Interview with Don Worth about Beneath Apple Manor

Raw text from my interview with Don Worth, the creator of Beneath Apple Manor.

By Joachim Froholt

This is the raw text from an interview I did back in 2014 with Don Worth, creator of what was (probably) the first «roguelike»-game, Beneath Apple Manor. I decided to publish it all here so that it can be of use to researchers, historians and others interested in the history of computer games.

Unfortunately I don’t have access to my original correspondence, so this is a bit random and most of the time my questions are missing. The photos and scans were all supplied by Don Worth.


Printout from FRON. Image credit: Don Worth.
Printout from FRON. Image credit: Don Worth.

At the time I was a systems programmer (software engineer/systems administrator in modern terminology) for UCLA on its big IBM mainframe. I had written some games on the mainframe that were written in IBM 360 assembler language. Most notably a multi-player space game that involved exploration of a random map of stars. Players could only see spaces where their ships were, so it was partial information – meaning things could change in various parts of the map but you only saw the areas you occupied. Beneath Apple Manor built on this concept for a single-player game by randomly generating the dungeon levels and only allowing you to see the area next to your character. Monsters could be moving all over the place outside your immediate vicinity and you wouldn’t know it.

I was an avid Dungeons and Dragons player from about 1976 with other programmers at UCLA. So when the first Apple IIs came out I immediately bought one and the first thing I thought to do was to write a simulation of the D&D rules. I made a list of the actions you could have in D&D (eg. listening at doors, bashing down doors, opening treasure chests, etc.) and implemented them one by one in BASIC. I wrote some things in assembler to speed things up (eg. the display updates, sounds, etc.)

When it came to marketing the game, there wasn’t much available for the Apple at that time – just a Star Wars shoot-em-up and a port of the old Star Trek textual game. So I took my game around to the computer stores in the Los Angeles area and gave them a copy for free and asked them to let me know how many copies they wanted. There were quite a few orders from doing that. Then we sent out copies to stores that advertised in Creative Computing and lots more orders rolled in. After that, the burden of producing the product (I was hand assembling the little books and making the diskettes and tapes myself) and managing the sales got to be too much since I kept my day-job at UCLA. So I took the product to Quality Software who sold BAM from then on. Later I did an updated version of the game with better graphics for Quality Software.

Don Worth circaa 1975. Image supplied by Don Worth.
Don Worth circaa 1975. Image supplied by Don Worth.

Why make an RPG?

We played D&D a lot during lunch hour (which sometimes stretched into more than one hour). My boss called me into his office once and reamed me out for playing too much «Goblins and Ghosts» as he called it. LOL I loved D&D – you could use your imagination to visualize things (sort of like listening to an old-time radio drama) and you could interact with the dungeon master’s environment and everyone got to contribute to embellishing the story. I didn’t much care for the complicated rules and tables you had to use – so I thought maybe I could hide all the complexity with a computer simulation.

Does your earlier space game exist today?

No. I think the backup we had for it was on a 9-track mainframe tape. Long gone. I do have a file folder with some of the print-out maps from the game and the rules. It was very elaborate. We called it «FRON», a combination of my name and my friend Frank Wood’s name who wrote the user interface. I did all the internal game logic.

How would you describe the community around games on these big mainframe computers?

FRON. Image credit: Don Worth.
FRON. Image credit: Don Worth.

I wouldn’t say there was a whole lot of «community» w/re to mainframe games. Some of them surfaced on the Arpanet (later to be called the Internet) in the mid 1970s. In particular I remember playing Start Trek (the text-based game) on TI Silent 700 printer terminals that had a roll of thermal paper over a dialup modem (the kind you plugged the telephone handset into cups for the ear and mouthpieces.) There was also the colossal cave ADVENT text game (a genre that went on to fame with Roberta Williams of Sierra Online). But mostly I didn’t know who wrote those games and we never communicated about them. This was long before there was email or bulletin boards. Each of us was pretty much in isolation.

So most of my game development community was the other programmers at UCLA. We did programs that created huge mazes on the plotter and then held contests to see who could solve the maze fastest (dragonmaze on steroids). And I wrote a Kalah (Owari) game where the computer played against you by examining every possible counter move to your moves going 3 or 4 moves ahead. I couldn’t beat it. There were probably a lot more I don’t remember.

I would say that several of the programmers there loved games. Some board games (Rail Baron, Triplanetary, etc.) as well as role playing games like D&D. We had a few games on the mainframe (like Kalah) that people could play alone, but playing together was preferred.

