Frozen Fish 1: Amiga

home *** CD-ROM | disk | FTP | other *** search

/ Frozen Fish 1: Amiga / FrozenFish-Apr94.iso / bbs / alib / d4xx / d495 / analyticalc.lha / AnalytiCalc / SrcAndDocs / AnalyDocuments.LZH / AnalytiCalc.Doc < prev next >

Wrap

Text File | 1989-08-03 | 253.6 KB | 5,498 lines

A N A L Y T I C A L C The Analyst's Tool User Manual and Reference Copyright (C) 1985, 1986, 1987, 1988 G.C.E. AnalytiCalc Reference Manual Page 2 AnalytiCalc Overview AnalytiCalc AnalytiCalc is an electronic spreadsheet program and integrated system which was written in "portable" Fortran and runs on your AMIGA, PDP11, VAX or MSDOS systems. It has a variety of commands and operates on a "command verb" syntax. That is, everything you tell it is a command. Some commands enter numbers or formulae; others manipulate them or change the display. Online help is available (though limited) via the Help command (or the HELP key). While the primary functionality provided is that of a spreadsheet, AnalytiCalc contains means for integrating word process- ing documents, running other programs under its control, integrating its graphics output, and accessing data files created either by selective writes from its own screens or by other programs. Unlike single huge programs which attempt to provide every function known to Man in one image, AnalytiCalc is a "second generation" integrated en- vironment which provides the "glue" that allows you to use the best other specialized programs you can find in a smooth way. It is also constructed to work smoothly with other programs, so that you need not give up the communications programs, editors, word processors, or database managers you already have. AnalytiCalc has 5 major ways of addressing cells, over 85 func- tions (more depending on how you count them), over 70 commands (with more variations), unlimited windowing and programmability at the cell and the sheet level, 16 digit precision, and a built in online calcu- lator. It is designed to handle traditional spreadsheet problems, data access problems, equation solving, engineering applications, and finance, and is designed to be easy to extend both by adding to the program itself and by adding commands externally. The entire program can be driven from a command file, and individual cells can be driven by one or more command files in addition to formulas; commands in these command files allow programmable control of operations with very high flexibility. Interfaces exist to graphics outputs, to word processing in- tegration with the spreadsheet (or AnalytiCalc may be used as a word processing integration tool as a kind of framework document integra- tor), to data access from data files separate from the sheet, and to saved sheets permitting access to systems of spreadsheets. Simple methods of moving to other system commands make this program a power- ful command interface as well as a unique analytic tool. AnalytiCalc is command driven, though a number of these commands are automatically generated by function keys. Its commands are mnemonic and intended to be extremely fast to use once learned. AnalytiCalc Reference Manual Page 3 Unlike nearly all other spreadsheet type products, AnalytiCalc delivers what it promises. If a cell can be addressed, it can be used, subject only to limitations on the amount of disk storage you allocate for backing store. You can use any "n" cells (where "n" is 18000 on the Amiga or PC) fully with enough disk backup. Contrast this with memory only systems which may address 500,000 or more cells, but which permit only a few percent of these to be filled be- fore running out of room, even on large machines. (For example, the popular 1-2-3 spreadsheet on IBM PCs is reported to hold at most around 12,000 simple formulas on a 544K machine, or around 39,000 9-digit numbers. (Our tests have run out of room at less than 7000 cells on a 512K AT.) In contrast, AnalytiCalc can hold 32,000 17-digit numbers, long formulas, or text strings in its cells, using disk store as backup for memory. The 8088 version currently will run in 256K bytes of memory and handle up to 18,000 cells with up to 109 characters of formula in each. Thus the 8088 version can manage something over 2.1 megabytes of information, while programs like 1-2-3 V1A are restricted to 0.3 megabytes approximately on normal memory PCs.) The Amiga version is similar in space to the 8088 ver- sion but since it is built without overlays occupies more space. The Amiga version will run in a 512K machine with the non-interlace screen and default (4000) stack. Use with interlace screen in 512KB requires a very minimal environment. The "new CLI" functions will not fit in 512K, but everything else should work. Since the format of a spreadsheet sometimes needs to vary, no simple fixed size can satisfy every need. For that reason, AnalytiCalc allows its storage to be repartitioned so that up to 18,000 columns or 18,000 rows may be used. This is done under user control (with normally suitable defaults) and allows use of cells anywhere in the 18,000 by 18,000 space. You cannot fill more than 18,000 cells (chosen anywhere in the region, since there is no space penalty for not using the upper left corner), but you have an address range of 324 million cells. Legal column names range from A to ZPH, and legal row names range from 1 to 18,000, and your display can show any of this range in as many parts as there are cells shown on your screen. (One application of this would be to use columns JAN, FEB, MAR, APR, MAY, ... and TOT for month oriented applications, using D## forms of cell addressing to do cross-column sums, if you want to avoid a line of labels. This is more of a feat of virtuosity than a production technique, but it will work and can be used as you like.) INTRODUCTION AnalytiCalc's command structure is somewhat unique but simple and logical once mastered. Commands you learned for the spreadsheet system of your choice generally are NOT the same as here. But you CAN perform all the same functions. Remember two things first: in the default mode, you need an Enter command ("E ") to enter text or numbers into the sheet, and AnalytiCalc Reference Manual Page 4 floating point numbers (most are floating point rather than integers) need a period (.) in them to get AnalytiCalc to recognize them as numbers rather than text. If you forget the period, the numbers will be displayed, but only as labels, and the cell will have a numeric value of zero. If you want a more "traditional" command interface (entering just numbers or formulas or labels/text by default and starting all commands with a "/", use the command /; to get into that mode. The prompt will change from > to : then as a flag and all commands will need to be entered as /command where the command text is as described herein. The descriptions here apply exactly to the "command-mostly" mode, which is the initial default. The /; mode is sometimes called the "enter-mostly" mode, as it is convenient for entering large amounts of textual or numeric information. The command "//" gets back to "command-mostly" mode. You can have as many windows onto the sheet on screen as you care to define. Default operation will allow screen redrawing as you move beyond the edges of the display. This is more or less a stan- dard type scrolling operation, and windows are preserved during scrolls, but all move together. Separate scrolling can be handled easily via function keys if desired however. The NS (NoScroll) com- mand can disable scrolling if desired. Windows on the screen are just regions of the display which point to areas on the "physical" spreadsheet. To conserve pixels, standard Amiga windows are not used; the spreadsheet display is in one single Amiga window, with an auxiliary window below for some line oriented operations and the calculator. If the program's display is not as expected, use a V command to redraw the screen and/or a R command to recompute to ensure the dis- play is really incorrect. Automatic recomputation occurs normally when data is entered, so most invalid conditions due to forward references in equations or lack of an extra recalculation after copy- ing is done are corrected in the normal course of use. Because of the slowness of the cursor controls, AnalytiCalc attempts to update only regions of the display requiring it. It does so by redrawing only cells that change value. Occasionally a cell which is modified from a TEst command or another cell, or a textual cell, is not up- dated. A View command (V) and/or a RF (Recalculate with Force) com- mand generally clear up any glitches. Getting Started with AnalytiCalc The AnalytiCalc spreadsheet asks some questions when you run it, to set up its working storage. Initial Questions (Startup) Before AnalytiCalc begins with its screen-oriented operation, it looks for a file (in the current directory) called ACINIT.PRM. If it exists, it is read for replies. If not, AnalytiCalc asks a question AnalytiCalc Reference Manual Page 5 about which screen resolution to use. If you are using a non interlace screen (640 by 200 ordinarily), specify this. The alternative is the interlace (640 by 400) screen. AnalytiCalc will size its displays to fit the alternative chosen. If you tell AnalytiCalc to use the 640 by 400 size on a non interlace (640 by 200) Workbench screen, it will cause AmigaDos to crash in some cases. AnalytiCalc just uses windows on the (booted) Workbench screen. It also assumes you have already selected the Topaz 8 font (80 columns, not 60) in Preferences. If you must use a 60 column screen, you will have to manually resize windows and some displays after startup. The S command to relocate the bottom two rows (command and current contents displays) and DB commands to reset the number of rows and columns displayed will be useful in this. (See below.) If you want a pre-prepared set of replies for AnalytiCalc's ini- tial questions, just create a file ACINIT.PRM. (You need to reply to the first page worth of questions in an ACINIT.PRM file, so the fol- lowing example shows sample replies to the others needed.) A simple way to set up replies would be to edit in a file ACINIT.PRM as follows: N Any title desired - use ZA command to reset this. 22 22 1 1 The file just contains the replies as though they were entered on the screen. If the file exists, the screen prompts will not appear however. After these commands are put in, any other desired commands can also be included to further set up operation. If you prefer to come up in "enter-mostly" mode, for example, you could put in a command /; in this file. Initial Screen When you run AnalytiCalc, it first asks you whether to change the default floating point format. Most numbers are "floating point", which means they can have fractional values. You will seldom need integer numbers (which can take on only values -2,000,000,000 ... -3,-2,-1,0,1,2,3,4,... 2,000,000,000), but will use normal "floating point" ones in which it is legal to have values like 1.35, 2.5, and so on. Just think of "floating point numbers" as a synonym for "numbers". The display format for floating point numbers is ini- tially set to the Fortran format F9.2 (9 characters wide, 2 decimal places), but if you reply Y, it lets you enter any format you like as the default format. It will try to use it, so if you enter an ille- gal format, AnalytiCalc will keep asking for a format until it finds one that can display pi (3.14159) correctly. Note that the default sets the format when you first put a value or formula into a cell to be displayed numerically. Once you fill a cell, its display format stays until you alter it with the DF command. Thus if you enter some values, then use the S command (Setup) to change the format default, and enter some more cells, only the second group of numbers/formulas entered will have the changed display format; the first group will not be changed. You can have up to 75 separate formats on a sheet, AnalytiCalc Reference Manual Page 6 chosen from among anything that Fortran can display. (A minitutorial appears later in this manual.) AnalytiCalc then asks you to enter up to 80 characters as the Title of the spreadsheet. This is displayed at the top of the screen and is used as a label for the sheet when printed out. (The space is made available for this rather than being wasted on a copyright statement on your screen.) To reply, enter any title you like, up to 80 characters long, followed (as all AnalytiCalc commands) by a carriage return. This will appear across the top of the sheet on screen and in print images of it. You may include any desired leading spaces in the title. It will be printed on screen printouts, so make it descriptive of the sheet you'll be using. (Note it can be excluded from printouts if desired when row and column headers are being skipped.) This title remains until an S command is given (at which time you can change the work file also), though reading in a saved sheet also reads in the saved title. Discussion of Scratch Files The Amiga and PC versions of AnalytiCalc use memory arrays backed by temporary disk files for values and formulas. These ask how many K (bytes) to reserve on disk for the value file and the for- mula file. Generally there should be somewhat more formula space than value space if many long formulae are used; otherwise pick roughly equal values. These scratch files are deleted on exit from AnalytiCalc-68, but the sheet will NOT run out of room provided suf- ficient file space is allocated. If it runs out due to lack of room, a message on screen will tell which file is out. Save the sheet and restart then with a bigger file. The files are allocated on the default disk when you run AnalytiCalc. Detail Four questions are asked. First, AnalytiCalc asks how many rows and columns you expect to use on the sheet. Try to be accurate in the replies you give, but great precision is not necessary. It will then compute the necessary sizes of disk files assuming all the area you gave will be solidly filled in and display these sizes for ad- vice. Then, it asks how many K to allocate to value and formula backing storage. Reply with what you want to use on disk; the files will be allocated on your default disk and erased on exit. For a discussion of the issues of storage allocation, see the Appendix on this at the end of this manual. When these questions are answered, AnalytiCalc will display its Calculation Screen. AnalytiCalc Reference Manual Page 7 Calculation Screen Display The sheet is displayed at the top part of the screen and a com- mand cell in Row 23 is shown with the current position encoded. This starts off looking like A 1> (if in "Command-mostly" mode) or A 1: (if in "Enter-mostly" mode) and then AnalytiCalc awaits your commands. Note that if the cell pointed to by the command (which is always a Physical sheet address) is occupied (i.e., has ever had any numbers or text entered into it), that cell is displayed in the display in reverse video when you are positioned there. If the cell is totally empty, it is not displayed, however, in any reverse video. A pattern of blanks is displayed, wherever the cell is null. This pattern is in reverse video. (In Amiga, the row is not usually Row 23; it is approximately either row 17 (non interlace) or row 44 (interlace).) CONTROL In "Command-mostly" mode (the default), AnalytiCalc is a COMMAND DRIVEN spreadsheet system. This means that you control the sheet by entering a command followed by Return. These commands are in the first few columns of the command line and are the first few characters (often the first 1 or 2) you type, and they are REQUIRED for every command. AnalytiCalc does NOT assume any command names. AnalytiCalc will translate inputs to upper case unless a " character exists in the input string. Many of the most-used commands are however implemented as single stroke function keys; the ALT-F1 key displays a diagram of these. The numeric keypad has been implemented in as mnemonic a way as possible also, so that arrow keys, home, end, page up and down, and control left and right arrows all have their expected meanings. (For example, End goes to the end of the active region, and the INSert key gets into "Enter Mode" (F9 exits that mode) to allow continuous entry of data.) (In "Enter-mostly" mode, AnalytiCalc is much more similar to "traditional" spreadsheets in appearance. If a / is seen to start a command, a one line prompt will appear in line 25 giving some initial character mnemonics for commands. It is not meant to be exhaustive, but just a reminder.) AnalytiCalc Reference Manual Page 8 COMMANDS The commands available in AnalytiCalc are each described separa- tely. The following is a brief listing alphabetically of their names. Some infrequently used commands are omitted. +J filename Start journaling to filename +N Close journal file < Rewind input file %prompt%cmd%key% Issue prompt and do cmd depending on key $ or } Perform operating system command > or >>pattern Search for formula containing or starting with pattern * Comment line -prompt Load arguments after prompt // Use "Command-Mostly" mode /; Use "Enter-Mostly" mode /# Swap current mode with save mode (command-mostly or enter-mostly) 1,2,3, or 4 Move cursor Up, Down, Left, or Right @file.typ Read file.typ as input instead of console AA nn {R/C} Add absolute nn rows or columns AR nn {R/C} Add relocating nn rows or columns (nn may be neg) CA in-range out-range Copy All (Absolute) CV in-range out-range Copy Value CF in-range out-range Copy Formula (and Format) CR in-range out-range Copy Relocating (all) DB ncol,nrow Display Bounds (no. cols,rows on screen) DF range [format] Display Format of range to format DL range {R/C}n:m Display Locate range as Row/Col to dis- play col:row n:m DS{R/C}{A/D} n Display Sort row/col, Asc/Desc row/col n DT range {F/I} Display Type range as Float or Integer DW ncol,wid Display Width of col "ncol" to "wid" chars E expression Enter expression into cell E" expression Enter expression as text, no case translation ED 'oldstring'newstring' EDit cell ET expression Enter expression as UPPER CASE text EV expression Enter expression as computable formula F filename/nskip File read from filename onto display G Get saved sheet. Many variants. Hn HELP and show page n IR inrange outrange In Place Relocate inrange as if moved to outrange K Go into interactive calculator (need *V 3). *E goes back. L cell Go to cell AnalytiCalc Reference Manual Page 9 M{0/1/2/3/4/5} Move - set move direction after enter MS or MH Macrocell Show or Hide - show or hide in- cluded files. OA cell Origin Absolute cell - Map Screen with cell in upper left OAD cell Origin Absolute Displace cell - Map Screen with cell in upper left leaving old windowing alone OR cell Origin Relative cell - map screen down/right of cursor with cursor=cell ORD cell Origin Relative Displace cell - map screen down/right of cursor with cursor=cell leaving old windows OV + OVerride absolute refs, make relative OV - OVerride off - let abs refs be absolute P Put (save) spreadsheet. Many variations. R Recalculate sheet RB cell Set Relocate Boundary at cell RF Recalculate, Force recalc of constants RM Recalculate Manual - no recalc until R cmd S Setup - Global mapping/width/title setups TE expression TEst - Evaluate math expression in cmd mode - many special variants. V View - Redraw screen VF View Formulas - Draw with Formulas in- stead of numbers VM View Manual - no screen repaint until a V cmd VH+, VH- View Hack on/off. Controls display of row being calculated during recalcs. W Write display to file or printer. X eXit from AnalytiCalc. Asks for confir- mation. ZE range ZEro range of cells ZA Zero All of sheet. Asks for confirma- tion. Also used for reinitializing. AnalytiCalc Reference Manual Page 10 INDIVIDUAL COMMANDS Moving the Cursor CURSOR MOTION (Commands to move around the sheet one cell at a time) 1 (Move up) (or uparrow) 2 (Move down) (or down arrow) 3 (Move left) (or left arrow) 4 (Move right) (or right arrow) The cursor motion keys move the cursor on the display in the direction indicated. Note these keys are in the same order on the keyboard as the arrow keys of the VT100. NOTE THAT THE KEYS 1 TO 4 ARE INTERPRETED IN THE SAME WAY AS THE UP, DOWN, LEFT, AND RIGHT ARROW KEYS ON THE VT100. THEY APPEAR IN THE SAME ORDER AS THE ARROW KEYS ON THE VT100 ON THE KEYBOARD. The X and Y accumulators are set to the column number and the (row number + 1) respectively during commands. This allows pro- cedures to detect cursor position or motion where useful. These ac- cumulators may be used within cells for anything; resetting them does not move the cursor. If the arrow keys move to the edge of the screen (or if auto mo- tion does this), an automatic OA command is done to effectively scroll the entire viewing screen by the number of rows or columns minus 2. The command NS (meaning No Scroll) will inhibit this redrawing. The scrolling can be re-enabled by giving the command SC (SCroll). These commands do not redraw anything nor do they cause recalculation; they merely toggle an internal control. The initial state is SCroll mode. AnalytiCalc Reference Manual Page 11 Insert or Delete Rows or Columns ADJUSTING PHYSICAL LOCATIONS (Adding or Removing Rows or Columns) AA number R Add "number" rows ahead of the current physical position, moving all cells below the current cell down by "number" rows and losing the physical sheet's bottom "number" rows' contents, NOT adjusting variable references where the variables are in the moved range and are not the position independent (P## or D## forms) type. AA number C Add "number" columns ahead of the current physi- cal position, moving all cells right of the current cell right by "number" columns and losing the physical sheet's bottom "number" columns' contents, NOT adjusting variable references where the variables are in the moved range and are not the position independent (P## or D## forms) type. AR number R Add "number" rows ahead of the current physical position, moving all cells below the current cell down by "number" rows and losing the physical sheet's bottom "number" rows' contents, adjusting variable references where the variables are in the moved range and are not the position independent (P## or D## forms) type. AR number C Add "number" columns ahead of the current physi- cal position, moving all cells right of the current cell right by "number" columns and losing the physical sheet's bottom "number" columns' contents, adjusting variable references where the variables are in the moved range and are not the position independent (P## or D## forms) type. These commands permit the operations known on other spread sheets as "insertion" or "deletion". Deletion takes place where "number" is negative and the sense of the motion is reversed. These commands change the physical sheet's contents and permit wholesale modifications to variable names (other than P## or D## forms which are position independent already and are left alone for simplicity and to preserve some of the flexibility of the D## forms which can be used for projections of 3D sheets onto the display and math with ranges of these). To save a sheet for merging such selected values, both forms, or at least the numeric form, of the Put commands should be used. Note that the previous contents of the rows or columns inserted is RETAINED, though it can be zeroed by the ZE command if desired. It will have been copied also to the lower or further-right row/column also, but do not be alarmed; the space is free to re-use. You may however elect to use this feature as a shorthand for copy. The AR/AA operations take place on the physical sheet and do not affect the display sheet mapping to the physical sheet. AnalytiCalc Reference Manual Page 12 Copying and Replicating Cells COPYING CELLS CA V1:V2 V3:V4 Copy all cell attributes from V1:V2 to V3:V4. CV V1:V2 V3:V4 Copy numeric values only CD V1:V2 V3:V4 Copy Display Formats only CF V1:V2 V3:V4 Copy Formulas only (no relocation) CR V1:V2 V3:V4 Copy all cell attributes as in CA but relocate cell names to new location from old one. Only relocate names right of or below the cursor or on same row/col with cursor (cells above or left of cursor are NOT relocated). IR V1:V2 V3:V4 Relocate formulas inplace, computing displacements using distance "moved". The Copy commands let you reproduce cells' contents in other lo- cations on your sheet without re-entering them. The CA form is most useful, though the CD part can be used to change default formats and the others may be handy. Note that formulas are NOT changed by the copy operation. If they should function correctly in their new loca- tion relative to different cells, the cell identifications should be position independent forms (P#n#m for forms relative to physical sheet, D#n#m for forms relative to display sheet) so no modification is necessary. CA will normally save all information about a cell in the new cell. Unlike the other C class commands, the CR command WILL relocate any variable names that are below or to the right of the cursor at the time the command is given. Names above or left of the cursor are not changed. The row and column the cursor is in is included in the relocation region. Thus, to be sure names are relocated, use a com- mand like L A1 first to position yourself at the top left part of the physical sheet. BE SURE YOU DO THE REPOSITION IF YOU USE THE CR COMMAND!!!!!!! Note however that if you need certain cells NOT to be relocated, the ability to place them above or left of the cursor when copying or replicating cells around and having them left alone can be useful. Normalization factors, for instance, can be left alone (e.g., at the top of a column) and cells referring to them be copied without having them point to the "wrong" normalization cell. When you specify one variable to the C class commands in the V1:V2 range location and give a range V3:V4, the command will copy the named V1 into all cells in the V3:V4 range. AnalytiCalc Reference Manual Page 13 The IR command allows inplace relocation of formulas. This operates from the cursor to the Relocate Boundary (see below). The *U HERE function returns the current maximum column used in the W ac- cumulator and the current maximum row used in the Y accumulator to permit the repeated use of IR from a command file to simplify moving sections of the sheet. Setting Size of Display on Screen DISPLAY BOUNDARIES SETTING (Number rows/cols) DB n,m Set number of columns displayed on screen to n, number of rows displayed to m. The DB command sets display size. It also effects the number of cells modified by the OA and OR commands, so setting the display small, doing an OA or OR, then setting it large, leaves the part om- itted as it was before the OA or OR. The displayed window also con- trols