The Stack

The Wizardry magic system has a skewed symmetry to it. Spellcasting enemies and monsters pull their powers from the same spell list as the player. That may not be so notable today, but it was not something you took for granted in CRPGs of the time. Mainly what it means is that anything you come to rely on to give you an edge in fights will eventually become something the enemy can do as well. Until you can defeat the final boss, there’s no perfect, unassailable supremacy. There’s just an arms race.

I call it “skewed” because, despite perfect symmetry in what spells are available, there’s asymmetry in how spells are used, and it’s produced by the basic asymmetry of a singe party going up against an entire dungeon full of monsters. The player has to worry about wasting spell slots that they might need later. For the monsters, there is no later. They literally only exist for the duration of an encounter. As a result, it makes sense for them to use spells that I’ve declared “not worth it”, especially when they outnumber you and can cast them in quantity. I particularly noticed this back with the single-target instant-death spells used by the Priests of Fung back in Wizardry III, but I’m seeing it more and more with Wizardry V‘s expanded spell list. There’s a spell HAKANIDO that “drains magic”, expending some of its (single) target’s spell slots. This can be devastating when cast on your own casters, who have to continue to deal with the loss after the fight is over, but doesn’t seem worth using on monsters in a normal encounter. It’s a spell that’s in the spell list mainly just so enemies can use it against you, putting it in the same category as things like Animate Dead in D&D. Nonetheless, it’s there for you to use if you want to, and maybe it’ll become more useful at some later point, when I’m facing enemies with high-level spells and preventing their casting is of paramount importance.

Plus of course there’s some trivial asymmetry in what spells are even useful to monsters. There are a number of utility spells that are only used outside of combat, which means combat encounters never use them. There are also combat spells that are only useful against specific types of target, like ZILWAN (does massive damage to undead) and MAGATO (banishes demons). But that’s a relatively trivial matter.

I had a bit of a surprise recently. Remember how I said that there was an NPC (named the Ruby Warlock) blocking a passage, and he wouldn’t move unless I gave him something to drink, and after a few iterations of this I got tired of fetching drinks for him and decided to try fighting my way past him? Well, once you’ve killed him, he doesn’t show up at his post any more, the encounter replaced with a pack of “demon imps”. Before this, it had been a rule throughout the Wizardry series that enemy deaths aren’t permanent — that anything you kill in a fixed encounter comes back when you leave the dungeon level. Even Wizardry IV, the one game in the series to have any persistent effects on game logic that weren’t embodied in inventory, kept all deaths temporary.

That isn’t the surprise I mean, though. That came later, when I had occasion to go to the Temple of Cant in town for the first time in a while — my Priest had gotten paralyzed, and my Bishop hadn’t learned how to cure that yet. The UI for the Temple is a menu of all the dead or otherwise disabled characters who you can pay the priests to help. And there in the list was the Ruby Warlock.

I’ve confirmed since then that other NPCs show up at the temple when killed, and can be resurrected. (In a way, it seems a little unfair. When my guys die, I have to drag them out of the dungeon before the temple lists them.) So it’s basically an elegant general way to accommodate the player’s desire to solve problems through violence, and still provide an undo button in case you kill someone with important dialogue. Wiz5 is the first game in the series to need such a system, because it’s the first one to have killable NPCs who have functions other than being killed.

A few other random observations on this:

• If you’re the kind of person who does genocide runs in Undertale, you can use the temple as a sort of trophy case, preserving the bodies of those you’ve offed.
• The whole Wizardry system supports passwords on individual player characters, presumably to support multiple players playing from the same disks. It’s easy to imagine two players getting into fights over whether a given NPC should be dead or not, one player repeatedly killing them and the other repeatedly resurrecting them.
• I’ve talked with enough NPCs to get the impression that the final boss is a demonic being known as the Sorn. I wonder if the Sorn is resurrectable?

OK, let’s take a good hard look at the new spell names and see what they tell us about the implied magical language. We’ve got fully 29 new spells — nearly half of the list of 63, mostly for Mages rather than Priests. A couple of them — LADALTO, LABADI — are just an obvious extension of what was already there, adding the prefix LA-, as previously seen in LAHALITO, to other existing spells. (LA- just means “the effect is even stronger than you get with the prefix MA-“.) Similarly, MAMOGREF is immediately recognizable as the common intensifying prefix MA- applied to the existing protection spell MOGREF, and this is borne out by its effect.

