|
The title "screen" in an interface that recreates the appearance of the original teletype. Note that Garriott must have assumed I'd play the game, and thus included code to boot out any player named "Chester."
|
Dungeons and Dragons #1
United States
Independently developed by Richard Garriott; never published except recently, as source code
Written with a teletype interface for a CDC-6000*, reaching its final version in 1977
Re-written from printed source code and given
a new interface by Santiago Zapata in 2014
Date Started: 18 June 2021
Date Ended: 18 June 2021
Total Hours: 2
Difficulty: Easy (2.0/5) unless you hit a wrong key
Final Rating: 4
Ranking at Time of Posting: 1/428 (0%)
*Thanks to El Explorador de RPG for determining that it was a CDC-6000 and not a PDP-11 as commonly reported.
In its efforts to catalogue everything--even unplayable, unfinished, unreleased, and just rumored--MobyGames has included
Dungeons and Dragons #1 by Richard Garriott. (For anyone very new to this world: Garriott is one of the most revered CRPG developers, responsible for the
Ultima series, in which he appears as "Lord British.") The story as it exists on the
Shroud of the Avatar blog is that Garriott's father promised him a new Apple II if Garriott could write a working RPG. Garriott used the only tool available to him, a teletype at his high school connected by modem to a CDC-6000, and wrote
Dungeons and Dragons #1. Over the years, he tried 28 more variations, until attempt #29 became
Akalabeth: World of Doom (1980).
|
The type of machine on which Garriott programmed this game, from the Shroud of the Avatar blog.
|
As often happens in this era, the title of the game has been confused on multiple sites, including MobyGames, with its program name, which was DND1. Let's be clear about this once again: The name of a game is what's printed on the title screen, not what you type to get it to run. It's The Dungeon, not "pedit5"; The Game of Dungeons, not "dnd"; Pool of Radiance, not "poolrad"; and Dungeons and Dragons #1, not "DND1." And yes, I verified that was the title in the original code.
In any event, Garriott printed the program before he left high school and had it in a closet for years. In 2014, he
scanned and published it on the
Shroud of the Avatar site, offering a "Citizen Level Pledge Reward worth approximately $550" for the programmer who could create the best modern version that remained faithful to the original code. Two runners up got rewards at a lesser level. There were two winners, one (Mundi King) for a web version and one (Richard Flemming) for a Unity game engine version. Of the two, only
King's is still online, but King changed a lot of stuff from the original code, including the name of the game. Like my colleague Ahab ("The Data-Driven Gamer"),
who covered the game in 2019, I preferred
the contribution from one of the runners-up: Santiago Zapata, AKA "Slashie," a Colombian developer. His interface is brilliant, recreating the look, feel, and timing of a teletype printing the game line by line on a piece of off-white paper--which would have been necessary, of course, for any teletype interface without a monitor. He also produced a plain version (to which I switched after the gimmick wore off) and an "enhanced version" which fixes typos and various oddities.
The original version (or port of the original version) is pretty rough. After the title "screen," it asks if you need instructions. If you say "Yes," it says, "Who said you could play," and quits. There are, in fact, no in-game instructions. If you enter any player name other than "Shavs," it quits. You can't backspace if you make a spelling error (remember, this is all being printed as you type). If you get past that, it automatically rolls 0-18 values for strength, dexterity, constitution, charisma, wisdom, intelligence, and gold--no re-rolling allowed--and then forces you to pick from fighter, cleric, and wizard classes. Charisma always rolled 0 for me, but a video recorded by Zapata shows him getting values in that attribute. Fortunately, the reference card that comes with the game indicates that "nothing happens" based on charisma.
|
The title screen and character creation in Zapata's "plain" version.
