Despite this being a direct plot continuation of Brian Howarth’s Arrow of Death Part 1 (which featured here last year) this marks a transition point in his work.

From Mobygames.
Specifically, his first three games in his Mysterious Adventures line were originally done for TRS-80 only. All further parts (picking up in 1982) were written using the Scott Adams database format, matched exactly enough that the same interpreter can be used to play Scott Adams and Mysterious Adventure games. Essentially, this meant removing “room descriptions” and relying on room names and items (and later, graphics) to create the environment. While it seems something of a loss the portability was part of what led Mysterious Adventures to be successful in the first place. The TRS-80 never lit the UK market on fire.
As far as how this happened: based on a note in this interview, Howarth had seen an article by Mike Woodroffe of Adventuresoft asking for programmers to port the Scott Adams games over to British computers.
Quoting the interview:
A non-programmer friend of mine was very keen to be part of creating adventures and had come across an editor that could compile/create and interpreter that could digest Scott Adams’ adventure data files. It became clear to us that if I could adapt my code to be able to interpret the Scott Adams data files, we would also be able to use the editor to allow non-programmers to write our own adventure data files, then package them up into my new engine and supply Molimerx with their voracious demand. My code was pretty compatible to the way the Scott’s code worked and only required some massaging to be compatible.
So the adventure series was now expanding nicely, but apart from the TRS80 platform, Molimerx only wanted me to port the code to IBM PC. He had no interest in any of the flood of new machines that were starting to saturate the market in the UK at the time. My new targets for porting the engine were machines such as Atari 400/800, Sinclair Spectrum, BBC Computer, Commodore 64, Oric Atmos (seemed like a new machine appeared in the UK each month). From this, I became so embroiled in porting that forward motion on creating new titles ground to a halt.
Something feels a little off to me here — supposedly Howarth’s attention on the Scott Adams format came from Woodroffe’s article, but the actual block quote seems to imply “a non-programmer friend of mine” had “come across an editor” (I would guess The Adventure System). My best reconciliation of the two stories is that Howarth learned about The Adventure System first, saw the Scott Adams format was well-documented, then saw the Woodroffe article, then used his memory of the documentation to be able to make Scott Adams ports.
Also, what does Howarth means about his code being adaptable? If he’s referring to the first three TRS-80 games in the old format, they are rather different and using the database required a complete ground-up rewrite of not just the game system itself but the games themselves. Scott Adams also said (without being specific) that “my recollection of a few of the items may be a bit different” so I suspect there might be some fudging of the sequence of events (perhaps unintentionally, we’re talking about events in 1982). Gareth Pitchford’s run-down of events comes with receipts but I’m still not confident on how things really happened.
Nevertheless, all we’re really worried about in the end is the game itself; to continue from Part 1 I’m going to stick with the ZX Spectrum.

A weirdly existential opening.
To recap, several games ago we found a golden baton, but in Arrow of Death Part 1 we found the baton to be corrupted by some sort of distant evil named Zerdon. In order to defeat Zerdon we needed to get the parts intended to form the Arrow of Death. We successfully found all the parts before the story cut off but had yet to make the arrow.