what is read in with the F command. Note that for automated text retrieval, the VM command (to turn off redisplay during com- mands) may be useful in connection with using DB commands, F com- mands, and then V commands to turn the display updating back on from command files invoked either by auxiliary keypad keys or from @file commands. For an example of DB, if you want a display of 14 columns by 20 rows, the command DB 14,20 would be the way to achieve it. When a narrow screen (e.g. a 40 column screen) is used, the DB command may be used to reduce the number of columns shown to allow AnalytiCalc to operate. If your screen supports more than 80 columns, AnalytiCalc will allow their use by increasing the number of columns seen. The maximum bounds for DB are usually larger than a real screen will hold, but the W command will write out up to these larger bounds to allow long sheet printouts. On the Amiga version, up to 75 lines length are permitted per page. AnalytiCalc Reference Manual Page 14 Setting Display Format and Changing between Formula and Number Display DISPLAY FORMAT SETTING DF V1:V2 [format] Set Display Format AND set up whether to display formula or number The DF command sets the display format for the variable range V1 to V2 inclusive to be whatever is in the square brackets. This is expected to be a valid FORTRAN FORMAT specifier for display of the number and may be up to 9 characters long. If the format specifier entered is however just A or L in the first character, the program will indicate the text is to be displayed literally, not as a number, so titles and such can be shown. To display numbers in Fortran A or L forms, just use formats like 4A1 or similar; as long as there is something ahead of the A or L it goes to Fortran. The program will try your format once before writing to disk to ensure against any crashes that will damage your file. Otherwise, the format is up to you. You can even display in octal or hex if you wish! The DF command is used to switch a cell from display of numeric information to formula/text display or back on a cell by cell basis (the VF command globally switches cells from numeric to formula and back, though a cell set to display text with DF always displays text). To change a cell to numeric, merely use the DF command to give it a numeric format (the default is normally F9.2 so the command to set, say, cell G3 to be numeric (possibly after entering a formula not containing +, -, ., (, or [ characters) would be: DF G3 [F9.2] for the standard Fortran F9.2 format (2 decimal places, 9 digits wide display). In AnalytiCalc-VM, every cell may have an entirely unique format. In AnalytiCalc-PC, however, only a finite number of unique formats is possible (usually about 45) since storage is in memory. The design limit is 255 unique formats, but very few spreadsheets will need more than the 45 normally provided. Typical ones only have 2 or 3 unique formats. The formats may be ANY Fortran formats desired. The DF command will also set the type to Floating if the format begins with F or E, and will set the type to Integer if the format begins with I. Beyond this (e.g., for octal or hex) you must use the DT command (below) to set the storage to floating (8 bytes) or in- teger (4 bytes). AnalytiCalc Reference Manual Page 15 The DF command will attempt to avoid erroneous format inputs by trying the format out on your current location. If errors occur, the DF command will be quietly ignored, and the program will not crash. This makes it difficult to switch to integer displays. To switch a cell or range to integer, you must use the L command to move to a cell not shown on the screen, then use the DT command to change the desired cells to integer type, then use the DF command to change the display format to an integer display format. The cells will then work correctly as integer cells. A cell may be switched back to* its previous numeric format by using the command DF v1 [*] where v1 may be a cell or range. This allows temporary examination of formulas and resetting to numeric without losing the format. What Is Displayed NOTE: There are 2 items in any cell which may be displayed. One of these is the number in the cell, which takes 4 bytes if an integer, 8 if floating point, and which is displayed nor- mally using a Fortran FORMAT statement type declaration. The other is the FORMULA associated with the cell, which may contain just text or labels. This is selected by the A or L formats. Thus, the declaration DF A1:A64 [8A1] would display the NUMBERs in column A as 8 characters (as- suming this makes sense), and would cause strange errors if the number did not make sense as characters. The declaration DF A1:A64 [A] would display the FORMULA TEXT in column A on the screen. This normally will always make sense, and is the normal way to display labels. FORMAT microtutorial To reset FORTRAN Formats, the Fw.d format is useful. In that format, w means an integer giving the width in characters of the out- put number, and d is the number of decimal places to be shown. For example F9.2 has 9 places total, 2 to the right of a decimal point. A valid number in this format might be (supposing its' exact value is 123456.782) 123456.78 The format F9.1 would give 123456.8 (note rounding) The format F10.3 would give AnalytiCalc Reference Manual Page 16 123456.782 You may devise whatever formats desired. In addition, AnalytiCalc will only display the number of characters that fit in the column on screen, so display width can be set to remove undesired decimals. Other FORMAT types The default F9.2 format occupies 9 of the default 10 characters of column width. Where additional labels are needed, the 1Hc or 2Hcc formats may be used. For example, to show dollars, one might write a format as [1H$,F9.2] in the DF command. If as an alternative one wanted to show thousands of dollars with one decimal place, a format like [2H$K,F9.1] would place "$K" ahead of each number. Thus, a value of 345.67 in the first format would look like $ 345.67 and in the second format it would look like $K 345.7 (since the spreadsheet only knows numbers, not their dimensions). Where it is desired to display large numbers, FORTRAN offers the E and G formats (the form of such a format is Ew.d or Gw.d where w is the width of the number as printed and d is the number of decimal points). The E format displays the number as a fraction followed be E followed by a power of 10. The G format chooses a floating point display or an E format depending on the number's size. Note that in E and G formats, there need to be 4 spaces in the decimal field just for the exponent. The format you display in may be wider than the column; the first "n" characters of the number are shown, where "n" is the width of the column. Thus, if you use a format like F10.0 (which displays 10 digits followed by a period), the period will not show in the default 10 character column width. Available format designators for floating numbers (the ones you'll use almost always) are: Fw.d - Simple display with w characters of display (counting the decimal point) and d decimal places after the decimal; Ew.d - Display as a fraction followed by Enn where nn is the ex- ponent of 10 to multiply the fraction by. The format kPEw.d causes k digits to be displayed to the left of the decimal place, so that a format of E10.5 might display the number 73.453 as ".73453E+02" and a format of 1PE10.4 would display it as "7.3453E+01". The w is the to- tal number of characters in the display, and the d the number of decimals shown. Note that the w must have 4 extra characters to han- dle the "E+nn" part of the number. Very large or small numbers can be handled by this format where they would be too wide in F format. Gw.d - This format works like the Fw.d format provided the number AnalytiCalc Reference Manual Page 17 is small enough to fit in the widths provided, and like the Ew.d format otherwise. The display will include only d digits (or k+d di- gits if using kPGw.d), so for cases where it works, the Fw.d format is more space efficient. Several "modifier" formats are available. nHxx allows characters xx in the output. nX includes n spaces. TRc causes the next output to appear c spaces right of the current one, and 'xx' acts like the nHxx in outputting characters xx. The 'xx' or nHxx designators must be separated by comma from following format designators. Also Dw.d is equivalent to Ew.d. The optional forms Ew.dEe and Gw.dEe act like Ew.d and Gw.d except that only e digits of exponent are displayed. Fine Control Remapping Displayed Cells DISPLAY LOCATE (Map physical sheet variables onto Display Sheet) DL V1:V2 Rn:m point at display coordinates (n,m) (col,row) and, going left to right along a Row, copy in variables V1 through V2 (actually, point those display sheet cells at variables V1 thru V2). DL V1:V2 Cn:m point at display coordinates (n,m) (col,row) and, going top to bottom along a Column, copy in variables V1 through V2 (actually, point those display sheet cells at variables V1 thru V2). The DL command lets you associate any variables in the physical sheet with any location on the display sheet (the display sheet coor- dinates always count from 1 as the topmost row or leftmost column, through the maximum number there). This flexible association means that in principle every cell on the display sheet can be a "window" in the sense of other sheets, though that would be quite hard to fol- low. If the :V2 part of the command is left off, only one display cell is changed. This command is present to give "fine grained" dis- play control. Most of the time the OA and OR commands will be used instead, but this command allows you to tailor a display as needed. It is recommended that the journalling be turned on before you issue many DL commands to capture them into a file that can be played back (via @file) to recreate the display mapping. The +J file and +N com- mands turn journalling on and off. AnalytiCalc Reference Manual Page 18 SORTING DISPLAY ROWS or COLUMNS SORTING DISPLAYS DSRA#nn Sort display on row/col nn (display coord) The DS command sorts "normally" ordered display sheets. It operates something like a sorted OA command, the sort key being the number in a column or row. The command looks like this: DS Command specifier R or C - Row or Column specifier (i.e. whether to sort a row or a col.) A or D - Ascending or Descending order sort nn - row or column number to sort on. The command letters must all appear together. Thus a command to sort Column 4 on the display in Descending order would be: DSCD 4 and a command to sort Row 3 in Ascending order is: DSRA 3 The sheet should be in "normal" order (as from an OA command) prior to the sort for use. The area sorted is the part of the sheet dis- played (set by the DB command, so it may actually exceed what shows on the screen. The Amiga version of AnalytiCalc normally supports very large display sheets to permit printout of large displays which are not all visible on the screen; the screen display discards all the un-displayable points, but the print routine, and the sorts, do not.) That is, the display sheet size may be given as larger than your screen will display, and the sort will cover that area. Note that ONLY display sheet locations are sorted. A sort on the PHYSICAL sheet may be done via a cell or sheet program, but this primitive does not. AnalytiCalc Reference Manual Page 19 DATA TYPE control DATA TYPE SETTING DT V1:V2 F Set V1 thru V2 to Floating data type DT V1:V2 I Set V1 thru V2 to Integer data type The DT command sets Float/Integer attributes on physical cells. Initial default is F (8 bytes) but this can set it to I (4 byte in- teger). Computation is done in floating point always, but I type variables will be converted to integer when stored and may be dis- played in valid Fortran formats. This information is stored/restored by the P/G class commands. The type flag may also be set with a DF command, which recognizes I formats as implying that data is of in- teger type (if the first character of the format is an I) and treat other formats as floating point (real) data. The integer data can hold up to + or - about 2 billion. The floating point representation permits nearly unlimited size numbers but only has 16 (roughly) di- gits precision. Integer arithmetic is, within its valid range, exact. When integer data is retrieved, it is converted to floating point for calculations and converted back upon storage; this avoids many potential internal problems of function validity. Setting Column Width DISPLAY WIDTH SETTING DW N,M Set column N on display to be M characters wide DW lets you set any column width. They are independent of each other, and may be set in any order. Maximum width supported is 132 columns. Note that when many widths are to be varied, the VM mode is useful to prevent redrawing the screen between settings; the V mode can then be reset once the widths are as desired. Setting column widths is expected to be frequently done via a command file (see the @ command below). The column number is normally displayed across the top of the screen along with the physical cell the display column currently is pointing to on the top row. This is shown in a form like "AF = 4" in reverse video (or different colors). You use the 4 in this command. Because some commands (notably this one and the DL command above) use DISPLAY coordinates, the columns are labelled in this way to make it easy. Rows are not labelled because the labels take too much space on screen. Note that the width is set in the DISPLAY in AnalytiCalc, not on the PHYSICAL sheet. Thus, if you set the third column width to 20 (so column C initially is 20 characters wide, let's say) and then scroll over so column H is the third over, column H will have the 20 character wide display, because the width is part of the DISPLAY mapping and does NOT attach to individual cells. This is done because cells in a display column can be remapped in any way the user wants and no other consistent method is available. AnalytiCalc Reference Manual Page 20 EDIT command EDIT FORMULAS ED 'string1'string2' Replace string1 with string2 in current cell formula. Either string may be null if desired. The ED command allows editing of formulas with some extra bells and whistles. Basically, the new string replaces the old one, EXCEPT that it is possible to do some limited argument replacement. The constructs &1 through &4 will be replaced by arguments 1 through 4 as last entered by the "-" command (see below). These may be from 0 to 51 characters long each, and multiple & constructs may appear in a substitution. No & constructs may recurse, however. A second ED command can have the effect of the recursion in some cases however. The construct &5 will evaluate accumulator Z and, if it is in a range where it may be so represented, take its integer part and con- vert it to a numeric string (up to 10 characters) in the place of the &5. The construct &6 will replace the cell with the contents of the integer part of accumulator Z provided that that number is between 1 and 256. Thus, any single non-null ASCII code can be entered. Ac- cumulator Z is used to avoid conflicts with %, which is used too fre- quently to reserve for this purpose. To set up Z interactively to a constant, the command "TE Z=nn." will set Z to the value nn. The EDited formula is re-entered into the cell after the edit. If the initial string is null, the EDit command appends the replacement string onto the contents of the cell. If the replacement string is null, the corresponding contents of the cell's formula are deleted. Note that ALL matching strings in the formula are replaced, not just the first. If automatic motion has been specified after an Enter, it also occurs after an EDit. This allows a "dummy" EDit (e.g. ED / / /) to simply move in the default direction without changing any formulas. AnalytiCalc Reference Manual Page 21 ENTER (Entering Data and Expression Formats) E expression Enter expression and set as formula or text based on contents ET expression Enter expression and set as text E" expression Enter expression and set as text. Note this form does NOT translate to upper case. EV expression Enter expression and treat as computable. This command enters the given expression at your current cursor position, and may optionally move in some direction that has been set up by the M command (Move) after the entry is made. The Expression may be text or one or more numbers, functions, or equations (separated by the \ (backslash) character if more than one per cell) in fully parenthesized infix notation. The expressions en- tered resemble normal algebraic expressions, with the usual operator precedence. There are 26 global accumulators named A through Z (plus a special one named %) available; the result of a calculation is always left in the % accumulator as well as the cell. Operators available include addition (+), subtraction (-), multiplication (*), division (/), and exponentiation (** or !), with normal precedence. Many functions are also available, described below, for other opera- tions. The operator evaluation order is functions, exponentiation, multiplication and division, addition and subtraction, just as in Fortran and most programming languages (and on most calculators). Parentheses may be used to group calculations to override these defaults. Simple examples of Enter are: E 523.458 Enter the number 523.458 into the current cell E SUM[A3:A15] Enter the formula to compute the sum of cells A3 through A15 and produce the result in the current cell E MONTHLY REVENUES Enter the label "MONTHLY REVENUES" into the current cell. Notice that the string does NOT have any of the characters .[(+_ in it, so it is treated by default as a label. E (A2+A4+A6+A8+A9*A10+A11*A12)/(SUM[A2,A4,A6,A8]-A13) Places the above formula into the cell AnalytiCalc Reference Manual Page 22 so that it will be evaluated. The multiplications and divisions are done before additions and subtractions, as is the normal mathematical standard. The formulas may of course contain references anywhere on the sheet; they do NOT have to all refer to the same row or column. The text entered becomes the Formula for that cell, and AnalytiCalc takes a guess as to whether that text is just text or a number to compute. If the text contains any of the characters (, +, -, ., or [, it is assumed to be a formula, and the cell is marked as containing a computable formula. Otherwise it is just treated as a label and the formula itself is displayed. If you find the guess is not to your liking, the DF command below lets you alter the format and tell AnalytiCalc that the entry is / is not a numeric. An addi- tional special character is the '"' (double quote) character. If this character appears any- where in the entered formula, AnalytiCalc will take the formula as text, regardless of whatever else is in it. You can also combine comments and formulas by using a separate state- ment and the *C construct (e.g. SUM[A3:A12]\SQRT(%)\*C SQUARE ROOT OF THE LENGTHS) of Calc. Once a cell is set up as numeric or text, it remains as it was originally set until changed (using the DF command, described below). The ET and E" forms of the Enter command force the expres- sion to be entered as non computable so that it is displayed rather than computed. The EV form forces it to be computed, regardless of whether any special characters exist. Numbers MUST have decimals to compute properly however. Thus "EV 45" is INCORRECT, where "EV 45." is CORRECT (assuming you want the value 45. to be computed; the first will not). If the decimals are omitted, the expressions may be considered integer. These will be converted, but the conversion has some limitations and is advised against. Also, integer division truncates results to produce an integer result (so that 7/2 would evaluate to 3, while 7./2. would evaluate to 3.5, for example). Thus, to avoid surprises, use decimals. Commands will be translated to upper case unless a " character appears anywhere in the command. If a " character is found, only the first 2 characters are converted. Most keys map to indirect files to make customizing AnalytiCalc easier. These files must be in your current directory to be found AnalytiCalc Reference Manual Page 23 (or a path search utility must be used). The system help file is called PCCHELP.HLP and must also be in your current directory to en- able the HELP command to work. However, where filenames are prompted for, AnalytiCalc can accept full pathnames. If the program cannot open a file for reading which is requested, it will attempt to create an empty file (and warn you of this on the top line of the screen). Within command files, it is sensible to permit entry from the console of commands after prompts. Therefore the % command has been added to facilitate such entry. There is an alternative method of handling prompts (the "-" command) but this one is somewhat more flexible. You will find examples of this in some of the .CMD files supplied with AnalytiCalc, and may use it in your own if desired. The operation is as follows: a line beginning in % is assumed to have several fields: %FLD-1%FLD-2%FLD-3 in which FLD-1 if present is typed out to the screen first. Then FLD-2 is passed as the command. If FLD-3 exists and starts with a ? or \, then the terminal is read and the result is added onto the end of FLD-2 before the command is passed back to the spread sheet. If the command begins with a space or control character or a \, then it is commented out and the command file falls past the % line. If however FLD-3 started with ? and the initial read-in character is not \ or a control character, the command file is rewound. This feature is used in the enter command file (AKD.CMD) to implement an "enter mode" somewhat akin to that of Visicalc, Supercalc, or others of that ilk by automatically inserting the "E " of the AnalytiCalc Enter com- mand until a \ is seen. If FLD-3 should be present and start with a & character, the auxiliary file is closed there and console input done as in \ input. (Note that a control-Z may be used to terminate Enter mode also, and F9 is preset to terminate it.) This func- tionality is probably less convenient than just using the /; command to go into "enter-mostly" mode now, and should probably be ignored. It is handy in templates to be used by novices though, since it gives extremely easy console data entry in a controlled fashion. Examples The expressions that are entered can be numbers or formulas re- lating numbers and other cells. They may also consist of several statements separated by a \ character which will all be evaluated at that cell, up to 110 characters. Thus, valid commands might be E 55.234 Enters value 55.234 at current cell E B4+H2+D5 Cell will contain the sum of cells B4,H2, and D5. E Totals Cell contains literal text "Totals". FORMULAS AnalytiCalc Reference Manual Page 24 Formulas may be used too. These include 2 classes. The first are the multi-argument functions, which either appear at the start of their statements or substatements (alone), or inside expressions. The difference between the classes is that the multiargument func- tions must have cell NAMES or RANGES as their arguments, while the single argument functions can have arbitrary Expressions in their ar- guments. Otherwise they may mix freely. All functions listed below may appear in command files or cells, and arguments must be cells, not expressions or constants. The IF statement, while it is treated as a function syntactically, may NOT appear inside general expres- sions, but must begin a cell equation or sub-equation (i.e., it may begin after a \ in a cell containing several operations). It cannot be used inside command files driving cells, but its functionality is supplied there by the "*J label" function, so no real limitation ex- ists on capability. The available multiple argument functions are: SUM[variables] Sum of all arguments MAX[variables] Max of arguments MIN[variables] Min of arguments AVG[variables] Average of arguments AVE[variables] Average of arguments ignoring cells containing 0.0 (e.g., labels) STD[variables] Standard deviation squared IF [V1.rel.V2] statement | else-statement AND[variables] Boolean AND of all variables in list IOR[variables] Boolean inclusive OR of variables NOT[variable] Boolean complement of variable XOR[v1,v2] Boolean exclusive OR of v1,v2 EQV[V1,V2] Boolean "equivalence" of V1,V2 (complement of exclusive OR, true if bits have the SAME value) CNT[variables] Number of nonzero variables in list MOD[V1,V2] Returns V1 modulo V2 (i.e., remainder of V1/V2 division.) SGN[v1] Returns 1.0 times sign of V1 LKP[var,variables] Lookup variable in "variables" range greater or equal to var, return its index (starting with 0) into variables range. LKN[var,variables] Lookup variable in "variables" range less than or equal to var, return its index (starting with 0) into variables range. LKE[var,variables] Lookup variable in "variables" range strictly equal to var, return its index (starting with 0) into range. Note all LKP, LKN, LKE return the last variable index if no satisfactory value found. NPV[disc,vars] Net Present Value of vars (equal time AnalytiCalc Reference Manual Page 25 interval numbers), at discount rate disc where disc is a fraction (e.g., .12 for 12%). Be sure the arguments are all cells, not literals, including the disc argument. Set a cell to the fraction desired for the disc argument and use its name in the arguments to NPV, rather than attempting to use a literal (e.g., 0.12) directly. IRR[PV,FV,returns] Internal Rate of Return. Will compute internal rate of return on up to 20 periods, returning rate per period. The returns are expected to be at equal time intervals. PV and FV are initial and final values of investment and the result is computed via Newton approximation. As with all functions here, be sure all arguments are cell names, not literal numbers. PMT[princ,inter,nper] Payment function. Generates payment given principal amount, interest per period (as a fraction), and number of periods. PVL[pmt,inter,nper] Present Value function. Generates present value of a loan given payment amount, interest rate, and number of periods. All financial functions assume that payments are for equal time intervals. RND[V1] Random number. Generates a random number distributed uniformly in the interval 0.0 to 1.0. An argument is required but never touched. CHS[Idx,Range] Choose cell from range based on value of cell Idx. If cell IDX contains 1, first element of range is chosen; if 2, second, etc. Zero is returned if out of range. Range can be any comma-delimited list of ranges also. ATM[A1,A2] Arc Tangent of A1/A2, but four quadrant version. Note these are single cells. In all cases except IF, arguments are