|
You then purchase equipment from a shop of 15 items, including such adventuring staples as rope, spikes, flasks of oil, and rations. There don't seem to be any class restrictions on armor based on class. If you screw up and type the wrong number while shopping, you get booted out of the shop and into the dungeon. When you're done, it's off to the dungeon. I forget to mention that earlier in the game, you've selected the dungeon number you want to explore from, I guess, 1 to 6. Those numbers are supposed to access dungeon files that no longer existed by the time of the contest, so the contestants just invented their own dungeon layouts.
|
The equipment list.
|
Once in the dungeon, gameplay is quite a bit like the later Wizard's Castle (1980) line. The character is in a gridded dungeon in which numbers represent the character, floors, walls, doors, enemies, and gold. You input numbers from 1 to 11 to act, with the numbers corresponding to commands like move, open a door, search for traps and treasure, use a weapon, fight, look, cast a spell, save the game, buy a spell, and buy hit points (I don't know who you're buying them from). That the save game feature actually works was a big surprise given how much of a house of cards the rest of the code is.
|
The dungeon. "9" is me; 6s are treasure; 5s are enemies; 1s are walls.
|
Combat is fought by trading blows with a list of 10 enemies, including goblins, trolls, skeletons, gnomes, and balrogs--the latter correctly spelled in this game. If you're a spellcaster, you're in for a tough time until you find some treasure, as spells have to be purchased individually and the cheapest is 75 gold pieces. Wizard spells seem to be "Push," three levels of "Magic Missile," "Kihl" (I don't know if that's an error or "phantasy spelling"), "Find Traps," "Find Secret Doors," "Teleport," and two spells to "change" what I assume are attributes. Clerics oddly get most of the same spells but in a different order, plus two levels of "Cure Light Wounds."
|
I cracked up at "Good work you just killed a man."
|
|
A menu of available wizard spells.
|
I lost more characters to traps than anything else. You want to have a healthy supply of ropes and spikes, because falling in a trap requires one of each to get you out.
|
Not once is "you're" spelled correctly in this game.
|
The game has no point beyond this. You don't earn experience for kills, though you do make money and can purchase extra hit points as a kind of "character development." There are no items to find in the dungeon. From inspecting the code, there's no winning condition. You can't leave the dungeon. When you inevitably die from combat or traps, you don't even get a score to compare with other players. The best that you can do, I believe, is kill all the monsters on a level, at which point the game says "All monsters dead" and "Reset," but then it just kicks you out. You have an option at the beginning of the game to set the "reset" to yes or no; I assume if you say "no," it just says "All monsters dead" and kicks you out.
|
Screw it. I'm calling this a win.
|
Although gameplay isn't much like Ultima or even Akalabeth, there are a number of things that foreshadow Garriott's commercial games. The first is the treatment of spells as purchased inventory items
instead of innate abilities, a mechanic we saw through Ultima II. The second is a randomization of the
player's experience based on a seed number that is otherwise meaningless
to the player (this would be re-cast as the player's "lucky number" in Akalabeth).
The third is the ability to purchase hit points directly with gold,
although in later games, the player has to find the sovereign first. Fourth, although food doesn't seem to have any point here, it does exist in the game. Food remains crucial in all of Garriott's titles, all the way through the 1980s and 1990s, when most developers weren't including it as a mechanic. (It's not a "first," alas, as the PLATO Moria had hunger and food in 1975).
A GIMLET is rather pointless for this one, but the game does have a few RPG elements. It gets 0s for the backstory, NPCs, encounters, and graphics, sound, and interface, and 1s for everything else, giving it a final score of 6. But I can't leave it there because a 6 is what I gave Dungeon (1979) and The Devil's Dungeon (1978), and while neither of those were excellent games, they were at least finished programs and relatively bug free. Subtracting 2 points for bugs, spelling errors, unimplemented features, and forcing me to play a character named "Shavs," that puts us at 4 points, meaning Richard Garriott now has both the highest-rated and lowest-rated games on my list.
