Beta Testers Needed for classic 80's mainframe text adventure

I’ve been working for decades (literally) to get this game into the modern age so it’s not lost forever,
but I’ve had a really hard time finding any testers in order to find any possible bugs.

This may be considered pre-release, as it’s publically posted for decades, anyone can try it,
but with near-zero feedback I don’t know if it has any bugs.

I need testers to try it out, find and report any bugs,
so that I can finally release this fully, and finally retire :slight_smile:

The Game is called “The New Castle”

the game can be found at Index of /files/castle

there is zip, rar, gzip, with installer and “raw” versions for Linux and PC.
My Mac died so I can no longer port it to that platform, hopefully someday I can.

There is debugging and logging capabilities built-in to help with bugs

Please try it out and report any bugs/etc to me (the author) at:

Would appreciate any help getting rid of any remaining problems.


The 1983 Barry Wilks game, Castle? Rather than a 1970s game? Unless you’ve uncovered more background on the original game.

Might want to update the IFDB page to link to the latest version, anyway…

It originally started on a PDP system in the 70’s,
but the code this is based on is from the 80’s.

I’ve changed the top of this thread in accordance with the basis for this code,
but it’s really beside the point.

The link on IFDB is to the Sourceforge page, which is where the game ‘lives’ and will always have the most updated version of the code, which is the same link as I’ve provided here.

It was recommended to post here to get testers,
I just need someone to actually test it and report bugs and maybe work with me to get this finished.

The question about 1970s/1980s was because I, like many here, are really into the whole history/research side of things of text adventures and I hadn’t previously come across any references about any 1970s roots. Just looking to learn more and add sources to its entries on the various databases.

The actual “download” link there isn’t working, which is why I mentioned that. The website link, as you say, does work.

Good luck with getting this finished. I know that Canalboy has spent quite a bit of time playing it already… He’s really experienced with these sort of huge ex-mainframe games, that scare off most lesser players! :slight_smile:

I finally figured out how to change that download link, so that’s been updated I hope correctly
I also copied the v21 release to that same old filename, so in case people download the old one will get the v21 updated version.

the Index of /newcastle link given will also get you the latest version

I haven’t had any response from Canalboy in many months, and no one else has tried it, or at least, no one has reported any problems.

Why would an ex-mainframe game scare off most players? they are among the easiest to play,
far simpler, and smaller than most if not all infocom games. Not sure why you call it “huge” either.

Castle is nowhere near as big a game as Zork/Dungeon or Adventure/Colossal Cave if we’re doing comparisons. Maybe even less than 1/3 the size of it.

As far as the history goes, I didn’t deem the PDP 1970’s beginnings to be of any relevance, it’s not even close to the same game as it’s early days back then.
Originally it was written in Fortran on the PDP, that’s late 1970’s.

But very “quickly” rewritten (from scratch) on the Vax (VMS) started probably in 1980, with Barrys final, completed version in or about 1983. I started working with the code around 1980.

This is why I say “circa” 1983, because the timeframe in around 1978-1985 gets really messy.

The original Vax (VMS) Basic (I think it was Basic) punch cards have been long lost to the winds of time.

Someplace I believe I have a binary of the Vax EXE file, but I haven’t seen any sign of it in decades, despite my efforts to track it down. So that may be lost as well :frowning:

The only place it was known to survive was on my one of my magtapes but some idiot decided to be a jerk and erase my tapes.

I currently have someone doing an audio reconstruction of that tape, but the odds of getting anything from it are slim to nil.

I also wrote a version of Castle early on for the Commodore-64, but I’ve never been able to find a copy of that code in any of my archives, though I know it exists.

The earliest code I have is what I wrote for version 2.0 way back on a 486 using Turbo Pascal.

I’ve lost two modules of code in all the years, one I don’t even know what it was for any more.

Be that as it may, the game is complete, as per my printouts of original play from the 1980’s
I played Barrys original version on a vax using a teletype and saved those printouts.
I also took copious notes, and played and replayed it many dozens of times, ensuring nothing was missed.

it’s all contained in my version of the game.

Anyhow that download link works now.


Well, I know that Canalboy has mentioned it in various places to people who like this sort of thing, so he has been helpful there too.

Sometimes it helps to share a little of the history of a project like this to get people interested, especially with so much competing for their time and attention. It is intriguing to hear about how you’re attempting to piece together a game from printouts and memories. It’s obviously a deeply personal project, given that you’ve been going back to it for thirty years or so. Playing the game originally must’ve been an important or formative experience for you, to want to spend so long recreating an adventure from its shadow.

