This is a multi-part message in MIME format.

------=_NextPart_000_0040_01C013F3.ED6517E0
Content-Type: text/plain;
        charset="iso-8859-1"
Content-Transfer-Encoding: 7bit

Oops!  I accidentally hit send for the previous message before attaching the
FP routine snippet as promised.  It is attached to this message instead.

------=_NextPart_000_0040_01C013F3.ED6517E0
Content-Type: application/octet-stream;
        name="fp24flt.ins.aspic"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: attachment;
        filename="fp24flt.ins.aspic"

;   Subroutine FP24FLT
;
;   Convert the signed fixed point number in REGA to 24 bit floating =
point
;   in the low 24 bits of REGA.  REG8 is a signed value indicating the =
number
;   of fraction bits included in the fixed point REGA value.
;
;   24 bit floating point format:
;
;     24 bits are used to describe a floating point value using 1 sign =
bit
;     7 exponent bits, and 16 mantissa bits as follows:
;
;     |         byte 2        |         byte 1        |         byte 0   =
     |
;     |           |           |           |           |           |      =
     |
;      23 22 21 20 19 18 17 16 15 14 13 12 11 10  9  8  7  6  5  4  3  2 =
 1  0
;     =
------------------------------------------------------------------------
;     |  |                    |                                          =
     |
;     | S|      EXP           |                   MANT                   =
     |
;     |  |                    |                                          =
     |
;     =
------------------------------------------------------------------------
;
;     S  -  Sign bit.  0 for positive or zero value, 1 for negative =
value.
;
;     EXP  -  Exponent.  The overall floating point value is the =
mantissa
;       value times 2 ** (EXP - 64) when EXP is in the range from 1 to =
127.
;       The special EXP value of 0 is only used when the overall =
floating
;       point value is 0.0.
;
;     MANT  -  Mantissa.  Except for the special case when the overall
;       floating point value is 0, the mantissa represents a fixed point
;       value such that 1 <=3D mantissa < 2.  This means the integer =
part of
;       the mantissa is always 1.  Since this integer part is always the
;       same, it is not stored.  The MANT field contains the 16 most
;       significant fraction bits of the mantissa value.  Therefore
;       MANT =3D (mantissa - 1) * 65536.  An overall floating point =
value of
;       0.0 is indicated by EXP =3D 0.  In that case MANT is reserved, =
and
;       should be 0.
;
;    Consider the following examples:
;
;      0.0 --> 000000h
;
;        S =3D 0  (positive or zero)
;        EXP =3D 0  (special case for 0.0)
;        MANT =3D 0  (special case for 0.0)
;
;      1.0 --> 400000h
;
;        S =3D 0  (positive or zero)
;        exponent =3D 0, EXP =3D 64 --> 40h
;        mantissa =3D 1.0, MANT =3D 0
;
;      -3.141593 --> C19220h
;
;        S =3D 1  (negative)
;        exponent =3D 2, EXP =3D 65 --> 41h
;        mantissa =3D 1.570797, MANT =3D 37,408 --> 9220h
;
;   Unless otherwise specified, overflow and underflow values are =
silently
;   clipped to the maximum magnitude (7FFFFF for positive, FFFFFF for =
negative)
;   and zero, respectively.
;
------=_NextPart_000_0040_01C013F3.ED6517E0--


*****************************************************************
Olin Lathrop, embedded systems consultant in Devens Massachusetts
(978) 772-3129, olin@cognivis.com, http://www.cognivis.com

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.