Check out
Ahab's coverage for a few bugs and foibles that I didn't experience, and
Nathan Mahney's coverage over at "CRPG Adventures," particularly notable because he played a version that is no longer online. The contestant seems to have added a few things. Starting characters get 1,000 gold pieces instead of a much lower random value, and "Mjolnir" is in the equipment shop. The dungeon is more interestingly represented with asterisks for walls, and your character sheet and inventory remain active. None of those features were in Garriott's original code. I should also mention that Zapata's "
remake" is significantly cleaned up. You can give yourself any name; you select class before attributes; you can re-roll attributes; the dungeon map prints more like a roguelike; and various bugs and misspellings are corrected.
|
The dungeon map in Zapata's remake.
|
I don't mean any of this entry to sound like I'm bashing Garriott. He was 16 when he wrote this game, on less-than-ideal equipment, never intending it for publication. It's fun to see where a famed developer got his start. I think
the contest was a fun idea. It's worth reading the comments on that page. There's a clear wheat-from-the-chaff separation going on there as some commenters complain about the code and the conditions and others just roll up their sleeves and get to work. I think I'm at the point where I could replicate this game in Python. Maybe I'll give it a try.
I played one of the browser-based remakes upon discovering their existence a number of months ago. I guess it would have been fun enough to pass a few minutes back then if you didn't have access to any better computers or games, and if I were Garriot I would have been quite pleased with myself for such a first attempt.
ReplyDeleteFirst map is nice
ReplyDeleteWhat a time to be alive that must have been.
ReplyDeleteHave you ever tried postal roguelikes, waiting two weeks for the next turn to result in something?
DeleteMy friend played a multiplayer Empire clone (the wargame) by post. Back then I was jealous, but these days I see the thing for the scam it was. You had to pay a monthly fee to play something like 3-4 turns a week of a quite simple war game, where any player might drop out anytime. But at the time it felt quasi-magical.
DeleteNot to mention the games where you has to snailmail disks. The postage itself was quite brutal.
Well, I wrote worse games when *I* was sixteen...
ReplyDelete"its efforts to catalogue everything--even unplayable, unfinished, unreleased, and just rumored--MobyGames"
ReplyDeleteJuts to clarify, Mobygames is very strict in regards to insisting that the games it documents were definitely officially released in some provable form. There are other sites that document unreleased games, leaked prototypes that never made it to market and other vaporware, but Mobygames is having none of it.
I don't think that policy explains this game, or really any of its cataloging of PLATO games, which were all amateur student efforts never "released." It particularly doesn't explain Pits of Baradur, Bugs N' Drugs, or Think15, for which the only evidence is recollections on some message boards.
DeleteI think it's because on mainframes there weren't published games, so they have to make exceptions. Otherwise, on most non-digital platforms a newly scanned cover is almost required - you can't even use images from the internet.
DeleteI don't agree that they "have" to make exceptions. The number of amateur games that were written on school or work computers, played by a few people, and then lost must number in the hundreds of thousands. As far as I can tell, the only difference between them and Don Daglow's Dungeon is that their authors haven't bothered to mention them publicly. I can't tell MobyGames what to do, but if I could, I'd tell them that nothing belongs in their database for which we can't at least verify a title screen.
DeleteWhy does it bother you so much?
DeleteSigh. I suppose it shouldn't. It's just that every time I think I've covered a set of years, I go back to MobyGames, and someone has added three or four more games. Now if I don't make them a part of my master list, someone will inevitably e-mail me about them. So now I've got to check them out, and half the time they turn out not to be RPGs (by MobyGames's own definition, not mine) and the other half they turn out not to actually exist. Check out the recently-added Trilogy for another example.
DeleteBut I'm the one who chooses to use MobyGames as a primary source.
Don't worry, be happy :)
DeleteI've been a Mobygames contributor for over a year now and joined their Discord server so I know a little about how their process works.
DeleteThere are several approvers (all volunteers) who look at user submissions and check if the information is correct before approving the entry for the database. When you enter a game into the database, you have to deliver a short description of the game, the name of the company or individual who developed and published it, and the year it was published in.
