RPN V1.00 900131 - Release notes by MLO Excuse my bad english, folks ... RPN is a computer program that simulates, on the AMIGA screen, an Hewlett-Packard pocket computer. A first version of RPN was written several years ago in FORTRAN for an HP21MX mini computer, then moved to a VAX; a BASIC version for IBM PC's was also written. Now RPN is translated in C (the better human creation after italian people) for Commodore Amiga; the compiler is Lattice C V5.04, and the original environment an Amiga 500 with 2.5 MBytes of Fast Ram, but the program must run on every configuration and should (with minimal changes) be compiled and linked using Manx's Aztec-C. The source files, the objects, and the MAKEFILE used to compile and link RPN are included: you are welcome to modify the source, if you will find some error (although tested, this program could contain errors - this is human) or if you do not like the aesthetics. But, please, send me your changes at the address that you will find at the end of this Read.Me file. The simulated pocket computer was originally an HP-32, then some features of HP-11C were also included; I assume that you are familiar with these pocket computer families, or at least with the  Reverse Polish Notation (RPN) logic they use: RPN was introduced by Lukasiewicz in 1924, and is based over the statement "every function must be introduced immediately after the variables that are its arguments". All pocket computers use this logic for functions of a single argument, e.g. the square root: you type the number, then you select the "square root" function. RPN computers use the same logic for all functions, the four basic arithmetic operations included; in this way you do not need a key to start the computation (the "=" key), you do not need parenthesis, and you can inspect, access and modify every intermediate result. RPN has a four registers stack, and sixteen memory registers: six of them, from A to F, are used for temporary storage in statistical computations; but you can use all of them to store your temporary results. Stack and registers are always visible in a dedicate window. To start RPN simply type RPN from CLI; or double-click over the RPN icon (beware: real men do not use icons). You should obtain a main window with key gadgets to be selected using the mouse, and a string gadget (the Input Field - IF); and a secondary window showing the values stored in the stack and in the sixteen registers. You can use also pull-down menus to perform some operations. To input numerical values, select the IF gadget - or type a character that is suitable for numerical input, both in fixed point notation or in scientific notation, and the IF gadget will be selected for you. You can modify what you typed, because the IF will be read only when you deselect this string gadget (hitting the RETURN key; the value will be stored in the X register) or selecting a key gadget or a menu option (the numeric value will be stored in the X register and then the function you required will be performed on it). Every operation is tested against arithmetical errors, and the stack or the registers will be modified only if no error is detected; if you by mistake select a wrong, but legal, operation, you can - 1 - recover the stack and registers status selecting the Last X menu option and then the inverse of the last function.You can with RPN  perform: - The four basic arithmetic operations. - Inverse, square root, square, powers, percentage. - Exponential and logarithmic functions, both natural and decimal. - Trigonometric or hyperbolic sine, cosine and tangent and their inverse functions. - Convert angles to radians, sexagesimal and centesimal degrees; convert numbers from decimal to sexagesimal system, or the inverse; convert coordinates from rectangular to polar, or the inverse. - Statistical computations: mean, standard deviation, linear correlation coefficient, linear regression, coordinates of points on the linear regression line, errors on the ordinates with the Fisher's formula. - Direct arithmetic operations on memory registers. For trigonometric functions, you can select angle measurements in radians or degrees, both sexagesimal or centesimal; all computations are performed in double precision (64 bits). The Inv key must be followed by Sin/Cos/Tan and indicates their inverse function; Hyp introduces the hyperbolic functions, and Inv Hyp, or Hyp Inv,  their inverse.Reminder: - The % key computes in X the X percentage of Y. - The sequences Recall "Operation" Register perform an Enter, then show in X the result of X "Operation" Register. - The sequences Store "Operation" Register store in Register the result of Register "Operation" X. - For the mean and the standard deviation: after hitting the Mean  key you will find the mean in X and the standard deviation in Y. - Hitting the L.R. key, the linear regression line A+Bx is computed; you will find A in the X register, and B in Y. The memory registers A..F contain: A Number of points B  Sum of x values C  Sum of the squares of the x's D  Sum of the y's E  Sum of the squares of the y's F  Sum of the products x times y - Polar coordinates: the radius is in X and the azimuth angle in  Y. - 2 - About the author: please notify all modifications to: Maurizio LORETI (or MLO, or I3NOO) Via G. Donizetti, 6 35010 CADONEGHE (PD) ITALY Also reachable with E-Mail at the address you will find in the source headers. LORETI at IPDINFN (BITNET) Have fun! MLO - 3 -