Atari users have a surprisingly wide
selection of programming languages from which to choose. We've got three
dialects of BASIC, four C compilers, eight or nine FORTHS, a pair of
Pascals, PILOT, Logo, WSFN, a Lisp interpreter, numerous 6502 assemblers
and a couple of hybrids like BASM and Mirth. Not bad for a "game
Leave it to Optimized Systems
Software to come up with yet another way to tell your Atari what to do.
OSS has been the leading purveyor of alternative operating systems and
languages for the Atari since before I can remember. Action!
is only the first of a whole new line of OSS products that's been causing
quite a stir in the Atari underground. It's been touted as the first
programming environment developed specifically for the 6502, and the
fastest high-level language available for the Atari. These are pretty
strong claims which, after playing with the system for several weeks,
appear to be totally justified. As you are about to read.
In syntax and overall structure, Action!
bears a strong resemblance to Pascal, C and other members of the Algol
family. It's a procedure-oriented language featuring global and local
variables, user-definable functions, parameter passing and powerful
structures like DO loops, FOR-TO, WHILE, UNTIL and IF-THEN-ELSE. Three
basic data types are recognized: 8-bit BYTEs (or CHARacters), 16-bit
signed INTegers and 16-bit unsigned CARDinals. The system also supports a
variety of extended data types including pointers, subscripted arrays,
strings and records.
Listing 1: Reserved Keywords
1 includes all of the keywords reserved for use by the Action!
system. These are used to declare variables, define new procedures and/or
functions and to control the operation of the compiler. BASIC veterans
will note with alarm the total lack of keywords that do interesting things
in and of themselves, like SETCOLOR or DRAWTO. They're missing for a very
good reason. Unlike BASIC, Action! does not
limit your programming to a limited number of safe little commands. It
invites you (indeed, forces you) to invent the commands you need to solve
problems yourself. The keywords in Listing
1 are the tools the system gives you to, in effect, write your own
language. If this prospect doesn't excite you, maybe BASIC has been
holding your hand for too long.
Listing 2: Library Procedures
Don't get the impression that Action!
leaves you completely on your own, though. The cartridge includes a
library of useful I/O, graphics and system-level routines that you can use
to start building more elaborate programs. Listings
2 and 3 will give you an idea
of what's available. The resemblance of many Action!
library words to Atari BASIC commands is intentional; the kindly folks at
OSS want to make your transition from BASIC to Action!
as painless as possible. This concern for familiarity unfortunately
extends to the Action! graphics library,
which offers exactly the same (limited) access to the hardware as Atari
BASIC. Other weak points of the cartridge library include inadequate
control over memory allocation and a mysterious lack of support for the
Atari's built-in floating point math package.
Listing 3: Library Functions
Most of the elements in an Action!
program are delimited by space characters - as many as you like! You don't
have to keep track of line numbers, semicolons, brackets or any other
nuisances that can make you feet more like a bookkeeper than a programmer.
Just follow a few simple rules regarding commas and parentheses, and
you're all set. Action!'s modern design
encourages a wide-open style of program composition, with plenty of
freedom regarding the use of blank lines, upper and lower-case characters,
indentation, comments and other flourishes that improve readability and
make coding more fun.
A four-part system.
Internally, the Action!
system consists of four distinct modules. There's an editor for creating
and modifying program source text, a compiler which translates source text
into executable machine code, a run-time library that supports the
compiled code (described above), and a monitor which acts as a switchboard
between the other three modules and (if you're using a disk drive) DOS.
A very important distinction between
Action! and every other compiled language
for the Atari is that these modules do not have to be loaded in separately
from disk. All four are tucked away inside the SuperCartridge, safe from
accidental erasure and ready whenever you need them. Further, the system
is arranged so that your source text and compiled code can reside in
memory at the same time. This self-contained design combines the
performance of a compiled language with a degree of interactiveness
usually associated with an interpreter. A stroll through the modules will
show you what I mean.
Somebody at OSS once told me that
the text editor in the Action! cartridge was
originally going to be marketed by itself as a word processor. It isn't
hard to believe. There are so many features and options in the Action!
editor that I can only touch on the most interesting here.
editor uses your TV as a virtual window into a text area that can extend
well beyond the edges of the screen. Unlike the standard Atari screen
editor, you can type up to 240 characters on a single line with no cursor
wraparound. How? When your cursor reaches the right edge of the screen,
the line you're working on (and only that line) starts to coarse-scroll to
the left. You can keep right on typing until a buzzer informs you that
you've reached the rightmost position in that line - the right
"edge" of the text window. Move your cursor back towards the
left, and the line scrolls to the right until you hit the left edge of the
window. This design neatly eliminates the usual confusion between
"logical" and "physical" lines of text.
