FCALC.DOC

			  Freyja Calculator
			HP-45A-like Calculator

DESIGN

This calculator is a non-programmable, scientific calculator, roughly
equivalent to an HP-45A or a 41C with some functions missing.  It
offers some extra functions (e.g., binary math).  The internal
arithmetic is done using the C language "double" arithmetic supplied
with your C compiler.

This documentation assumes that you already know all about RPN
calculators in general and the HP calculators in particular.



ENTERING NUMBERS

Numbers look pretty much as you would expect.

- Real numbers can consist of the digits 0-9, a radix mark ('.' or
','), digit separators (the non-radix mark '.' or ','), an exponent
character ('e' or 'E'), and an exponent.  Use the '~' or '`'
characters (or F9) for entering negative numbers or exponents.

- Binary numbers start with a # sign, then contain the digits 0-9,
a-f, or A-F.  They are interpreted and displayed according to the
current default base.  They will also be masked to the current word
size.

In all cases, number termination is by means of a space, ^G, ^M, ^J,
^[, $, %, *, +, -, or / character.

Some operators include the above termination characters as part of
their names (e.g., "1/X").  You can enter these by starting the
command with a single quote ("'").  Doing so disables the "immediate
action" of the %, *, +, -, /, `, and ~ characters.

Numbers are converted between real and binary as needed (and expected)
by the operators.  There are no illegal combinations, although there
may be illegal values (e.g., square root of a negative number).



DISPLAY

Numbers are displayed according to their type and the current display
settings.  These settings are:

- Only Std display format is supported for now.  There are two other
settings:

	radix mark		. or ,
	digit separators shown	Y or N
	 (if Y, the separator is the one of , or . that is not the current
	 radix mark)

- Binary numbers: always displayed in the current base #XXXY, where Y
is 'b', 'o', 'd', or 'h', designating Bin, Oct, Dec, and Hex modes.



STACK LIFT

The calculator handles stack lift more like an HP-28C/S or 48S/SX
than a "normal" HP.  Number (and command name entry) is handled in a
separate place and not the X register.  The entered number (or
command) can be edited arbitrarily before it is terminated by a
closing function that is actually executed.  Hence, the concept of
stack lift does not play a very important role.  All operations leave
stack lift enabled except for:

	CLX
	ENTER^ (or T)

No operations are neutral.


COMMANDS

^G	Exit calculator w/no action.
^H	Delete the preceeding character or clear X if no preceeding character.
^J	Exit calculator, inserting X into the buffer at the point.  Leave
	the mark at the start of the number and the point at the start.
^M	Finish the current number or operator.
^[	Exit calculator w/no action.
SP	Finish the current number or operator.
$	Return to the editor.
%	Percentage.
'	Start a quoted command.
*	Multiplication.
+	Addition.
-	Subtraction.
/	Division.
?	Help.
`	CHS or NEG.
~	CHS or NEG.
^?	Delete the preceeding character or clear X if no preceeding character.

F1	Help.
F9	CHS or NEG.
F10	Exit Freyja.
Alt-B	Last X, immediate execute.
Alt-C	X<>Y, immediate execute.
Alt-L	CHS, immediate execute.
Alt-M	RCL.
Alt-N	STO.
Alt-V	RDN, immediate execute.
Alt-X	1/X, immediate execute.
Alt-Z	SQRT, immediate execute.


COMMAND LIST

These commands are all executed by typing their name, then completing
them with a space, ^M, or ^J.  If a space is used as the terminator,
it does not do an implied Enter.

	%%%	register designator:
			#		numeric
			X,Y,Z,T,L	stack
			A		alpha
			\GS,~		summation
			.%%%		indirect
			IND%%%		indirect
	###	number
	&&&	label:
			#		numeric
			"string"	alpha label
			.###		indirect
	@@@	buffer name

name		does

%		percent
%CH		percent change
%TOT		percent of total
*		multiply
+		add
-		subtract
/		divide
1/X		inverse (not typable: must use INV)
<%%%		recall
>%%%		store
AND		bitwise and
B		set binary mode
CF###		clear flag
CLRG		clear registers
CLST		clear stack
CLX		clear x
D		set decimal mode
DEFAULT		restore default settings
DIGSEPOFF	set the digit separator mark to not displayed
DIGSEPON	set the digit separator mark to displayed
ENTER^		enter
FACT		factorial
H		set hexadecimal mode
HELP		help
INV		inverse
L		recall last x
LASTX		recall last x
MEMVIEW		view interpreted calculator memory in %calc% buffer
MOD		modulus
NEG		negate
NOT		bitwise not
NULL		no op
O		set octal mode
OR		bitwise or
R		roll down
RADIX,		set the radix mark to ,
RADIX.		set the radix mark to .
RCL%%%		recall (try rcl~78)
RDN		roll down
R^		roll up
S		swap: x<>y
SF###		set flag
SQRT		square root
ST%%%		store
STO%%%		store
T		enter
WSIZE		set the word size
WSIZE?		get the word size
X<>%%%		swap with
XEQ@@@		switches input to the specified buffer (sort of like
		executing a program)
XOR		bitwise xor
XRND		takes an argument from the stack (small integer) and
		rounds the number to that number of decimal places
X^2		square
\GD%		delta %



CALCUALTOR-RELATED FREYJA COMMANDS

^[ ^A	Mark the current number (mark to start, point to end).
^[ ^B	Move backward one number.
^[ ^E	Mark the current number and enter it into the calculator.
^[ ^F	Move forward one number.
^[ ^M	Insert a printed copy of the number in the X register into the
	buffer.  If an argument is supplied, do ^[ ^A ^W first.
^[ ^U	Invoke the calculator.
^[  ^]	Load the program specified by the argument into the keyboard macro.



SYSTEM BUFFERS USED

%calc%		miscellaneous calculator output (e.g., status dump)



FREYJA-LOCAL FLAGS

start	size	default	use

65	8	BINSIZE	word size for binary integers

73	2	hex	binary number mode:

	flag	74	73
		0	0	decimal 
		0	1	octal
		1	0	binary
		1	1	hexadecimal



PROGRAMS

A program is created by means of the ^X ( and ^X ) commands.  This
program can be saved the ^[ ^] command.  This command inserts the
current keyboard macro contents at the point.  You can then use normal
editing commands to manpulate and save it.

The typical sequence is:

	- create a keyboard macro
	- create an empty buffer
	- give the ^[ ^] command
	- save the buffer

You now have your program saved.  You can execute it in one of two ways:

1:	- find the file with your program
	- XEQ<name>

2:	- find the file with your program
	- give the ^U ^[ ^] command to copy the buffer to the keyboard macro
	- execute it with ^]

The second method allows you to give a repeat count.