MatrixMul

[dch2]

I'm trying to use MatrixMul in the advanced analysis package.  It works fine if all my arrays are of type 'double'.  But if they're all of type 'float' I get an error.  The function prototype indicates the arrays are of type 'void *', which suggests to me that the arrays should be allowed of any type, as long as they are consistent.  Is it possible to use floats in this function?  The standard help file isn't very useful for answering this question.  I'm working on an application that requires high speed, so I don't want to take the time to cast an array from float to double.

[msaxon]

The function prototype is misleading. Since the function does not allow you to specify the datatype of the data, it will only work with double[][] arrays.

[Eric_B.]

As Martin said, the function prototype is a little misleading (or at least unclear), but depending on what version of CVI you have, the help file may actually specify that the arrays have to be doubles.  CVI 8.5 and later mention this in the help file if you search for "matrixmul".  You can also see it in CVI 8.5 and later if you right-click on one of the Matrix controls on the function panel for the MatrixMul function.