Well, it wasn’t all that different than playing on a screen. You would type in a move and it would print out what would be on a screen. Sector maps in the Star Trek game for example. And for text games like ADVENT it was natural. You type in «pick up wand» and it says «wand picked up» Of course you had a lot of paper to throw away after a session! lol Computers didn’t really have graphics back then and the first joystick I saw was in a UCLA engineering research lab. (There was a professor who was examining the effects of marijuana on the brain and had his subjects play a very early version of Asteroids on a vector graphic display that looked like an oscilliscope with a joystick. We thought that game was very cool! I had never seen anything like it before that. Was probably in the mid 1970s.)

I think one of the graduate assistants in the lab might have written that game. I don’t remember much about it anymore – it was a raster graphics display – not pixels. You had a little space ship that could fly around on the screen and shoot at another space ship.

Original notes for Underground Adventure (Beneath Apple Manor). Image credit: Don Worth.
Original notes for Underground Adventure (Beneath Apple Manor). Image credit: Don Worth.

And then later on, what was the community like around the first home computers?

Again, it was a bit early for bboards and such. I knew of Bill Budge and people like that but never had occasion to run into them. My publisher (Quality Software) knew their competitors personally I think. But I was an independent software developer and book author, so I didn’t get out that much. I attended user group meetings, which I think the other developers did too, but we were in different areas. I also knew a lot of the employees at local computer stores (which is how I met my co-author Pieter Lechner – he worked in a Lawndale, CA computer store that I frequented.)

I would say so, yes. I taught a 6502 assembler class at the local store (see attached flyer) and met a lot of people that way. I know a lot of kids would hang out in the computer stores after school and play the demo games. Most of the people my age were working full-time, so I didn’t spend much time at the stores myself.

Yes, I think that was the point of having an Apple II in the first place. It was a hobbyist’s machine. So you were expected to use it to learn how to program I think. Later, when they put Visacalc on it it became more of an appliance for a lot of people.

Steve Wozniak / Beneath Apple DOS

Don Worth with his game. Image supplied by Don Worth.
Don Worth with his game. Image supplied by Don Worth.

No, it had nothing to do with Beneath Apple Manor – he was contacting me about my disassembly of DOS. I had disassembled and annotated all of DOS and had copies made and comb bound and mailed one up to Apple. He called me worried that I would publish their source code and imperil their copyright to DOS. I assured him I wouldn’t do that. So he was grateful. And later, when I had published Beneath Apple DOS, he thought it was a great help to the flourishing developer community and helped us get access to new Apple versions to test. (See attached)

What was it like to get a home computer for the first time?

I thought it was my dream come true. I had been programming on the mainframe so long, I never thought I would have my very own computer. The first thing I wanted to do was reverse engineer all the software to figure out how it worked. That’s why I disassembled DOS (and then wrote Beneath Apple DOS). The Apple was so cool because they documented so much of it. It wasn’t a «black box» like the Commodore or Atari.

Could you tell us a little about the challenges of making a game for such an early platform?

Well, since I was using BASIC, the only challenge was squeezing as much game as I wanted to write into such a small amount of memory. And sometimes, when BASIC was too slow, I would have to rewrite parts of it in Assembler to speed things up (eg. screen updates). Saving and loading from tapes really was a pain – it would only work some of the time and you had to fiddle with the volume and tone controls on the tape player a lot. And I didn’t have a printer yet so I had to write out all my listings on paper and when I changed something in the program, I would have to cross-out and update the handwritten «listing».

Source code on paper. Image credit: Don Worth.
Source code on paper. Image credit: Don Worth.

Well, now that you mention it, I guess so. And it allowed me to look at the whole game at once (there weren’t a lot of rows on the TV screen at any one time.) I was used to having printouts («listings» as we called them on mainframes) of big programs. Also, it helped me with my design process to write down ideas. See attached pages. (Before I had a name for the game it was «Underground Adventure»)

I think I had 20 or 30 yellow pad sheets with the source code on them. I would write out new sections of code in pencil and then, when debugging, make changes so that I kept the «listing» in synch with what was in the program.

Original save mechanism

On the original BAM if you accidentally typed END you could type GOTO xxx (forget what statement number) and the game would resume where it left off. The original way you saved the game was to exit the program, write down the contents of a bunch of variables on a piece of paper, then when you wanted to play again you would load the game, set those variables, and resume.


Oh, yes. I suppose I could have gone on from there – I had plans for an «Outdoor Adventure» that would use fractal landscapes to create random geographies that you could explore in much the same way you could explore the underground areas in BAM. I had read about how the aerospace industry used fractals to construct landscapes to fly over. I had some test code that plotted random surfaces that looked a lot like real-world land masses and such (see attached). But I over designed the game to the point that it collapsed under its own complexity and I lost interest in writing that much code. And by that time I was busy with Beneath Apple DOS and Beneath Apple ProDOS anyway.