I wonder what the people who started with Wordier Part 1 thought when they got to this game.
You start out at the edge of the marsh things left off on last time, next to a plain and a chasm. The area appears to be empty for a shrub which is a “bundle of fluffy leaves”, and trying to DIG at the plain (unprovoked, I just had my tingly Adventurer Sense going) reveals a FLINTSTONE. There’s also a “narrow gorge” with “water at the bottom” yet I think this part may be meant just as scenery.
The chasm (see above) can be JUMPed into although it’s a one way trip. This suggests a linear structure but the game subverts that later; still, for a little while only one or two puzzles are presented at a time.
You land near a rope bridge; if you try to cross it a bird flies overhead and drops an iron helmet (which you can retrieve on a lower ledge). The lower ledge is also next to a crevice with a lamp.
Going past the rope bridge leads to a “straggly weed” you can pick up, and then an “iron grille” with “machinery”, which I have yet to be able to do anything useful.
My guess, based on my inventory (which includes a sword from the last game, as well as the pieces of the Arrow) was that I’m supposed to use the leaves and weed somehow to make a fire that connects with the grille? No verb I’ve thrown at the items gets recognized, neither MAKE FIRE nor MAKE SPARK work, so I may just be barking up the wrong tree.
(It took me later in my playthrough before I tried it, but I might as well spoil now what the weed is really useful for is eating; it increases your strength, rather like in Katakombs.)
Going back to the rope bridge, the room is described as having “ropes holding the bridge up”. You can cut the ropes and die. I originally thought perhaps this was a trap or gag, and if this was Acheton that’d be the case, but it doesn’t quite fit in with the Howarth style: once he switches from minimalist to ultra-minimalist, everything is important.

I didn’t find out until much later that if you’re not holding the fluffy shrub, this maneuver kills you; the shrub acts like a pillow.
It turns out this lands the player in a hub of sorts, although it isn’t obvious at first. Here’s the map as it looks on the first pass:
There is a locked door that blocks one route and a “heavy door” that blocks another. Otherwise, there’s a dead guard that can be searched for a uniform, a “wheel” that can be turned (next to a guard; you should disguise yourself with the uniform first)…
…and a “heavy kite” by a platform and the place where the dead guard with the uniform was.
If you wear the iron helmet from the bird you can safely JUMP while holding the kite and float down to the start of the game.
So that means the structure is (so far):
Area 1 -> Area 2 -> Area 3 -> Loop back to Area 1
It took me a little while to realize that the wheel turning is what makes the loop useful. By turning the wheel, you shut off a source of water in a gorge, leaving only mud that can be jumped into.
Searching the mud yields a lever that can be pulled, yielding a passage to a new area I haven’t fully reckoned with yet. There’s a dungeon with a prisoner locked behind a grating:
A “grotesque animal” at a “guard room”:

There’s no guard mentioned in the room, so it is strange the illustration shows a guard rather than the grotesque creature.
A “starving mule” at an “underground stable”:

Feeding the mule some of the strength weed doesn’t work.
And a “temple” with a “tapestry”:
There’s a button hiding behind the tapestry. Pressing it reveals an altar. Looking at the altar reveals a candle. Lighting the candle summons a “column of flame”. Once the column of flame is summoned you can PRAY, and the game claims “Something happened!”
What that something is, I don’t know; this is as far as I’ve gotten. I will say this has been more enjoyable that Part 1, not so much in the puzzles, but in the map. In addition to all the things found in the new area, there’s a bolted door. The bolted door leads back to the wheel/kite area!
Area 1 -> Area 2 -> Area 3 -> Loop back to Area 1 -> Area 4 with Area 3
To be clear with a map:

Red = 1, Orange = 2, Yellow = 3, Green = 4. Green connects back up with Yellow.
This kind of unexpected map interconnection I’ve found to be one of the most satisfying element of adventure games (or really, any games). It does make things slightly more complicated on the linearity front; when you know a particular obstacle will never again be seen, that restricts what objects might be helpful to that obstacle quite seriously, whereas if an obstacle can be returned to much later, it potentially opens up any object in the game. This is the type of tradeoff I’m willing to make for the feeling of a world with more depth than it has at first appearances.
The most curious thing is that the part that I had to skip while still baffled — the mysterious grille with machinery — is the only part I can’t return to. While you can go from Area 1 back to Area 2 by jumping down again, the bridge is destroyed so there’s no way to get across to where the grille is. The fortunate thing is that this isn’t Hezarin; the move count is low enough it won’t be hard to repeat everything, if that turns out to be necessary because of some clever trick missed earlier.
For once, I’m not “stuck”, just “stopped”, but historically I’ve hit a wall with Howarth before, so we’ll see how far this goes.