variable names separated by commas, or variable ranges, or mixtures separated by columns. Note too that ALL THESE FUNCTIONS REQUIRE CELLS OR ACCUMULATORS AS ARGUMENTS; literals and single argument functions may NOT be used here. (That is why it is easy to place multiple commands on a line; the ACs or cells may be loaded in "separate" steps.) The IF [] state- ment is a partial exception since a range may be used to replace ei- ther variable inside the [] condition brackets. The sum of the ranges will be used in that case. A possible IF statement might look like IF [S2.LT.S5] R=2. | R=0. The Boolean variables are computed by taking the integer values of the variables and performing the appropriate boolean masking opera- tions on them. The values are returned as whatever type (floating or integer) the cell is. 32 bit integers are used. The CNT operation AnalytiCalc Reference Manual Page 26 will (generally) omit cells that have not been initialized, since these have a value of 0.0 and are not changed unless modified expli- citly. It may be used to compute a more accurate average than AVG where many uninitialized cells add zero values in. The quotient of the SUM divided by the CNT of a range will give an average of nonzero cells only. The LKP function quickly locates a cell in a list where the value is in a desired range; the *G function may be used to ex- tract other cells close to the addressed cell (e.g., beside it or below it). The LKN and LKE functions are exactly the same but search for lower or equal values instead of greater than values. This makes varied search strategies easy to do. These functions are most useful in user command files for selecting parts of the sheet according to criteria, as they allow very fast searches of large parts of the spreadsheet from inside command files. In this application the LKP class functions would be accessed via the TEst command. The NPV function is the standard financial Net Present Value function, assum- ing the discount rate is the fractional interest percent per period for the amounts in the range given, and that the range of numbers represents equal time periods. Remember to give a FRACTION rather than a Percent, so a 5% interest rate per period should be given as .05 for example. The IRR function will compute IRR given present value, future value, investments per period (for up to 20 periods), and assuming that the periods are equally spaced. The IRR value re- turned is per period and must be multiplied by periods per year to get a yearly rate. Note that other financial functions like PMT may be easily obtained by simple cell programs. The RND function is provided to give random numbers. This is useful in some types of Monte Carlo financial modelling. If distributions other than uniform ones of random numbers are desired, they may be obtained via standard mathematical transformations from this uniform distribution. The random number sequence is identical every time the spreadsheet is run. BE SURE IN USING THESE FUNCTIONS THAT ALL ARGUMENTS ARE IN SPREADSHEET CELLS OR ACCUMULATORS. LITERALS ARE NOT PERMITTED AS AR- GUMENTS. Thus, a command like "SUM[A1:A5,5.5]" is illegal. Instead, use a command like "A=5.5\SUM[A,A1:A5]" or "A=SUM[A1:A5]+5.5". The section on the interactive calculator (CALC) mode also men- tions the single argument functions available. These include normal trig and math functions like SQRT, SIN, COS, ATAN, AINT, ALOG, EXP, SINH, COSH, and some others. These single argument functions may be used in normal infix notation expressions, and do not have restric- tions on the types of arguments permitted them. See the section on Calc mode below for the full list. There are also a large number of functions that are syntactically commands, but which perform calcula- tions. These are also described in the Calc section and include date and matrix arithmetic and some addressing arithmetic functions. AnalytiCalc Reference Manual Page 27a Summary of Variable Addressing Variable names may be of several forms; examples will be given first to illustrate. D5 - Cell is in column D, 5th row down. (Col. 4, Row 5) D$5 - Cell is column D, 5th row down, but in relocating copies the column (D) is not altered (unless the global override is set). D5$ - As above but the row is not alterable. D$5$ - As above but neither row nor column may be altered...an "absolute" reference. P#4#7 - Cell addressed is 4 columns to the right of and 7 rows down from the current cell. (P#0#0 is the current cell). These are computed on the base spreadsheet. This form of addressing is not subject to relocation, ever. D#4#7 - Cell addressed is the cell DISPLAYED 4 columns to the right of the current one and 7 rows down. Since the display map can be anything, this means the actual cell referenced can be anything, depending on the current display. This form of addressing is not subject to relocation, ever. P#%AB - Supposing accumulator A contains 4 and accumulator B contains 7, the cell referenced is 4 columns right of the current one and 7 rows down on the physical sheet. D#%AB - Supposing accumulator A contains 4 and accumulator B contains 7, the cell referenced is 4 columns right of the current one and 7 rows down on the display. P#_AB - If accumulator A contains 4 and accumulator B contains 2, the cell addressed is D2 (4th column over, 2nd row). P#$D3:C2: - if cell D3 contains 4.0 and cell C2 contains 2.0, the cell addressed is D2 again. AnalytiCalc Reference Manual Page 27 Variable Names A valid variable name might be H14 or N22 or A3, or the position independent forms P#n#m and D#n#m where n and m are offsets in Physi- cal or Display sheets from the current column and row. A valid range is 2 variable names separated by colon (:). Thus the range B2:B12 is valid and means B2 through B12 inclusive. The range expression D#0#1:D#0#12 is also valid and means the cells on the display sheet in the same column but 1 to 12 rows down from the current cell in- clusive. (The displacements may be negative too, with - sign to in- dicate it, as D#-3#2 for the display cell 3 columns left and 2 rows down from the current position.) The P# and D# forms are valuable since they work wherever you move them. Other variables may be relo- cated when moved (renaming them to correspond to their new loca- tions), but the P# and D# forms need never change and allow working with projections across the physical sheet in various ways. There are a few restrictions with these forms. They usually need to be en- closed in parentheses (e.g. (P#3#5) rather than P#3#5) inside ex- pressions. For most purposes they do work interchangeably with other name schemes however, though the "normal" naming scheme is generally easier to use. Copying with relocation works with normal cell names, provided that the cursor is above and left of cells to be renamed (see below). The keypad copy command normally moves to A1 during a copy for this reason. The D# forms do not always work in the ex- tended address areas of the sheet, depending on recalculation mode. The special construction P#%ab or D#%ab allows relative address- ing relative to current physical or display location using accumula- tors named a or b (where a or b may be any accumulator name from A through Z but upper case is REQUIRED). This computable relative ad- dressing affords greater flexibility in command procedure construc- tion. In Amiga AnalytiCalc, the cell address form P#_cr is available. This address provides absolute cell addresses using the numbers in the two accumulators c and r for the column and row of the cell to be accessed. Since accumulators run out somewhat quickly in AnalytiCalc, an alternate addressing form is also provided. If cell Cn holds a column number and cell Rn holds a row number, the cell name P#$Cn:Rn: will address the cell whose column is in Cn and whose row is in Rn. Such addressing may be indirect to one level only; that is, the Cn and Rn may be any legal address form for a cell EXCEPT the P#$cell1:cell2: form. These address forms permit a degree of data driven computation unavailable in other spreadsheet type programs. If a formula contains the construction {VN (where VN is any cell name), the FORMULA for cell VN is inserted into the formula being evaluated at that point, when the cell value is to be computed. This may be done several levels deep, but is stopped after 10 translations to avoid infinite loops due to circular references. Some care must be used here since the final translated formula must fit in 110 characters and will be truncated if it gets too long. This construct is useful where formulas are to be tested. These cell names may (in AnalytiCalc versions for 320K and over) be replaced by alphanumeric names of your choice; this will be discussed below. The formula AnalytiCalc Reference Manual Page 28 replacement feature provides a handy inline macro capability permit- ting complex formulas to be entered more easily. This feature may be used with the P#%ab form of addressing to provide runtime addressing of formulas anywhere in the matrix. It may also be used (with the position relative P## forms of cell names) to conserve storage where a long formula must be repeated many times, at the cost of some speed. The command processor may use these inline macros in commands. If a command line contains 2 successive underscores (__) anywhere in its length, the underscores will be stripped out of the command line and the command line will be scanned for {Vn forms. If any are found, they will be substituted for prior to interpretation of the command. This will occur AFTER any console text is read if console text is required by one of the %%% forms of command line. This per- mits formula cells to be used as a convenient place to store inter- mediate answers in models being driven by command files. It is not particularly useful from the console, but works, and may be used to reenter a formula without moving it explicitly if there is a reason to do so. (ENTER __{Vn). Since it is sometimes useful to have mixed formulas which allow certain cell references to be absolute and never relocated, with other references which are relocatable, cell names may be specified absolute. If the character "$" (dollar sign) is placed after the letter, the number, or both, parts of the variable name, the part of the variable name followed by the dollar sign is absolute. Thus, the variable AB30 is relocatable in both row and column. Variable AB$30 is relocatable in row only (the 30 is not followed by $), and the AB column will not be changed in any relocate. Variable AB30$ is relo- catable in column, but the row 30 reference does not change. Vari- able AB$30$ is totally absolute. Note that these absolute variables are harder to move around even on adding or deleting rows, so they should be used with care. This relocation protection can be tem- porarily inhibited with the OV + command and reinstated with the OV - command. This will allow copy of cells with relocation if it is needed, without laborious editing. The program starts in OV - mode. If the OV + mode is set, the program will remain in OV + mode until OV - is set up again. The commands may be given in the form OV-, OV -, OV+, or OV +. Variable Ranges Variable ranges must lie in a single row or column. Erroneous specifications are rejected and ignored generally. For versions of AnalytiCalc supporting matrix operations, matrix dimensions are however given by specifying the upper left and lower right corners of the matrices. Thus the range A1:B2 is the 2 by 2 matrix containing cells A1, A2, B1, and B2. For operations using full rectangular re- gions, support is offered via auxiliary keypad commands which allow filling of large regions or moving data into them. This is done via command files automatically invoked, but the methods used are tran- sparent to the user. Other operations on rectangular regions not confined to a single row or column may easily be added in the same way. The internal implementation permits copying rows to columns and vice versa as well as rows to rows and columns to columns; this is why regions must be one or the other. It is expected that regions are specified starting at top left and ending at bottom right. This AnalytiCalc Reference Manual Page 29 must be done; regions will not be completely handled otherwise. By using in-cell loops for their innermost computations, these macros are quite quick in execution and offer few disadvantages. They do however ask separately for the bounds of the areas being modified. Function Arguments Note that the square brackets ([]) are needed to delimit arguments in multiple argument functions. The results are left in accumulator % and in the cell. The IF form will compare 2 variables and execute either the statement or the else-statement (which must NOT be another IF). The valid relations are: .EQ. Equal .NE. Not Equal .GT. Greater than (V1 greater than V2) .LT. Less than (V1 less than V2) .GE. Greater than or Equal to (V1 >= V2) .LE. Less than or Equal to (V1 =< V2) and must appear in this form. (Note that your formulas must generally be in upper case only; text can be anything.) Unary Functions The second class of formulas permit fully general arguments of expressions, cells, and the like (and are evaluated AFTER the multi argument functions lexically). These are more conventional, use parentheses, and include single argument functions like SQRT, ALOG, EXP, SIN, COS, AINT, ABS, ATAN, and so on. See the built in interac- tive calculator mode manual section below for more detail. However, they may use variable names as anywhere, not just single letter ac- cumulators. Thus, SQRT(A5) is a valid number. There is a function list in the documentation of Calc mode later in this manual. Accumulators (Scratch Storage) The single letter accumulators of Calc mode (letters A through Z and %) are available globally and may take on all Calc mode data types (see the description of calc mode below) including multiple precision. They are usable globally on the entire sheet and may serve as arguments to command files or return cells for results. In- side command files, Calc mode can perform any of its usual calcula- tions. The multi-argument functions listed above are not available there (though they can be simulated by enumerating cells), but every- thing else is. The letter accumulators are available in all cases in addition to cells. Of course, any unused cell may be used as an ac- cumulator too, but cells in the spreadsheet are at most 8 bytes long and cannot hold multiple precision arguments. AnalytiCalc Reference Manual Page 30 To get, e.g., the square root of the sum of several variables (in a row or column; ranges may only be in rows or columns together) one might have a formula line like: SUM[A1:A20,B4,F2,C5:C20,J3:M3]\SQRT(%) which computes the sum of the 2 cells and 3 ranges and leaves it in %, then computes the square root of that and leaves it in the cell. The formulas to Calc may be complex but MUST NOT be longer than 78 characters. Break up longer ones so they can be fed to Calc in smaller chunks. Notice that the expression given could have been written as SQRT(SUM[A1:A20,B4,F2,C5:C20,J3:M3]) but was broken up for tutorial purposes. Auxiliary Commands and Functions The commands of CALC mode should be examined carefully for usefulness. You can enter programs into CALC mode which can be used as user-defined functions from the sheet if need be. See the do- cumentation about the Calc mode (interactive calculator) for full lists of functions and further information. Several commands available in cell calculations which affect computation exist; several will be described here as well as in the section on calc mode (which describes single cell calculation facili- ties in addition to those described here). It should be noted that these functions start with the * character which also starts comments in command files to AnalytiCalc. Because of this, the "Enter-mostly" mode treats any text that starts with * as a comment. To enter a FORMULA beginning with an * character in "Enter-mostly" mode, you must use the syntax /EV formula where formula might be something like "*U IDATE" (without the quotes), i.e., something that may start with a * character. This method allows the formula to be entered correctly. Notice this is the "Command-mostly" command for entering a formula. It ALWAYS works, regardless of the type of formula. One of these is the *G command. Its syntax is *G V1,V2 which has the effect of evaluating variables V1 and V2 as column and row numbers on the physical sheet and returning the physical sheet variable addressed by those variables. The ranges are clamped to fit on the sheet. This gives indirect addressing, permitting lookup tables to be made up by creating the index you want and then AnalytiCalc Reference Manual Page 31 accessing the addressed table element. Typically the *G command would be the last on a formula line, though it need not be. Its value is returned in the % accumulator. Another is the *W command, which simply takes the number in % and places it in the formula for the current cell on the spread sheet. This may be used to "freeze" a value, replacing a formula by a constant for future use. Note that the previous contents of the formula entry are lost. Thus, a *W command in the sheet will nor- mally cause itself to be overwritten after one use. The number is stored in a format D35.24 in the worksheet. To implement user functions, if Calc sees an expression of the form "*@filename.ext", it will execute that file as a command at that cell. Any filename may be used and as many functions as desired can be handled in this way (with obvious overhead), with nesting depth of up to 3 permitted. These command files correspond to "macros" in other spreadsheets, and exist outside the sheet, so a command file can be used by many worksheets. AnalytiCalc has also a command called *P which allows resetting of the physical coordinates while executing a cell's formulas, FOR THE DURATION OF THAT CELL'S FORMULA ONLY or while in K mode (calc mode) only. The syntax of the *P command is either just *P (which causes AnalytiCalc to prompt for column and row), or *P Variablename (which moves the current location to the named location), or *P@ variable1,variable2 (which moves to the location which is defined by variable 1 and 2 as col. and row). There are several other useful calculation commands, a few of which are summarized here. The command *F LABEL will evaluate the % variable. If it is positive and nonzero, then the current input line will be rewound and the program will (silently) seek a line of the form *CLABEL. This allows loops to be defined with % as the counter. This command acts on the unit used by the AnalytiCalc @ command, NOT the unit for the Calc *@ command, and is a way that the Calc formula conditionals may be reflected back into its command processing. The TE expression command should be used first to set the % variable as needed to any arithmetic expres- sion to control the loop. The command *J LABEL will act as does *F LABEL but its search will be in the current file opened by a Calc *@ command (i.e., an *@file inside a cell on the sheet, or given from the console after a K command [see below] to enter bare interactive Calc). AnalytiCalc Reference Manual Page 32 The command *QF or *QW file ?searchkey? <cc> ("Query Float" for getting a real ("floating point") number into the sheet from a file, or "Query Write" to write part of a file into the sheet) will open file and look for lines starting with searchkey as given (with _ as a wild character). Upon finding this, it will look for the <cc> and if it finds it, it will select the part of the line in the file between the 2 characters given (which may be alike) as they are first found, and will do one of 2 things. If the command was *QF, then Calc will Float the value found there (attempting to use a large D format) and return it in %. If the command was *QW, then Calc will Write the selected part of the record into the formula at the current physical position on the sheet (which may be reset by an IMMEDIATELY PRECEDING *P command). The record may contain the search key anywhere. It is up to the user to make sure the formulas copied make sense. This however allows access to sequential files of data which can be created by an editor, word processor, etc., and which might hold keyed information that the spreadsheet should be able to manipulate. As a user convenience, there may be a second key specified (also inside a pair of ?s). If two keys are specified, Calc will search for and require that BOTH be somewhere in the record accepted. Note too that the files opened may not have any records longer than 128 characters per line. There are 2 variant forms, in- volving the 4th character (i.e., the next one after the F or W of *QF or *QW). These may permit sequential access to a data file. The *QFK command is exactly like the *QF command (ditto *QWK and *QW) BUT it will not close the data file after finding the desired cell unless it reached EOF. This permits subsequent commands of the *QFN command (for "Next") to continue reading the file. The *QFN command will continue reading the file (the file spec must be there but is ig- nored) from the next record and will also not close the file. (Ditto *QWN.) The *QFC command will act exactly as *QFN BUT will close the file when the request completes. Thus, a file may be opened with a *QFK (or *QWK) command, subsequent data retrieved by *QFN commands, and finally the last data retrieved with a *QFC command. Also, *QF commands will close the file prior to opening another. The command mnemonics here may be read as Query File [Next/Close/Keepopen]. Formulas entered may have the following constructs which will be edited into the formula as entered (so a command file can enter them): _@V1,V2 Means get the values stored in V1 and V2 and use them as column and row locations pointing at some cell in the sheet. Replace the construct with the name of that cell. _#V1 Means take the real number in cell V1 and unpack it as if it had been a packed value from a formula with 8 characters packed; then convert it back into ASCII and place in the formula in place of this construct. This construct is intended to be used with the *U STRVL AnalytiCalc Reference Manual Page 33 function to allow retrieval and edit of formulas. The *U XQTCM function permits use of the EDit command (as well as others) within a cell for string manipulation. Only 8 characters at a time may be represented (the limit imposed by cell size), but repeated use allows handling any desired length. An application would be setting the cell used as V1 to a packed representation of either of two strings inside an IF [] and then EDit- ing the string into the formula as desired where one of the 2 strings should be made to appear. This capabil- ity is crude but provides the bare bones functionality. Additional USRFCT calls to give simpler (but more ver- bose) string manipulations are not too hard to add should the user need them. Formula Length Formulas have 109 characters allotted to them for use inside the spread sheet. It may happen these are not enough for use. In other spreadsheets, the technique is to put the formulas in other cells in- side the sheet and possibly use screen windows to avoid display of the intermediate results. While you can use that technique here too, you have an alterna- tive. It is to use an indirect file in that cell and inside the cell in the spreadsheet just have a command "*@FILE.TYP" where the file is named "FILE.TYP" (you pick the real name). Inside this file may be whatever calculations you like. So long as the file is to operate in a fixed cell location in your spreadsheet, you may use variable names like A3, D1, etc. in- side the file. Your command files may include any multiple argument functions except the IF function. (The *F and *J functions can be used to provide equivalent functionality inside command files however.) The only restriction on multiple argument functions is that their arguments must be cells, not general expressions. Once you add or delete rows or columns, or copy a formula, though, the file may not work as expected if the cell calling it has moved. Since the relocation done by AnalytiCalc only operates on formulas stored directly in the spreadsheet, it cannot relocate vari- ables in the indirect files. However, you can make the files "posi- tion independent" so they WILL work wherever they are called from. (This has the side benefit that a single file can be called from many cells.) The way to make the files position independent is to use location-relative variable names instead of absolute ones. That is, inside cell B4, the cell immediately above can be named B3 (the ABSOLUTE name), or it can be relocatably named P#0#-1 (the relative AnalytiCalc Reference Manual Page 34 name). Likewise the cell one to the right would be named either C3 or P#1#0, and so on. The cell two down and one left would be either A5 or P#-1#2 (you can begin to see how it goes). A file containing only location-relative variable names will work even if called from a different cell, since the references will be figured out when used. These references are more difficult to write than the absolute ones, but since they work without renaming, they provide a level of flexi- bility needed where long formulas must be used. These forms should be used wherever indirect files are used to hold cell formulas. The computed cell name forms like P#%ab can be used also provided the a and b accumulators are initialized properly, or the command file does its own relocation using the X and Y accumulators set on entry. By using command files, it is possible to add your own functions to the spreadsheet. Just devise a command file that performs the arithmetic and addressing you need and use *@file.typ where you want to evaluate the function. Arguments can be passed in accumulators and P## or D## forms can be used for cell addressing (note that it's P#col#row and D#col#row). The *U HERE function can be used to return the current location in two accumulators to allow address arithmetic. Linking Multiple Sheets To enhance AnalytiCalc, it is possible to have a cell refer to cells from another, saved sheet anywhere in the system. The sheet must normally have been saved NUMERICALLY (see PPN command below) starting at the upper left (i.e., cell A1) as would be the convention for using with ACGRAF, the associated graphics output utility. The command *XV filename V1 will eXtract the Value of cell V1 (V1 is any variable name, e.g. C15, AA3, F9, B34, etc.) from the saved sheet saved in filename. The filename is any legal host system file specifier. Note that there may be as many files as desired here, but only one cell is returned. If it is desired to