or "<" instantly moves you to the rightmost or leftmost
character in the current line, respectively. You can also change the
maximum width of the text window to any convenient value, such as how many
characters will fit on your printer.
editor allows you to create a second text window, co-resident in memory
but otherwise completely independent from the main window. The 2-window
editing mode is represented visually by a split screen, with the bottom
half of the image devoted to the auxiliary window. You can jump back and
forth between the two windows and transfer blocks of text if desired; the
editor remembers where you were working in each window and automatically
returns you to that point when you return. Additionally, you can save,
load or delete text in one window without disturbing the contents of the
other. That means, for example, that you could load a library of routines
into the auxiliary window, review them and copy the ones you need into
your main program, which has been in full view the whole time! Sure beats
LISTing and ENTERing lines of BASIC, doesn't it?
Other noteworthy capabilities of the
Action! editor include global search and
replace, instant access to the beginning or end of a file and the ability
to delete, move and copy, selected blocks of text. The block move and copy
functions are implemented so nicely that I have to tell you about them.
When you hit the SHIFT/DELETE keys, the line you're working on disappears,
just as with the Atari screen editor. But the line isn't gone forever.
It's being held in a buffer, waiting to be moved or copied to anywhere
else in your text window(s). Simply move the cursor to a likely spot and
hit CTRL/SHIFT/"P" (for paste) to dump the contents of the
buffer. Several adjacent lines of text can be sent to the buffer by
repeatedly "deleting" them with SHIFT/DELETE. Action!'s
method of picking up and dropping blocks of text feels very natural if
you're used to the Atari screen editor, and it also eliminates the
annoyance of losing a line of work by accidentally hitting SHIFT/DELETE.
Incidentally, you can automatically undo any changes you have made to a
line of text by hitting CTRL/SHIFT/"U".before leaving the fine.
Before you toss out your AtariWriter
cartridge, let me point out a couple of small but irritating problems in
the Action! editor. There's a feature called
tagging which allows you to mark any location in your text by assigning it
a unique one-character identifier. You can later return to that point in
the text at any time by calling its ID code. It's a good idea that,
unfortunately, isn't pulled off particularly well. If you set a tag in a
line and change even a single character in that line, the tag disappears.
This restriction (which is documented) considerably reduces the usefulness
of the tagging option, to say the least.
My other gripe is with the way the
cursor appears to flash and jump around the screen when it is being moved
up or down, as if it isn't sure where to go next. The solid command line
on the bottom of the screen also seems to jerk occasionally as you cursor
around. Minor cosmetic points, perhaps, but an unstable cursor seems out
of place in this otherwise superb little text editor.
After you've put the finishing
touches on an Action! program and saved it
out to disk, what next? Press the CTRL/SHIFT/"M" keys
simultaneously and you'll find yourself staring at a barren white bar
across the top of your screen. This is Action!'s
monitor, the central interface between the editor, compiler, machine and
Monitor functions are invoked by
typing a one-character code letter. You can select various compilation
options, save and load compiled programs, examine the values of variables
and memory locations and trace the execution of your programs. You can
even use the X (execute) directive to interactively test almost any
procedure or function. This capability is very unusual (and useful) in a
compiled programming language.
Unlike Atari BASIC, which compiles
each line of program text as it is typed, Action!
requires that your program be explicitly translated into machine code
before it can be executed. This isn't nearly as formidable as it sounds.
All you have to do is type the letter C from within the Action!
The compiler accepts source text
from either the editor (default), or from a text file saved onto cassette
or disk. If you've been using both text windows, Action!
will compile only the text in the window you last edited. Compilation is
almost unbelievably rapid, especially when the source is the editor. I've
never seen Action! take more than a few
seconds to compile even a fairly large program that was in the editor.
Small programs are compiled before you take your finger off the RETURN
key. You can optionally instruct the compiler to list each line of source
text to the screen or a printer as it is being compiled. This slows the
compilation considerably, however.
A compile error causes the system to
display the line where the error occurred, along with an error message
number. Surprisingly for an OSS product, there are no English error
messages. If you re-enter the editor after a compile error, you'll find
the cursor obligingly positioned over the questionable spot in your text.
Successfully compiled code is
executed by typing the letter R (run) from within the Action!
monitor. If you're accustomed to the leisurely pace of Atari BASIC, get
ready for a shock. OSS isn't kidding when they say Action!
How fast is fast?