You may not get the best response for your playtesting request at the moment, as the big annual IF competition is soon to launch; which ties up authors, play-testers and reviewers for at least a month or so. But hopefully some more will give the game a go towards the end of the year.

Who was Barry, by the way? A fellow student/follow staff member?


Yeah, I was going to post to say the same – hopefully you get a good response now (my sense is the old-school text-adventure aficionado community is orthogonal from the mainline IF Comp one) but either way I’d bump this thread in late November to see if you can get some more attention then!

Barry was a fellow student and friend of mine,
And although our lives have diverged greatly over the decades, I would still consider him as such.

I played all the classic mainframe (text) games back in the day,
Colossal Cave (Adventure) and Zork - both on the Vax (VMS) System were the first.

I also found, and played Empire a lot, and a little bit of Rogue.

Then I found “The New Castle”, as it was known, and spent many hours over many months, and years actually, playing it and mapping it out.

Many other students were playing as well, it was different from those other adventure games, and different from Infocom.

If you look at the PDF (Manual called the castle user guide) I did for the game there’s a hint just by looking at it - I had some fun and the watermark is an image of the “Castle” which the game is based on. I thought that was a nice addition.

So a very long time and lots of fun playing and mapping it out, and discovering little secrets and mysteries, and I solved it dozens of times so that I could be absolutely complete in my documentation.

By this time Barry had moved on to greener pastures and had lost his punch cards, so I had to be certain I was covering everything.

Having an eidetic memory doesn’t hurt :slight_smile:

Those still were the best times of my life, being in University, day to day with students.

There’s something special about Castle in comparison to everything else, and even all the infocom games like Leather Goddesses of Phoebos, HitchHikers Guide, Bureaucracy and the like.

Douglas Adams was probably a bit of influence, having humor in scifi takes quite a bit of skill.

This also may explain my fixation on “trek7” which to me has always been the best Star Trek game ever made, even if it is text-only. A multi-player star trek game with some really nifty additions.
You won’t find Moonbase Alpha (Space 1999) in any other Trek game, that I can assure you.

It was a simpler time, but maybe a bit why books are sometimes better than movies, the images are created by your imagination, and no movie can stand up to that.

Interactive text adventure where you can explore, learn, it’s a fine art which I’m glad places like IFDB haven’t lost. But a game like Castle couldn’t be created with inform (z-engine) or any other adventure creator, it’s got too many specialty things in it.

Eg: the original Trek7 had a “time of day” and “day” lock on it, so that students couldn’t play during study hours without a secret code (which eventually got passed around) but that section of code was commented out (still there, just commented) before I first looked at it. Such a thing doesn’t make sense in today’s world.

In the Castle documentation, you’ll see I mention the maze is just a new creation I made up,
well as it turns out I found the section of my original printouts of the game which covered the maze,
so as of v21 the game has the original maze path included as an alternate solution.
So there are two ways out of the maze - the original path or the new path which I coded.

It was, and is, one of my favorite games, but as I mentioned, I’ve solved it too many times, and with an eidetic memory it means I have a hard time finding possible bugs. I can’t think of things other people would try.

Save/restore options were never part of the original game, I added them as a whim but have since disabled them in order to focus on bugs in actual game play.

Once the bugs are eradicated, and if there’s enough interest, I’ll re-enable the feature.

V21 is about as close to pristine original as is physically possible, and that’s the way I want it.

Yes, it means a great deal to me, and I’d like to get it done while I’m still capable.
After almost 40 years, it’s about time to retire. lol


Edit: forget it

1 Like

Yeah, that was me! I’m probably the only person posting about Castle :slight_smile:

Someone there directed me to post here, so here I am.

I will appreciate any help with testing I can get,

And I’m dedicated to working with someone with debugging/logging and other things built into Castle to help resolve any problems.

It’s been a very long road since I started playing this game, and I’m hoping the last few bugs will get crushed then it can be released fully, I can retire.

But mostly I think people should try the game, it’s very different, I’m hoping future generations can enjoy it, once it’s on the Internet, it lives forever…


I was the one that suggested you ask for testers here. I see that you’ve already generated a bit of interest. Unfortunately, it’s a busy time of year in the IF world. Submissions for IF Comp are due in a day or so and ECTOCOMP starts in four days, so most of the regulars will be busy playing, judging and reviewing IF Comp games or writing ECTOCOMP games.