extract a formula from a saved sheet saved NON-NUMERICALLY, the command form is *XF filename V1 and this will cause the formula in "filename"'s cell V1 to overwrite the current physical sheet. To avoid this, one might write a formula like *P V2 \*XF myfile.typ V1 which would cause current sheet variable V2 to have its' formula loaded from saved sheet "myfile.typ"'s variable V1. Note too that the GPR or GDR commands will load saved sheets onto the current sheet (starting at the cursor) with all formulas relocated, so that entire saved sheets can be merged quickly where the one by one operations just described would be too slow. AnalytiCalc Reference Manual Page 35 In general, save your sheets symbolically (non-numerically) for archival purposes; save numerically for post processing or handling of data by "master" sheets or for graphics. You may want to edit AUXKPD.TXT so that F4 saves the entire sheet both symbolically and numerically if you frequently do both and want the two to be in syn- chronism. The default save, PPX, now saves both numbers and values so it is normally all that is needed. Loading Non-AnalytiCalc text files onto the Sheet FILE LOADING F filename/nnnn Read filename, skipping nnnn lines if the "/nnnn" is there, and load contents onto the currently displayed screen, broken into columns as the current screen is set up. This command is designed to make it easy to load document files into AnalytiCalc in quasi-image mode. Once loaded, the characters are in formula cells and may be moved around the sheet. The display format must be reset to get them to be treated as non-text; this command will reset all loaded cells to text format display. It repaints the screen upon loading. This feature makes it easy to cap- ture data in AnalytiCalc from other unrelated programs. You can com- pute with cells that are loaded in this way but contain valid numeric data by changing them to numeric cells by the DF range [*] command. If you need to add decimals to the end of the cells' numbers, the ED //./ command will add a decimal to the end of a number and the ED / // command can be used to get rid of extra space characters in a formula. AnalytiCalc Reference Manual Page 36 Getting Saved Sheet Data from Files RESTORING YOUR SPREADSHEET GD Reload sheet down/right onto display coordinates. GP Reload sheet down/right onto physical coordinates. GB Reload sheet like GP but from sheet saved in binary mode with PB command. GPR Reload sheet onto physical coords and Relocate formulas as if the saved A1 cell were moved to the current cursor. GDR Similar to GPR but on display coordinates. Gxx+ Add saved sheet (if default save) values to current values; enters Recalc Manual mode. Gxx- Subtract saved sheet (if default save) values from current values. Also enters Recalc Manual mode. "xx" can be PR,BR, PX, etc. The GP/GD commands are the inverse of the PP/PD commands and load those files to potentially new locations. This is the simple way to merge sheets. In these cases, you are separately asked for filenames and limits on what part of the sheets to load/save. Only cells that are marked in use are saved, so sparse sheets take less space than saving the direct access files. Note that on save or re- store the ranges to restore are asked for. To restore the sheet starting at the cursor, specify the lowest row/column offset of 1 and the highest either 0 or a very large number. Any rectangular region can be restored by giving the X and Y coordinates of the lowest and highest offsets from the upper left to restore, counting from 1. If the P or D is omitted from the GPx or GDx commands (where x is either R or anything else), the action will be determined by whether the sheet was saved by a PP or by a PDx command. REMEMBER: THE LOW OFFSET SHOULD BE 1 OR LARGER, NOT 0. The GB commands restore spreadsheets that have been saved with the PB commands. The save formats of the PP and PB commands are NOT compatible, so sheets saved with PB must be loaded with GB. The *X and graphics routines assume the PP form save, but the PB form is faster. Spreadsheets saved with PB cannot safely be transported to other machines, either. Spreadsheets saved with PP can be moved from any AnalytiCalc to any other. GB commands take the same modifiers as GP commands do. The GPR or GDR commands relocate a saved sheet to a new loca- tion. If for example a sheet is saved from A1 on down and right, if it is loaded into another sheet when the cursor is at another cell, all formulas are relocated on load so that they are correct for the new location. (This follows the normal relocation control rules so that the $ forms of names may inhibit relocation.) The intent of this is that it makes it trivial to merge multiple sheets and make summary sheets. In AnalytiCalc/Amiga, the logic to write to cells involves checks of the built in symbol table to ensure that cells are not pre- viously there and needing to be replaced. This slows the Get opera- tion considerably, and is normally bypassed. However, where a saved sheet is being merged into an existing sheet (rather than restoring into a previously empty region of the sheet), the full logic may be AnalytiCalc Reference Manual Page 37 needed for correct operation. To get it, enter the filename followed by "/M" (so that to read SAVESHEET.PCC in the full-update mode you reply SAVESHEET.PCC/M to the "Enter Filename>" prompt, rather than just SAVESHEET.PCC, as an example.) (M stands for Merge.) The + or - modifiers in the 4th character of the command allow you to add or subtract multiple saved sheet values. These require that the normal default save (which includes both numeric and formula data) have been done to save the sheets being added. The values in the sheet have the saved values either added (if +) or subtracted (if -) from the current values. The manual recalculation flag is set to prevent any automatic recalculations also, since formulas are loaded from the last sheet loaded and overwrite other formulas in the same cells. (This is generally not a problem.) To manipulate these values if there is a need to, the TEst command and the matrix math functions (which include multiply matrix by constant) allow two dimensional areas to be moved around or multiplied by constants without turning on the sheet recalculation. Thus, averaging 4 sheets could be done by a GP command, three GPx+ commands, and then a TE using the *U MMPYC function to multiply the whole area by 0.25. (One could use TE A=0.25 to get a "cell" A0 to use in the MMPYC function if no 0.25 value existed to use already.) The AnalytiCalc HELP System HELP Hn Display a screen of help. (n from 0 thru 9) The H command will display a screenful of help giving a summary of many AnalytiCalc commands and syntax options. When you type a second return, it will redraw the screen. The F1 key also means 'Help', and the ALT-F1 key draws a diagram of the function key assignments. Control F1 will draw a one line reminder HELP message on line 25. There are multiple help pages. They are selected by commands H0 through H99, the number defaulting to 0 and selecting which help screen will appear. The PCCHELP.HLP file must be in the default directory for this to work correctly on Amiga currently. AnalytiCalc Reference Manual Page 38 MOVING TO LOCATION (GOTO anywhere) LOCATE CURSOR (GOTO somewhere on the sheet) L variable-name Position cursor at variable-name The L command moves the current position to the variable named on the Physical sheet (and the display sheet if the variable is on the display). Examples might be SKIP 1 L B3 Move to cell B3 L D6 Move to cell D6 Setting Direction for Motion after Entering Cell Contents MOTION DIRECTION (Set Default Direction to move after an Enter) THIS COMMAND SETS DIRECTIONS FOR AUTOMATIC MOTION AFTER ENTER ONLY. SEE BELOW FOR HOW TO MOVE THE CURSOR AT OTHER TIMES. M0 (Restore last auto move direction) M1 (Move up) M2 (Move down) M3 (Move left) M4 (Move right) M5 (No auto motion) The M command causes automatic cursor motion after an E command in the direction indicated. This only occurs after an E command. The default condition is not to move at all after an E command, and this may be reestablished by a command of form M5 (actually Mc, where c is any character other than 1 to 4.) The M0 command restores the last automatic motion direction and saves the current one as the new last direction. This is useful where a command procedure needs to turn off auto motion but wants to turn it back on at the end; a dummy EDit may be used if a single motion is needed. AnalytiCalc Reference Manual Page 39 Including Multiple Files into Reports MS Macrocell Show (handles &% and &&% files) MH Macrocell Hide (no special & treatment) &%file,nskip,nshow overlay file on screen &&%file,snskip,nshow Include file for output The MS and MH commands turn on or off the handling as print time macros of the &% and &&% forms of filenames. These permit you to use AnalytiCalc for report generation or outline handling more easily. In MS mode, if a cell has an &%file form in it, that file is read when that cell is displayed. The first "nskip" records are skipped in the file, and up to "nshow" lines of the file are displayed on null cells of the spreadsheet. Up to 4 files may be open at a time here. The left margin of the file text is made to coincide with the location of the cell containing the reference. You can use the $ (spawn) command to edit these files with your favorite editor or word processor on the system and they will automatically be updated on screen. The &&% form is used to include (possibly lengthy) files in output. The files are included in spreadsheet output at the point they are seen, indented by the amount of the indenting of the &&% command form. Therefore, you can use AnalytiCalc to organize a mul- tilevel hierarchy of printout of reports. If a &&% form is seen in one of the files being included, that file is opened and included. This nesting may go to a depth of 4 levels. This permits a real "frame editing" using AnalytiCalc and your favorite editor or word processor (WITHOUT requiring you to learn yet another editor.) In general, AnalytiCalc is not supplied with a text editor. Most people will already have one and it was thought a waste of time to ask anyone to learn another. Rather, the program allows you to use whatever editor you like (via the $ or } commands to exec your own editor) even under the sheet. What is added is rather the abil- ity to ORGANIZE documents flexibly and easily. AnalytiCalc graphs also are written in a format that permits this treatment. AnalytiCalc Reference Manual Page 40 Resetting Displayed Part of Sheet (Windowing and Scrolling) ORIGIN RESET (Windows onto physical sheet) OA variable Reset origin of display to variable OR variable Reset display down and right to start with variable OAD variable Reset origin of display to variable leaving windows intact. ORD variable Reset display down and right to start with variable leaving windows intact. The OA and OR commands let you do quick scrolls around the phy- sical sheet by resetting the upper left corner of the display sheet to point at any variable on the physical one and make the rest follow suit (OA), or will reset the display sheet starting at your current cursor location to point at the physical sheet variable and make all parts of the displayed sheet down and right from your current loca- tion correspond likewise. Thus, if you start with a normal initial display and execute the commands L C1 OR H1 you wind up with columns A and B still on screen, but now the remain- ing columns are H, I, J, K, L instead of C, D, E, F, and G. This happened because cell H1, down and right were pointed to by the dis- play sheet starting where you were, which was cell C1, initially the 3rd column and first row. Had the command been OA H1, the column la- bels would be H-N rather than the initial A-G. If the OAD or ORD forms are used, the area remapped is simply moved by the offset between the old origin of the region you're remapping and the new origin, so windows stay relatively placed as they were. Scrolling now uses OAD and ORD forms so windows stay in- tact. The difference between OA, OAD, OR, and ORD may be illustrated in the following figures, which represent the screen mapping after a command sequence AnalytiCalc Reference Manual Page 41 OA A1 L C1 OR AA1 L A11 OR A100 which produces a screen mapping like this: (note only part of the map is reproduced...) A B AA AB AC 1> A1 B1 AA1 AB1 AC1 2> A2 B2 AA2 AB2 AC2 3> . . . . . 4> . . . . . 5> . . . . . 6> 7> 8> 9> 10> A10 B10 AA10 AB10 AC10 100> A100 B100 C100 D100 E100 101> A101 B101 C101 D101 E101 102> 103> 104> and so forth. Note that columns 105> AA, AB, and AC are now mapped 106> for only 10 rows. Thus we have 107> 3 screen regions, not 4. While after a sequence like: AnalytiCalc Reference Manual Page 42 OA A1 L C1 OR AA1 L A11 ORD A100 the resulting screen mapping looks more like this: A B AA AB AC 1> A1 B1 AA1 AB1 AC1 2> A2 B2 AA2 AB2 AC2 3> . . . . . 4> . . . . . 5> . . . . . 6> 7> 8> 9> 10> A10 B10 AA10 AB10 AC10 100> A100 B100 AA100 AB100 AC100 101> A101 B101 AA101 AB101 AC101 102> 103> 104> and so forth. Note that columns 105> AA, AB, and AC now extend to 106> the full screen. This gives 107> 4 screen regions. These figures illustrate the usefulness of the OAD and ORD com- mands. AnalytiCalc Reference Manual Page 43 Putting Data to Files SAVING YOUR SPREADSHEET PD Put out current Display sheet (Formulas) PP Put out current Physical sheet (Formulas) PDN Put out current Display sheet (Numbers) PPN Put out current Physical sheet (Numbers) PB Save spreadsheet in Binary form PxxM Put out display to physical map as well as worksheet information. Saves displayed mapping only. PxxMA Put out display to physical map. Saves entire mapping, whether within DB command bounds or not. S Startup questions. Allows global width and format settings and new title entry. The S command will lead to questions being asked about whether to update the display mapping. If you reply Y, you will be able to reset the default display column width to whatever you like between 1 and 120 characters and to say whether to reset the display sheet to point to the upper left of the physical spreadsheet. In any case you will be able to reset the default format and the spreadsheet title. In AnalytiCalc-Amiga, this is all the S command does. The PD and PP commands write Display or Physical sheet cells that are active to a specially formatted sequential ASCII text file, always starting from the current position of the cursor and saving right and down. Some questions will be asked so you may enter a max- imum displacement to save (so not all of a large sheet need be saved) in columns and rows. On reload (the GP/GD commands), these questions are also asked, and minimum displacements are asked also to allow you to enter basically any region of the saved sheets into the current sheet starting at the current location and moving down and right. This gives a large scale move capability and a sheet merge capabil- ity. The PPN and PDN commands save numbers instead of formats to permit loading of isolated numeric values from other sheets by using the GD or GP commands (below). They record only numbers and not for- mulas, but this allows any cell's numeric value from other sheets to be loaded into another sheet without loading all formulas and repeat- ing calculations. Also, the PPN/PDN commands are needed to use the ACGRAF program to graph parts of a sheet if this form of output is desired. The PP/PD commands are suitable for archival storage of spread sheet data, or for its transportation to differently compiled ver- sions of AnalytiCalc. The PB command works like PP EXCEPT that it stores in "binary" form. This is considerably faster than PP but not as general. In particular the GB command must be used to reload the sheets so saved and the *X class commands cannot use binary sheet data; nor can the supplied graphics routines. The final M or MA qualifiers allow you to save the display map for the sheet in addition to the data and formulae. Where your map is nontrivial this is a good idea, though it consumes considerable disk space and its' use must be considered on a case by case basis. AnalytiCalc Reference Manual Page 44 Recalculation Control RECALCULATE R Recalculate all of the sheet. RM Recalculate manually only until R command given RE Recalculate Entries as made but nothing else until R command given. RI Recalculate Incremental - Calculate displayed cells only. RM Recalculate manually only until R command given RAF Recalculate sheet, without constant recalculation RF Recalculate, Force constant recalculation. RR Recalculate, Force, but stay in RE or RI mode if in it already. The R command forces a recalculation of the sheet. This will evaluate all formulas on the sheet and update the display to show the new numbers. Normally, recalculation is done after each number is entered to update the new cell, but the RF command forces total re- calculation in case anything was missed. The standard keypad's Re- calculate key generates the RAF command automatically for ease of use. Since these calculations can be time-consuming, you can suppress them while doing input by entering the RM command (Recalculate Manually) which suppresses recalculation until an R command is en- tered. This command's format is just RM For further speedups, AnalytiCalc can be put into RE or RI mode to recompute only the cells just entered after each entry, or to com- pute cells on the display only. It starts in RI (Recalculate In- cremental) mode until an R command is given. After an R or RF command is given, AnalytiCalc goes into a mode where it will compute ALL cells that have anything in them at each sheet recomputation. Another RI or RE command will return to the previous mode. The RR command does a RF (Recalculate, Force constant recomputation) but stays in Incremental or Entry mode. Whenever a new sheet is loaded, a single R is done automatically regardless of the current mode. To speed normal recalculation, AnalytiCalc only computes con- stants once. The normal RF command will force them to be recomputed also. AnalytiCalc normally will keep track of the lowest and furthest right cells in use and only ever try to recompute cells that have active values there. AnalytiCalc Reference Manual Page 45 Setting Relocation Boundary RELOCATE BOUNDARY RB* Set maximum row and col for relocation to end of sheet RB Var Set maximum row and col to be relocated to coords of Var (cell name) The RB command affords a way to specify a lower right corner to the area relocated by any of the copy and relocate operations. Below or to the right of the relocation boundary, moves are all treated as absolute even if specified as relative. Therefore the area below this boundary will not be relocated by the AR commands or the CR com- mands. This allows the AR operation to relocate only a finite max- imum number of rows or columns, or permits cells moved beyond the boundary specified to be unrelocated. The upper left corner of the relocation area is defined by the cursor position, so this permits any rectangular area to be moved without relocation while the rest is relocated. Combination of this feature with the AR and CR commands permits moving rows or columns (or parts of them) via command files. This moving would generally be done with the IR command to relocate a region in place to be moved to another site, then actually copying it (absolutely) to the desired destination. Alternatively, the AR/AA/CR commands can be used to move things around. For example, a rectangle could be moved away from the origin further by using the cursor and the RB command to delimit the area to move, then using a command file to cycle through a set of CR commands to copy all cells down the desired amount. The AA command could then be used to delete the un- desired top cells and the area to be moved now moved into its new ad- dress, all other cells having been relocated. The functionality of moving cells is however much better served on AnalytiCalc by the OA, OR, and DL commands to reset the display mapping (possibly while log- ging keystrokes so the mapping can be easily reproduced) to effect the desired move. For simply reformatting the screen, DL affords control at the level of any regions desired. A command file should be devised for moving cells but only used when really needed badly, if such actions are needed for other reasons. There will be very little need for them under most circumstances however. Where cells must move closer to the origin, an AR command at the origin can be used to provide temporary space in which to move the cells; at the end, they will be moved back further. The IR command (Inplace Relo- cate) can be used to avoid major moving around the sheet. The *U HERE function will return the current max column in accumulator W and the current max row in accumulator Y to facilitate this. These maxima can be used to write an efficient command file to copy only the areas in use. Internal loops in a formula (with the *U XQTCM function) are the fastest way to implement such a loop. AnalytiCalc Reference Manual Page 46 TEST expression command TEST TE expression Evaluate expression, save in % The TEst command allows command level calculations, for use in controlling loops. It evaluates the expression (up to 79 characters long) and stores in the % accumulator. Useful for *J type commands to control branching in AnalytiCalc. Any expression may be handled in a TE command, and forms of the sort Z=expression are allowed. NOTE however that it is frequently desirable from command files to add a C to the end of the formula since sometimes junk from preceding lines is not zeroed. The \*C causes it to be treated as a comment and ignored. The TEst command will calculate formulas even in restricted re- calculation modes (RM mode, RE mode, or RI mode) since it ALWAYS evaluates the expression regardless of mode. The construct to evalu- ate a cell with a command like TE V1=__{V1 (which sets a cell value to the computed value of its formula) is ex- plicitly legal and can be used where it is desired to perform irregu- lar partial recomputations while in RM mode. VIEWSCREEN Control VIEWSCREEN UPDATE V Redraw screen with normal formats VF Redraw screen showing all formulas VM Don't redraw screen until a V or VF is given. The V and VF commands redraw the display screen showing numerics as numerics in the selected format (V) or displaying whatever part of the formulas fits on screen in the column (VF). This is used wherever the picture on the screen is damaged, e.g., by operating system broadcasts. VM is used to inhibit redraw while numerous changes may be made, to save time. A V or VF command will re enable display updates. AnalytiCalc Reference Manual Page 47 Writing ASCII images of Displays WRITING SCREEN TO HARDCOPY OR FILES W Write screen to file The W command lets you write a hard copy of the screen as it ex- ists on display to a file. AnalytiCalc will prompt you for the file name; the device specifier should be included. Note that if the filename you enter (or the last character after the device or file specifier) is a "/" character, AnalytiCalc will omit the column and row labels in the output file or printout so that you need not edit them out afterwards. They will always be displayed on the screen however. To output to the lineprinter, the file specifier to give is PRT:. The output file in any case has no strange escape sequences and may be edited with any normal system editor. To junk any output you don't want, use a filename of NIL and to go to a generic printer you may use a filename of PRT. If the character "%" appears in the file specifier, the title will not be displayed on printout. This is designed to help those producing paste-together printouts from several sheet areas. EXIT command EXIT X Exit to operating system. The X command exits from AnalytiCalc to the operating system. In general, this is the ONLY way (short of an abort command to the operating system) to get out of AnalytiCalc, and is the ONLY WAY TO LEAVE GRACEFULLY. You should use the PP or PD command to save your sheet prior to XD if you want to save your sheet; otherwise it will be lost. Be at least that cautious; the program will NOT keep track of whether you've saved the sheet or not. Please note and use this exit method; it allows AnalytiCalc to clean up and close windows, which otherwise may not be done. AnalytiCalc Reference Manual Page 48 ZEROing cells ZERO VARIABLES ZA Zero whole sheet ZE V1:V2 Zero variables in a row or column in range from cell V1 to cell V2 (V1,V2 = names of cells) The ZEro commands clear out a sheet. The ZA command will also ask if you mean it and needs a reply of Yes to actually clobber the whole sheet and return it to pristine purity. The ZA command occasionally takes a noticeable time to clear the workfiles out; do not be alarmed at this. It will ask afterwards for a new title, new display format default for floating point, etc. If you give the ZA command, AnalytiCalc will ask you to confirm that you indeed want to zero the entire sheet. A reply of Y will proceed to execute the command; any other reply will abort the clearing operation. TEXT SEARCH OF FORMULAS SEARCH for formula substring >STRING Find cell whose formula contains STRING >>STRING Find cell whose formula starts with STRING The > command (greater than sign) causes a search to be made from the current position down and right to the relocate boundary for a formula containing the string after the >. The cell found (the first one) will become the current cell. The origin is not moved, but the command OA P## will cause the cell found to become the upper left cell on the screen if desired. This is a handy tool for locat- ing sections of a large sheet. The comparison is between the entered string and the formula, but terminates with the shorter. Therefore, aliasing occasionally occurs. Just skip past the offending cell and redo to continue to search. The >>STRING form anchors the search at the start of the string so this will not occur, though very short formulas whose beginnings match the search string will match. The underscore character (_) is a wildcard character for formula searches. AnalytiCalc Reference Manual Page 49 Driving AnalytiCalc from Command Files EXECUTING COMMAND FILES @filespec Read filespec and execute commands as though typed in. The @ command lets you treat a file as though it were your console input (returns on endfile) to let you execute saved sets of commands to set up your sheets. Note that if you issue the K command (below), those inputs come from a separate file on logical unit 1. Recording Commands Interactively JOURNALING +J filename Opens filename as a journal file, recording commands to that