LabVIEW 2013 FPGA Module Help
Edition Date: June 2013
Part Number: 371599J01
»View Product Info
Use the smallest data type possible for nonconstant values to decrease the size and increase the speed of an FPGA VI. Consider the following guidelines when creating an FPGA VI.

Numeric functions—Numeric functions coerce all inputs to a largest data type. For example, if you wire a 16bit and a 32bit integer to the inputs of the Multiply function, the function coerces the 16bit integer into a 32bit integer. If you want to avoid coercion to a larger data type, use the To FixedPoint function to convert the integer inputs into equivalent fixedpoint data types and configure the Multiply function output for the desired fixedpoint representation.

Loop Timer, Tick Count, and Wait VIs—You can use the smallest Size of Internal Counter possible for the FPGA VI.

While Loop—You can convert the 32bit iteration terminal output to the smallest data type that still supports the maximum number of iterations the loop executes. You also can avoid using the iteration terminal by using shift registers instead.

Index Array function—The Index Array function uses a 32bit integer as the default for the index input. If the array you wire to the array input contains less than 256 elements, you can change the representation of the input from a 32bit integer to an 8bit integer.
You also can save FPGA resources by choosing the smallest data type possible for the following block diagram objects:

Comparison functions
 Wires to Case structures case selectors
 Controls and indicators