Archive for the ‘kaves-of-karkhan’ Tag

Kaves of Karkhan (1981)   10 comments

Via the Kaves of Karkhan manual. (Museum of Adventure Game History.)

Five years ago when I was writing about Treasure Hunt (1978) I remarked on a lack of deviation from adventure-genre norms; the Crowther/Woods version of adventure was essentially so good (and already in computer form) that most that immediately followed just copied the model, rather than approach their own way.

CRPGs, by contrast, were trying to adapt a tabletop game, and it wasn’t terribly obvious what form an adaptation would look like, so there were lots of early experiments.

Kaves of Karkhan feels like a game from a parallel universe where the standard text adventure format never dominated. I reckon the reason why it exists in the first place is that it comes from a series which started with a CRPG: Dragon Fire (1981), which was covered by The CRPG Addict in detail here.

Screenshot of Dragon Fire from Mobygames.

According to the manual, the second game, Kaves of Karkhan, uses the same characters as the first: an unnamed warrior, dwarf, huntress, and elf. (The first game had a wizard but you can’t control him in this game for reasons you’ll see in a moment.) The manual tries hard to build lore around these characters, even though they are unnamed:

There is little traveling in this time before the harvest, and a new face arouses much suspicion. Some say the barbarian seeks revenge upon a man with a quarter-moon scar on his left cheek. Others say he’s a professional bandit specializing In the exotic: the left hoof of the centaur, the lost crown of the Faerie King, the eye of the stingbat, and the like. And still others say he seeks to give up his present occupation as fighting man and find something more peaceful, perhaps as an artisan’s or baker’s apprentice. A few insist he flees memories of a lost love.

The story starts directly after the first, where the party defeated an evil dragon and received bucketloads of treasure. The dwarf is busy showing off in a tavern, including a jewel he found “outside one of the rooms on the third level”.

A hairline fracture suddenly appeared in the jewel’s surface.

The dwarf leaned forward anxiously. The crack seemed to be branching off, dividing, but silently. He was amazed. His jewel was crumbling right before his very eyes, but completely without sound.

A shadow suddenly obscured the crack. The dwarf looked up, but there was no one standing over him. He looked down and the shadow was still there, in fact had spread; the shadow crept across the surface of the jewel as If it were liquid. Upon closer examination the dwarf could see that the shadow had issued from the crack.

The gem was a container for a demon named Maldameke who is now breaking free. The wizard manages to contain the demon, for now

“Take the jewel . . . the pieces . . . return them . . . to Maldamere’s home … the bier … the top of the mountain … even one piece … will draw him … back there … trap him … in the Kaves of Karkhanl Hurry! Hurry! Cannot … hold him … long … but beware … beware … his influence … is still … felt … in those underground … realms … “

but the rest of the party now needs to “find your way through the maze of hallways within the crags of Karkhan, solve the traps, and then deliver your piece of the gem into the bier at the top of the mountain”. Each of the four original characters (warrior, dwarf, huntress, elf) picks a team to take along. In actual gameplay, I found no difference between the choice of main character (and you have no interaction with the characters you don’t pick), so the “team” is what’s important.

Yes, ten characters, and you need to keep track of their names and occupations (only in the manual). I used a spreadsheet.

After starting the game, you are told to open the entry doors you must solve an anagram.

It’s always two four-letter words jumbled together, but the words used are random from a fixed list. This one was STEMROPE. There’s lots more valid two word combinations here (like MORESTEP or MOSTPEER) but none of them work.

Then you’re dropped into a randomly generated first-person perspective, and the pain begins.

This incidentally means Kaves of Karkhan is the first 3D-perspective adventure by someone other than Med Systems.

The game moves sluggishly (especially at authentic 1981 Apple II speeds!) and the maze is so random it seems to have no logic at all. You can go down a dead-end hallway only to turn around and find a stairway up has appeared.

The main “gameplay” is a set of randomly appearing traps and encounters, and again, there seems to be no logic to their placement or appearance. A hall with a chasm one moment might turn into quicksand in another. (Only after defeating the obstacle the first time, though — you can’t switch which obstacle you’re looking at just by going back and forth.)

In order to get by an obstacle, you have to type a two-word command. Most of the time it’s USE (character) or USE (item) although there are a few exceptions. Quite often you can lose an item or die by getting it wrong; here’s a transcript of the water obstacle above.


Alana was my (now-expired) sorceress. I quite often would burn through my entire party (ending the game) while trying to get by a single obstacle.

Occasionally there is enough logic to passing an obstacle that I was able to do it first try; when encountering some weeds I tried USE MILES, my farmer…

…but for the most part, on each obstacle, I had to lawnmower down through my entire list of available objects and people.

Here I am getting by a mystic portal by using THROW BUCKET.

While there are some multiple solutions to puzzles (THROW SWORD also works on the above puzzle), I knew if I lost a character or item I could potentially get stuck, so I made generous use of save-states while I took notes on how to defeat each obstacle. My “favorite” piece of absurdity was using my acrobat to defeat a lake of fire.

Oh yes, the game is timed. If you switch emulator speed to “fastest” in order to avoid sluggish walking you get an immediate game over.