I like the conciseness and innocence (or raw brutality) of the old mainframe games, but I’m not keen on the sudden death, guess-the-verb and pointless mazes. My most recently finished mainframe games were probably ‘Castle’ written by Peter Langston using the Wander system, and ‘Castlequest’ written by Michael S Holtzman and Mark Kershenblatt in Fortran. It’s interesting that all three games (including your ‘New Castle’) have the word ‘Castle’ in the title. I presume yours is unrelated to the other two.

As I pointed you here, I now feel compelled to test ‘New Castle’. I’m currenty busy on other projects, but I’ll try to squeeze in some time over the coming weeks…unless I decide to write something for ECTOCOMP. Are you open to suggestions for improvement (like tweeks to puzzles and verbs that aren’t accounted for) or are you only interested in out and out bugs? I can cover both and let you decide what to include or exclude. That’s your prerogative, anyway.

1 Like

Tweeks to puzzles I would say “it depends” and that’s the only honest answer I can give.
As long as it’s something that doesn’t drift too far from the “essence” of the game, it would probably be ok, but the likelihood of anything like that I’d say is astoundingly low.

“missing verbs” probably though I’m hard pressed to think of any that would apply, surprise me! lol
There’s only so many ways you can say “get” and “put” (ie: “take” and “drop”)

None of those games are in any way related to this one even the slightest tiniest bit. that’s for sure.

The game can be solved even if you never enter the maze, it’s not mandatory, Barry & I aren’t that cruel. I recreated the original solution and added it to the current version.
I also added a little bit to the maze rooms to make them at least a tiny bit more palatable.

All text adventures are “guess the word”, Castle might be easier this way because it reduces it all to just two words, and it only looks at the first five letters of each.
that limits the possibilities, makes it more strict maybe, but should make it easier to.

As my highschool computer teacher often said: Keep It Simple Stupid (heh)

The Castle itself, as the watermark in my user guide lets on is based on Middlesex College building at the University of Western Ontario in London - but that won’t help you at all. lol.

There’s one “easter egg” I really love in this game, it’s one of my favorite parts,
it “breaks the fourth wall”, very few (none I can think of) do this.

As I said, I hope it lives on, on the internet, enjoyed by many long after I’m gone.


UPDATE: I’ve updated both the PC and Linux versions of Castle to make a few minor changes at the request of a user.

Still in critical need of a tester for two major things:

  1. is the game solvable?

And 2) are there any bugs which prevent the game from being solved?

I can of course solve it, but then I pretty much have everything about the game memorized, which doesn’t present the way anyone would normally play it.

PC and Linux versions updated to v21.0.2 here:

I want to focus solely on practical playability - not on possible expansion of nouns/verbs, descriptions or responses. The core game play is the most critical piece that no one has play tested in probably 20 some years. :frowning:

I removed the save/restore feature a while back, it was never part of the original, and just compounded bugs, so maybe once all the bugs are ironed out I might add it back in. But for now it just makes bug fixing too difficult - end up trying to resolve two bugs instead of one.
And it was never meant to have such features. but, if it can be stable, it’d be something I’d look at.

I hope I can get someone to look at the core playability

1 Like

I’ve rebuilt the Mac version of the game! and updated both the PC and Linux versions of Castle

Still in critical need of a tester for two major things:

  1. is the game solvable?

And 2) are there any bugs that prevent the game from being solved?

You can find PC, Mac, and Linux versions updated to v21.0.14 here:

1 Like

I’ve rebuilt the Mac version of the game!

I get a bad CPU type in executable error when trying to run the Mac build on my M1 MacBook. I think that means it is a 32-bit binary.

1 Like

Yeah, unfortunately all I have available to build the Mac code with is a MacBook Pro 17 that runs OS X version 10.8.0 (Mountain Lion) - this thing is about 14 years old.

I’m sorry it’s the best I can do with the resources I have at the moment.
I am working on getting a more modern Intel 64 bit Mac but not sure when (or even if) it’ll show up.
I guess I was a bit overzealous and excited at getting my Mac version rebuilt :slight_smile:
If it shows up, I’ll get it built as 64 bit and posted as soon as I can.

For now I’ve renamed the Mac files to denote “32bit” to prevent that from causing anyone else issues.

The PC version should run under DosBox, and definitely under VirtualBox/Parallels/Fusion.

