Renga in Blue   Interactive fiction and the All the Adventures project

Via an old eBay auction.

The obsession to find the first instance of something can occasionally make sense. We can trace patterns that stem all the way to the beginning. For example, the first Sierra game (Mystery House) established a penchant for instant-death that became nearly a Sierra trademark for a decade and a half, and the attitude was one that their competitor Lucasarts specifically was in opposition to.

As a rule, adventure games should be able to be played from beginning to end without “dying” or saving the game if the player is very careful and very observant. It is bad design to put puzzles and situations into a game that require a player to die in order to learn what not to do next time.
— Ron Gilbert, Why Adventure Games Suck And What We Can Do About It (1989)

Even in the case where the first instance of something was never followed as an example (due to obscurity of the work itself, the idea being too far ahead of its time, or random luck), it’s interesting to see as a proto-concept free from outside influence. A good example of this would be how Mystery Fun House manages participatory comedy even when the text is typical 1979-level sparseness.

Alternately, we can see early ideas that die as object lessons, getting a good notion of why nobody desired to copy a particular concept and if those conditions still hold now. The lack of compass directions in Empire of the Over-Mind come to mind; we can see the concept doesn’t mesh well with large-scale maps, and this also suggests that a map without much traversal (like some modern interactive fiction) would fare a lot better.

However, I don’t think it’s useful to think of being first as some sort of trophy, a historical totem to claim person Q’s biography is superior to person R’s, or to insinuate group X is better than group Y. First, this suggests history itself is some sort of competition. Also, convergent evolution can lead to entirely separate people coming up with the same thing (see how Wander did adventures before Adventure) and just because something made it to market in July rather than August doesn’t mean it is a superior. In fact, due to the factors of obscurity/luck the second or third to arrive at an idea can be much more influential than the first.

This is my rather long-winded way of introducing Atlantean Odyssey as likely the first full graphical adventure game, ahead of Mystery House.

Some caveats:

“Likely the first”: We have the source code above, solely credited to Teri Li (which includes lots of POKE statements to make the graphics, basically making assembly language in BASIC source) although I haven’t been able to find it published anywhere except The Captain 80 Book of Basic Adventures which came two years later. Based on the history of how Spider Mountain was written, there is strong reason to suspect this version had a 1979 release in stores. However, the only extent tape copy I have been able to find (see eBay picture above) is a machine language version released a year later which adds Mark Robinson to the credits. It’s still possible with a 1980 release it came first, but I haven’t been able to dig up any records.

“Full”: One could reasonably point out that earlier mainframe work like Zork and Stuga had the occasional ASCII art, which arguably counts as “graphics.” However, the use was intermittent; it’s not like every room had an illustration.

In any case, with the fussy details out of the way, how does Atlantean Odyssey play?

The goal, as usual, is to find all the treasures (6 of them).

The very first room has a sailboat you can board with a knapsack, a speargun, a flashlight, and scuba gear. All of these items are entirely unnecessary. The scuba gear does let you dive underwater, but the capacity runs out fast enough I believe it won’t last the whole game (there’s a magical solution the game gives you right away that lets you avoid the issue). The flashlight is rusted and broken. The speargun can be used on a nearby shark but the shark just kicks the player to “Davy Jones Locker” (aka death).

Nearby you find a medallion:

A FLAT HEXAGON WITH A RUBY-SAPPHIRE-BLUE DIAMOND AROUND A GREEN OPAL

and a hint on a temple wall:

I SEE A BAS-RELIEF SCULPTURE OF A MAN WEARING A MEDALLION. HE IS UNDERWATER, AND THE MEDALLIONS RUBY IS GLOWING!

By typing “PUSH RUBY” you are able to dive underwater, finding a second temple.

There’s another mural, which hints that “PUSH OPAL” will be useful; it takes you to another portion of the ocean where the game continues.

I’m not going to walk through every event that follows (it’s straightforward looking at everything in the environment, and further use of the medallion) but I did want to point out the main difference between this game and Mystery House: the game is entirely self-sustaining without the graphics (there’s even a later C64 version which doesn’t have graphics). Items are described in the text, rather than drawn in the room, so you don’t have the Mystery House situation of trying to guess what sort of object a squiggle is indicating. When there’s a door, the “open door” and “closed door” variations of the room look exactly the same.

Atlantean Odyssey’s method became the mainstream for text adventures; Infocom, Magnetic Scrolls, and Legend games generally had all the necessary content in the text, and even allowed for turning the graphics off if desired. (Zork Zero is the only exception from those companies I can think of, and that’s simply on particular puzzles like a peg-jumping one where graphics are needed.)

Mystery House’s method, on the other hand, led directly to the thread of interactivity directly dealing with with graphics instead of text, resulting in 1984’s King’s Quest I and all those games that followed.

My usual online-Javascript site doesn’t do well with this game, so if you’d like to try it out, you’ll need an emulator and a download. It’s a little less ambitious / revolutionary than Mystery House and consequently a little more playable.