Gear Design In Early Video Games Vol. 2(Ultima, Wizardry, and Bard’s Tale)
Hello friends. You’re reading the second in a series of articles I’m doing on gear design in early gaming. If you haven’t read part 1 then go do that before reading this.
This is a screenshot is from the seminal classic Ultima, Richard Garriot’s 1981 follow-up to Akalabeth.
Ultima (1981)
Ultima saw nearly three times as many weapons as its predecessor, including a pistol, phazer and blaster… Ultima gets weird.
You obtain most of the weapons by buying them, the same way you did in Akalabeth, but there is a twist.
Ultima’s shops, and I think this is a first in gaming, actually grow alongside the player, with more powerful gear being added to the inventory over time. Pretty groundbreaking stuff. Here’s how it works: there are only a handful of weapons available at the start of the adventure — the shop’s inventory increases after you’ve taken so many steps in the game world, unlocking better weapons at 1500 steps and 3000 steps. It sounds painfully simple, but in practice it instills a sense of progress that Akalabeth and its contemporaries were lacking.
How shops work in terms of gear progression is an important part of a game’s identity and an often overlooked one. Critiques will focus on character development and loot drops, but rarely if ever examine how a game’s economy progresses along with the player. And it’s probably inching towards absurdity to call Ultima’s shop system an economy, but you get my point.
The strongest weapons in the game are actually not found in stores at all. Instead you can find them in these castle ponds. If you drop ten pence in the northern one here, you’ll get a random weapon in return with a chance of the weapon being one of the few you can’t purchase.
This is what I was talking about when I mentioned the balance between deterministic and random rewards. Most games of the era, and even the games of today, use a mix of deterministic and random rewards, but Ultima does something really interesting by combining the two elements in a single reward loop.
If you know how the pond works, then you probably know what weapon you’re looking for, namely the blaster which is the strongest weapon in the game by far. If you get it on your first attempt, then congrats, you beat Ultima. I mean it’s that freakin’ powerful. So I think it’s safe to assume that anyone going to the pond is going after that space gun. That intentionality behind the player’s action is really powerful. But to be clear, not as powerful as the blaster. Nothing is.
I want to illustrate this idea a bit more and jump away from classic gaming for just a second. A couple years ago I dug into Diablo III for a month or two; I’m not sure how much has changed in the game’s meta since then, but I remember set bonuses were usually the way to go…
And in order to effectively use set pieces you needed a ring called the Ring of Royal Grandeur, which you could get by grinding bounties in Acts I and IV specifically. I don’t remember a whole lot from Diablo III’s endgame, but I do remember grinding Act I to get my ring and how happy I was when they finally started dropping for us. I also don’t remember any of the other items aside from that one ring… and that might be because of how darn powerful it was (and again not as powerful as the blaster — nothing is), but I also think it’s because I enjoyed farming for it… I had a goal… I wanted that ring, not the hundreds of rings that drop from random cultists. But that dang ring.
At the end of the day everything I’m talking about here also lines ups with the works of B.F. Skinner and his work on operant conditioning. There are plenty of other essays and wikipedia articles that can explain his work better than I can. The best explanation is one from Extra Credits:
If I have one major critique of Ultima it’s that damn blaster. Once you get it there is almost no reason to ever use any other weapon ever again. And the emergence of that dominant tactic removes a significant amount of strategy from the game.
It is a common mistake. Even today. So how do you avoid making it? One game that figured it out, at least to some degree, was Wizardry, which released later the same year.
Wizardry (1981)
Wizardry took two and a half years to complete, with a whole year dedicated to play-testing before release. And it shows. Especially compared to Ultima and Rogue, the latter of which only came out a year earlier.
This is the first computer game, at least that I know of, where certain weapons will do more damage to certain enemies. Meanwhile certain armor pieces will provide better resistance to certain elements. Though not very deep, I do think this kind of design offers a layer of strategy that we often take for granted in the 21st century.
There is one weapon I want to talk about in detail because its design answers that question I asked a bit ago. That weapon is the Dagger of Speed which swings a total of 7 times per attack but has one major drawback — it is the only weapon in the game that offers any change to a character’s defense, and it actually makes it worse. By equipping it, you’re giving up defense for the sake of your offence.
And you’ll see this concept a lot in modern titles, where designers balance a powerful weapon by saddling it with major drawbacks.
By saddling a high damage weapon with weak defense, you’re basically asking the player if they’d prefer risky situations where the reward is high or if they would prefer safe situations where the reward is low. This is one of the most common choices in all of gaming.
Of course I’m not sure if it really works in Wizardry; I imagine most players get through the game and don’t even notice the decreased armor. I’m not even sure if it was done intentionally — it might have been a bug. But… It’s still here, it’s still in the game, and it’s the one of the first instances of this concept that you can find.
Wizardry does one other thing that is worth a brief mention.
Every level in the game has a number of fixed rooms where treasure could potentially spawn. Wizardry randomly picks from those designated rooms and a large number of them, maybe around 80%, gets a treasure chest. That means that, while the game appears to randomly reward the player, it’s actually on a pretty tight leash. The designers know about how much treasure a player will have by the time they reach a certain floor, and that gives them a little wiggle room when designing enemy encounters that are both challenging and fair.
There’s a lot to love about the game. And that’s probably due to the year the game spent in playtesting. But that year doesn’t seem to have been enough to catch many of the game’s bugs. The worst of which causes some items to never spawn. The long sword, short sword and helm+1 never spawn in a treasure chest. And the only reason I bring that up is to illustrate how error-prone these games sometimes feel in retrospect. So when I said earlier that i’m not sure the dagger of speed was intentional or not, I really meant just that.