file. (Note: auxiliary inputs are not so recorded, where asked for as separate questions.) +N Closes any open journal file The journaling commands allow typed commands to be saved into a file later usable as an indirect command input to AnalytiCalc to al- low recreation of screen formats, data, formulas, etc. Extra ques- tions (e.g. those used for filename and areas to save or restore in P or G class commands) are NOT saved by the journal. They will be read by @ commands however and may be edited into a journal file with any editor. If this is desired inside a AnalytiCalc session, the Spawn command (below) is recommended after the +N command is used to close the file. The VM and RM commands are recommended inside @ files (with R and V commands at the ends) to suspend display updating and recalculation until all commands in the journal file are done. The Journal command is the way to get AnalytiCalc to remember your keystrokes when making up any command so that you can tell what you're doing. Later if you don't need to watch the screen update, use the VM command (View Manual) to turn display update off during the command files' operation, and the V (View) command to turn it back on at the end. The major uses of journaling are to save screen formats as they are set up and to allow capture of keystrokes for use in command macros. Executing Other Programs or Commands from AnalytiCalc $ command-line EXEC command line to the OS (minus the $ sign). May need extra CR to free console. } command-line EXEC command line (minus the }) AnalytiCalc Reference Manual Page 50 and await a return on console before return to sheet. The $ command permits AnalytiCalc to spawn commands to the host OS. The command is whatever follows the $ and results in that com- mand being run as an AmigaDos RUN type command. This permits use of other system utilities during a session with AnalytiCalc. The vari- ant using the } character rather than the $ character will print the message ENTER RETURN TO REDRAW SPREADSHEET> at the bottom of the screen and will not repaint the screen until the return key is pressed. This allows you to examine the screen as long as needed be- fore erasing and redrawing the sheet. NOTE that in a minimum 256K system you will NOT HAVE ENOUGH MEMORY to execute other commands. The size program you can run depends on the amount of memory you have left. The commands are passed to COMMAND.COM and executed if room enough exists to do so. The graphics tasks are designed to be invok- able this way. Interactive Calculator Mode DROPPING INTO INTERACTIVE CALCULATOR K Drop into interactive Calculator The K command gets you into Calc, described below, and lets you do computations. When you give the interactive Calc program the *E command to exit, you return to the spreadsheet. This may be useful where you want to do auxiliary computations not in the spread sheet or try out formulas that you may want to debug interactively before entering them onto the sheet. This calculator knows decimal, hex, and octal radices and can be used to debug indirect command files written for use within spreadsheet cells; the *V 3 command can be given and then the *@file can be given in Calc mode, providing a trace playback of the actions of a command file to see what it is do- ing. AnalytiCalc Reference Manual Page 51 Conditional Restart of Command Files LOOPING COMMAND FILES _< Rewind the current AnalytiCalc input file This command will allow a command file to restart itself from the beginning. It does so only if the % variable is positive (per- mitting some control over it). A command file may use the TEst com- mand to set the % variable prior to this command. An < found in a cell equation causes the cell equation to be re-evaluated from the start of the formula, PROVIDED that the "<" is seen at the start of one of the cell's equations, AND that the % ac- cumulator is greater than 0.0 at that time. This is provided to al- low use in conjunction with *WF, *P, and *U XQTCM so that multi-equation formulas can loop, doing jobs like selecting rows from regions, entering sequences of numbers, and the like. This provides a compact and fast looping structure that does not depend on rewind- ing files externally. The *U XQTCM TE {V1 construct is not supported however. The looping is limited to at most 100 passes to prevent in- finite loops. COMMENT commands COMMENTS * any text Any line beginning with * is treated by AnalytiCalc as a comment (except when in the *F mode when it is sought). This allows you to comment your indirect AnalytiCalc command files. AnalytiCalc Reference Manual Page 52 Prompting for Data to be EDited into formulas READING ARGUMENTS OFF CONSOLE - Prompt The "-" command will print the prompt text on the screen and await entry of a line of text. This text is stored as up to 4 argu- ments for possible later use in EDit commands. Spaces separate input arguments, and no argument may be longer than 51 characters. This facility is intended for use with indirect command files (@ files) which may want to be able to prompt for parts of formulas and then edit them into the command formulas for use, or to prompt for parts of later-used Calc commands. Note that Calc never sees any formulas that are given the A format, so they can be prepared in this way, to be used when they are given a numeric format. A suitable - command might be -Enter net pay> which would produce the printout (in row 23 of the screen) Enter net pay> and to which one might reply with a number to be entered onto a "template" worksheet. Application Notes In entering numbers, use a decimal (.) to get them to go in as floating point. If numbers are entered with no decimals, they are considered integers. The system will convert the values to floating point for storage, but incorrect results have occurred where formula ranges have been mixed since multi-argument functions assume the con- versions are all alike in their ranges, and binary formats of in- tegers and floating point numbers differ greatly. Note that the default format for numeric output is F9.2 and that it may be changed if desired. If you set up the format of a cell, that marks the cell valid and the format is not altered by input defaults, nor the numeric/text interpretation of the cell. To put in a lot of labels, then, set up format A using DF and then enter the labels. If the sheet's calculations contain forward references (down or right), you may need to give the R (Recalculate) command a few times to get proper values. If the values on the sheet do not stabilize, you probably have a circular reference in your formulas and should examine the cells that do not stabilize for the circularity. Calcu- lation is done across first (along rows first). This allows any file references to be as nearly sequential as possible. AnalytiCalc Reference Manual Page 53 If you enter numbers with exponentials, use +/- in front of the E so that AnalytiCalc can tell that the "E4" type strings in numbers like "1.3E4" are really exponents, not cell names. That is, Don't enter 1.35E4 Enter 1.35E+4 to avoid confusion when cells are relocated. Filenames that look like cell names should be avoided too. That is, a sheet that con- tains a formula like @D5.FRM might, upon getting moved from D5 to F3, be changed to @F3.FRM and IF this is not what is desired, it could cause confusion. Note that column widths on the display are properties of the DISPLAY columns, not the PHYSICAL columns. This arises because a display column may cover potentially 20 different physical columns on the sheet due to mapping. That is, column 3 may have cells from column C of the physical sheet, but can be set to have cells also from columns R, AZ, Q, and W (for example), with different display formats for each. Rather than choosing some width a function of the physical columns which could cause confusion, the width is up to you. The columns are labelled with the physical cell column of the adjacent cell in the top row and also with the display column number to make it easier to enter DL commands, which use display coordinates (pure numeric, column:row) rather than physical coordinates (alphanumeric, e.g. AK15). Interactive Calculator Section THE CALC MODE OPTION (K) and Functions Available: This describes the CALC option (K) of AnalytiCalc which is also the mathematical engine used for expression evaluation except for multi-argument functions and IF statements. NOTE: You will need to use the command *V 3 to get any numbers displayed initially; AnalytiCalc initially sets the display View flag to zero for speed reasons. OVERVIEW CALC mode is a calculator designed to evaluate arithmetic ex- pressions. In its basic form, expression evaluation is similar to that used by ANSI FORTRAN with calculations performed on INTEGER*4 AnalytiCalc Reference Manual Page 54 and REAL*8 constants. Variables may also be invoked. Standard al- gebraic operator precedence is used, and accumulators (A through Z, %, and all spreadsheet cells) may be assigned values by Calc state- ments of the form: Accumulator=expression Additional features include octal and hexadecimal arithmetic capabilities. Commonly used commands and expressions can be placed in a file and executed when convenient. How to Enter CALC mode To enter Calc, use the K command of AnalytiCalc. Calc mode will prompt as follows: CALC> Before starting in this mode, type the command *V 3 to get the interactive portion to display results. Now try typ- ing 123+456 followed by a carriage return. CALC will evaluate the expression and output the answer 579 It then prompts for further input. Try other expressions such as 12.0 - 99. (answer=-87.00000000000000) -(-32767+(6-2)**8-(512/(409-401))) (answer=-32705) 3*5/7 (answer=2) 3*(5/7) (answer=0) Mixed mode is legal, for example 1977/50. is evaluated as 39.54000000000000 Reals may be expressed using D or E format. For example 1.2E10*2.D0**3-1.D-8 is evaluated as 0.95999999999999992D+11 Accumulators AnalytiCalc Reference Manual Page 55 Variables may also be used to retain values for later use. Variables may be the letters A through Z, or any AnalytiCalc physical sheet cell name. All accumulators default to REAL type. To set I to a value use the usual FORTRAN syntax, for example: I=2**10-1 Try typing the single character 'I'. CALC will respond with its value. We can now use I in various expressions such as J=I-I/3*3 % is a special variable that retains the value of the last ex- pression evaluated. For example, to successively add up the numbers 1, 2, 3, 4, 5, and 6 we could enter 1 %+2 %+3 %+4 %+5 %+6 Note that you can examine the value of the variables by typing the appropriate single character followed by a carriage return. Such an examination does not change the value of %. Getting Back to the Spreadsheet To exit from CALC, type *E (or *EXIT) or *S (or *STOP) This will return you to AnalytiCalc. (To exit AnalytiCalc, use the X command.) Mathematical Functions for Calc and AnalytiCalc -SPECIAL FUNCTIONS- CALC recognizes a variety of special functions. For example, to calculate the square root of 2, we can type SQRT(2.) CALC responds with the value 1.41421356237310 Each function may have an expression for its argument. For ex- ample, AnalytiCalc Reference Manual Page 56 A=2.0*SQRT(ALOG(9.)+3.) sets A to 4.55948443459838 The following special functions are available: FUNCT NAME ARG TYPE FUNCT VALUE DESCRIPTION ------------------------------------------------------- ABS REAL REAL absolute value DABS REAL REAL absolute value IABS INTEGER INTEGER absolute value IFIX REAL INTEGER REAL to INT conv. AINT REAL REAL REAL truncation INT REAL INTEGER REAL to INT conv. IDINT REAL INTEGER REAL to INT conv. EXP REAL REAL e**X DEXP REAL REAL e**X ALOG REAL REAL natural logarithm DLOG REAL REAL natural logarithm ALOG10 REAL REAL logarithm base 10 DLOG10 REAL REAL logarithm base 10 SQRT REAL REAL square root DSQRT REAL REAL square root SIN REAL REAL trigonometric sine DSIN REAL REAL trigonometric sine COS REAL REAL trig. cosine DCOS REAL REAL trig. cosine TANH REAL REAL hyperbolic tangent DTANH REAL REAL hyperbolic tangent ATAN REAL REAL arc tangent DATAN REAL REAL arc tangent Note that the multi argument functions (e.g. SUM[range], AVG[range], etc., are all also available in CALC mode with the same restrictions mentioned above. The IF [expression]yes-expr|else-expr command is however NOT available in CALC mode; use the *J expression instead (see below). -WORKING IN OCTAL AND HEXADECIMAL- You may change the base used to specify constants by using the *B command. Legal forms are command action ------- ------ *B displays current default base *B 8 changes default base to octal *B 10 changes default base to 10 *B 16 changes default base to 16 AnalytiCalc Reference Manual Page 57 Suppose we have changed the default base to octal. Then adding 7 + 1 we obtain the result 00000000010 (BASE 8) If the default base is hexadecimal, we can enter 9 + 1 which is evaluated as 0000000A (BASE 16) Suppose we have assigned A=1 then 1+A gives 2.000000000000000 even when the default base is 16. If we wish to add the hexadecimal digit 'A' to 1, enter 1+0A We now obtain the desired 0000000B (BASE 16) This leading 0 is only necessary when the first hexadecimal digit is greater than 9. If constants are entered with digits that are not legal for the base being used, the entire number is converted using a more appropriate base. For example, if we have set the default base to octal and type 1+9 the 9 is not an octal number so it is converted to base 10. If a base 16 number is involved, the result will be in base 16. You may temporarily change the base for a single integer con- stant by preceeding it with ^8 for octal ^10 for base 10 ^16 for base 16 AnalytiCalc Reference Manual Page 58 For example, if the default base is 10, 100+^840 gives 132 a base 10 integer. I=100+^1610 gives 116 also a base 10 integer. Note that the '^' can only be used to specify the base of con- stants and that expressions such as ^16I are illegal. To declare variables to be integers of a specific base, we can use the commands *INTEGER (base 10) *OCTAL (base 8) *HEX (base 16) for example, *INTEGER A declares variable A to be a base 10 integer. *HEX B,Z,F declares variables B, Z, and F to be base 16 integers. *DECIMAL lists all the variables that have been declared to be of type DECIMAL. To summarize, there are three distinct ways of making base declarations when using CALC. The first is to use the *B command to designate the base default value. This is used to determine the base for constants when they occur in expressions. It does not in any way influence the type of any variables found in an expression. The only way to change the type of a variable is with a specific CALC command such as *INTEGER A,B Note that the *INTEGER command can be used to set types for AnalytiCalc cells. However, exercise care in doing so that your dis- play formats match the new type. Suppose for example that the default base is 10 and we enter *OCTAL A A=100 then CALC responds with AnalytiCalc Reference Manual Page 59 00000000144 (BASE 8) Finally, the last way to change a base is to use the explicit base specifiers for a constant, for example ^10 123 ^8 777 ^16 AB Commands and Other Functions for Calc and AnalytiCalc All commands to CALC (as distinguished from expressions to be evaluated) begin with an asterisk. To obtain a list of all possible commands, type a question mark followed by a carriage return. Most of the commands have already been described. The following section gives an explanation of the remaining commands. Many of these have the role of functions too, and may change the value of % or other ac- cumulators. COMMAND DESCRIPTION ------- ----------- *@filename Where filename is the name of a file of CALC com- mands. CALC reads the file and executes the commands. Up to 5 nested calls can be made. Recursive calls are not allowed. CALC prompts with CALC<n> before each command line is executed, where n is the calling level. You may optionally follow the file name with a blank followed by a single variable name (a single alphabetic character or %). CALC will then execute the file until the value of that variable is zero or negative. The test of this variable is made before the file is exe- cuted and not during execution of commands within the file. If the variable's value is not positive when the command is initially encountered, the file will not be opened for execution. This facility allows any or every cell to be driven by command files while in spreadsheet mode. *ASCII Declares a list of variables to be of type ASCII. Useful when decoding ASCII characters. For example, if we set A to be of type ASCII, then typing A=77 results in the character 'M' being output. The inverse operation is the single quote. It allows us to specify a single ASCII constant. For example, if we type 'M then the character 'M' is echoed and indicates that % holds that character and has data type ASCII. Suppose that the variable I has data type INTEGER. Then we can output the base 10 code for the ASCII character 'M' by entering AnalytiCalc Reference Manual Page 60 I='M which results in 77 being output. Notice that you may not be able to enter certain control characters that are intercepted by your operating system. Characters whose value is less than 32. are output by printing the character '^' followed by the equivalent ASCII character of that number plus 32. For example, A=10 results in ^* being output since 42 is the ASCII code for the character '*'. See Appendix A for a table of the characters output by CALC to represent such non-printable characters. *C COMMENT line. The characters that follow are ignored by CALC. This is useful when documenting files containing CALC commands. *N NOVIEW. Prevents CALC from outputting the value of the ex- pressions evaluated. This is especially useful when executing files containing CALC commands that initial- ize variables to special values. Equivalent to *V 1 *V VIEW. Controls CALC's printing options: command output class -------- -------------- *V 0 error messages *V 1 error messages command lines read from files *V 2 error messages value of expressions evaluated *V 3 error messages command lines read from a file value of expressions evaluated *V same as *V 3 The default setting is *V 0. Notice that other legal forms are *VIEW 1 and *V2. You may want to include a command *V 0 somewhere before using any command files to avoid diagnostic printouts if you use indirect cell files much. *R READ. Allows a single line to be read from the terminal. Useful in files of CALC commands to allow additional commands to be entered (like *S to exit from that file) AnalytiCalc Reference Manual Page 61 or simply as a way to halt terminal output until the carriage return key is pressed. *REAL declares specified variables to be REAL*8. When the value of such variables are output, FORTRAN's D format is used. *DECIMAL Declares specified variables to be REAL*8. When the value of such variables are output, FORTRAN's F format is used. Variables A-H and O-Z default to type DECIMAL. *S STOP. Same as *E *E EXIT. Terminates CALC session unless it is used within a file of CALC commands. In this case, CALC closes the file and continues with the next command. *Z ZERO. Zeroes all variables except %. Data types are not changed. *G *G V1,V2 (where V1 and V2 are cell or accumulator names) will evaluate V1 and V2 as the column and row numbers, on the physical sheet, of the desired cell. The ad- dressed cell's value is retrieved and used as the resulting number. This is used for lookup tables. *W Takes the value at the current cell and writes it out to the formula as a numeric (floating) value. The current cell may have been modified temporarily by the *P com- mand. This will overwrite the formula at that cell. The *WF command will work the same as *W except that the format used for rewriting the number will be the current cell's display format instead of a pre-fixed D32.25 format. Some extra spaces may be introduced which may be removed by the command "ED / //". This allows freezing values to whatever precision is desired. *P The *P command resets the current cell coordinate from within a cell (until the next cell is evaluated only). It has several forms: *P - By itself, *P causes Calc to prompt for the new phy- sical column and row number. *P V1 - This moves the current location to the named cell where V1 is the cell name (e.g. A5, H2) *P@ V1,V2 - This uses V1 and V2 (cell names) as column and row numbers and changes the current physical cell position to that defined by the contents of cells V1 and V2. This gives complete addressing of the sheet from within any cell. *F *F Label - If the value in % is positive and nonzero this command rewinds the input file for the AnalytiCalc @ command (not the Calc @ command) and seeks a line beginning with the characters *CLabel (where "label" is what you put after the *F command). This allows Calc commands inside a cell to direct outside command entry. *J *J Label - This command behaves as the *F command but AnalytiCalc Reference Manual Page 62 operates on the file used by the Calc @ command rather than the AnalytiCalc one. Thus, conditions in Calc program files may be implemented. Data File Access *QF *QW The *QF (Float) or *QW (write) commands are used to examine sequential files created outside AnalytiCalc and return values or formulas. Their syntax is *QF filename ?key1? ?key2? <lm> or *QW filename ?key1? ?key2? <lm> where ?key2? is optional l and m are delimiter characters for start and end of the parts of the selected records to extract (defaults to first part of the record) filename is just the file specifier in the host OS. operation is that the file is searched from the beginning for records containing key1 (and key2 if specified) and then the part of the record between character l and character m (delimiter characters, which may be the same as for example 2 com- mas) is extracted. That part of the record is then ei- ther Floated by converting to a floating point number using a wide Floating Point format specifier (in *QF) or read in and made the current cell's formula between those delimiters (in the case of *QW). Records may not be over 128 characters long. This command would be used to extract data from a database or word processing files. The first key will begin at the start of the line for each file if it's first character is the ` (accent grave) character. Then the key searched for is sought always at the start of the line, minus the ` character. The delimiters may contain 3 characters to specify positional fields. In this case, the first delimiter character is ` (again) and the second 2 are encoded start location and length fields. They are just the binary values of these + 32 (starting from 1) to turn them to printable ASCII. Variant forms are *QFK, *QWK (which keep the file open after opening it so that subsequent *QFN or *QWN or *QFC or *QWC commands can read sequentially down the file), or the mentioned *QFN and *QWN commands, which read the file from the Next record (leaving it open for more Next searches) and the *QFC or *QWC variants which perform the *QFN type action but close the file on completion. The file spec must be present in *QFN, AnalytiCalc Reference Manual Page 63 *QWN, *QFC, and *QWC commands, but is ignored there. If the search fails, the % accumulator is returned as -999999. and on *QW type operations where there is no failure, the % accumulator returns a value of 1.0 to indicate success. In the *QF commands, the value re- turned is in the file searched. These return values occur regardless of the variant. Note that the same logical unit (4) is used for all *Q class operations, so a subsequent *QF or *QW will close the unit after a string of *QN searches of indeterminate length. A re- quest that searches the null device is a simple way to ensure the unit is closed. It is possible to leave it open and interfere with some other operations, so the user is cautioned to pay attention to this aspect of program operation when designing a sheet that uses it. *U funct args The *U command is a means to allow easy user ad- dition to the AnalytiCalc function repertoire. It passes the command line to the USRFCT subroutine which is set up to recognize up to 5-letter function names (edit it to add what you need) and to process the func- tions called. The parsing is up to you. Note that the VARSCN routine may be called to convert variable names into coordinates in the XVBLS array. Functions supplied include: DATE functions operating on dates: IDATE, YRMOD, JDATE, JTOCH, and DATE which all return or convert between Julian dates (day since a fixed date) and year/month/day binary, and ASCII dates. MATRIX functions which handle matrix math: MTXEQ, MOVEV, MDET, MPROD, MADDV, MSUBV, MMPYT, and MMPYC. These handle all variants of normal matrix algebra. MMPYT is particularly handy for producing the dot product of 2 rows or columns (sum of products of corresponding ele- ments). Note that matrix math works only in the prime area however. That is, matrices must reside above and to the left of cell BI301 (so the legal range for ma- trices is A1 through BH300). EQUATION SOLUTION via the VARY function, which allows up to 8 accumulators to vary to iteratively search for solu- tions to equations anywhere on the sheet. UTILITY functions, including STRVL, which converts formula text to numeric values, XQTCM which executes commands from inside formulas, and HERE which returns current location and some extra sheet status information. DETAILS: AnalytiCalc Reference Manual Page 64 A version of USRFCT is supplied which can call the system to return the system month, day, and year in ac- cumulators T, U, and V respectively. The % returns with a value of yymmdd to permit easier sorting too. However, any functions desired may be added in this way and results extracted in whatever way is convenient. The yymmdd value returned is a Julian date based on 1/1/1980, in the % accumulator. This is the IDATE() function. There are 4 other date functions supported in the most capable AnalytiCalcs. These permit conversion between ASCII dates displayed as YY/MM/DD and Julian dates, or back. The call *U YRMOD VY,VM,VD returns the Julian date (in %) computed from the Year (in VY), the month (in VM) and the day (in VD), where VY, VM, and VD are sheet cells. These may be the result of date arithmetic. The call *U JDATE Var assumes the formula in cell Var (any spreadsheet cell name) contains a date string in the format YY/MM/DD. It reads this formula and converts the date to a Julian date, returning it in the % accumulator. The call *U JTOCH Jul,Var assumes that variable Jul (any spreadsheet cell) con- tains a Julian date and changes it into an ASCII string in the cell whose name is in the Var position here. This