A few of the new names just confuse me. DESTO opens locks, but previously -TO only ended attack spells. Maybe it’s conceived as attacking the lock? I guess in a way it’s overcoming defenses? LITOFEIT is obviously connected to the previously-existing LOKTOFEIT, but LOKTOFEIT is the “teleport out of the dungeon” spell, whereas LITOFEIT is levitation. I suppose the common factor is rising up, Wizardry III notwithstanding. I’d feel a lot more comfortable about this if LOK- or LIT- appeared anywhere else. MOLITO, which does group electrical damage, seems to have been replaced with a weaker, lower-level version called MELITO. This is a pattern not seen elsewhere, and it frankly seems a little grammatically implausible to me.

We know there’s a prefix LO- from the existence of MILWA (temporary light) and LOMILWA (permanent light). We now have a spell LOKARA that looks like it has the same prefix, but although KARA looks reasonable, it isn’t on the list. What’s more, what LOKARA does is attempt to make the earth swallow up foes (an instant-death effect to which flying creatures are immune). What would KARA do, make the earth swallow them up for a little while?

Both Mages and Priests get a spell that conjures elementals: SOCORDI and BAMORDI, respectively. I note the root DI (life) in there, which seems fitting. -ORDI could mean elemental, in a sense like “brought to life” or “artificial life” or something, the remainder of the word indicating two different ways of going about it. And the BAM in the Priest version could shed light on BAMATU. I had wondered about that — it made it look like BAMATU should mean “opposite of MATU” when in fact it’s a stronger version of it. What if the prefix is actually BAM? We see -ATU in a couple of other new spells: BOLATU (petrify one creature) and KATU (charm NPC). This does not convince me it’s a morpheme.

CORTU and BACORTU are an interesting pair. CORTU creates a “screen” that blocks any enemy spells cast at the party. BACORTU creates a “fizzle field” that prevents a single enemy group from casting spells, offensive or otherwise. These are both new concepts, and they’re distinct things, mechanically. But they’re hardly opposites. Maybe I’ve got entirely the wrong handle on BA-. Maybe instead of “opposite”, it means something like “counterpart”, or even “enemy”: just as DIOS is a spell you cast on your friends (to heal them) and BADIOS is a presumably similar spell you cast on your enemies (to hurt them), so too is CORTU a protective effect on your friends and BACORTU a limiting effect on your enemies.

Mainly, though, I’m probably trying too hard to make sense of things. The new spells, for all that they try to harmonize with the legacy ones, are the product of a different designer’s mind. Who knows how much of the intent behind the names was communicated to him and how much he made up?

Progress continues apace, although the sheer amount of dungeon to explore means I’m still on level 2. I’m getting a strong sense that each level is divided into multiple distinct areas, in irregular shapes, where each area has either one or two major puzzles in it. Wiz4 was kind of like this too, but there the areas were entire levels, and here they’re just not-quite-isolated sections of grid that fit together in the overall map like jigsaw pieces. It’s tempting to declare that the grid doesn’t matter, that all we really need to do is identify the areas and how they relate to each other, but the grid is still important for identifying empty space where undiscovered rooms can be found.

A lot of the puzzles are lock-and-key in nature, but slightly disguised: a lock in the form of a machine with a token slot and a key in the form of a bag of tokens for it, or a door held shut with chains that you need a hacksaw to open. That one counts as a case of Caesar’s Ladder, I suppose: you have to find a hacksaw somewhere in the dungeon simply because the world outside the dungeon is unreasonably small and shallow. That’s something that hasn’t changed.

At one point, there’s a wizard who blocks your way and demands alcohol. This is a twist on the lock-and-key puzzle you only get in RPGs: ones where if you don’t have the key, you can beat up the lock instead. I had what he wanted, but unlike most keys, giving it to him uses it up, forcing you to fetch it afresh from a different area (guarded by a hurkle beast) if you ever want to go that way again. So on my third pass, with considerable trepidation, I tried fighting him. This is the sort of thing that some previous titles would punish with hard and sudden death, but we’ve got a new designer on board, and it wasn’t too bad. My party lost more health than I liked, though. I’m not trying it again any time soon.