Fractal landscape for the sequel. Image credit: Don Worth.
Fractal landscape for the sequel. Image credit: Don Worth.

How did you decide to turn it into a commercial game, and how did you deal with the actual production?

I think I always had in mind selling it – I don’t really remember now. The production was pretty primitive. I had a little hobby printing press in my garage so I set lead type for the labels on the diskettes and cassettes and hand printed them on labels. I pasted up the rule book using a hand waxer and cut out bits of paper that I had typed on a Selectric typewriter and took the paste-ups to a local offset printing company who made 100 or 200 of them at a time. Then I hand collated the pages of the book and bound them with a spiral binding device, all on my coffee table, put them in zip-lok baggies, and then copied the disks and tapes myself one-by-one. I drove them around to local computer stores and gave the stores a free copy to try out. Later on I hired my mother-in-law to assemble the books and package them for me. I was glad to unload all that on Quality Software!

What kind of feedback did you get, and how did you get it?

Every once in a while I would get a postcard or letter care of my publisher. But I probably only got a dozen or so. I didn’t get much feedback really. There were some reviews in Creative Computing and such. I even got a review in a Christmas issue of Playboy magazine.

Screenshot from Beneath Apple Manor.
Screenshot from Beneath Apple Manor.


There was a product called the Galfo Integer BASIC compiler which took an Integer BASIC program and tokenized it (turned it into numeric representations of the various BASIC commands) and then ran it with a runtime assembler program that interpreted the tokenized program. It was very similar in concept the the way PASCAL was implemented later on. Anyway, it sped up execution speed dramatically because the interpreter didn’t need to do a lot of string lookups and such to convert «PRINT» into a branch to the right location to execute and didn’t have to look for variables since their addresses were already stored in the program.

So I compiled BAM and got it working on the Apple. Then we reverse engineer the Galfo runtime interpreter and rewrote it in Z80. Then we could move it to the Atari and IBM PC with minor changes. The compiled/tokenized BAM could be simply copied to the new machines and it ran.

What kind of influence do you think BAM had on the evolution of crpgs?

I’m not sure. I didn’t sell that many copies, and Rogue came out two years later and was much more widespread, so it had a bigger influence. A lot of people send me emails these days and say they remember playing BAM when they were a kid, so I guess the game touched a lot of lives. But I don’t think it had much influence on the industry.

Screenshot from a later version.
Screenshot from a later version.

I don’t know. Probably not – but who knows. I have never talked to any of them. But there aren’t too many ways to implement a D&D simulation I guess.

I was aware of Plato but didn’t use it myself. I wasn’t much into higher-level languages (other than BASIC) – I liked programming in assembler since I could do things in that that weren’t possible in Fortran, PL/I, APL, LISP, SNOBOL, RPG, etc.

Can you remember what kind of sales numbers you had (before and after handing the game over to Quality Software)?

I think I had sold about 500 copies prior to going to QS. I don’t know how many they sold – probably in the low thousands. My books (Beneath Apple DOS, Beneath Apple ProDOS and Bag of Tricks) outsold the game by a long shot. The income from them is now a big part of my retirement investments.

Did you ever consider becoming a full-time game developer?

Office, cirka 1982. Image credit: Don Worth.
Office, cirka 1982. Image credit: Don Worth.

A friend of mine who worked with me at UCLA left UCLA and went to work for Exidy up in Silicone Valley. He introduced me to the president who wanted to hire me as VP of Software Development but I was comfortable at UCLA. I really considered going for it – and I don’t know what would have happened if I had gone – but things works out well for me at UCLA in the end (much nicer pension!!) so I’m glad I didn’t jump ship. And making games wasn’t really what I loved the most. I liked doing all sorts of things in programming and I got more exposure to new and interesting things at UCLA. I was one of the earliest developers on the Internet afterall!

Did you have any idea about what the Internet would eventually become?

No idea at all. I thought it was kind of a waste of time at the time – kind of a government-funded experiment that would never work out. LOL It was a government research project funded by the Advanced Research Projects Administration (ARPA). I had a hard time understanding why you would want to use a network to connect all the mainframes together – we used coaxial cables to connect terminals to them and leased lines to exchange punched card images to submit batch jobs. Also, the original ARPAnet was VERY slow. When you were typing in Telnet it would take a second or two for each letter to echo back on your screen. And file transfers took forever. I wrote the first implementations of File Transfer Protocol and User & Server Telnet for UCLA. I remember having to rewrite them several times because I would get it almost done and then the RFC committee would come up with a new set of specifications and I would have to start all over again. I had no idea that I was doing anything significant. Now it’s something to be proud of! I wish I had known that at the time.

Stuff from FRON: is a Norwegian website mostly dedicated to indie-, niche- and retro games. If you liked this article, we have some more content in English.