My verb list, for reference. Orange are verbs recognized by the game. It’s a fairly generous spread; the “cut rope while holding it” maneuver I figured out much more quickly from knowing that HOLD was a possible verb.
Oh, there’s definitely more to be discovered/cleared up around the engine change to a Scott Adams compatible format. (I think that might be down to an issue with the CASA interview itself, but I’ll come to that in a minute)
When you’re interviewing someone about something from forty years or so ago it’s incredibly apparent how common it is for dates to be misremembered, time to get compressed and events get switched around, totally unintentionally changing a narrative. It’s rarely intentional but it is amazing what gets confused with time or even forgotten.
In the case here I guess we have to go back to the contemporary sources. Mike Woodruff’s plea for a programmer (to port the Scott Adams games to the micros favoured in the UK at the time) came when his Calisto Computers was acting as the UK agents of Adventure International. It was published in Computer & Video Games issue 25, page 147… in November 1983. That’s well over a year after Arrow of Death (Part 2) and Escape from Pulsar 7 were released. And well after Brian set up his own Digital Fantasia label.
So, yeah, on face value the whole timeline needs re-examining. But I think that it is the structure of the interview is being misleading rather than Brian. The interview on CASA is punctuated by some italicised “framing” paragraphs that aren’t part of either the questions or Brian’s answers. They’re what are implying that Brian saw Mike’s plea first etc. Brian doesn’t mention that seeing the news piece with Mike led to the engine switch. He is saying that was done to satisfy Molimerx’s demand.
I know I’ve misread that interview myself and duplicated that information elsewhere, so it’s definitely confusing. It seems clear, re-reading it again and looking at the sources, that switching to a Scott Adam’s compatible system was nothing to do with porting them to the UK machines (that came much later) but merely down to a desire to simplify the process churn out adventures more quickly and (it’s implied, I think) to try and get someone else to help write them.
Do you have a copy of the Retro Gamer one that you linked? It has been taken down from Internet Archive. (I think I used to own a physical copy of that one but I had to cut down — I just was curious if there was anything different in that interview.)
One other element I didn’t toss in (the intro was messy enough already, and I need to investigate this aspect further) is that the TRS-80 version uses what appears to be regular Scott Adams interpreter straight up. No porting done. Not even the Adventure System remake.
I’ll have a check but I’m not sure if I saved that one offline. It is visible (for me at least) through PressReader… https://www.pressreader.com/uk/retro-gamer/20180614/281547996584333
Re. the TRS-80 version using the Scott Adams interpreter… that is interesting. Although, given the Mysterious Adventures are some of the most ported and re-released adventures ever made, I guess it might be hard to know for sure where that archived version(s) originated and if it was the initial Molimerx release. There seems to be a lot missing from the Molimerx catalogue (such as the IBM PC ports).
Looking back at that article… it does mangle the timeline in a similar way to that unintentionally done by the CASA interview. There’s no way those events happened in that order with those dates.
The article is still helpful, insofar as it is really explicit that Howarth had the documentation but not the Adventure System itself. The timeline also works out for when the Adventure System started to get sold (end of 1981) so he would have been able to get a hold of the document between games 3 and 4.
Pingback: Arrow of Death Part 2: The Fire-Walker | Renga in Blue
I’ve recently played Adventureland for the first time (12 treasures so far; trying to figure out what to do in the lava room) and I just realized that the version I’m playing (C64 via Retroarch) was ported by Brian Howarth. I think he did a pretty good job. The text is apparently identical to the original and the images are quite nice. For such tersely written games I think there is no reason not to play the graphical versions. The only boring thing is that you can’t have the image and the room description up at the same time. They fixed it for #2 (Pirate), and also added items to the images, when they are on the ground.
I’ve been pretty loose about which version to play — I’m happy to pick graphics even when the first version doesn’t have it. (Enough of these games have missing versions I can’t afford to be picky anyway.)
after you’re done check out a video of the PC-88 japanese version, it has pretty interesting (and different) graphics
Wow!