Some years before embarking on my current Wizardry run, I randomly found a nicely-bound quadrille-ruled notebook in a garage sale. It had just over 20 squares on the horizontal axis, so I declared it to be a perfect medium for 20×20 Wizardry maps, and tucked it away in anticipation of the day that I got back to the series. The satisfaction of having all my maps officially squared away together like this has been a significant part of my motivation for continuing to play the series over the last couple of months.

So imagine my dismay on discovering that Wizardry V breaks the bounds! Dungeon levels here are significantly lager than 20×20 — I don’t know what the maximum size is, and the shape of my maps so far doesn’t give me much of a hint. What’s more, the levels extend in all directions from the origin. In Wiz1-4, once you could find map coordinates with DUMAPIC, you could confidently place (0, 0) in the lower left corner of your paper. You can’t do that here. Or, well, you can, because it doesn’t really matter where you start; you’re probably going to go past the edge at some point no matter what you do. You just have to accept that your maps will be in multiple pieces and try to find good places to break them.

Despite my complaint, I do think this is an overall good thing for the game. For one thing, recall all the complaints I’ve made about how short the mapping and exploration phase of each level was relative to the amount of grinding needed to survive on the next. Making the exploration phase last longer is certainly one way to solve that. But I think the larger impact is simply the lack of any sense of finality. When you’ve seen every square of a 20×20 grid, you know you’re finished. When you don’t know how big the map is, there’s always a possibility that there’s more you haven’t discovered.

Indeed, I’ve already hit a couple of points where I thought I was probably done exploring dungeon level 1 (apart from a few locked doors that I don’t seem to be able to get through yet), then thought of something else to try, and was rewarded with new territory to explore. This is something that this game is doing a lot better than its predecessors, and it requires not just the map size but multiple ways of locking off content to enable it. Hence all the changes to how doors work.

Well, I’ve found the plot. In the north of dungeon level 1, there’s a temple full of mad priests, some of whom attack you, but the head priest is willing to parlay. He tells you about the bad guy du jour and his his plan to unleash massive destructive forces. Ho hum. I suspect that this is connected to the cataclysms in Wizardry III (which, remember, didn’t go entirely resolved in that game), but I haven’t seen this stated explicitly.

The reason I’m devoting an entire post to this dialogue is that it’s interactive. That’s a first for the series — we’ve had riddles before, but not conversations. It’s essentially what IF fans call an ask/tell system: you type in topics to ask the priest about, in the hope that they’re in his dialogue table. From other games, I know this is a system prone to problems. It’s all too easy to author the content in such a way that essential information is either easily missed, locked behind topics that the player doesn’t guess, or accessible out of sequence, easily-guessed topics referencing things you haven’t been told yet. But we’ll see.

I suppose it’s a case of Wizardry expanding its ludic vocabulary and experimenting with what Wizardry can be, but it feels fairly mild in that regard. Wizardry IV was all a grand experiment, twisting both form and content in unanticipated directions. Key-word dialogue is new territory for Wizardry, but it’s already well-trodden ground. Ultima had been using it for ages by this point.

Going into Wizardry V immediately after the first four games feels a lot like switching from Wizardry to Might and Magic did. There’s just a lot of little changes. The wireframe dungeon graphics are unaltered, but the text font is different and the monster pics are bigger and more finely detailed. Secret doors work completely differently and apparently locked doors are a thing now, although I haven’t seen any yet. Thieves have new Hide/Ambush actions to make them more useful in combat. There’s a whole system of weapon ranges to complicate the front row/back row stuff. There are bows.

The spell lists keep mainstays like MAHALITO, DIOS, KATINO, and TILTOWAIT, but throw out a lot of the useless and redundant stuff and add in new things with a greater variety of effects. There are spells that add new concepts like summoning elementals and making screens that block enemy magic — way too high-level for me at the moment, but they’re something to look forward to. There are spells to deal with the new dungeon mechanics, like one that unlocks doors. There’s a levitation spell that sounds like it duplicates the effect of the winged boots in Wiz4, but in a more baked-in, systematic way. And with all the new spell names, we get a bunch of new morphemes that I’m going to have to analyze at some point. (I don’t think they’re being entirely consistent with their word production, but that’s okay. Natural language isn’t entirely consistent either.)