Then you progress to a checklist where you can enter genre information. The list contains all the main genres (action, RPG, strategy, adventure) as well as a lot of subgenres, thematic elements, etc. You enter whatever you think fits the game.
When you're done, you have to add a comment citing your sources. This is mandatory, you cannot submit before writing something into the textbox. I tend to put a link in there that leads either to the developer's official website (or an archived wayback machine link if the website is no longer active), or a store page if the game is still actively sold (itch.io, Steam, Gamersgate, etc). If there is a third party website with valuable information, I post a link to that, too, like a fan page that contains release info.
While the Mobygames approvers do have standards and asked me to edit my submissions when I've made mistakes, those edits are usually just about the easily provable facts. I got a year wrong once, and the approver noticed when he checked the info contained in my links. Another time, I picked Windows as a platform when the game only had a DOS release, because I remembered playing it on my dad's Win 95 laptop back in the day. The approvers are good at picking up mistakes like those.
But when you claim something is an RPG and nothing in the sources you provide clearly contradicts that statement, the approvers will believe you.
So the problem isn't with Mobygames' standards, but with submitters getting the genre wrong.
Also, Mobygames approvers go through massive lists of submissions on a daily basis, so you can't expect them to personally hunt down a working copy of every game and play it long enough to check if all the genre tags were set correctly.
That's what corrections are for. Anyone with a Mobygames account can submit a correction if they notice a mistake in an entry. I submitted corrections for a title before (Teudogar and the Alliance with Rome was just called Teudogar, but it was sold under its full title on the official website, and the title screen also showed the full title, so having it as just "Teudogar" in the database was wrong), and corrections to a category (Assassin's Creed Origins was in the category "Setting: City - Cairo" when it's set centuries before Cairo was even founded), and they were swiftly addressed by the approvers.
So if you notice anything wrong with a Mobygames entry, go ahead and submit a correction request! Mistakes in the database can only be corrected when the approvers are made aware of them.
For the record, it was me who added DND1 to Mobygames (my sole contribution). I figured it was of enough historical significance that it should have an emtry.
DeleteI submitted a correction on the genre of Make Your Own Murder Party back on 15 January 2021. It is still "pending." That really doesn't encourage me to submit more requests.
DeleteBut I'm not trying to bash MobyGames's volunteer admins or contributors. I experience personal frustration with some of its practices, but only because I've chosen to use the site a particular way.
However, I still don't understand the disconnect between what Rowan and JarlFrank say are the site's policies and its clear acceptance of games that never had any release nor, in some cases, any clear proof of their existence.
Mobygames approvers take a long time to go through submissions because of how huge the lists are they have to go through. The longest it took for an entry of mine to be approved was about 3 months. Remember that MG is intended to be a database for all games ever released on any platform, so they get a LOT of submissions.
DeleteThe only thing I dislike is that sources cited in the comment section of a submission aren't released, they're just for the approvers' eyes so they can check if everything is correct.
I got my entry on "The Earth Lords" approved recently. I included links to the "Really Obscure RPGs" thread on RPG Codex, where a user discovered the game. A link to a reddit thread where someone uploaded the game. And an archive.org link to the now defunct website of the developer, which contains a lot of information about the game's development.
Neither of these three links appear in the database entry. Go ahead, search for The Earth Lords on Mobygames. There's no cover, no screenshots, nothing that proves the game's existence... because the sources you include for the approvers don't become part of the database entry.
So if the only proof for a game's existence is an obscure old website on the Wayback Machine, you'll have a hard time finding it unless you ask the submitter of the entry himself.
You are doing the lord's work. Well done.
DeleteSo if you notice anything wrong with a Mobygames entry, go ahead and submit a correction request! Mistakes in the database can only be corrected when the approvers are made aware of them.
DeleteThey approve YOUR requests because you're an insider and you particpate in their chat and they know you personally. Please don't think everyone else gets the same treatment.