And that element of these games sometimes makes it difficult to analyze them properly. Armor in Ultima, for instance, doesn’t seem to work the way it should. At least I don’t think so? It didn’t feel like it made much of a difference to my playthrough, and when I looked it up online I only found conflicting information. Some research told me that armor worked like in DnD where it helps with evasion, but actually doesn’t prevent any damage, while other sources told me that armor was bugged — it didn’t work on the overworld at all, but it should work just fine once you’re in a dungeon and a third source told me that armor was indeed bugged, but that it didn’t work in the overworld or in dungeons.
I’m not really sure what was going on and since the game only had, like, three pieces of armor in it, I didn’t spend a whole heck of a lot of time figuring it out. I think this little aside is a good example of why games need to over-communicate information to the player. If armor does actually prevent damage or increase your evasion rating, you gotta let your player know it
If I had to guess, Ultima’s armor probably increases evasion and that’s why it sometimes seems bugged. I have no evidence of this aside from the fact that the game already takes so many cues from D&D, and other video games of the 1980s inspired by the same property handled armor the same way.
Okay, so let’s talk about the elephant in the room. Armor working this way doesn’t make any sense. I suspect, but have no proof, that the only reason it works this way in D&D is because of the limitations of a dice-based tabletop game created in the late 1970s. The mechanic has no place in the world of computers, where we can do more or less whatever we want behind the scenes.
The reason I bring it up, and talk about it at length, is because I feel like it encapsulates another really important idea. And that’s the idea of affordances.
Affordances are the properties of an object that show the possible actions a user can take with it. The properties that tell you what to do with a door handle are its affordances — its shape, its color, the material it’s made with. And affordances are, at least in part, informed by a user’s history. A beanbag chair, for instance, only affords its usage through past experience.
If you give the user an item, or a mechanic or system, and they have no idea how to use it, then part of the problem might be that it has bad affordances. I’ve always felt Armor Class in D&D was somewhat confusing because of its poor affordances. The real world use of shields, the image of platemail, even the word armor communicates a different function than what AC ends up performing.
A lot of early video games followed D&D’s lead when it came to how armor worked, and I suspect that’s what led to the confusion around how it worked in the original Ultima. Or maybe not, I don’t really know.
Ultima IV: Quest of the Avatar (1985)
One of the first games where armor worked as you’d expect was Ultima IV. Fans will remember the game as the first in the “Age of Enlightenment” trilogy that continued in Ultima V and Ultima VI. Even at the time of its release, the game was praised for the way it pulled the series away from hack and slash dungeon crawls towards a more nuanced and narrative-focused adventure.
And it was, by most accounts, a nightmare to produce. When it came time to choose between thorough play-testing and a holiday release window they hunkered down and tested the hell out of the thing.
Nah, just kidding. They rushed through play-testing, and the game released with a slew of bugs. And remember, this is before the days of reliable internet and day 1 patches. Whatever shipped with the game in 1985, stuck around, and you can still see remnants of its troubled production in the game’s gear design. The Halberd, for example, is a great weapon, not because of its attack damage or anything like that but simply because it can attack enemies through walls. And although it’s never been confirmed, I imagine that’s a bug, not a feature.
But honestly, while Ultima IV’s scope is huge, I was most impressed by how modern it felt. And I think that’s in large part due to the game’s more modern take on gear design. Armor works like you think it would. So does combat. There are even side quests, which often have more of a hand in defining the gear they reward than any stat boost.
Bard’s Tale (1985)
Bard’s Tale came out the same year. It’s a seminal game, but, like Ultima it’s also a messy one.
Our old friend armor class shows up again. Except this time Armor Class not only helps you avoid hits, but also helps you land attacks as well. So a spectral mace +5 won’t help you hit, but in an increase in armor would. I’m not sure how you’d know it worked this way unless you read it in reference docs or on the internet. Modern games get around issues like this by clever naming. You want a stat to have sway over getting hit and hitting enemies? That’s not armor class. That’s dexterity.
Weapon damage in Bard’s Tale is all kinds of screwy too. It seems like it’s built around a dice roll similar to classic D&D, but how that system gets its final number is a bit… strange. First off, weapons straight up do random damage. Remember that trident from AD&D that did 3d4 damage? Like I said, that kind of design makes the weapon damage a little bit more predictable, enough to plan around, but still leaves plenty of room for randomness. Bard’s Tale throws all of that out. You have just as much a chance to do 3 damage as you would 6 damage.. But wait! It’s actually even a little weirder than that… So here goes:
Say you have a weapon that does 2–8 damage or 2d4. Bard’s Tale will pick a random number between 0 and 8, if the value is lower than the number of dice, in this case that’s 2, then it will increase it to the number of dice. So if it rolls a 1, or a 0, doesn’t matter, as far as the game is concerned, they’re both considered 2. So you’re three times more likely to do 2 damage than any other amount. And this issue is exacerbated as you increase damage and dice rolls since more and more possible numbers are actually just the base damage. It’s somewhat predictable, but not really in the way you want it to be.
Aside from that mess, Bard’s Tale item design is… Pretty excellent. The luckshield makes hiding succeed 100% of the time; the conjurstaff halves all spell costs — and in both cases no other items in the game can do either of those things. The important point here, one that still survives in modern day design, is that items with special capabilities, capabilities you can’t find anywhere else, create interesting questions and interesting scenarios. I think a key piece of good item design is creating limits in your game and then creating items that exceed those limits. In Bard’s Tale you have MP like in most games, and you can’t restore it until you go to town, UNLESS you have the mage staff, which will help you restore MP while exploring. Or as Bard’s Tale’s own Michael Cranford said a couple of years ago at GDC, “rules in gaming are meant to be broken”.
See you next time.