Things are so changed that the game doesn’t even try to support importing characters from previous titles. For the first time since Wizardry I, the game has its own character creation system. It seems to be basically the same as in Wiz1, with the same classes and races and stat ranges, but the end result of creating a character is changed in one significant way: instead of each character getting a small amount of gold to buy equipment with, they all get some basic starting equipment, which can’t be sold for cash. In other words, you can’t prepare for your first delve by pooling the party’s gold and buying good gear for the front line. Possibly for the first time ever, I’ve sent fighters into the dungeon wearing nothing but leather armor.

Fortunately, the first dungeon level is gentle enough that leather is all you need — I’ve gotten my entire party up to experience level 2 without any deaths at all, just like back in Wiz1. Just one more sign that, in contrast to 2, 3, and 4, this game was intended as a jumping-on point for new players. I’d be thinking of it as a reboot, if it weren’t for those fan sites that described it as part of the Llylgamyn Saga.

Is it connected to the rest of the Llylgamyn Saga in plot? I don’t know; I haven’t seen any plot yet. Not even the manual contains any indication of why we’re exploring this dungeon or what the Heart of the Maelstrom is — a message in the game seems to suggest that the dungeon is the Heart of the Maelstrom, but that’s all I know.

What I am getting is a strong impression of refinement, of the designers adjusting the rules to make it a better expression of what they intended all along, in ways that they couldn’t do while it was all technologically linked to Wiz1. Even the starting equipment is part of it, forcing players to not skip the initial part of the intended upgrade chain.

Heavy spoilers for the endgame here.

When I saw Hawkwind of Skara Brae in this game for the first time, I wondered about him. I had previously seen the name Hawkwind and the town of Skara Brae in the Ultima series, and possibly also knew Skara Brae as the setting of The Bard’s Tale. There had to be a connection, but it was harder to get information about such things in the days before the Web. About all I could learn was that Skara Brae was the name of an archeological site in Scotland.

Things are easier now. The common factor is Roe R. Adams III, who worked on all the games in question. Hawkwind is his gaming persona, and thus his appearance here is a self-insert, and a self-insert of the most comically aggrandizing sort. Hawkwind is completely unaffected by all attacks, physical or magical, and in battle he just ignores you: instead of attacking, he performs such actions as brewing a cup of tea, reading the newspaper, and taking a nap. His only weakness is the weakest monster in the game, which makes for a substantial and satisfying puzzle, because there are several other encounters between Hawkwind and the nearest pentagram, making it difficult to keep the fellow alive.

Hawkwind is far from the only bit of inside-jokery here in the final chapter. A party called “The Softalk All-Stars” is obviously a shout-out. There’s a whole slew of encounters with various Orders and sundry nobility, all of whom are individually named not just in combat, but in the text that appears before, just to make sure you see everyone’s names even if you don’t fight them. According to sources online, all these are the personae of Adams’ friends from the Society for Creative Anachronism.

These various Orders and Barons and so forth provide one route to victory: by means of a bunch of optional tasks in the dungeon — including, but not limited to, that witch’s brew puzzle I’ve mentioned before — you can gain the favor of the groups, one by one. They’re not loyal to Trebor at all, it turns out, and will gladly make you king if you can give them all what they want, a touch that comes off as satirical. (Heck, the password to enter the castle is “Trebor Sux”.) One group will support you in return for a million gold pieces, which was a bit of a shock — you get gold from each encounter, but it’s mostly useless, and the few places where you can spend it ask for such paltry amounts that I hadn’t been keeping track of it at all, but it turned out that my entire ascent of the dungeon yielded less than half what the Captains were asking. Fortunately, there’s a fixed encounter in the Training Grounds that yields gold way out of proportion to its difficulty, presumably included just for this puzzle. But for the most part, this route is the least fighty and most puzzly way to end the game, and doesn’t involve challenging Hawkwind or the All-Stars at all.

It’s also in some ways the least satisfying: becoming king means abandoning your quest for the Amulet. Is it really absolute power if you can’t do the one thing you’ve had your heart set on from the very beginning? No, let us not be satisfied with being a mere king when godhood is within our grasp.