Execution speed is very important to
Atari programmers. Why? Because much of the software written for the Atari
relies heavily on graphics, where a few extra machine cycles in the wrong
place can make the difference. between a spectacular special effect and an
interesting but unmarketable demo. High speed isn't likely to hurt a
non-gaphics program, either. This is in accordance with Moriarty's Maxim:
It is much easier to slow down a computer program than it is to speed it
A number of attempts have been made
to devise a universal method for comparing the speed performance of
computer languages and hardware. In September of 1981, Byte magazine
published an iterative number-crunching algorithm called the Sieve of
Eratosthenes, which calculates all of the 1,899 prime numbers between 3
and 16,384.* The Sieve has since become the informal industry standard for
clocking the speed of microcomputer languages.
*Jim Gilbreath, "A High-Level
Language Benchmark." Byte, VI, 9
(September 1981), pp. 180-198.
4 is an implementation of the Sieve in Atari BASIC. It requires 19,490
jiffies or approximately 5 1/2 minutes to execute on an unmodified 48K
Atari 800 system. I recognize that Listing
4 is not the most efficient way to write the Sieve in Atari BASIC, but
it is the clearest and most portable way, and that's what counts in this
application. You might like to try rewriting the Sieve for better speed
performance. I've achieved improvements of better than 30% with tricky
10 DATA 941,347,921,5,772,308,90,393,3
25 DATA 623,689,395,581,796,17,419,352
Although I love standards, I don't
like the Sieve. It's not easy for beginners to understand, it takes too
long (in BASIC, anyway), and it doesn't test the Atari under real-world
conditions, with lots of 6502 processor time being "stolen" by
Antic for video DMA. I wanted a benchmark that anybody could appreciate,
operating under the kind of DMA conditions an Atari program is likely to
find itself up against.
Back in Issue 11, I devised a little
program that fills a GRAPHICS 24 screen with color, one byte (eight
pixels) at a time. It was used to compare a couple of BASIC compilers at
the time, but it's equally valid in any run-time environment. My
definitive BASIC implementation of this test appears in Listing
5. Screen Fill, as the program shall henceforth be known, executes in
4025 jiffies or about 67 seconds on a 48K 800. (Again, improvements are
possible, but for the sake of clarity let's stick to Listing
5.) I'll be using Screen Fill in conjunction with the Sieve to judge
the performance of every new language I review from now on. So let it be
written; so let it be done.
10 REM * SCREEN-FILL BENCHMARK
11 GRAPHICS 24
12 POKE 19,0:POKE 20,0
14 FOR I=0 TO 31
15 FOR J=0 TO 239
16 POKE SCREEN+J,255
17 NEXT J
19 NEXT I
21 GRAPHICS 0
22 PRINT TIME;" JIFFIES"
10 DATA 206,5,2,185,233,103,695,394,78
OSS includes a implementation of the
Sieve benchmark in their Action!
documentation. I rewrote the code slightly to make it match my BASIC
implementation more closely; the modified program is shown in Listing
6. It executes in 89 jiffies or just under a second and a half. I'll
save you a calculation by pointing out that the Sieve runs about 219 times
faster in Action! than it does in Atari
BYTE RTCLOK=20, ; addr of sys timer
SDMCTL=559 ; DMA control
BYTE ARRAY FLAGS(8190)
SDMCTL=0 ; shut off Antic
RTCLOK=0 ; only one timer needed
COUNT=0 ; init count
FOR I=0 TO 8190 ; and flags
FOR I=0 TO 8190
IF FLAGS(I)='T THEN
TIME=RTCLOK ; get timer reading
SDMCTL=34 ; restore screen
PRINTF("%E %U PRIMES IN",COUNT)
PRINTF("%E %U JIFFIES",TIME)
7 is an Action! implementation of
Screen-Fill. This demanding little gem executes in 32 jiffies (slightly
more than half a second), or 126 times faster than its BASIC counterpart
under maximum DMA handicap. And if you cheat by replacing ing the nested
FOR-TO loops with an Action!, SETBLOCK
procedure in the form:
you'll obtain an execution time of
just five jiffies. This is essentially the same amount of time it takes
the equivalent machine-language code to do the same job. No other
high-level Atari language that I am aware of can match this kind of speed
BYTE RTCLOK=20, ; addr of sys timer
SAVMSCL=88, ; lsb of screen addr
SAVMSCH=89, ; msb
I,J,TIME ; declare variables
FOR I=0 TO 31
FOR J=0 TO 239
PRINTF("%E %U JIFFIES",TIME)
Pulling the wings off a butterfly.
Once I got a taste of Action!'s
dizzying speed, I had to find out what was going on inside that demonic
little cartridge. So I used the W (write object code) option of the Action!
monitor to send a copy of the compiled Screen-Fill benchmark to a disk
file. Then I read it back into Ralph Jones' Ultra Disassembler (published
by Adventure International), massaged the labels and commented the code to
make it correspond to the Action! source
text, line by line. The result appears in Listing
Assembly programmers will appreciate
the extraordinary efficiency of the Action!
compiler. The code in Listing 8 is
totally non-recursive. It uses no special stacks or indirect pointers to
control the flow of execution, just pure in-line machine code with an
occasional JSR into a cartridge library routine. This is "native
mode" compilation at its best: simple, clean, and very, very swift.