That is wrong, Harland. I had been submitting thing to the MG database for almost a year before I joined their Discord server. I joined when I found an issue for which I didn't find a report button on the site.
DeleteUp to that point, I already had over a dozen new game entries approved, as well as one correction. In fact, my longest waiting time for an entry was *after* I joined their Discord.
It all depends on which lists the approvers currently go through. If the corrections list currently has 1000 submissions to verify, but the new game list only has 100, approvers tend to gravitate towards the shorter list first. Submissions appear in the list based on the order of submission: if you submit something before me, it has a higher place in the list.
Sometimes approvals take longer because they are elevated to an additional reviewer. I've had items get approved nearly a year after submitting them because of this. But typically when I don't screw up my entry, things are approved after about 1 month.
DeleteI've added over 2,000 games at this point, but mostly I annoy the approvers, because I submit things before going to bed for the night and make stupid mistakes.
I also annoy the CRPG addict because I find old TRS-80, Atari, and Apple II games with mild RPG credentials.
I spent a decade or so lobbying heavily for Mobygames support for terminal and mainframe games, but I must confess that since they were finally allowed in, I haven't followed that beat very carefully.
DeleteI do know that their main contributor vedder is somewhat of a completionist (currently working through the entire library of Commodore PET games, year by year -- every Simon, Star Trek and Mastermind clone) and he is pretty rigorous about not making unsourced claims. Absolutely he is in many cases attempting to document games that can no longer be played (eg. games formerly hosted on online services such as Compuserve, GEnie or Prodigy), but if he has good, vetted period sources for the information on the games he is attempting to document, he will faithfully recount every alleged fact included therein.
But you will have a devil of a time confirming or correcting any of it without knowing his sources. (If you like, I bet Mobygames admins would be happy to share his sources with you. I would happily go to bat for you in the forums and introduce a motion to expedite any of your inquiries to the front of the line, in the interest of the important work you're doing here.)
Completely untrue, Harland. I have contributed to MobyGames, and yet I've never talked to any of the staff.
DeleteIt's nice of JarlFrank to be so polite while he calls you endless names for YEARS on RPG Codex
DeleteHuh, wasn't aware this still existed. Now I wonder if this'll be the ratings floor or if there's anything that'd manage to rate even lower than this
ReplyDeleteI can see Ultima IX getting a 3.
DeleteAs bad as Ultima 9 is, it's still better than some of the shovelware we've already seen here.
DeleteYeah, my experience with 9 was that it was garbage for Ultima but more or less fine overall. Not great by any stretch, but not the worst thing in the world either.
DeleteUltima XI was the first - and for a long time the only - Ultima game my wife ever played in her life. She was astonished that I (whose first Ultima experience was U4) and others considered it a bad game. She had never been exposed to Ultima lore and experience before and honestly enjoyed the game for what it was.
DeleteWasn't U9 also infamous for being extremely buggy, regardless of how well it fits continuity?
DeleteBugs were certainly an issue, but I believe the bugginess of the experience also hinged strongly on which 3D accelerator card you were using... I believe if you had a VooDoo 3 card it was mostly fine, but on anything else it ran horribly and was prone to crashing.
DeleteUIX is bad, but as others note, not as bad as it's made out to be. It's not even the worst Ultima (I'm looking at you UII). I just liked the joke (and the thought of Richard Garriott having the *two* highest and two lowest rated games on the blog).
DeleteFrom a gameplay point of view I think U8 is worse.
DeleteMan, that second picture brings back memories. My first exposure to computing was with a model 33 ASR connected to an HP 2000C computer via a timeshare network in 1976. I was a senior in high school, my friends and I had a blast writing code for that system.
ReplyDeleteThank you Chet, for all the work you do on this blog, I've read it from the beginning and enjoy reading your work.
What a neat bit of history, and the fact that he encouraged/enabled others to have a go at it.
ReplyDeleteAny idea if anyone got their start from this competition?