wipes out any previous formula in that cell, though it does not alter display format or type infor- mation. The T, U, and V accumulators return with the month, day, and year on return. The call *U DATE VY,VM,VD,Var uses VY, VM, and VD as year, month and day, and AnalytiCalc Reference Manual Page 65 computes a Julian date from them. It then composes an ASCII string of form YY/MM/DD for that date and stores in the formula for Var. The use of Julian dates is made because the dif- ference of two Julian dates is precisely the number of days between the two dates. These functions are designed to make date labels easier to do. There are actually 4 supplied USRFCT routines. The larger of these are also able to solve matrix equa- tions (this demonstrates how other programs may be in- tegrated with the package). The function solves the matrix equation AX=B where A is a square matrix and X and B are rectangular ones, A being n by n and X, B be- ing n by m. The call formats for IDATE and MTXEQU are *U IDATE() *U MTXEQ(AA:AA,XX:XX,BB:BB) where the notation AA:AA means two cells at the upper left and lower right edges of matrix A (e.g. B2:C3 for the 2 X 2 matrix so defined), and the XX:XX and BB:BB notation means the same for the X and B matrices. This capability is used to solve systems of linear simul- taneous equations and does so with extreme speed. Note however that errors can occur if the A matrix is singular. The cell returns 0 if the matrix was singular and does not solve the equation then; 1 is returned if the equation was solved. Note that matrix A is modified by the call, so a reasonable precaution is to ensure that the matrix is reset by AnalytiCalc prior to each computation. This may be done by: 1. Placing the MTXEQ call in a cell left and down from the A matrix and 2. Ensuring there are some alphas in each entry in matrix A. This may be done by placing a \*C FRC RECALC into each cell. This is a comment and will not affect the number. However because it contains alphas, AnalytiCalc will then re-compute each number whenever the sheet is computed. or AnalytiCalc Reference Manual Page 66 Using the MOVEV function to copy the real A matrix onto the working one before the MTXEQU call every time. (This is the efficient way to do it, as recalculating the constants every time wastes resources.) To have the X matrix cells displayed without being modified, it is necessary to have the cells' formulas contain a self-reference. The easiest way to achieve this is to place the "formula" P 0 0 into each cell. This effectively will cause display of whatever number is there. A similar technique may be used to set up matrix A, where the real matrix is elsewhere and formu- las in A are just the P 0 0 type. The real matrix may contain formulas like "34.65\D12=%" to set up the desired contents here AND to reset cell D12 (in the work area) to the entered value. The B matrix is not modified by the MTXEQ routine. A simpler way to handle the problem is to use the MOVEV function (below) to copy the desired cells to a scratch A matrix from wherever it was, so the scratch matrix may be destroyed without impact on display. If this is done, the source of the A matrix need not be specially bollixed up. Several additional functions exist in the larger ver- sions of USRFCT, the largest of which is designed for use with VAX due to space limitations on PDP11. *U MOVEV mtxa,mtxb where mtxa and mtxb are matrix specifiers. This com- mand moves values from mtxa to mtxb (useful prior to calling MTXEQ). *U MDET mtx This function computes and returns the determinant of matrix mtx (specified as v1:v2 where v1 and v2 are upper left and lower right corners of the matrix). The next functions are implemented in the largest of the USRFCT routines, and presently are limited to VAX implementations, in which they are present by default unless the build files are changed. *U MPROD A,B,C This function will multiply matrix A by matrix B giving matrix C, provided that their dimensions are compati- ble. Note this and later functions may be available only on VAX versions of AnalytiCalc. AnalytiCalc Reference Manual Page 67 *U MADDV A,B,C This function adds matrix A to matrix B and stores in matrix C. All matrices must have the same dimensions. It can be used for adding values in any pairs of rec- tangular regions much faster than a series of sum func- tions would permit. *U MSUBV A,B,C This function subtracts matrix B from matrix A leaving the result in matrix C; as above, matrix dimensions must be the same. It also is far faster in subtracting rectangular regions than numerous separate equations would be. *U MMPYT A,B,C This function multiplies matrix A-transpose by matrix B and stores the result in matrix C; dimensions must be compatible. Also, as in MPROD, matrix C must not be the same as either A or B. In ordinary use, this can be handy for dot products. That is, suppose you have two column parts, G10:G50 and R20:R60 and you want to form the sum (G10*R20 + G11*R21 + G12*R22 + ... + G50*R60) for later use. In many systems, you need to make another column of the G10*R20, etc. terms and add it up. Here, you can use the MMPYT function as *U MMPYT G10:G50,R20:R60,G51:G51 to form the sum and store it in cell G51. (It can be stored anywhere; G51 is chosen just for illustration.) If you have a row and a column to multiply, you use the MPROD function in- stead of MMPYT for this. Since this is a built in function, it can be orders of magnitude faster than the separate - column approach. *U MMPYC A,B,K This function multiplies every element of matrix A by constant K (a cell or accumulator) and stores the result in matrix B. This function is FAR faster than separately multiplying many cells. If you want to scale a rectangular region of the sheet by some con- stant, this function can be used to quickly do so, mul- tiplying each cell in the region by the constant in cell or accumulator K (K can be ANY cell or accumula- tor). Wherever you need to adjust data by some frac- tion, this function should be used as the fastest way of doing it. AnalytiCalc Reference Manual Page 68 *U VARY X,A,W,I,P;Q;R;S;T;U;V;W (Equation Solving by Iterative Search) This function allows AnalytiCalc to automatically search for solutions to equations over up to 8 dimen- sions. The operation is that the accumulators named in the fields shown as P;Q;R;S;T;U;V;W (one to 8 may be specified, only one is required) are varied by a frac- tion W about their initial values (later scaled down by the gradient of the change in X) to attempt to get ac- cumulator or cell X to equal accumulator or cell A. This is done for I iterations, where I is another ac- cumulator. Care should be taken that I is not initial- ized in every pass of the spreadsheet, but at most 20 iterations will be done in any one pass as a safeguard in any case. The accumulators given must be indepen- dent variables; set dependent ones or normalization conditions elsewhere on the sheet for use with calcula- tions. Any function or set of functions can be entered and the program will step the accumulators to attempt to find a (local) minimum of the value of (X-A) where X and A may be either accumulators or cells. The other arguments must be accumulators (i.e., named A through Z, though prior automatic use of X, Y, and Z may con- flict if these are used). This function will allow iterative solutions to be found for situations not soluble directly, with stops for inspection every several iterations. The TE command may be used con- veniently to set the accumulator designated as I here without need to modify a cell to do so. The VARY function is initialized by a negative or 0 value in the I accumulator (the LOGICAL one, not I itself). The equations controlling the variation may appear anywhere in the sheet, and variation begins at the initial values set into the varying ACs (P-W in the illustration). Both the logical X and A accumulators may vary, as well. As an example, let us suppose we want to solve the equation A=SQRT(B) for B, given A. Humans would of course just square both sides, but in an iterative solution, we would vary B until we found A. The method used here performs the variations intelligently so that we actually approach the correct value asymptotically; the signal we have a solution is that the variation stops. To set this up, we might enter the following AnalytiCalc Reference Manual Page 69 equations in a row or column (or anywhere actually; order doesn't matter) in the sheet: I=-1.\A=55.\P=12.\*C Set up to initialize with negative I, initial guess 12., and desired value 55. W=1.\*C Set "step scale" P+0.\*C Display P accumulator on screen B=SQRT(P)\*C Enter equation relating B, P *U VARY B,A,W,I,P Now we go back to the first equation, since the desired value in A (55.) has been set and the initial guess for SQRT(55.) has been entered (12.) and we may now set the number of iterations desired to drive the calculations to solve the equation B=A (which amounts to B=55.) (which is what the VARY function does), by varying P to control B. The idea here is that we can control P, but B is an ARBITRARY function of P (and possibly other AC's) and we will vary the independent AC's like P to find what THEY need to be to arrive at the desired dependent "goal" values. Here we are solving B=A, or B=55., or SQRT(P)=55. and finding P in this way. In the first equation we now enter I=10.\*C Enter some number of iterations to search for between displays. and for every display, we will see how P varies since we are displaying P in one of the cells. It will con- verge on the square root of 55. (which is 7.416) after some iterations. We may just use the TE command in- stead to set the I accumulator to avoid some of the complexity of messing with cells on the sheet. We then use the R (Recalculate) command after the TEst command to cause the solving to begin. Note that we could have more than just accumulator P varying and could have much more complex functions. Also, both A and B could be functions of any or all of these accumulators. Matrix computations may be used, or any other functions available through the spreadsheet, and the equations may be entered anywhere on the sheet. In this way, some quite powerful sorts of solutions to equations may be found. The philosophy of this is that if you know enough algebra to set up the equations for a problem (the hard part), you can set them up so the spreadsheet can solve them directly AnalytiCalc Reference Manual Page 70 yourself, if there is a simple way to do so. If there isn't, and if you can express the desired result as an equation, the AnalytiCalc program can do iterative searching to try and solve the problem for you, varying up to 8 accumulators to do so. If you really want to search for a situation in which several cells are close to several others, just make up an additional cell that is the "summed distance" between the desired targets. For example, if you want to arrive at a solution where cell B20 should approach B30, D25 should approach D35, and cell X15 should approach cell AQ5, make 2 new cells. One should be zero, and the other will have a formula that looks like (B20-B30)**2+(D25-D35)**2+(X15-AQ5)**2 or a simple variation, perhaps using multiplying in- stead of **2 as was done here. This will produce the desired results, PROVIDED that the cells depend somehow on the accumulator values you are varying. There are 2 more functions for the USRFCT.F40 sources which permit access to formulae and allow more flexible com- mand procedures. The function *U XQTCM command will execute the command (terminated by the end-of-line), with any com- mand except X or K permitted, from inside a cell. This allows command files driven from cells to control mov- ing data, etc. It is suggested that cells containing these references in command files have the last command in the file set the cell format to [A] so that the com- mand execution will be once only, unless repeated com- mands are desired. The function "*U STRVL V1,start;len" will return a value that is made from up to 8 characters in the FOR- MULA of cell V1 (where V1 is any cell name), starting at character "start" and for "len" characters. The value is computed such that a numeric sort on these values is equivalent to an alpha sort on the formulas; the characters are multiplied from the left by 128 and the next ASCII value is added. This function may be combined with the DS commands to sort on text, or with the XQTCM functionality to drive a slower but more thorough sort on values. The function "*U HERE" will return the current lo- cation on the matrix. The normal result is a hash-coded address of the physical cell pointed to, in the % accumulator, and the column and row in AnalytiCalc Reference Manual Page 71 accumulators T and U respectively. The row number is up by 1 (as the X,Y accumulators are at start of cell evaluation) to account for accumulator storage. Also, the flags for manual display update and manual recalcu- late are returned in accumulator V. The 4 bit being set means manual view is set on, and the 2 bit means manual recalculate is on. The W and Y accumulators re- turn the current maximum column and row used also. These are designed to make it easy to use the IR com- mand (Inplace Relocate) to go over only cells in use. The function *U FFTFW and the function *U FFTRV perform Fast Fourier Transforms in the forward and reverse (inverse) directions on the given range of data. The largest power of 2 of data points less than or equal to the given range is used and a one-dimensional real FFT is performed. This is useful for digital filters and the like. The input values are replaced by their FFT so it is best to work with values copied from the real inputs (use the *U MOVEV function to move the values before use). The function *U LINEF Vy:Vy,Vx:Vx (with the Vx range optional) fits a line to the input range. It will use the Vx range as the X coordinate if supplied or will compute a range if none is supplied using the cell coordinates. It computes the line equation Y=mX+B and returns the slope m in accumulator U and the inter- cept B in accumulator T. The value returned in accumu- lator % from the function is the Y fit error of the line fit. The W accumulator gets the correlation coef- ficient of X and Y. In the VAX/VMS and Amiga versions of AnalytiCalc, a call of form: *U STxxxxxx range;range;...;range>outrange;outrange so that the "xxxxxx" part is the function name to be called will allow any function to be called. These functions must be linked with AnalytiCalc, but they need not be modified. Input ranges of the spreadsheet are copied to an internal scratch area and passed to the function, which is called. On return, the output areas of the spreadsheet are filled in with return values in the scratch area which were passed to the functions. This facilitates adding new functions to AnalytiCalc, since pre-existing code can be incorporated fairly painlessly via modification of one "glue" subroutine's tables. There is no requirement for either input or output ranges, so constructs like plotting packages and the like could easily be added via this mechanism. *XV filename V1 *XF filename V1 The *X class commands are for sheet linkages. *XF loads a Formula from another saved spreadsheet, while *XV loads a Value (normally from a NUMERICALLY saved sheet). They look in the given file for the named variable in the saved sheet and place the formula or value in the current position of the current sheet. Note the *P commands may be used to reset the position if it is desired not to overwrite the current formula in the case of the *XF command; the *XV command as- sumes the sheet was saved numerically and does not mod- ify the formula. Note that each *X class command gets 1 number or formula. This command will NOT work with sheets saved with the PB commands. AnalytiCalc Reference Manual Page 72 AnalytiCalc File Interface Functions Command level functions are given as FILcccccc where ccc is the actual command. Equivalent functions are available from cells using the syntactic form *U DBcccccc arguments (where the "arguments" part is any arguments, generally some required.) Some extra optional VAX functions exist that allow regions of the sheet to come from or go to sequential or random files to or from some range of cells. The following commands exist (the "*U DB" or the "FIL" is omitted:) OPINS range filename Open input sequential on filename for cells in range When the input file is open and any operation causes a read of the cell, the FORMULA will be taken from the file and used. If the file is opened with the OPINU option (for Update), then when the FORMULA of the cell is written, it will also write to the file. The OPINU option applies only to random access files. Initially both input and output reads are disabled (ENAINP and ENAOUT enable them). When a sequential file (or device such as a mailbox, pipe, or communications line) is used for input, it is advantageous to read data into a range once, then disable input/output again, to allow it to be handled between these commands. The EDTINP and FMTOUT commands are designed to do this to make it easier to use AnalytiCalc for a filter between sequential files. OPINR range filename Open input random on filename for cells in range OPINU range filename Open range for update on filename for read and write as ran- dom access. CLSINP Close input CLSOUT Close output OPOUTS range filename Open output sequential from range cells to filename OPOUTR range filename Open output random from range on filename ENAINP Enable input file readin (initially disabled) ENAOUT Enable output write (initially disabled) DISINP Disable input area readin AnalytiCalc Reference Manual Page 73 DISOUT Disable output write EDTINP range Enables input and output and, for each cell in the given range, reads and writes the cell, allowing the file read/writes to take effect. Each cell is flagged as valid but of text type; the DF command must be used to reset any that should be treated as computable. Input and output are disabled on completion of the command. FMTOUT range VALOUT range Enables input and output and for each cell of the range takes the VALUE of the cell, reads the cell, sets it of text type, and writes the text equivalent of its value to the cell. In the FMTOUT command the cell's display format is used for the conversion. In the VALOUT command a large builtin format is used to preserve all significant digits. Spaces are discarded prior to output. Cells are left con- taining textual data corresponding to their numeric values, stored in the formulae. The DF command may be used to reset these cells to numeric types if desired. Conversion errors are ignored. Input and output are disabled on completion of the command. CMPFRM V1:V2[,V3:V4] Compares two formulas. It returns, in the % accumulator, the index of the formula in cell V2 in the formula in cell V1. Lengths used are those of both formulas UNLESS the V3 and V4 cell arguments are seen. In that case the value of V3 will be used as the length of the formula for V1 and the value of V4 will be used as the length of cell V2. If ei- ther value in V3 or V4 is outside the range 1 to 109 both values in V3 and V4 will be ignored. Also on output the W accumulator will be set to -1. if V1 is lexically earlier than V2, 0. if they are lexically equal, and +1. if V1 is lexically later than V2. LENFRM V1:V2 returns the length of the formula for cell V1 in the % ac- cumulator and in cell V2 IF CELL V2 IS VALID. Otherwise the cell specified in V2 is ignored, though it must be present in the command or function. TRMFRM V1:V2,V3,V4 Reads the formula in V1 and uses V3 and V4 as start and end byte numbers within it. It returns to V2 the formula that is between the start and end bytes, trimming the V1 formula into V2 by chopping out the undesired parts. The find sub- string function CMPFRM can be used to find delimiter bytes if absolute columnar formatting is not desired. AnalytiCalc Reference Manual Page 74 Associated DBxxxxxx and FILxxxxxx commands both exist and have the same syntax. These commands give a means for having a spreadsheet template operate on files (treated as formula inputs or outputs) from other independent programs. Random files have 128 byte record length and unique indices computed from the relative position in the range being used. Sequential files are variable length but never have more than 110 bytes (formula size) in them. Interpretation of formulas is up to the spreadsheet. On sequential reads, data is read until EOF and then the file is closed. Input is treated as a "last step" before returning the for- mula to the main program and just overwrites any internally stored formula. A dummy edit will enter the file data into internal storage. On output, file write and internal storage both occur. Open failure just leaves the files closed. "range" means v1:v2, a full 2 dimensional matrix range with upper left corner and lower right corner cells specified. Calcula- tion order is unaffected and generally is across fast, down slower, like a TV raster scan. The functions for CMPFRM and LENFRM are provided in addition to the other functions here to allow the spreadsheet to more easily lo- cate and parse text in data coming from generic applications. Note that the {cell constructs can be used (possibly with an *U XQTCM E function) to move pieces of text around in conjunction with copies. The CMPFRM function can help find out if the reply from the applica- tion is what is desired and already built-in conditional handling can be used for acting on this information. As in all these functions, CMPFRM and LENFRM have the forms FILCMPFRM and FILLENFRM interactively, and *U DBCMPFRM and *U DBLENFRM in formulas (with the remaining arguments included.) These functions are meant to allow some "connection" of AnalytiCalc with arbitrary other applications by batch files or EX- ECed subprocesses, where AnalytiCalc can produce inputs for another application's console input and accept another program's output as its input in an intelligent and controllable way. Also the string functions provided can be useful for certain ap- plications independent of file I/O. AnalytiCalc Reference Manual Page 75 -ADDITIONAL FEATURES- CALC is similar to FORTRAN with respect to operator precedence. Blanks may occur anywhere on a command line without effect except after a single quote mark used to specify a single ASCII character constant. CALC extends the ANSI FORTRAN syntax by allowing the following: 1. multiple assignments on one line, for example I=J=K=812 2. Unary + and unary - are allowed, for example 2*-3 +2+-7 -2**4 are all legal. The last expression evaluates to 16 because the unary - has a higher precedence than exponentiation. 3. exponentiation may be indicated by using ! as well as ** If any of the declarations are entered (such as *INTEGER) and no argument to this command is given, then CALC will print out the vari- ables that have been assigned that data type. Note that a variable can be assigned to different data types using such commands and still not be assigned a value. If you attempt to output the value of such a variable, an error message will result. USAGE NOTES 1. When you iterate on a file by a call such as *@REPEAT X then note that A) X must have been set to a positive value when the command is executed or else the file will not be executed. B) If the file of commands does not change the value of the vari- able X you will enter an infinite loop. You can explicitly set X to a non-positive value, use the *Z command to zero it (if it is not %), or include a *R command to give you a chance to reset the variable and get out of the loop. C) *E and *S will allow you to exit from the command file REPEAT but will not of themselves prevent repetitions. D) Entering constants echo on the terminal (assuming *V is properly set) and can change the value and type of the vari- able %. This is important to remember when using % to con- trol the iteration of a file. 2. In FORTRAN, -A**2 is the same as -(A**2) with CALC, AnalytiCalc Reference Manual Page 76 -A**2 is the same as (-A)**2 (just like SNOBOL!) 3. If R and A are positive reals and I is a positive integer, some compilers like RSX-11M's F4P won't allow (at run time) evaluation of (-I)**R (-A)**R 4. Under RSX-11M F4P you will find 2**.5 to have value 1 while 2.**.5 has value 1.41421356237310 5. 10E10 is a hexadecimal constant (integer) while 10.E10 is a real. AnalytiCalc Reference Manual Page 77 Appendix A ASCII CALC ASCII CALC ASCII CALC ASCII CALC CODE PRINTS CODES PRINTS CODE PRINTS CODE PRINTS ---- ------ ----- ------ ---- ------ ---- ------ 0 ^ 32 64 @ 96 ` 1 ^! 33 ! 65 A 97 a 2 ^" 34 " 66 B 98 b 3 ^# 35 # 67 C 99 c 4 ^$ 36 $ 68 D 100 d 5 ^% 37 % 69 E 101 e 6 ^& 38 & 70 F 102 f 7 ^' 39 ' 71 G 103 g 8 ^( 40 ( 72 H 104 h 9 ^) 41 ) 73 I 105 i 10 ^* 42 * 74 J 106 j 11 ^+ 43 + 75 K 107 k 12 ^, 44 , 76 L 108 l 13 ^- 45 - 77 M 109 m 14 ^. 46 . 78 N 110 n 15 ^/ 47 / 79 O 111 o 16 ^0 48 0 80 P 112 p 17 ^1 49 1 81 Q 113 q 18 ^2 50 2 82 R 114 r 19 ^3 51 3 83 S 115 s 20 ^4 52 4 84 T 116 t 21 ^5 53 5 85 U 117 u 22 ^6 54 6 86 V 118 v 23 ^7 55 7 87 W 119 w 24 ^8 56 8 88 X 120 x 25 ^9 57 9 89 Y 121 y 26 ^: 58 : 90 Z 122 Z 27 ^; 59 ; 91 [ 123 { 28 ^< 60 < 92 \ 124 \ 29 ^= 61 = 93 ] 125 } 30 ^> 62 > 94 ^ 126 ^ 31 ^? 63 ? 