/*****************************************************************************
*                                                                            *
*   BITSCALE.SPL Copyright (C) 1998 DSP Development Corporation              *
*                               All Rights Reserved                          *
*                                                                            *
*   Author:      Randy Race                                                  *
*                                                                            *
*   Synopsis:    Convents raw counts to engineering units                    *
*                                                                            *
*   Revisions:   28 Jul 1998  RRR  Creation                                  *
*                                                                            *
*****************************************************************************/

#include 

#if @HELP_BITSCALE

    BITSCALE

    Purpose: Converts raw AD counts to scales engineering values

    Syntax:  BITSCALE(xi, numbits, yl, yh)

              xi      - input series or scalar
              numbits - an integer, the number of AD converter bits
              yl      - a real, low value output range
              yh      - a real, high value output range

    Returns: A series or real

    Example:
             Bitscale(4096, 16, -2.5, 2.5)

             returns 0.312543, the corresponding output for input 4096
             of a 16 bit converter with an output range of +-2.5

             Bitscale(-128..127, 8, 0.0, 10.0)

             returns a series ranging from 0.0 to 10.0


    Remarks:
             Bitscale assumes offset binary input data, i.e the input
             data ranges from -(2^numbits)/2 to (2^numbits)/2 - 1.
             Bitscale does not automatically clip out of range values.

    See Also:
             Linscale
             Quantize
             Rescale
#endif


/* scale bits to output range */
bitscale(xi, numbits, yl, yh)
{
        return(linscale(xi, -(2.0 ^ numbits) / 2, (2.0 ^ numbits) / 2 - 1.0, yl, yh));
}