I absolutely LOVE entries like this! I really can't be bothered to Google endlessly for RPG history, and this site is a veritable gold mine of information on games that I never would have known about otherwise. As far as I'm concerned, this site is THE definitive RPG reference. Well done.
ReplyDeleteWhat are we doing with our lives right now. I mean...really...
ReplyDeleteHaving fun. What more do you need?
Deletehere is something you can't understand... good work you just killed a man!
ReplyDeleteI was waiting for this one.
DeleteTalk about a dummy prize for a contest...
ReplyDeleteThe greatest prize was actually the opportunity to create it, I guess :) also hello long time no heard about you, Knight Burzmali :D
DeleteI'm glad you liked the port and that it served a purpose for videogame conservation :) I must also note that the "Reference Card" is something I put together (trying to make fun / pretend something young Garriott would do).
ReplyDeleteJimmy! Thanks for coming by. When are we going to see a modern RPG from Slashware?
DeleteWell... we've been working for quite a bit on NovaMundi which is already out in Early Access (and I'm planning the second game in the series already), and Ananias has been out for a while now (and I see it's already on MobyGames) but not sure if you'd count it as "modern". Both lean more into roguelikes than plot-heavy cRPGs, but I see these are welcome here too :)
DeleteI had somehow missed Ananias I like the way it looks in the screenshots!
DeleteThank you! those are a bit outdated since development went all the way thru 2018. But in any case it's good that someone went thru the hassle to submit it there :)
DeleteThis comment has been removed by the author.
ReplyDeleteI believe the king port of DND1 won the contest because it was made to work somewhat on a phone which was one of the stipulations of the contest. The other ports are much nicer, visually, but tend to get cut off on a phone display.
ReplyDelete>[...] Nathan Mahney's coverage over at "CRPG Adventures," particularly notable because he played a version that is no longer online.
ReplyDeleteI believe the version he played is still online, but under another directory in the same website:
https://www.kirith.com/dungeon-adventure/
Despite how well-known Richard Garriott and Ultima are, internet is full of inaccuracies in this regard, but I recently came across one that seems to come from Garriott himself and for which I cannot find an explanation.
ReplyDeleteI know that he was born in 1961, and he started going to the University of Texas at Austin in the second half of 1979. At the beginning of that same year he finished his last year of high school in Clear Creek, where they supposedly had the terminal connected to the PDP- 11 with which he created DND#1.
The fact is that the copy of the DND#1 source code that Garriott published is dated February 1979 in the document itself, but the BASIC language used is not that of a PDP-11, but that of a CDC, such as the University of Texas at Austin's CDC 6600... how is this possible?
Did the high school's terminal actually connect to the university's CDC 6600 and not to the PDP-11, or did Garriott have access to the university's CDC 6600 before finishing high school and there he translated his game into the new BASIC?
If the first option is correct, everyone, starting with Garriott himself, has mistakenly classified DND#1 as a PDP-11 game.
But there is still more, if Garriott started in 1977 with version 1 of his game and Akalabeth, a variant of version 28, was already finished at the end of 1979, why did he bother to print version 1 in February of that last year, let alone translate it into a new BASIC, instead of doing it with the most current version?
DeleteThis can be easily explained if you assume the published 1979 version is the result of Garriot learning a new computer and porting his earliest and simplest game as an exercise.
DeleteI asked Garriott himself, and he says he didn't program anything in the University of Texas computer, he had switched to Apple II by then, so the strange BASIC used in DND#1 had to be the one from the computer he connected through his High School's teletype... and the only BASIC I have found that uses the BASE, CLK and FILE statements or functions with the same exact syntax as DND#1 source code is CDC Cyber/6000 series BASIC v2.0 from 1973.
DeleteIt was written more than ten years ago, so I'm not sure he'll recall the source or if he confused something, but the Digital Antiquarian stated that the (Texas) high school Garriott attended starting with his second high school year had a teletype terminal connected remotely to a CDC Cyber mainframe (i.e. not a PDP-11).