95 _ 127 Interactive Calc in the AnalytiCalc environment Calc runs as part of AnalytiCalc to do much of the math. The K command drops into straight Calc and it works as described here. There are a few additions however. First, Calc can access all the variables on the sheet (names a1 and up), which may be up to 8 bytes long. However, while it can examine or set any of these, the spreadsheet recomputes them from its AnalytiCalc Reference Manual Page 78 disk based formula file when re-entered by the *E command. To modify this file, the "current position" in the physical display sheet may be set by answering the prompts after giving the *P command. Then the current display value (contents of the "%" variable) may be en- tered into the disk file by executing the *W command. Numerous other manipulations are possible, and by setting the view level to 3 (*V 3) it is possible to debug a command file that drives a spreadsheet cell from interactive mode (by tracing through its' operation to see what it is doing). All interactive Calc commands work also from indirect files. The physical location is clamped to legal values. Note too that any unused cells of the sheet may be used as scratch accumula- tors. However, they may not assume any types of multiple precision numbers, as they are all 8 bytes long. The internal expression stack is 40 tokens deep, so very complex expressions may have to be split into pieces to work. Error messages will be shown if this happens; they are placed on the top line of the screen. The AnalytiCalc GRAPHICS System ACGRAF GRAPHICS INTERFACE AND OUTPUT FROM ANALYTICALC Introduction The AnalytiCalc system uses a separate program to provide its graphics output, permitting easy tailoring to different graphics devices. The routine supplied uses a character graphics subroutine (GHASP) to produce histograms and scatter plots from any rows or columns of a saved (numerically saved, that is) sheet which has been saved via the PPN or PDN commands. The output is printed normally. However, users with other devices may easily modify this program to insert their own favorite graphics output device output into the pro- gram. There are several options to govern plotting, and one plot per command is produced. Once a saved sheet is generated, many plots can be made from it without reentering the sheet. An advantage of this approach is that the address space consumed by the graphical routines is not taken from the spreadsheet. Graph- ics may be handled with a fair degree of transparency by command files, or even by spawned commands. Another advantage is that addi- tional graphics devices may be added easily without modification to the spreadsheet. The current version uses character graphics to permit usable displays even on monochrome monitors. (Note: there are several additional graphics programs supplied which are given in AnalytiCalc Reference Manual Page 79 source in BASIC. These provide additional functionality not described here but allow line charts, bar charts, piecharts, and numerous other options.) Getting Started: Run AnalytiCalc (Type ANALY) and create your spread sheet to your liking. When it is set, use the following commands: L A1 go to cell A1 (top left of sheet PPN Put out physical sheet numerically. When AnalytiCalc asks for a filename, give one which describes your file well (with a file type) followed by return. Example: My- file.num When AnalytiCalc asks you for maximum displacements, enter 0 in both cases (equivalent to telling it to save everything). When done, you may use the X command to exit AnalytiCalc. Now run ACGRAF by typing ACGRAF When ACGRAF first starts, it asks for an output dataset. This may be any file or device on the system. If CON: is used as output (the terminal), or if the filename chosen begins with the letters "TT", the graph will be scaled to fit on a terminal screen. Otherwise it will be sized to fit a sheet of (normally 132 column) printer paper and will have some extra statistical information about the plot. When ACGRAF runs, it will ask for a command line. A typical command to give a scaled plot of column H from row 1 to row 30 would be Enter plot command>NN myfile.num H1:H30 [S] Note that the rows and columns are referred to exactly s on the sheet itself. The initial NN could be LL or LN or PP or other letters, but L in column 1 takes the logarithm of the numbers in the first (in the example the only) range of numbers selected, and L in column 2 takes the log of the numbers in the second range. A P is used instead of an N where the values displayed should start from 0. rather than from the lowest value in the region selected. Should the range to be plotted include only POSITIVE numbers, make the first 2 letters P's. This will ensure the scale starts at 0 rather than at the minimum positive value selected. The command line would then look like: Enter plot command>PP myfile.num H1:H30 [S] AnalytiCalc Reference Manual Page 80 Scatter plots plot entries in the first range against cor- responding entries in the second range selected. One dimensional plots are up to 100 bins wide and scatter plots are up to 50 bins by 50 bins. Height of histograms is Scaled to fit a page if the [S] form is used and will Vary to whatever is needed if the [V] character is used. Widths of plots are less than 100 (or 50) if the number of entries plotted are less than those dimensions; if the number of en- tries plotted is greater, it will be scaled to the 50 or 100 bin maxima (which derive from paper size mainly). If the option [H] is chosen, the plot will be Hacked off at one page, but scaled as if the [V] Vary option were specified. SYNTAX AND USAGE: This program is designed to allow an interactive user to enter a single command line to the program which it will parse and allow graphic output from AnalytiCalc saved spread sheets. The assumption made is that the sheet has been saved with the PPN or PDN command. The filename must appear in the command line and variables in the file (named AS THOUGH THE CURSOR HAD BEEN IN CELL A1 WHEN THE PPN / PDN WAS DONE) may be histogrammed or scatterplotted against each other. A FORTRAN plot package for ordinary printers will be used for this version. Other routines may be readily used as replacements for specific graphics output devices. There is a document supplied PCSAVFMT.TXT which will give the format used to save AnalytiCalc/Por- taCalc files to simplify reading them in. The popular VisiCalc (tm) DIF format was NOT used in this program because it cannot encode the variety of display formats available. The format chosen was rather a straight ASCII format permitting almost any language to read the saved files with a minimum of grief, and which permits editing saved sheets with a normal text editor. (The fields are even comma separated to make reading in with BASIC easier.) Input syntax: PP or NN or LL filename.ext V1:V2 [c] V3:V4 +switches where an L in columns 1 or 2 takes log of 1st or 2nd range numbers (base 10), a P indicates Positive numbers (so 0 is included in the range to be plotted and the plot starts at 0. and increases rather than starting at the minimum observed value for an initialized cell in the saved sheet), and N means Normal, i.e., neither of the above; the minimum and maximum coordinates are taken as min and max values in the set of values to be plotted, filename.ext appears at the start of the command line after a space and with a space following it and is a valid RSX or VMS file AnalytiCalc Reference Manual Page 81 speciciation (the same as was given to AnalytiCalc in the PP or PD command that saved the sheet normally). V1:V2 and V3:V4 are ranges. V3:V4 is optional and its presence implies a scatter plot. These ranges must be either a row or a column or part of them. If only range V1:V2 is present, a histogram will be done using the Scale option of GHASP to fit the plot onto a page. The plot will be set up for 100 bins horizontal, 50 vertical, or the number of entries to be plotted, whichever is less. If the V3:V4 range exists, the character Q in the [c] position (the [] are required) will result in a "density" plot in which the program will attempt to print darker in filled bins. This is crude and the default is to use a 2 digit number. Again, plot size will be scaled to 50 by 50 bins. For histograms, the character S implies Scale the plot to fit one sheet of paper, the character H implies Hack the plot off at one sheet, and the character V implies Vary height, using as many sheets of paper as needed to plot the data. The character S should be used for "normal" scatter plots (in which number of entries in a scatter plot "bin" is represented as a number or character). To use ACGRAF from inside AnalytiCalc, use the $ command to pass the command line to the system; this will allow graphics to be done without leaving the AnalytiCalc environment. This can be done from an indirect file driving AnalytiCalc also, without special interven- tion, using normal DOS redirection commands. Two switches are implemented. +Hnnn sets Height of the result plot to nnn characters. +Wnnn sets the Width of the result plot to nnn characters. Both may be specified for a plot; they override default sizes. They may be used to produce smaller graphs than usual for inclusion in documents or displays. AnalytiCalc Reference Manual Page 82 Appendix B Discussion of Storage Allocation Issues Storage Allocation Discussion If you use AnalytiCalc, the storage needed per cell in the work- file is 10 bytes per cell, and for formulas, 12 bytes of formula go into every 16 bytes of workfile (the remainder is for accounting overhead). Thus, for an 18,000 cell sheet, 180K bytes is the maximum useful value file size, and 288K bytes would permit a 12 character formula for each cell, while 2592K bytes would give every cell 108 characters and is the maximum useful size. (You can allocate more memory to allow 109 characters per cell, but this usually will be mostly wasted.) Since a symbol table algorithm is used, sizes may be as small as desired. If the size chosen is less than the memory al- location built in (1K is guaranteed to be), no files will be opened and cells will be stored in memory only. Note that if value and for- mula file sizes are given negative, a slower but sometimes more effi- cient packing will be done of the file. The workfiles are opened in one's default device at the time of running the program. The "slow" algorithm picks a diskfile page to use by taking the cell number mod- ulo the number of pages in the file. The "fast" one picks a diskfile page to use by taking the position in the total sheet (columns in- creasing fastest) and making the number of cells in a diskfile page vary as number in the sheet divided by the size of the diskfile. Thus, cells close to one another on the sheet are kept close on the disk file. This can overflow a disk page if too many cells for a page are used close to each other. Therefore, the program will ask you how many rows and columns you propose to use for purposes of disk clustering. If the "fast" algorithm is used, the disk page will be chosen as (cell address)*(number of disk pages total)/(size of the total sheet you said you'd use, clamped to the max physical size). If you say you'll use a small area, cells will be scattered better. If you use beyond the area you said, disk pages are chosen in modular fashion. Therefore, if you run out of room in a file, try either varying the size for allocation, or adding space to the files. If you give the exact size you're really using, the "fast" algorithm will equally divide disk pages among cells on the screen you're using for rectangular use. The size question is ignored if the "slow" al- gorithm is used, since that algorithm equipartitions without needing to know the size used. Typical values to use for the number of rows and columns you intend to use for small problems might be 16 by 16. The program asks how big to make the disk based files next, given that it now knows how to allocate cells to the files. There are two effects of importance here. First, if you don't allocate enough space, scratch files may overflow and you'll have to save the sheet and rerun with bigger files. Second, if the scratch files are TOO big, the program will waste time switching pages in AnalytiCalc Reference Manual Page 83 memory unnecessarily. Therefore, when you tell AnalytiCalc how much space you intend to use, it prints out the sizes of files needed AS- SUMING that you will fill in EVERY cell in the range indicated, and that formulas are 12 or less characters long. If you use less cells, both files may be shorter. If formulas are longer, the formula file needs to be made bigger. The program won't actually go out to disk unless the files are bigger than its in memory buffers, so don't worry about that condition. The suggested sizes are usually a good guess. (These are printed on the screen after you tell it how many rows and columns you intend to use.) AnalytiCalc Reference Manual Page 84 Appendix C AnalytiCalc Logical Names AnalytiCalc/Amiga has a symbol table and allows you to define symbolic names of up to 16 characters length, containing letters or numbers, and use them inside equations. To define a symbolic name, use the command TE {SYMBOLICNAME=Vn where Vn is a cell name (like A5, R17, AB122, or the like), and "SYM- BOLICNAME" is the name you wish to define. The name is now asso- ciated with the CONTENTS OF the cell named (as Vn). When you use the construct {SYMBOLICNAME in an equation, the contents of the cell as- sociated with that name replace the name prior to evaluation. Any contents may be used, though the total resulting string must not grow to over 109 characters. Recursive substitutions are permitted to a depth of 20, so name definitions may use other name definitions. To remove a name definition, use the command TE {SYMBOLICNAME=0 You should remove definitions explicitly since they are not automa- tically removed by adding new definitions. Up to 301 names may be defined in the sheet. Note that scratch cells A0 through approximately BK0 may be used to hold definitions if desired. To store cell definitions for later use, the special command must be used as follows to save them: TE {*= This will cause AnalytiCalc to ask for an Output file. You reply with a filename to store (or show) the definitions. The program will then dump all known symbol definitions in the same format needed to do the definitions of the names. If a disk file is specified, it can be read in via the @file command so the symbols may be redefined to a later AnalytiCalc run. Storing definitions on a CON: window can give a quick look at them interactively, but usually it is better to print the definitions (which can be done by giving PRT: (or your printer's real name if that's not it) as the output file name. AnalytiCalc Reference Manual Page 85 This facility can save considerable formula typing time. Where you have a very large formula to be duplicated many times, if it can be specified relocatably, a reference to a short name (or a cell, since {name and {cell function the same) can save storage and allow you to get by with smaller disk work files. EXAMPLE RUN: L D0 ; go to phantom cell D0 ET SUM[A1:A4] ; fill in a macro definition as text TE {SALES=D0 ; define "sales" as this sum M2 ; set automatic motion down L A1 ; go to A1 E 2.3 ; enter some junk data just to show this E 4.86 E 1.2647 E 8.3454 EV {SALES ; Enter sales value as a numeric. ; the cell will display the sum of A1 thru A4. An example position independent formula might be: SUM[ P#0#-5:P#0#-1] which means the sum of the cells from 5 above the current physical cell to 1 above the physical cell. This form if used in a name will work wherever it is used, regardless of location. AnalytiCalc Reference Manual Page 86 Appendix D AnalytiCalc Function Summary The available multiple argument functions are: SUM[variables] Sum of all arguments MAX[variables] Max of arguments MIN[variables] Min of arguments AVG[variables] Average of arguments AVE[variables] Average of arguments excluding zero args STD[variables] Standard deviation squared AND[variables] Boolean AND of all variables in list IOR[variables] Boolean inclusive OR of variables NOT[variable] Boolean complement of variable XOR[v1,v2] Boolean exclusive OR of v1,v2 EQV[V1,V2] Boolean "equivalence" of V1,V2 (complement of exclusive OR, true if bits have the SAME value) CNT[variables] Number of nonzero variables in list MOD[V1,V2] Returns V1 modulo V2 (i.e., remainder of V1/V2 division.) SGN[v1] Returns 1.0 times sign of V1 LKP[var,variables] Lookup variable in "variables" range greater or equal to var, return its index (starting with 0) into variables range. LKN[var,variables] Lookup variable in "variables" range less than or equal to var, return its index (starting with 0) into variables range. LKE[var,variables] Lookup variable in "variables" range strictly equal to var, return its index (starting with 0) into range. Note all LKP, LKN, LKE return the last variable index if no satisfactory value found. NPV[disc,vars] Net Present Value of vars (equal time interval numbers), at discount rate disc where disc is a fraction (e.g., .12 for 12%) IRR[PV,FV,returns] Internal Rate of Return. Will compute internal rate of return on up to 20 periods, returning rate per period. The returns are expected to be at equal time intervals. PV and FV are initial and final values of investment and the result is computed via Newton approximation. PMT[princ,inter,nper] Payment (mortgage payment per period) function. Will compute payment per period for principal amount "princ" with interest AnalytiCalc Reference Manual Page 87 per period as "inter" and number of periods as "nper". All arguments must be cells. The formula is the standard ordinary annuity formula. Interest rate must be a fraction so that 14% would be 0.14, for example. PVL[payment,inter,nper] Present Value formula. Computes present value of an annuity given "payment", the payment per period, interest rate per period (as a fraction, so 12% is 0.12) in "inter", and number of periods as "nper". All arguments must be in cells. RND[DUM] Generates a random number between 0. and 1.0. An argument is needed but it is NOT touched by this function. .INDEX Random numbers .Index RND function IF [V1.rel.V2] statement | else-statement Compares two variables and executes either "statement" (if the relation is true) or "else-statement" (if the relation is false). Valid relations (in the place of the .rel. above) are: .EQ. Equal .NE. Not Equal .GT. Greater than (V1 greater than V2) .LT. Less than (V1 less than V2) .GE. Greater than or Equal to (V1 >= V2) .LE. Less than or Equal to (V1 =< V2) The following special constructs in a formula act as functions: _@V1,V2 Means get the values stored in V1 and V2 and use them as column and row locations pointing at some cell in the sheet. Replace the construct with the name of that cell. _#V1 Means take the real number in cell V1 and unpack it as if it had been a packed value from a formula with 8 characters packed; then convert it back into ASCII and place in the formula in place of this construct. This construct is intended to be used with the *U STRVL function to allow retrieval and edit of formulas. The *U XQTCM function permits use of the EDit command within a cell for string manipulation. The following single argument functions are available: FUNCT NAME ARG TYPE FUNCT VALUE DESCRIPTION ------------------------------------------------------- ABS REAL REAL absolute value DABS REAL REAL absolute value AnalytiCalc Reference Manual Page 88 IABS INTEGER INTEGER absolute value IFIX REAL INTEGER REAL to INT conv. AINT REAL REAL REAL truncation INT REAL INTEGER REAL to INT conv. IDINT REAL INTEGER REAL to INT conv. EXP REAL REAL e**X DEXP REAL REAL e**X ALOG REAL REAL natural logarithm DLOG REAL REAL natural logarithm ALOG10 REAL REAL logarithm base 10 DLOG10 REAL REAL logarithm base 10 SQRT REAL REAL square root DSQRT REAL REAL square root SIN REAL REAL trigonometric sine DSIN REAL REAL trigonometric sine COS REAL REAL trig. cosine DCOS REAL REAL trig. cosine TANH REAL REAL hyperbolic tangent DTANH REAL REAL hyperbolic tangent ATAN REAL REAL arc tangent DATAN REAL REAL arc tangent ASIN REAL REAL arc sine DASIN REAL REAL arc sine ACOS REAL REAL arc cosine DACOS REAL REAL arc cosine TAN REAL REAL Tangent DTAN REAL REAL Tangent The following "equation commands" also act as functions with the effects described: *@filename Where filename is the name of a file of CALC com- mands. CALC reads the file and executes the commands. *ASCII Declares a list of variables to be of type ASCII. *C COMMENT line. *N NOVIEW. *V VIEW. Controls printing options in K mode *R READ. Allows a single line to be read from the terminal. *REAL declares specified variables to be REAL*8. *DECIMAL Declares specified variables to be REAL*8. *S STOP. Same as *E. Goes back to spreadsheet. *E EXIT. Gets out of K mode, back to spreadsheet mode. *Z ZERO. Zeroes all accumulators. *G *G V1,V2 (where V1 and V2 are cell or accumulator names) will evaluate V1 and V2 as the column and row numbers, on the physical sheet, of the desired cell. The ad- dressed cell's value is retrieved and used as the resulting number. *W Takes the value at the current cell and writes it out to the formula as a numeric (floating) value. The *WF command stores the cell similarly, but uses the format of the current cell instead of the D32.25 format used for *W. *P The *P command resets the current cell coordinate from within a cell (until the next cell is evaluated only). It has several forms: AnalytiCalc Reference Manual Page 89 *P - By itself, *P causes Calc to prompt for the new phy- sical column and row number. *P V1 - This moves the current location to the named cell where V1 is the cell name (e.g. A5, H2) *P@ V1,V2 - This uses V1 and V2 (cell names) as column and row numbers and changes the current physical cell position to that defined by the contents of cells V1 and V2. This gives complete addressing of the sheet from within any cell. *F *F Label - If the value in % is positive and nonzero this command rewinds the input file for the AnalytiCalc @ command and seeks a line beginning with the characters *CLabel (where "label" is what you put after the *F command). *J *J Label - This command behaves as the *F command but operates on the file used by the Calc *@ command rather than the AnalytiCalc one. *QF *QW The *QF (Float) or *QW (write) commands are used to examine sequential files created outside AnalytiCalc and return values or formulas. Their syntax is *QF filename ?key1? ?key2? <lm> or *QW filename ?key1? ?key2? <lm> where ?key2? is optional l and m are delimiter characters for start and end of the parts of the selected records to extract (defaults to first part of the record) filename is just the file specifier in the host OS. The *QF command gets a part of a record containing the keys specified and turns it into a number ("floats" it). The *QW command Writes that part of the record to a formula. The special characters in the <lm> part of the command are delimiters of the area to be picked out. Variant forms allow such choosing to be by column number instead of key character if needed. *U YRMOD VY,VM,VD returns the Julian date (in %) computed from the Year (in VY), the month (in VM) and the day (in VD), where VY, VM, and VD are sheet cells. These may be the result of date arithmetic. *U JDATE Var assumes the formula in cell Var (any spreadsheet cell name) contains a date string in the format YY/MM/DD. It reads this formula and converts the date to a Julian date, returning it in the % accumulator. AnalytiCalc Reference Manual Page 90 *U JTOCH Jul,Var assumes that variable Jul (any spreadsheet cell) con- tains a Julian date and changes it into an ASCII string in the cell whose name is in the Var position here. *U DATE VY,VM,VD,Var uses VY, VM, and VD as year, month and day, and com- putes a Julian date from them. It then composes an ASCII string of form YY/MM/DD for that date and stores in the formula for Var. *U WKDYS D1,D2 computes the number of workdays between Julian dates D1 and D2 just as taking the difference of two Julian dates gives differences between calendar dates in days. *U WKDIN D1,N1 returns a Julian date that is N1 work days after the date in D1. *U IDATE() returns the current date as a Julian day. *U MTXEQ(AA:AA,XX:XX,BB:BB) solves equation AX=B where A, X, and B are matrices, and where the notation AA:AA means two cells at the upper left and lower right edges of matrix A (e.g. B2:C3 for the 2 X 2 matrix so defined), and the XX:XX and BB:BB notation means the same for the X and B ma- trices. *U MOVEV mtxa,mtxb moves values from mtxa to mtxb (useful prior to calling MTXEQ). *U MDET mtx This function computes and returns the determinant of matrix mtx. *U MPROD A,B,C This function will multiply matrix A by matrix B giving AnalytiCalc Reference Manual Page 91 matrix C, provided that their dimensions are compati- ble. *U MADDV A,B,C This function adds matrix A to matrix B and stores in matrix C. All matrices must have the same dimensions. *U MSUBV A,B,C This function subtracts matrix B from matrix A leaving the result in matrix C. *U MMPYT A,B,C This function multiplies matrix A-transpose by matrix B and stores the result in matrix C; dimensions must be compatible. *U MMPYC A,B,K This function multiplies every element of matrix A by constant K. *U VARY X,A,W,I,P;Q;R;S;T;U;V;W (Equation Solving by Iterative Search) This function allows AnalytiCalc to automatically search for solutions to equations over up to 8 dimen- sions. The operation is that the accumulators named in the fields shown as P;Q;R;S;T;U;V;W (one to 8 may be specified, only one is required) are varied by a frac- tion W about their initial values (later scaled down by the gradient of the change in X) to attempt to get ac- cumulator or cell X to equal accumulator or cell A. This is done for I iterations, where I is another ac- cumulator. *U XQTCM command will execute the command (terminated by the end-of-line), with any command except X or K permitted, from inside a cell. This allows command files driven from cells to control moving data, etc. *U STRVL V1,start;len will return a value that is made from up to 8 characters in the FORMULA of cell V1 (where V1 is any cell name), starting at character "start" and for "len" AnalytiCalc Reference Manual Page 92 characters. *U HERE will return the current location on the matrix and the current maximum row and column used. T,U get col, row. W,Y get max col, row used. *U FFTFW and *U FFTRV perform Fast Fourier Transforms in the forward and reverse (inverse) directions on the given range of data. *U LINEF Vy:Vy,Vx:Vx (with the Vx range optional) fits a line to the input range. Eqn: y=Ux+T. % gets err, W gets cor. coef. *XV filename V1 *XF filename V1 The *X class commands are for sheet linkages. *XF loads a Formula from another saved spreadsheet, while *XV loads a Value. The following *U DBxxxxxx functions also exist as commands of form FILxxxxxx with the same results. *U DBOPINS range filename Open input sequential on filename for cells in range When the input file is open and any operation causes a read of the cell, the FORMULA will be taken from the file and used. If the file is opened with the OPINU option (for Update), then when the FOR- MULA of the cell is written, it will also write to the file. The OPINU option applies only to random access files. Initially both input and output reads are dis- abled (ENAINP and ENAOUT enable them). When a sequen- tial file (or device such as a mailbox, pipe, or communications line) is used for input, it is advanta- geous to read data into a range once, then disable input/output again, to allow it to be handled between these commands. The EDTINP and FMTOUT commands are designed to do this to make it easier to use AnalytiCalc for a filter between sequential files. *U DBOPINR range filename Open input random on filename for cells in range *U DBOPINU range filename Open range for update on filename for read and write as random access. skip 1 *U DBCLSINP Close input AnalytiCalc Reference Manual Page 93 *U DBCLSOUT Close output *U DBOPOUTS range filename Open output sequential from range cells to filename *U DBOPOUTR range filename Open output random