)$

$\mathrm{false}$

$\mathrm{type}\left(\frac{1}{2}+3\mathrm{I}\,\mathrm{complex}\left(\mathrm{rational}\right)\right)$

$\mathrm{true}$

$\mathrm{type}\left(a-\mathrm{I}b\,\mathrm{complex}\left(\mathrm{name}\right)\right)$

$\mathrm{true}$

$\mathrm{type}\left(2+\mathrm{I}b\,\mathrm{complex}\left(\mathrm{name}\right)\right)$

$\mathrm{false}$

$\mathrm{type}\left(2+\frac{4}{7}\mathrm{I}\,\mathrm{complex}\left(\mathrm{numeric}\right)\right)$

$\mathrm{true}$

$\mathrm{cf}≔1.+2.\mathrm{I}\:$

$\mathrm{type}\left(\mathrm{cf}\,\mathrm{complex}\left(\mathrm{float}\right)\right)$

$\mathrm{true}$

$\mathrm{type}\left(\mathrm{cf}\,\mathrm{complex}\left[8\right]\right)$

$\mathrm{false}$

$\mathrm{chf}≔\mathrm{HFloat}\left(\mathrm{cf}\right)$

$\mathrm{chf}≔1.+2.\mathrm{I}$

$\mathrm{type}\left(\mathrm{chf}\,\mathrm{complex}\left(\mathrm{float}\right)\right)$

$\mathrm{true}$

$\mathrm{type}\left(\mathrm{chf}\,\mathrm{complex}\left[8\right]\right)$

$\mathrm{true}$