For those unaware, Warren Robinett has been beavering away on a book about his masterpiece for the Atari 2600, the game simply titled Adventure. According to his website the e-book should be released this year. Last September I sent him an e-mail to have my name placed on the mailing list for the eventual announcement of the book's completion, and I couldn't resist adding some "fan mail" stuff to the message. Mr. Robinett was kind enough to send me a reply, some of which I will now share as I (and hopefully many other Adventure fans) eagerly await his first-hand account of the genesis of one of the most popular video games ever made. I hope these excerpts will serve as a sort of preview of the book (or perhaps a mini-interview). I gather he is putting the book together on the side as he stays busy with professional engineering work, so patience with the process has been my way of looking at it.
EBM: "Dear Mr. Robinett, I anxiously await your e-book The Annotated Adventure. I like your idea of having the C and assembly versions run parallel on the pages, for comparison's sake. Most important to me is your commentary: I have enjoyed your various writings and lectures concerning the game and I'm eager to learn more.
One of my favorite aspects of the game is the translational symmetry of two rooms in the grey dungeon (including the dot room), contrasted with the reflectional symmetry of the majority of rooms in the kingdom. An interesting choice... I wonder if there was a reason behind it.
I also enjoy exploring the way the rooms align for all the non-player objects as they travel unhindered by walls. It is fun to discover how these paths devolve into loops, and to map these paths using different methods: graphical, symbolic, numerical, etc. Starting in any given room, and determining which rooms lie along a path of travel in one compass direction, reveals many interesting aspects of the kingdom's layout. As of course you are aware of, some paths of travel are revealed to go from screen to screen in a "circular" pattern i.e. 01-02-03-01-02-03, whereas other paths follow an initially linear pattern that falls into a subordinate "circular" pattern i.e. 01-02-03-04-05-03-04-05. Some paths are lengthy, while others (originating in the gold castle room and the number room [game select screen]) are abbreviated greatly. What I am most eager to learn from you and your e-book is how you made decisions regarding the manner in which the rooms communicate in an architectural sense (ignoring the walls, of course).
Some of the reasoning seems easy for me to suss out, but some of it eludes me. Limiting access to the castle rooms is sensible for gameplay reasons. The horizontal loop of the main hallway directly south of the gold castle seems intuitive. Many areas of the kingdom constitute somewhat self-contained realms like this, lending a sense of place to a potentially bewildering layout. But there are some long routes, like moving south from the black castle through the blue labyrinth which ultimately results in a small "circular" route that doesn't revisit most of the screens that preceded it. Was every path like these strategically planned, or are there any "accidental" results that follow from other layout decisions in a natural way?"
WR: "One thing I can tell you is that you have analyzed this topic more deeply than I did when making the game. I did make the castle gates the only way in and out of the castle interior regions. And I did try to make regions like the Blue Maze mostly link to themselves (but there had to be at least 2 exits from the Blue Maze, since it stood between the Yellow Castle and the Black Castle). Beyond that, every room had to have 4 links that went somewhere, because a Dragon or the Bat was going to sooner or later cross every edge of every room. If there was an apparent problem, I fixed it. But beyond the foregoing stuff, the precise topology of the game world just sort of evolved, as I added new regions during development.
Regarding the 2 rooms that did not have reflectional symmetry in the Catacombs (dark maze) inside the Black Castle, I had an attribute bit in my room-list data structure that controlled which Playfield symmetry was used. I had never used anything but reflectional symmetry up to the point I added this attribute bit to the data structure. I didn't really need the one you call "translational symmetry". But the Atari VCS platform didn't give you much to work with. I was trying everything I could think of to make the game more interesting. It was really quite pitifully boring in its earliest stages. So I added the alternate symmetry in a dark maze so that it would not be instantly obvious that I had the broken the symmetry — it would slowly dawn on the player that those two rooms were different. So my reason was not all that deep. I was just grabbing some low-hanging fruit to add just another little piece of variety to the game."
I believe the forthcoming book has been discussed elsewhere in the forums, but I am writing this to remind everyone about the book and to encourage those who are interested to sign up to Mr. Robinett's mailing list and confirm interest in the project. As I said, it's not like it can be rushed (nor would I want it to be) but making it known that interest is there will no doubt help the work see the light of day in good time.
Edited by The Eyeball Mural, Tue May 10, 2016 8:44 PM.