from range on filename *U DBENAINP Enable input file readin (initially disabled) *U DBENAOUT Enable output write (initially disabled) *U DBDISINP Disable input area readin *U DBDISOUT Disable output write *U DBEDTINP range Enables input and output and, for each cell in the given range, reads and writes the cell, allowing the file read/writes to take effect. Each cell is flagged as valid but of text type; the DF command must be used to reset any that should be treated as computable. Input and output are disabled on completion of the com- mand. *U DBFMTOUT range *U DBVALOUT range Enables input and output and for each cell of the range takes the VALUE of the cell, reads the cell, sets it of text type, and writes the text equivalent of its value to the cell. In the FMTOUT command the cell's display format is used for the conversion. In the VALOUT command a large builtin format is used to pre- serve all significant digits. Spaces are discarded prior to output. Cells are left containing textual data corresponding to their numeric values, stored in the formulae. The DF command may be used to reset these cells to numeric types if desired. Conversion errors are ignored. Input and output are disabled on completion of the command. *U DBCMPFRM V1:V2[,V3:V4] Compares two formulas. It returns, in the % accumulator, the index of the formula in cell V2 in the formula in cell V1. Lengths used are those of both formulas UNLESS the V3 and V4 cell arguments are seen. In that case the value of V3 will be used as the length of the formula for V1 and the value of V4 will be used as the length of cell V2. If either value in V3 or V4 is outside the range 1 to 109 both values in V3 and V4 will be ignored. Also on output the W accumulator will be set to -1. if V1 is lexically earlier than V2, 0. if they are lexically equal, and +1. if V1 is AnalytiCalc Reference Manual Page 94 lexically later than V2. *U DBLENFRM V1:V2 returns the length of the formula for cell V1 in the % accumulator and in cell V2 IF CELL V2 IS VALID. Otherwise the cell specified in V2 is ignored, though it must be present in the command or function. *U DBTRMFRM V1:V2,V3,V4 Reads the formula in V1 and uses V3 and V4 as start and end byte numbers within it. It returns to V2 the formula that is between the start and end bytes, trimming the V1 formula into V2 by chopping out the un- desired parts. The find substring function CMPFRM can be used to find delimiter bytes if absolute columnar formatting is not desired. As in all these functions, CMPFRM and LENFRM have the forms AnalytiCalc Reference Manual Page 95 Appendix E NEW FEATURES The following features were added to AnalytiCalc after the last edit to the manual: 1. Extended cell addressing. This feature allows you to treat the spreadsheet as though it contained 18000 rows and 18000 columns. You can use any storage in any part of this range. Thus, columns from A through ZPH are legal, and rows from 1 to 18000 are legal. The actual storage in virtual memory is still treated as a prime area of 60 by 300 cells, but extended areas move down the sheet as you go further to the right of the 60 columns, and right on the sheet as you go further down. The increments on overflow start at 50 rows and 10 columns, but can be reset by questions in the "modify widths and mapping" section of the S (setup) command. The mappings chosen are saved and restored in PP and GP commands. This allows you to use the storage as a giant area, though it can run out. (No possible one-CPU machine could really compute a 325 million cell sheet, if all cells were full, quickly enough to be use- ful.) Because it can be confusing to figure out what is going on, the Add (AR and AA) commands that insert or delete whole rows or columns do not work in extended addresses (i.e., beyond 60 columns and 300 rows). When using extended addressing, and after you have cells set up in the extended area, you should NOT alter the mappings via the S command, nor should you use the AA or AR commands. All other com- mands work normally. 3. The sheet save and restore now saves values in a PPX command as well as formulas. It also saves column widths and the number of rows and columns being displayed. If you specify a 4th character M in the command (e.g., PPXM instead of just PPX), AnalytiCalc will save the display-to-physical mapping in its saved sheet as well as the other sheet information. This considerably extends the saved size, so it is optional. Setting up your mapping while journaling is turned on, and then playing back a journal file, will still use sub- stantially less disk space, though having all the information in a saved sheet makes reload simpler. The PP commands have now the following effects: PPN Save numerical values (and labels) only. Formulas are NOT stored; only numeric values are. PPX Save both numeric values and formulas. The output files contain two records per cell used, one containing the value and the other containing the formula. On reload, AnalytiCalc Reference Manual Page 96 the value record sets the cell value, and the formula record sets the formula (which may cause the value to change once a recalculation is done). The value record is first and it's first character is "p" (lower case) while the formula record's first character is "P" (upper case). See the BASIC graphics programs for how to read the format, or type a saved sheet out to see how it looks. Use this form of save for most purposes. PPF Saves formulas only, not values (though a single recal- culation generally produces values). Use this for tem- plates where the data values are not meaningful or to save room. This sort of save cannot however be used for graphics, and the *XV command will not find useful values everywhere in such a saved sheet. PDN Saves Display sheet numerically only. Like PPN except that the offsets used are in Display coordinates, not relative to the physical sheet. PDF Saves Display sheet, Formulas only. No display sheet save of "both" numbers and formulas is implemented since the display sheet is smaller than the physical one and two saves can be easily done when both are needed. PxxM Where "xx" can be PN, PF, DN, DF, or PX - Saves Mapping as well as the sheet. This saves extra records (with row and column numbers offset by 64000 to keep them from confusing graphics programs) that allow Get (GP or GD or GPR) commands to set up the display to physical sheet mappings automatically. Thus, you can save all your windows and have them back automatically on reload. The mapping is saved from the current cursor location down and right only. The G (Get) class commands are unchanged except that they now reset all the new saved quantities when they are seen. Old saved sheets receive default values for the column widths, numbers of rows/columns, mapping, etc. so their behavior is not altered. V18-04 has the following new features: 1. If you have an ACINIT.PRM file, it will be kept open as an initializer file until EOF and may contain any AnalytiCalc commands. HOWEVER, if you use the file, it MUST contain at least the replies to all questions until you arrive at the spreadsheet's "calculation screen". It can be used to preset cell contents or perform other setups. Note that the ZA command allows you to reset storage parame- ters and the like, so having them preset by a startup file in this way does not reduce the generality of the program. It is often a good policy to do this to get the program started, as the setup com- mand can contain typed strings to initialize ANSI.SYS or to set up internal function key meanings without having to rely on an operator AnalytiCalc Reference Manual Page 97 or an external batch file to do so. The command %escape sequence to type%*%*% will just type the escape sequence; it may be used in ACINIT.PRM (after the part that answers initial ques- tions) to type any desired escape sequences for setting up the screen or whatever if that screen control method is chosen. L AO0 ET 1 L AW0 ET 2 L AR0 ET 3 L AT0 ET 4 L A1 Note that AB0 corresponds to code 59, and so on. AO0 corresponds to 72, AW0 to 80, AR0 to 75, and AT0 to 77 here. The 1,2,3,or 4 that are entered are the cursor move commands in command mode, entered as text so they cause no calculation overhead. Note that IF YOU HAVE AN ACINIT.PRM, IT will be read, NOT your console, for initial questions. (You can get the initial queries back with the ZA command however.) New Features of V18-07B AnalytiCalc V18-07B supports use for project management. To make this run efficiently, the selection functions were modified to record which cell was selected. Thus, the MAX, MIN, LKP, LKN, and LKE functions now return the column and row (respectively) of the cell they find in accumulators P and Q respectively. The function "*P@ P0,Q0" may be used to directly access this cell in subsequent calculations, or the P and Q accumulators can be used in other forms of addressing to access the selected cells. The functions otherwise behave as they did in earlier releases. The project management software (AnalytiProject) will be distributed only to registered (i.e., paying) users of AnalytiCalc. New Features - V18-07C AnalytiCalc V18-07C adds two functions to support computations in "workdays". *U WKDYS V1,V2 takes Julian dates in variables V1 and V2 and computes the number of workdays between the two, not counting the initial date if it is a workday (to be similar to differences between Julian dates). AnalytiCalc Reference Manual Page 98 *U WKDIN V1,V2 takes a Julian date in V1 and a number of days in V2 and returns the Julian date that is V2 workdays after the date in V1. It is thus a sort of inverse for the *U WKDYS function. The resulting date is guaranteed to be in the Monday to Friday range. A workday is Monday to Friday. The functions know nothing of holidays. New Features - V18-08A In V18-08A, AnalytiCalc can compute line fits and Fast Fourier transforms on data. *U LINEF Y1:Y2,X1:X2 computes a line whose equation is Y=mX+B, where m is returned in the U accumulator and B is returned in the T accumulator. The % accumu- lator returns the fitting error in Y. The line is fit over the range Y1:Y2 in Y, with the X coordinates optional and if they are present, they must be in the cells in the X1:X2 range given. If the X range is omitted, 1. to the max. number of points will be used. *U FFTFW V1:V2 computes the forward (t to freq) Fourier transform of the range and *U FFTRV V1:V2 computes the reverse (freq. to t) transform. The range must be a power of two long (if less, the largest power of 2 available is used and extra points are ignored). Data is assumed to be the real coef- ficients in the first half and the imaginary ones in the second half. FFT data is returned in this layout also. Thus, the real data (which is frequently the part of interest) is returned first. The reverse FFT of the forward FFT returns the original data. The FFT is of in- terest for finding the frequency distribution of partially sinusoidal distributions. New Features - V19.01A A modified command structure is supported now. If you type the command /; then the commands to AnalytiCalc are modified as follows: 1. Any command beginning with a digit (except single digits standing alone) is treated as an EV followed by the text. 2. Any command beginning with / has the / stripped off and the rest is treated as a command. AnalytiCalc Reference Manual Page 99 3. Any command beginning with " is treated as an E" command (enter text) with the " stripped out. 4. Anything not starting with a digit, a " character, or a / is treated as an Enter command and is prefixed with "EN " and used. The old (default) command mode is restored with the command "//" so that /; and // switch between the two. These commands may be given in ACINIT.PRM, other command files, or from the console. They introduce incompatibilities in that in the new version commands like "DF range [format]" must be entered as "/DF range [format]" (for ex- ample) to work. To facilitate writing command files that work in both modes the command / switches the mode with a save cell. A sequence of commands: /# (save current command mode in save cell) // (enter old command mode) ... (do all of command file) /# (restore saved user command mode) will allow a command file to operate and be written to run predict- ably in either user environment choice. Please NOTE that special characters in the first character are interpreted as commands in /; mode. The * character is handled as a comment. To enter formulas with functions starting with *, use the COMMAND form, e.g. /EV *U IDATE. This always works and disambigu- ates syntax. When the system is in /; mode, the cell prompt at the 23rd line of the screen ends with : so that it appears like AB 21: When the system is in // mode (initial default) the prompt ends with the > character so it appears like AB 21> This should aid in distinguishing modes during use. New Features of V19.2 In V19.2, AnalytiCalc adds the OAD and ORD commands. These are like the OA and OR commands except they leave the relative mappings of the screen alone. Thus, if you already split your screen into 3 AnalytiCalc Reference Manual Page 100 parts and use the command "OAD cell" to reset the mapping, the resul- tant screen still has 3 regions at the same relative positions as be- fore. If you use the ORD command, then only the part of the screen right and down from the cursor is affected, but any screen partition- ing is maintained. If these commands leave some areas pointing to nonexistent areas, use a command OR or OA to reset the areas desired to something real again. V19.2 added the change from > to : at the end of the prompt depending on // or /; modes also. New Features - V19-03A The recalculation has added 2 new modes in V19-03, Recalculate Incremental (the initial default), and Recalculate Entry. The R command re-establishes the old default mode which recalcu- lates ALL formulas on the whole sheet for every entry of new material. The RI command sets Recalc Incremental mode which recomputes all cells on the display screen, but skips non-displayed cells. The RE command sets Recalc Entry mode which ONLY recomputes the newly entered formula, leaving the rest alone. (It is reliable in the "prime" region of the sheet but may fail in extended address areas.) The RR command is equivalent to RF (Recalculate, FORCE recompu- tation of constants), EXCEPT that it leaves the mode alone so if AnalytiCalc is in RI or RE mode, it stays that way. Note a full recalculation is done after Get commands (loading saved sheets) regardless of mode. The command .. (two periods at the start of a line) in a com- mand file closes the command file and returns to the console. This has not previously been documented though the function has been present for some time. It is recommended that RE or RI mode be used instead of RM mode (which completely disables recalculation) during data entry if other modes are too slow. AnalytiCalc may be obtained from Glenn Everhart, 25 Sleigh Ride Rd., Glen Mills, PA 19342, for $25.00 (I'll supply the disk for that; I ask only $10. to register it however.) AnalytiCalc Reference Manual Page 101 INDEX *XV Function AA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 AA Command . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 Access to Saved Sheets . . . . . . . . . . . . . . . . . . . . . 71 Accumulators . . . . . . . . . . . . . . . . . . . . . . . . . . 29 ACINIT.PRM . . . . . . . . . . . . . . . . . . . . . . . . . 96, 97 AR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 AR Command . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 Arguments . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 Auto Motion Availability . . . . . . . . . . . . . . . . . . . . . . . . . 100 AVE Function . . . . . . . . . . . . . . . . . . . . . . . . 24, 86 AVG function . . . . . . . . . . . . . . . . . . . . . . 24, 25, 86 Boolean Functions . . . . . . . . . . . . . . . . . . . 24, 25, 86 C Command Calculation Order . . . . . . . . . . . . . . . . . . . . . . . 52 Case Conversion . . . . . . . . . . . . . . . . . . . . . . . . 22 Cell Addressing . . . . . . . . . . . . . . . . . . . . . . . . 95 Cell Commands . . . . . . . . . . . . . . . . . . . . . . . . . 31 Cell Names . . . . . . . . . . . . . . . . . . . . . . . . . 32, 87 Cell Precision . . . . . . . . . . . . . . . . . . . . . . . . . 77 Column Width Set . . . . . . . . . . . . . . . . . . . . . . . . 19 Command Arguments . . . . . . . . . . . . . . . . . . . . . . . 20 Command File Limitations . . . . . . . . . . . . . . . . . . . . 33 Command Files . . . . . . . . . . . 31, 32, 33, 49, 52, 61, 87, 89 COMMAND MACROS COMMANDS . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Comments . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 Conditional Loops Conditional Motion . . . . . . . . . . . . . . . . . . . . . 61, 89 Console Arguments . . . . . . . . . . . . . . . . . . . . . . . 52 Copy Current Position . . . . . . . . . . . . . . . . . . . . . . . . 7 Cursor Motion Data Base Access . . . . . . . . . . . . . . . . . . . . . . . . 32 DB Command . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 DF Command . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 Display . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Display Format . . . . . . . . . . . . . . . . . . . . . . . . . 14 Display Type . . . . . . . . . . . . . . . . . . . . . . . . . . 19 Displayed Entity . . . . . . . . . . . . . . . . . . . . . . . . 15 DL Command DS Command . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 AnalytiCalc Reference Manual Page 102 DT Command . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 DW Command . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 E Command . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 EDIT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 EDIT, Special Constructs . . . . . . . . . . . . . . . . . . . . 20 EMPTY CELLS ENTER Command . . . . . . . . . . . . . . . . . . . . . . . . . 21 Equation Solving . . . . . . . . . . . . . . . . . . . . . . 68, 91 Exit Extended Area . . . . . . . . . . . . . . . . . . . . . . . . . 95 Extract Saved Value F Command FFT FFTFW function . . . . . . . . . . . . . . . . . . . . . . . . . 98 FFTRV Function . . . . . . . . . . . . . . . . . . . . . . . . . 98 File Loading File Merging FORMAT . . . . . . . . . . . . . . . . . . . . . . . . . . . 5, 14 FORMAT Microtutorial . . . . . . . . . . . . . . . . . . . . . . 16 Formula Indirection . . . . . . . . . . . . . . . . . . . . . . 27 Formula Length . . . . . . . . . . . . . . . . . . . . . . . . . 33 FORMULA SEARCH . . . . . . . . . . . . . . . . . . . . . . . . . 48 Formula to Text . . . . . . . . . . . . . . . . . . . . . . 14, 22 Formula Value Access . . . . . . . . . . . . . . . . . . . . 70, 91 Formulas . . . . . . . . . . . . . . . . . . . . . . 22, 32, 33, 87 Fourier Transforms Frame Editing Freeze . . . . . . . . . . . . . . . . . . . . . . . . . . . 61, 88 Function Arguments Function Summary . . . . . . . . . . . . . . . . . . . . . . . . 86 Functions . . . . . . . . . . . . . . . . . . . 24, 29, 56, 60, 87 FVL (Future Value) function . . . . . . . . . . . . . . . . . . 86 FVL Function . . . . . . . . . . . . . . . . . . . . . . . . . . 24 G Command . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 Goal Seeking . . . . . . . . . . . . . . . . . . . . . . . . 68, 91 GOTO (Locate) Command . . . . . . . . . . . . . . . . . . . . . 38 Graphics . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 Help . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22, 37 Hexadecimal . . . . . . . . . . . . . . . . . . . . . . . . . . 58 Histograms IF conditional . . . . . . . . . . . . . . . . . . . . . 24, 25, 86 IF statement . . . . . . . . . . . . . . . . . . . . . . . . . . 29 Indirect Addressing Indirect Cell References . . . . . . . . . . . . . . . . . . . . 30 Indirect Command Execution . . . . . . . . . . . . . . . . . . . 70 AnalytiCalc Reference Manual Page 103 Initialization . . . . . . . . . . . . . . . . . . . . . . . . . 96 Input Region . . . . . . . . . . . . . . . . . . . . . . . . 72, 73 Insert or Delete Rows or Columns . . . . . . . . . . . . . . . . 11 Integers . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 Interactive Calculator . . . . . . . . . . . . . . . . . . . . . 50 Internal Loops . . . . . . . . . . . . . . . . . . . . . . . . . 51 IRR function (Internal Rate of Return) . . . . . . . . . 24, 25, 86 Journaling . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 Keypad Command Files . . . . . . . . . . . . . . . . . . . . . . 22 Keypad Commands . . . . . . . . . . . . . . . . . . . . . . . . 22 Keystroke Capture . . . . . . . . . . . . . . . . . . . . . . . 49 L Command . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 Line Fits . . . . . . . . . . . . . . . . . . . . . . . . . . . 98 LINEF function . . . . . . . . . . . . . . . . . . . . . . . . . 98 Linking Sheets . . . . . . . . . . . . . . . . . . . . . . . . . 34 LKP, LKN, LKE Lookup Functions . . . . . . . . . . . . . 24, 25, 86 Loading Saved Sheet . . . . . . . . . . . . . . . . . . . . . . 36 Locate current . . . . . . . . . . . . . . . . . . . . . . . . . 38 Lookups Looping . . . . . . . . . . . . . . . . . . . . . . . . 51, 61, 89 M Command Matrix Algebra . . . . . . . . . . . . . . . . . . . . . . . . . 66 Matrix restrictions . . . . . . . . . . . . . . . . . . . . . . 63 MAX function . . . . . . . . . . . . . . . . . . . . . . 24, 25, 86 MEMORY USE . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 MH Command . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 MIN function . . . . . . . . . . . . . . . . . . . . . . 24, 25, 86 Mixed Relocatable Names Mixed Relocation Override . . . . . . . . . . . . . . . . . . . 28 MOD Modulo function . . . . . . . . . . . . . . . . . . 24, 25, 86 Moving . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 MS command . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 Multiple Equations . . . . . . . . . . . . . . . . . . . . . 23, 30 New Features . . . . . . . . . . . . . . . . . . . . . . . . . . 95 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 NPV (Net Present Value) Function . . . . . . . . . . . . 24, 25, 86 NS No Scroll (Scroll Disable) . . . . . . . . . . . . . . . . . 10 Number Format . . . . . . . . . . . . . . . . . . . . . . . 52, 53 OA command . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 OAD command . . . . . . . . . . . . . . . . . . . . . . . . . . 40 Octal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 OR command . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 ORD command . . . . . . . . . . . . . . . . . . . . . . . . . . 40 Origin Control Commands . . . . . . . . . . . . . . . . . . . . 40 AnalytiCalc Reference Manual Page 104 Other Features . . . . . . . . . . . . . . . . . . . . . . . . . 75 Other Functions . . . . . . . . . . . . . . . . . . . . 30, 59, 60 Output Region . . . . . . . . . . . . . . . . . . . . . . . 72, 73 P Command . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 PARAMETER SUBSTITUTION PMT Function . . . . . . . . . . . . . . . . . . . . . . . . 24, 86 Position Independent Cell Names . . . . . . . . . . . . . . . . 33 Position Independent Names . . . . . . . . . . . . . . . . . . . 27 PPF command PPX Command . . . . . . . . . . . . . . . . . . . . . . . . . . 95 Prime Area . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 Printing Large Reports . . . . . . . . . . . . . . . . . . . . . 18 Project Management Random Numbers Ranges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 RB COMMAND . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 RECALCULATE . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Recalculation . . . . . . . . . . . . . . . . . . . . . . . . . 52 Redraw Command . . . . . . . . . . . . . . . . . . . . . . . . . 46 Relations . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 RELOCATE BOUNDARY . . . . . . . . . . . . . . . . . . . . . . . 45 Relocation . . . . . . . . . . . . . . . . . . . . . . . . . 11, 33 Replicating RND Function S Command . . . . . . . . . . . . . . . . . . . . . . . 6, 43, 95 SAVE . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6, 43 SC Scroll enable . . . . . . . . . . . . . . . . . . . . . . . . 10 Scatter Plots SCRATCH FILES . . . . . . . . . . . . . . . . . . . . . . . . . 6 Scrolling control . . . . . . . . . . . . . . . . . . . . . . . 10 Self Location . . . . . . . . . . . . . . . . . . . . . . . . . 71 Set Title . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Setting Number of Cols . . . . . . . . . . . . . . . . . . . . . 13 Setting Number of Rows . . . . . . . . . . . . . . . . . . . . . 13 Solving For Variables . . . . . . . . . . . . . . . . . . . 68, 91 Sorting . . . . . . . . . . . . . . . . . . . . . . . . 18, 70, 91 Sorting, by ASCII . . . . . . . . . . . . . . . . . . . . . 70, 91 Special Accumulators . . . . . . . . . . . . . . . . . . . . . . 55 Special Cell Addressing . . . . . . . . . . . . . . . . . . . . 27 Startup control . . . . . . . . . . . . . . . . . . . . . . . . 97 STD (Std Deviation) function . . . . . . . . . . . . . . 24, 25, 86 Storage Allocation . . . . . . . . . . . . . . . . . . . . . . . 82 String Functions . . . . . . . . . . . . . . . . . . . . . . 72, 73 Subprocess Execution . . . . . . . . . . . . . . . . . . . . . . 50 SUM function . . . . . . . . . . . . . . . . . . . . . . 24, 25, 86 Symbolic Names . . . . . . . . . . . . . . . . . . . . . . . . . 84 AnalytiCalc Reference Manual Page 105 TE Command Temp. Escape to Monitor . . . . . . . . . . . . . . . . . . . . 50 Temporary Motion Temporary Move . . . . . . . . . . . . . . . . . . . . . . . 61, 88 TEst Compute Text Entry . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 Text to Formula . . . . . . . . . . . . . . . . . . . . . . 14, 22 TITLE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Value Freeze Variable Names . . . . . . . . . . . . . . . . . . . . . . . . . 27 VARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68, 91 VIEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 View Command . . . . . . . . . . . . . . . . . . . . . . . . . . 46 View Control . . . . . . . . . . . . . . . . . . . . . . . . . . 60 W Command . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 Weekdays Functions . . . . . . . . . . . . . . . . . . . . . . . 98 Where Matrices Can Be . . . . . . . . . . . . . . . . . . . . . 63 Windows . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 Word Processing . . . . . . . . . . . . . . . . . . . . . . . . 35 Workfiles . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 Writing Screen to File or Print . . . . . . . . . . . . . . . . 47 X Command ZAP Sheet Command . . . . . . . . . . . . . . . . . . . . . . . 48 ZEro Command . . . . . . . . . . . . . . . . . . . . . . . . . . 48