Of the four remaining endings, three are just minor variations on each other: the Amulet is in the hands of a statue of the god Kadorto, who you confront with a set of essential key items and one of the three swords from the altar atop the ziggurat. The choice of sword determines what happens, and the endings contradict each other about the nature of Kadorto and of the Amulet in ways I’d find troubling if the game weren’t doing so much to keep us from taking it seriously at this point. But all these endings, including the one where you ingratiate yourself with the nobles and become king, end with the same words: “Have you forgotten something?” It’s a clear ploy to make even the most satisfying ending unsatisfying, to let you know that you should really be going for the fifth ending, the Grandmaster ending.

The Grandmaster ending is another variant on the three sword endings, but using a secret fourth weapon, the Kris of Truth, that you get from a secret eleventh dungeon level underneath the spot where you start the game. The ending begs the player to not reveal the details, and I took that seriously for a time, but the statute of limitations has definitely run out for this one. Level 11 is a network of small rooms joined by hallways, with one bending path leading through all the rooms, at one point crossing a deadly void that can only be traversed with the right equipment. In each room, a person asks a riddle. For example, in the first room, you get:

You are in a yellow room. Before you stands a being dressed from head to toe in deepest yellow…

“I am Discrimination at my best, and Avarice at my worst. I symbolize the part of the body upon which all the rest stands. What part am I?

To which you answer “feet” and are allowed to move on. Every riddle is like that: some colors, some seemingly-disconnected attributes, and finally the bit that, weirdly, lets you actually guess a body part. This continues until the final riddle, which is just really beautiful:

The answer to the Greatest Question is also the simplest. Upon what paths have you trod? Where are you?

In other words, where all other riddles in the entire series are embedded in the environment, this one is about the environment. Answering it requires paying attention to not just the contents of the rooms, but the map, and how it’s connected. It’s a bit squashed to fit into a 20×20 grid, but once you’re looking for it, it’s clear: the rooms and pathways form the Tree of Life, the Kabbalistic diagram beloved of 20th-century occultists in the Hermetic tradition. The diagram originates from Jewish Kabbalah, but my understanding is that it plays an extremely minor role there, whereas Aleister Crowley and his ilk were fairly obsessed with it, drawing all sorts of things into correspondence with its nodes and pathways, including the various colors and body parts seen in the riddles here.

Wizardry IV was released in 1987, in the midst of the Satanic Panic. One of the more ridiculous claims of the anti-Satanists of this period was that the magic systems in RPGs such as D&D are based on real occult rituals. Here in level 11 of Wizardry IV, that’s closer to true than it’s ever been. Not the magic system per se, but the content, at least, is drawing from the writings of real occultists, and it’s doing it to add power to an experience, to a moment of realization, by associating it with ancient wisdom and divine revelation. But honestly, I think the moment would be pretty powerful even without the cod-hermetic stuff, just by being a question that provokes a realization about what’s all around you.

In the Grandmaster ending, the newly-enlightened Werdna, on obtaining the Amulet, sees it for what it really is: an object whose sole purpose is generating conflict. And so he discards it, having moved beyond its allure. Slightly ironic, then, that reaching this ending the first time provided me with a keepsake I’ve kept ever since: the ending provides a code that, once upon a time, you could use to send off for a special certificate of completion. I’ve seriously thought about getting this thing framed and hanging it on the wall like a diploma, even if I am somewhat less in awe of the game than I was back then, and more willing to call my inability to solve some of the puzzles a fault in the puzzles rather than a fault in myself.

But then, if I’ve relied on hints more than I’d like this time through, it was to some extent simply because my relationship to games has changed. One of Wizardry IV‘s main repeated tricks, starting way down at level 9 and continuing to the castle roof, is simply expecting the player to diligently check every single instance of a pattern to find the one exception. And that was a much more realistic thing to expect of me when I had fewer games, and needed to eke out as much play from each one as I could. I can’t really recommend the experience today, but it was amazing thirty years ago.

