Colossal Cave for MDS

[Jon Hales]

As recently discussed, Kevin Thacker has located listings of FORTRAN code. He has now forwarded the pages to me for scanning.

There are ~100 pages on individual sheets of listing paper with dot matrix print. I have scanned about 30 pages and have been testing OCR on that sample.

A couple of extracts confirm the documentation refers to an Intel FORTRAN version:

1, "This program is a demonstration of FORTRAN 77 on the ISIS-II Operating"
1, "System. It is taken from a game originally developed by Stanford"
1, "University Artificial Intelligence Lab, and derives from the obscure"
1, "game "Dungeons and Dragons". . . "

[...]

Colossal cave is *ALWAYS* open, provided this is your MDS-80.

Incidentally, the text above is uncorrected OCR. There appeared to be significant challenges, which include:

- lines at the page break,

- faint text (a well-used ribbon)

- pages previously printed on the reverse, where there is darker print.

None of these have been a serious issue so far. But there remains a good deal of editing to be done.

Dates on the reverse side suggest the printouts may have been made in 1977.

There are references in several of the 12 sections (each section was stapled together) to a series of INCLUDE files that don't appear to match the printouts.

No ISIS disk has a label indicating any link to the program.

This message is to ask:

- does any member of the group have a disk containing a version of 'Colossal Cave' for the MDS/ISIS?

- is anyone interested to edit (part or all of) the OCR text (which may be my outputs or your own preferred OCR based on scans I can provide)?

- is anyone interested to assess whether it should be feasible to compile the resulting material?

- is anyone interested to compare the resulting material with other versions (as there appears to be no documentation of the origin of this source with (presumably) edits to run under ISIS?

Best regards

Jon

[mark.p...@btinternet.com]

Jon

I am happy to help.

I suspect the changes to the original Fortran sources of the unported  game will be minimal, so it might we worth checking against the original before relying on OCR which is often challenging with old dot matrix listings. 

WRT OCR some training of FineReader might help. 

Mark 

Regards

Mark Ogden

---

From: intel-...@googlegroups.com <intel-...@googlegroups.com> on behalf of Jon Hales <jonh...@gmail.com>
Sent: Sunday, 28 June 2026 13:56:06
To: intel-...@googlegroups.com <intel-...@googlegroups.com>
Subject: intel-devsys Colossal Cave for MDS

 

--
You received this message because you are subscribed to the Google Groups "intel-devsys" group.
To unsubscribe from this group and stop receiving emails from it, send an email to intel-devsys...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/intel-devsys/CANPXPJ%3DaXhx%2B%3DvQuavZbaDVX%3Df5AYn6Y6FDAFyDNYAg%3DUch%3DZg%40mail.gmail.com.

[Jon Hales]

Mark

Many thanks for the offer of help. I'll complete the scanning during the coming week.

The problem with 'the original' is that there are very numerous variants. But as some of the variants exist as source files, I agree that this could be helpful to minimise editing.

Some searching led me to find a blog (in 2007) by a Cambridge resident who mentioned having run the Intel MDS version during a training session at the UK offices of Intel in 1977 - and who mentioned a pile of listings he had in a garage. It would be useful to find another copy of the Intel version, whether on disk or paper.

One of the points I have become aware of is that the original author based many of the locations/scenes/details on an actual 'Colossal Cave'.

Regards

Jon

To view this discussion visit https://groups.google.com/d/msgid/intel-devsys/PAVPR02MB9403C70F1B6060941A29EA33FCE92%40PAVPR02MB9403.eurprd02.prod.outlook.com.

[Mark Gent]

Hi Jon,

Happy to help with (manually) editing the OCR text.

Kind regards,

Mark Gent

---

From: 'mark.p...@btinternet.com' via intel-devsys <intel-...@googlegroups.com>
Sent: Sunday, 28 June 2026 15:42:35
To: intel-...@googlegroups.com <intel-...@googlegroups.com>
Subject: Re: intel-devsys Colossal Cave for MDS

 

To view this discussion visit https://groups.google.com/d/msgid/intel-devsys/PAVPR02MB9403C70F1B6060941A29EA33FCE92%40PAVPR02MB9403.eurprd02.prod.outlook.com.

[William Beech]

Jon,

The only Colossal cave was for CPM-80.

I have seen nothing on any disks I hold for this software.

I am willing to help you with this in any way you need.  I would like too see where this goes.

Bill

[Sid Jones]

Colossal Cave also known as ‘advent’ on the MDS?

 

I do believe that I have a copy!

 

Time to fire up the MDS-888 and set the flux capacitor to 1978...

 

Regards

 

Sid

 

 

 

 

 

 

 

From: Jon Hales

Sent: Sunday, June 28, 2026 1:56 PM

To: intel-...@googlegroups.com

Subject: intel-devsys Colossal Cave for MDS

 

[mark.p...@btinternet.com]

All

The source of the original game (not the MDS port) can be found at

https://github.com/wh0am1-dev/adventure

 

regards

Mark

To view this discussion visit https://groups.google.com/d/msgid/intel-devsys/9333FE4497F6468CABE155D3017009F8%40LM010.

[Jon Hales]

Bill, Sid, Mark (Ogden), Mark Gent

Thanks to all for your responses.

Bill's comment about CP/M led me check a directory of MDS disks for CP/M (disks I had read some years ago). This contains text strings from the Adventure system, together with a couple of files named Adventure.com.

