I wanted to throw one more example I didn’t see listed here of a Scott Adams Interpreter variant that depends on autoget - David Malmberg took the existing PET engine from 1979 and besides adding disk load/save functionality, implemented a version of autoget that checks both the noun array and all the “getable” items, then wrote “Castle Adventure” some time after December 1980 (based on the comments).
I had previously written my own game interpreter in 6502 machine code that played Adventureland and Pirate’s Adventure from the data statements in the original PET BASIC games. When I got a copy of Castle Adventure, it didn’t work and I had to re-implement the search loop for GET and DROP to also check the item strings.
There’s also a variant of Castle Adventure for the Commodore 64 in the Interactive Fiction archive, but it has a couple of structural changes (removing load/save routines and chaining to a menu program for all the programs on the disk) as well as some specific presentation changes for the Commodore 64.
Outside of a few differences, all the variations of Castle Adventure can be traced back to the original 1979 version of the game engine.
It’d be interesting to see if there are any references anywhere for a version of Castle that predates the Micro type-in version. I guess it might’ve been floating around all the various PET user groups.
I almost certainly got my copy for the PET from someone in the local user group (COPUG). My version is substantially like the one in Micro except mine has line numbers that match the earlier Adventure International version for the PET, which is why I think they renumbered with sequential-by-one line numbers for publication.
Certainly by 1981 he was mentioning writing adventures in his bio (such as that in the October 1981 issue of Micro). So there may be evidence out there to support a 1981 date for Castle Adventure. I think David’s PET version of Dog Star is generally attributed to be from around 1980/81.
Very cool! Do you think it’s possible to get Castle Adventure (or Crusade Adventure) working with modern terps (perhaps with minor rewrites to the game code)? I guess I’ll need to manually type it in and see if it works at some point
I’m having a hard time tracking down Crusade Adventure. The AGT version of Crusade Adventure the game says:
…which is the issue where Castle Adventure was published. Are Castle Adventure and Crusade adventure the same game?
Specifically:
Makes me wonder if mr Malmberg was an amateur coder at the time, or if “management systems” is what they called “computers” at that company.
Yes, Crusade Adventure is the AGT implementation of Castle Adventure; hence the reference to the original Micro magazine article.
Re. how early Castle Adventure might be, in a Brass Lantern interview, David said
Reverse-engineering could’ve technically preceded Scott’s magazine articles describing the system.
We’re veering off-topic here, I think, but here’s an article on David’s work for Foremost-McKesson, if you are interested…
David was a very early convert to microcomputers so, yes, I guess you could say he started off as an amateur coder, even if computers became part of his day job; his PET screen print routine appeared in the very first issue of Compute! in 1979.
Ah, that’s interesting. That might explain why I wasn’t able to get the data file to work with other Scott Adams interpreters.
Don’t type it in. I can send you a copy of the BASIC listing for the Apple II and the PET. Just send me a PM if you want either or both.
According to the remarks in both the Apple II and PET listings, David’s game is based on the data structure explained in Creative Computing August 1979, and the TRS-80 interpreter in SoftSide July 1980 and BYTE December 1980. The Micro article was published in November 1982.
EDIT: I just noticed that the original article also acknowledges the articles in Creative Computing, SoftSide and BYTE.
OK… I just looked at the first dozen lines of the Castle Adventure version I have for the PET and the version published in Softside. Again, they are largely similar but there some differences. For one, David Malmberg’s personal contact info (address and phone number) are in the PET version I have. For another, line 8 in the Softside version, mentioning Softside, isn’t in my version at all. Functionally, there are few differences in the code, but the version I have happens to have specific variable definitions near the top, x=y=z and k=r=v, for example, which serve to define those variables first which will optimize finding them in 8-bit Microsoft BASIC because it scans memory looking for variables in order of declaration. The Softside version just defines variables as they fall in the code. I didn’t see anything more substantial in the parts I examined.
Indeed, those references are clear credits in the published version and I don’t see any evidence at all for a date earlier than 1980. It just seems a little odd (particularly coming from a programmer) for David to talk in that interview about “reverse-engineering” Scott’s game to make his own Castle Adventure if, in fact, David just read the magazine articles. I guess he might have originally reverse-engineered the game to make an early version of Castle Adventure and then used the published articles to help him create the printed version we’re familiar with.