DeleteIn my understanding this would square with Garriot's own notes on DND#1 where on the cover page he apparently wrote "[Rosion? Region?] 4 Cyber acct. #" followed by what I suppose was an access code (?). Unless, of course, the access to a PDP-11 also / through a teletype generally was called a "Cyber" or that note refers to something else.
Another account of these early years recounted in an excerpt here mentions his first year high school in California (Gunn) having a single teletype, remotely connected to a CDC Cyber mainframe while at the Texan high school (Clear Creek) he started attending the following year the teletype dialed into a "minicomputer" located off-campus. This could refer to a PDP-11. But then, did the CDC Cyber line-up not include minicomputers, too?
I have no idea about the exact differences between these computers or BASIC for a PDP-11 as opposed to that for a CDC Cyber, though I have also not seen so far some of those playing with or porting the code mention it would not fit that used on/for the former (one of them wrote the original code "appears to be a BASIC language dialect for the OS/8 operating system running on a PDP-8").
As for why Garriott would print his source code for DND#1 in February of 1979, Ahab / The Data Driven Gamer had wondered about the same thing and the resulting questions as well and considered the story did not really add up (see the last paragraph in his coverage of DND#1).
By the way, there was also already some confusion about the date when Akalabeth was created and sold, based on Garriott's own statements, see this article by the Digital Antiquarian (also linked by Chet in his coverage), including Garriott's answer in the comments.
Thank you Busca for all this info!
DeleteI knew about the Gunn High School's teletype connected to a CDC Cyber (I think Jimmy Maher could have mixed this with Clear Creek High School's teletype in his article), but I didn't know about the hand-written notes by Garriott. I think his reference to a "Cyber account" is a strong point here.
About the BASIC used, I haven't found any PDP-11 or PDP-8 BASIC that fits DnD#1's source code. Some PDP-8 BASIC looks similar, but the CLK function uses 3 parameters instead of one (most versions of BASIC used CLK$ with no parameters), FILE statement uses ":" instead of "=" (and a different file name format), and there is no BASE statement (there is a BASE monitor command not related to BASIC).
I started to investigate this just because I encountered people trying to play the game in its original platform and having problems identifying the version of BASIC. In the contest page, someone says the BASE statement is not in PDP-11 BASIC manuals (it's true), and other one says it is PDP-8 BASIC (the one you cited), but it isn't (and PDP-8 is a totally different machine than PDP-11 with even different word lenght).
Then there is other people like this one:
https://github.com/royedmund/dnd-1
Or this one:
https://twitter.com/disdev/status/1370545254378598404
Or this one:
https://www.reddit.com/r/Basic/comments/y5qa4k/can_anyone_help_me_identify_this_variant_of_basic/
And finally there is the comment by Garriott himself in your last link to Jimmy Maher's site, where he says in 2011 (before the contest and publication of the source code): "It’s core code was in fact made on a PDP 11 (if memory serves)", which I think denotes he had some doubts about it.
PD: this is not related to the issue, but Garriott's hand-written numbered maps remind me of certain other ancient game...
I finally found the answer in The Official Book of Ultima, Chapter 1, page 6, from 1990:
Deletehttps://archive.org/details/TheOfficialBookOfUltima/page/n17/mode/2up?q=Cyber
Jimmy Maher didn't mix things up, he went to the oldest source, and he's the only one who told the story right. Much later Garriott misremembered the computer which his High School's teletype connected to, and everyone has mistold the story since.
Indeed DND#1 to DND#28 were games written on a CDC Cyber.
And it seems that also DND#29, of which Garriott himself remembers nothing, but seems to be an attempt to make a version of his game in FORTRAN (some sentences in FORTRAN can be seen in the light, but those on the last page are in BASIC and belong to DND28):
https://gallery.ultimacodex.com/richard-garriotts-dnd29-notes/
If this summer proved nothing else, it's that EEDRPG is one of the best paperwork detectives in the business.
Delete