Further checking of ISIS directories has enabled me to locate a pair of MDS disks with the ADVENT files, with the INCLUDE and TEXT files. These are disks I read in 2021.

This perhaps obviates the need for further OCR efforts. I'll spend some time to assess what I have located.

IF these files are 'the treasure', I'll place them on my Google Drive for others to access.

Best regards

Jon

To view this discussion visit https://groups.google.com/d/msgid/intel-devsys/001701dd07b2%2475b014f0%2461103ed0%24%40btinternet.com.

[Jon Hales]

All

The contents of the disk with files for the MDS ADVENTURE (Colossal Cave) game are saved on my Google Drive. The link to access the two files is:

https://drive.google.com/drive/folders/1w-hs5yp93O-hYlm0lEsr1c7FURjdsFwB?usp=drive_link

On this occasion, I'm going to set access as 'Viewer' (rather than 'Editor'). If you feel you need 'Editor' rights, please advise.

I'll leave the files in place until at least the end of July 2026. I have no objection to the files being made available elsewhere.

The two files (given the code number 8101 at the time of reading the disk) are:

CCH8101.zip - this contains the 77 flux transition files (suffix: 'raw'). This set of files allows a disk image to be written to a formatted disk with appropriate use of a Greaseweazle or Kryoflux.

advent8101.zip - contains the files extracted from an IMD disk image with Mark Ogden's 'unidsk.exe'. Martyn Vale's IMG2MDS allows the files to be written to disk via a serial link with a PC.

The files are compressed to (hopefully) enable me to upload the files to the Google Drive without being blocked. I checked that I can re-download the two files without complaint.

To clarify one point: I mentioned finding two disk images. The disk details created by 'unidsk.exe' include an MD5 string for each file. In the case of the second disk, two files were marked as 'corrupt'. A comparison of the two lists of files (@II-ADVENT-SYS-DB) confirmed that in all other cases, the MD5 string matched between the two disks.

I have not examined the files exhaustively. It appears that in some cases (e.g. where two files have a similar name), one is a text file and the other is a version that has been compiled (or similar) - recognisable as the same text data, but with additional codes.

I believe the set of files is complete. I would be interested to know whether anyone finds the disk allows the game to be played. 

The relationship to earlier versions is described in the file f1.1 - which seems to assume the user is going to run the game under RT-11 and lists the changes made for that context.

Best regards

Jon

[Sid Jones]

Well lookie there...

 

My copy of ‘ADVENT’ on an 8” DD ISIS-II floppy from 1978 appears to have the set of Fortran-80 source files, the resource files and a CSD file to build it...

 

(But I don’t have Fortran-80 for ISIS-II...)

 

Due to having to copy stuff out at a really slow rate, I’ll take a day or two to get set up and move them into the PC plane of existence...

 

And I can try running it on a PC using ISIM85.

To view this discussion visit https://groups.google.com/d/msgid/intel-devsys/CANPXPJmiy8Ntr76jJ3DHfyBG33etR0dhYCk6_ycmNzYenJg%2Beg%40mail.gmail.com.

[mark.p...@btinternet.com]

All

I have tried to build some of the code from source and it is clear that the source files are inconsistent. At some point variables were changed to be LOGICAL or CHARACTERS but not universally. This leads to COMMON area size mismatches.

The good news is the ADVENT program itself contains the debug symbols so there is a chance at some point we can match the code, assuming that the version of FORT80 is one of those generally available e.g. 2.0 or 2.1

To view this discussion visit https://groups.google.com/d/msgid/intel-devsys/9F589BA70190446092B24074B42828C2%40LM010.

[Herbert Johnson]

Mark, do you happen to know if the Microsoft CP/M Fortran-80 is
substantially different from the Intel ISIS-II Fortran-80? Did Microsoft
supply Intel with Fortran? MS seemed to supply everyone else with
Fortran (TRS-80, HDOS for Heath).- regards Herb

On 6/29/2026 4:27 PM, mark.pm.ogden via intel-devsys wrote:
> All
>
> I have tried to build some of the code from source and it is clear that
> the source files are inconsistent. At some point variables were changed
> to be LOGICAL or CHARACTERS but not universally. This leads to COMMON
> area size mismatches.
>
> The good news is the ADVENT program itself contains the debug symbols so
> there is a chance at some point we can match the code, assuming that the
> version of FORT80 is one of those generally available e.g. 2.0 or 2.1
>
> Mark

--
Herb Johnson, New Jersey USA
http://www.retrotechnology.com or .net
preserve and restore 1970's personal computing
email: hjohnson @ retrotechnology dot com
or try later at herbjohnson @ comcast dot net

[mark.p...@btinternet.com]

Herb
I suspect it wasn't supplied by Microsoft, the compiler appears to be written in PL/M and consists of multiple overlays. It also has various optimisation levels not present in the Microsoft Fortan compiler
Mark

-----Original Message-----
From: intel-...@googlegroups.com <intel-...@googlegroups.com> On Behalf Of Herbert Johnson
Sent: 30 June 2026 05:14
To: intel-...@googlegroups.com
Subject: Re: intel-devsys Colossal Cave for MDS

--
You received this message because you are subscribed to the Google Groups "intel-devsys" group.
To unsubscribe from this group and stop receiving emails from it, send an email to intel-devsys...@googlegroups.com.

To view this discussion visit https://groups.google.com/d/msgid/intel-devsys/2b03001c-3207-4749-83ce-46dd3f5defca%40retrotechnology.com.