DOUBLE

Back to Function List

Convert to double precision

Usage

Jacket for M Language

Supported Syntax

double(X)

GFOR Supported	No
GCOMPILE Supported	No
Types Supported	- GDOUBLE GSINGLE GLOGICAL GINT32 GINT8 GUINT32 GUINT8 COMPLEX
Types Not Supported	-
CPU usage	Jacket only contains GPU computing functions. To run functions on the CPU, try one of the MathWorks products: http://www.mathworks.com/access/helpdesk/help/techdoc/ref/double.html

DOUBLE can be used to convert GPU variables to CPU variables. Execution of this operation results in a memory transfer from the GPU to the CPU.

One bottleneck in achieving reasonable amounts of speedup is memory transfer. If you wish to optimize your code and achieve speed-ups, try to reduce the amount of casts to double especially on large variables, as each such operation can incur a costly memory transfer operation which would slow down your code.