You spend the bulk of Wizardry IV with one goal: reaching the surface and escaping the dungeon. But achieving that goal isn’t the end of the game. It’s just the start of the endgame, set at the castle that was the player’s home base in Wizardry I, with all its familiar features: the Training Ground, Boltac’s Trading Post, etc. I spoke of the lower levels in Wizardry II as a power fantasy where you wipe the floor with immensely strong foes. The castle levels have another kind of power fantasy: facing off against low-level characters for the first time in ages and seeing how far beyond them you’ve become.

Where the Cosmic Cube gave us the first multi-layered experience in the series, the castle takes things a step further by giving us a coherent three-dimensional space: three levels designed as a geometrically unified whole. Where Wizardry III gave us a castle that was obviously fake, just a flat diagram of a castle, Wizardry IV gives us towers that are actually towers. And it’s made all the more satisfying by the way it pays off the unfulfilled promise of the three previous titles, turning the least real of places, a menu tree that we had to pretend to believe was a place, into the most real.

But at the same time, it’s struggling with the limitations of the engine again. When you get out onto the castle walls, you can walk off them and plummet to the level below — that much is straightforward enough. (The winged boots mysteriously stop working here.) But because it’s using the same wireframe-dungeon-corridor renderer as ever, you can’t see where the walls end. You basically have to feel out the shape of the thing by falling off the edge a bunch of times. And that doesn’t really fit with the fiction.

Sunlight makes some differences to gameplay. You can now teleport freely into the dungeon and back, the better to complete the puzzles you left behind. There are no random encounters in the surface world, just fixed ones at specific points. What was previously a sort of mixed combat RPG and adventure game is now more of a pure adventure game — that is, there are still combat encounters, but they basically fall into two categories: ones that are trivial, and ones that are themselves puzzles. I’ll go into more detail in my next post, when I describe the endings.

This post is particularly spoilery.

I’ve tried every obvious potential “golden path” to the egress. There’s one route I particularly like that picks up both of the Cube’s key items and hits every single “This Way to the Egress” sign without revisiting anything. Finding that path was a nice bit of puzzle-solving, too, as it relied on a chute that I didn’t know about until I went looking for it. But it was to no avail: the egress was still a blank wall.

Finally convinced that I was barking up the wrong tree, I started looking at walkthroughs with increasing boldness. Although I don’t remember it, I think I must have done this the first time through as well. Solving this puzzle relies on having all of the following realizations:

• The key item called “HHG OF AUNTY OCK”, found in the lower parts of the dungeon, is a reference to the Holy Hand Grenade of Antioch from Monty Python and the Holy Grail. (This much I remembered from my first pass.)
• The “egress” wall, alone of all the walls in the dungeon, can be blown up with the grenade. (This I did not remember, but I did think of trying it.)
• To use the HHG, you have to not just invoke it but drop it. It will not explode until dropped. (This is where I needed help.)
• The above step is difficult, because the HHG is cursed, and equipping it so you can invoke it binds it to your weapon slot, preventing you from dropping it. However, one of the other key items can unstick it. (This I figured out on my own.)

The thing that makes the puzzle unreasonably difficult is that you have to have all the above thoughts without feedback that you’re on the right track. When I found that invoking the HHG at the wall wasn’t enough to solve the puzzle, I didn’t think “I must not be using it right”, I thought “I guess that’s not it. The HHG must be used somewhere else.” The thing that unstuck me was finding out that it was on the right track after all.

Wizardry IV is largely an attempt at building an adventure game in the Wizardry engine. Sometimes it does this pretty well, but the two worst puzzles in the game — the egress wall and the inaccessible room in the ziggurat — both suffer from that combination. Partly it’s the lack of feedback I just described: the system only provides so much output in response to your actions. And that’s a problem even for things that aren’t adventure-game puzzles: I’ve got invokable equipment that I have no idea what it does. But there’s one other factor that the two worst puzzles share: they both rely on one-off exceptions to general rules. Adventure games are based around exceptions, of course, but when they’re well-designed, the exceptions become rules unto themselves, and that helps guide and channel player behavior. And that’s what we’re missing here. There have been other adventure games in a similar format — Asylum, Deathmaze 5000 — and they’ve had similar problems. Like Wiz4, they took pride in being extremely difficult, even though that’s just about the easiest thing for an adventure game to be. But I honestly don’t think this is intrinsic to the format. It’s just a symptom of the time when the format was popular.