I wish my Mac hadn’t died :frowning:

I don’t know if it helps at all, but tomorrow I’m going to try and get the raspberry pi version working.

I’m also toying with a VMS (Vax) version as a special request I got, but not sure that’s doable, Vax Pascal is a bit weird.

1 Like

I’m sorry it’s the best I can do with the resources I have at the moment.

No problem. I have a Windows machine I can try it out on.

1 Like

Petter, I’ve posted a newly updated version for Linux and Dos (no Mac update yet)
but version 22 pre-release is now available on all sites I maintain.

I’m no longer too concerned with solveability since I can’t get anyone to test that,
Since this is final pre-release, I’m only interested in show stoppers.

I’ve added a sort of “anti-parser” or “reverse-parser”, which I hard-coded just under 6,000
player commands (and command combinations) covering absolutely everything possible in the game with all items, treasures, objects, etc and including stuff any normal reasonable person might consider (in direct reference to the game). ie: “eat” is perfectly ok, and “keys” are a valid item,
but “eat keys” is totally invalid, and I don’t even attempt to parse it. why bother.
I’m not trying to build a language parser.

I’m just trying to release The New Castle as it was originally intended to be,
and this is that, fully complete. Solveable, yes there is a goal!

1 Like

One word of caution about “anti-parsers”: they make it extremely difficult to know what the error is when a command fails. If UNLOCK DOOR WITH KEYS gets the same error message as PUT KEY IN KEYHOLE and USE THE KEY TO UNLOCK THE DOOR, it’ll be very difficult for me to figure out that UNLOCK X WITH Y is the intended solution, I just needed to type KEY instead of KEYS.

It also leads to bugs where a particular object can be parsed with some verbs but not others: maybe you can TAKE LAMP or TAKE LANTERN, and can LIGHT LAMP, but can’t LIGHT LANTERN.


Such cautions are unjustified in this case, I’m afraid.

The New Castle is a mainframe game from the late 70’s/early 80’s

As originally coded (and I have maintained it exactly) the games “parser” only handles two words.
that’s it, two word max. so “UNLOCK DOOR” or “USE KEYS” is about as close to that as you’ll get.
“PUT KEYS” would function identically to “DROP KEYS” as in PUT/DROP vs GET/TAKE
So the first word is an action of some sort, and the second would be the item/treasure/object.

With only two words, you’re extremely limited in what you can do and then only related to the items/treasures/objects within the game.

I have ZERO interest in a full parser, it would totally destroy the “FEEL” of the game,
and violate the original intentions of it, completely changing it’s essence, as it were.

So the game is coded to be a total 100% recreation of the 1978-era original.
That’s what the original author wanted, and it’s what I want, and intend.

There is no LAMP in the game, but if there were, then yes, I would have coded alternate terms for it.
I’ve done that in all cases that make sense in reference to all items, treasures, objects.

With a two-word limitation, and a small number of items, treasures and objects,
Having a “dictionary” of just under 6,000 commands (and command combinations) is staggeringly huge!

You can examine any of these things within the game, and I’ve expanded “examine” to be a bit excessive. such as “EX”, “EXA”, “EXAM”, “EXAMI”, “EXANIM”, “EXAMINE”, along with
“LOOK”, “L”, “LO”, and “LOO”
but “LOOK AT KEYS” would not be valid because it’s more than two words.

The two-word limitation is spelled out in the user guide, documentation, intro, and instructions, along with “HINT” within the game.
It can’t be more clear than this.

The game is meant to be period-specific and I’m not going to break that.
This limitation makes it more difficult perhaps, but in actuality makes life a lot easier.

Too many games make you spend hours, days, weeks, months trying to figure out the correct word combination, essentially playing a game of scrabble or “guess the verb”,
With this game there is no such concern - you can get/take and put/drop items/treasures
and USE and such but by making it limited makes it easier.

As the game states: if the command doesn’t fit, even loosely, it’ll give you an “error” message.

This is how we ALL played this game in the 70’s and 80’s, and that’s how it’s going to stay.
It’s an exact recreation of the original down to the last decimal. It’s perfect.

Play as it’s meant to be, enjoy it for what it’s meant to be,
and forget the headaches of “guess the verb”, really it couldn’t be more obvious.

To date NO ONE has play tested it to actually solve it! And that annoys me a lot.
They spend months wasting time trying to trip up the parser.
This is my way to put a stop to that nonsense.
Play the game!

1 Like