I’ve yet to beat the game — I keep wandering the maze in circles — and I may soon just call this one finished. I will still make one more post, because this game represents another stab at the ultra-rare adventure-roguelike genre (where puzzles form the primary gameplay, yet the environment is still highly generative).

I don’t know who to credit for this game other than the company (Level-10). The previous game in the series (Dragon Fire) was made by Rodney Nelsen. The follow-up game (which we’ll get to next, but is very different) was made by Gene Carr. I think it more likely Gene Carr was the author of Kaves (the 3D engine was in the latter game but not the former); however, at the moment I have no proof.

The one person involved with all three games was Steve Rasnic Tem, who did the manuals. At least with Kaves, the backstory is stronger than the game itself! Steve Tem later went on to write quite a few books and win a World Fantasy Award for a novella he co-wrote with his wife, Melanie Tem.

Here’s one last excerpt from the manual to close things out, for now:

Looking around him, once again the dwarf felt vaguely puzzled by the variety of types in the human community. No other race to his knowledge possessed such a range. Packed elbow-to-elbow in the tavern’s central room he could see a skinny youth carrying a rope looped over his shoulders, a short man carrying three companions twice his size, a tall man with his face covered by gray gauze — all shapes and sizes of humanity. The dwarf wondered how humans must keep track of them all; it seemed very confusing to him.

Posted February 3, 2020 by Jason Dyer in Interactive Fiction

Tagged with

Kaves of Karkhan: Finshed!   3 comments

I wish I could say I had some grand strategy, but I pretty much just grinded until the game decided to show me a win screen.

The only thing I did do extra (and I’m unclear if this is really helpful) is I tried my best to go up rather than down. I used ladders and up-stairs when I saw them, but avoided holes in the ground and down-stairs when possible (it wasn’t always possible).

I honestly suspect the bier (the place where a coffin is placed before being taken to a grave, although it doesn’t look like that from the picture) might just pop up randomly when you’re far enough in the game.

I did make some honest attempts at mapping, and it helped a little at the local level; within a particular area the level is at least somewhat consistent, and it’s possible to systematically eliminate corridors as you test them. One thing I only discovered very late is that encounters “eat up” the square of the map you enter, so if you successfully do an encounter, you “jump” to the next square. That means you may entirely miss a side path that would normally be in that square and you’ll only see the path if you turn back.

Above is a typical configuration. I was going “east” and I hit the spot marked “X” and there was a river of blood I used a plank to get by. In the process, I missed seeing the passage that went “south” and had to turn around and enter the X position to find it. It’s possible to “skip” the intersection multiple times if you keep finding encounters there.

One last discovery I made was regarding chests. I hadn’t been able to interact with them (“OPEN CHEST” didn’t work, and trying to use my thief or a hammer or anything like that led to nothing happening). I finally discovered just OPEN by itself works.

Another episode of Great Moments in Game Parser History.

Unfortunately, I discovered this when I nearly was at the end, too late to be useful; I had a method past every obstacle except, of all things, a walled-up corridor.

You’d think a hammer would help here, but no.

Just for reference, the only other games I’ve hit in my sequence so far I’d call roguelike-adventures are Mines and Lugi. While Kaves of Karkhan was bad for personal enjoyment, it’s still fascinating as an artifact of design. Some of the elements — like having a limited subset of available items, and randomized puzzle placement but consistent solutions — seem like they’d make a roguelike-adventure a success, but they fell down hard here.

First, keeping track of 10 characters and 10 items was excessive. It made getting used to the environment rough, and I only felt comfortable after about two hours of gameplay.

Second, it makes for overly simplistic gameplay when each puzzle boils down to finding the right object or character. This is similar to my complaints with Devil’s Palace and The Poseidon Adventure where the authors try for higher difficulty without an adequately complex world modeling system to match. By contrast, Lugi had some persistent effects (like being infected) and puzzles that needed to be solved with objects in combination.

Third, the map was too random to use geography in any rational way. To compare with Lugi again, in that game it was possible to encounter a puzzle in one location, find a helpful object in another, then loop back to the original location to solve it.

Fourth, the punishment of losing objects or characters for failed puzzle attempts was too harsh in context, and it was impossible to reliably survive a loss of resources without already knowing how to solve most of the puzzles.

Fifth, having almost no items found during the process of the game undercuts a lot of what makes an adventure game fun (having an adventure game without the ability to “increase power” with new discoveries is akin to an CRPG that doesn’t let you level up your character). Even an essentially item-less game like Myst at least contains a steady drip of new information and clues.

The only immediate “fix” I could see that would help the game without more extensive design changes would be to allow a lot more alternate solutions. As things stood I was jamming pipes with juggling balls and walking on lava with buckets, and not because I was being creative; they were the only solutions I could find via brute force testing everything I had.

We’re going to have at least one more adventure-roguelike in 1981 — Madness and the Minotaur — at which point I’ll try a grand recap and armchair design of How to Make Such a Thing Work (or possibly, instead, a cautionary warning that such experiments are best left in the early 80s).

Long before we get to that, we have another game from Level-10 that re-uses their 3D engine for a much different game; one only comparable to a handful of computer games across history, and with a mystery that has never been solved.

Posted February 4, 2020 by Jason Dyer in Interactive Fiction

Tagged with