The output of a typical C or Pascal compiler looks like spaghetti by
Because compiled Action!
programs refer to subroutines that reside inside the Action!
cartridge, you can't run a program without the cartridge in place. This
may come as a disappointment to users who want to give copies of their
latest Action! game to friends who don't
have Action! OSS plans to remedy this
situation by offering a Personal Run-Time Package to licensed Action!
users for around $30. It's a utility that will let you turn any Action!
program into a self-standing entity that will run with no help at all from
the Action! cartridge, thank you. A
commercial run-time package will also be offered for a one-time licensing
fee of approximately $300. Both may be available by the time you read
this; contact OSS directly for more information.
Another $30 will get you OSS's
programmer's Aid Disk (PAD), a collection of demonstration programs and
library routines that wouldn't fit into the already crowded Action!
cartridge. The libraries include badly- needed support for player/missile
graphics, memory management and floating point math, precisely the
weaknesses I noted above. The demo programs are very instructive and help
to clarify some of the obscure features of the language. You even get a
full-blown game program, written in Action!
by our very own Joel Gluck.
The PAD squarely addresses many of
the shortcomings of the Action! cartridge
and documentation, and is an absolute must for all serious owners of the Action!
system. In fact, this material ought to be included with every new system
sold, even if it means bumping up the price a bit.
You can bank on it.
The 16K Action!
"SuperCartridge" is a technically interesting device in and of
itself. It employs a hardware technique called bank-selecting to make
itself "look" like an 8K cartridge. This gives you access to the
8K of RAM between $8000-$9FFF that is de-selected and thus rendered
useless by a conventional 16K cartridge, such as AtariWriter.
The bottom half of the
SuperCartridge ($AOOO- $AFFF) is divided into three independently
addressable 4K banks of ROM, which are automatically switched in and out
depending on what part of the system is in use. If your Atari has 48K or
more memory, it's even possible to address the 4K bank of RAM that resides
"under" this half of the cartridge. OSS's new DOS XL operating
system takes advantage of this capacity in a most ingenious manner. Look
for a report in a future issue.
The bank-select cartridge is a
nearly ideal home for Atari software. It gives the cartridge designer a
full 16K to work with, enough room for plenty of bells and whistles. It
gives the user an instant-loading, highly reliable environment with up to
40K of workspace. And because three of the memory banks occupy the same 4K
address range, a bank-select cartridge is very difficult to pirate. Let's
hope that more manufacturers start taking advantage of bank-selecting to
enhance the value and security of their products.
Advice and admiration.
I'm sorry to report that the Action!
Reference Manual doesn't do the language justice. In a commendable
attempt to satisfy beginners and experts alike, the Manual suffers from
lack of confidence, uncertain organization and a shortage of good, hard
technical data. Thank goodness for the numerous sample programs, which
communicate a lot more about the system than the text surrounding them.
Having once written the manual for a
new (and mercifully obscure) programming language, I can appreciate the
difficulties involved in deciding how much needs to be said, to whom, and
in what order. Nevertheless, a new language can only be as good as its
documentation. Until somebody sits down, rolls up his or her sleeves and
writes an authoritative book about Action!,
it will have a hard time attaining the wide acceptance it so obviously
deserves. I conclude this diatribe by acknowledging that the latest
edition of the Reference Manual (in the small yellow
notebook) shows a marked improvement over the first release.
cartridge itself has gone through a couple of changes since its first
appearance in August 1983. You can tell which version you have by using
the "?" (display memory) command in the monitor to examine
cartridge address $BOOO. If this byte equals $31 hex, you have the
original Version 3.1. A value of $33 indicates Version 3.3, in which a
number of minor 3.1 bugs have been corrected. The final version is 3.6
($36 at $BOOO), which should be ready soon after you read this. OSS has
always been very good about maintaining their products, so you shouldn't
have any trouble getting an upgrade if you need one. Consult OSS for
prices and availability.
I hope my kvetching about the
documentation doesn't scare you away. if sensible, structured code and
edge-of-the-art speed are what you crave in a high-level language, Action!
is exactly what you need. OSS's hideous orange cartridge joins the ranks
of valFORTH, Omnimon!,
ABC and MAC/65
as one of the most valuable development tools ever published for the
Atari. Congratulations and thanks to Clint Parker and OSS for bringing us
such an advanced product. You can expect to see plenty of support for this
exciting new language in future issues of ANALOG.