Game Design: Character-Building Exercise

Posted by Rampant Coyote on October 11, 2010

Modern CRPG design philosophy, at least so far as I can grasp from interviews and games, seems to hold that it is a sin to make a player wade through the character generation process prior to letting them get to “the fun part.” Character generation should be as streamlined as possible, minimized, and preferably delayed until after the adventure started. Players want story and action, not to be confronted with statistics.

Hey, I get it. It makes sense. You are promised on the box a game of adventure and discovery. But it starts out with a screen of numbers and unfamiliar terms that you are somehow expected to assign, balance, and not shoot yourself in the foot by making a magic user with an intelligence so low he can only cast the beginner spells. Oh, and rolling dice to create a character? Letting chance dictate that your character may be underpowered? Extra bad with bad sauce!

With some designers being very vocal about eliminating these design horrors of a bygone era, it’s easy to forget all those times I used to keep myself awake in Mr. Droneburg’s Geometry class (no lie: That was actually his name) making Champions characters. Or the little feeling of glee I experienced going through Eschalon‘s old-school character generation system (I only really played one character, but I made five).

Last week, I picked up Icewind Dale Complete, a late-90s RPG using the 2nd Edition Advanced Dungeons & Dragons rules. It is a less story-driven, more hack-and-slashy, and somewhat less famous little brother to Baldur’s Gate. It lets you create an entire party of six adventurers. Actually, it kinda requires it (I forget if it provides pre-generated parties, like its sequel). I thought I’d play through the first twenty minutes of the game, just to get a feel once again for a new (to me) old-school Bioware-style RPG experience.

As often happens with a good RPG, my quick trip to the fantasy world lasted considerably longer than I intended. Like, closer to two-and-a-half hours. At least half of that time was spent creating characters. I had no idea that much time had been passing, but I was completely entertained. And the random dice rolls were a big part of the entertainment factor. I was exploring the possibility space, trying to decide on the best arrangement of stats which, with a little bit of tweaking, could help describe someone filling a key role in my awesome adventuring party. It was challenging and gave me a gambler’s rush. I was having a blast, and I hadn’t even started “playing” yet.

Okay. Maybe there’s only a niche geeky old-school gamers that find deep, involved character generation to be fun. And sometimes I’m not all that into it, either. But I suspect that it’s more of a case of presentation and expectations. We old school tabletop gamers immediately recognize the page of stats for what it is and represents, but other gamers may wonder if they’d accidentally ran accounting software instead of a game.

Making – and growing – a character is a key element of the RPGs genre. And it can be FUN. Rather than simply delaying, hiding, and minimizing it, turning it into just a bonus point to drop into a skill tree, maybe designers can focus on enhancing this aspect of the genre. Make it more fun, make it easier to use, make it more interesting, and make it deeper. It’s been done, to a degree. Fallout 3 at least tried something kinda interesting with it. In the dice-and-paper world, there have been several systems and supplements that created a very detailed character generation system that combined random dice rolls and personal choices into creating not just your character’s “starting stats,” but an actual skeleton of a history, waiting to be fleshed out by a little bit of imagination. Why can’t CRPGs do the same? If you played the Cute Knight games, that’s pretty much the core gameplay right there. Okay, it’s not so much character generation as building one over time, admittedly, but there are some interesting ideas there. And it was for anyone but the traditional, hardcore audience.

I’d like to see more of that.

And I’d like to see more building entire parties for adventuring rather than just one character, but that’s a whole ‘nother rant.

Filed Under: General - Comments: 18 Comments to Read

Greg Tedder said,

Well said. I have actually started up Multiplayer games in Balders Gate 2 just so I could create the entire party. Good character creation is a lot of fun, and party creation can really prove to be an enticing challenge.
Robyrt said,

I don’t mind a lack of character creation as long as I am given the first few levels like candy, so I can “roll my own” type of character by the end of the tutorial spaces. Mass Effect 2 does this well, but only if you import a character from game #1 – exactly the part of your audience that didn’t need a few levels’ worth of free XP to go with your pre-generated class.
Jay K. said,

I’m kinda torn between the two sides. I mean, creating a character is all about imbuing potential into the numbers and onto the paper. Still, the play is the thing. If people wanted to just make families of characters, it would be called Genealogy & Geriatrics or Allies & Ancestry or something like that.

What’s telling is that the new Red Box from WotC doesn’t use the traditional character creation at the outset. Instead it leads a player through a series of choices, kind of like Ultima IV, to help them establish their first character. After that though, it’s standard character building though in the other books.
Jay K. said,

I really need to try Eschalon I.
Rampant Coyote said,

My purpose here wasn’t to advocate one over the other (I mean, I adopt the more minimal approach with Frayed Knights, too). I really just wanted to point out that the whole character creation – and BUILDING – thing really can be a lot of fun for people, and shouldn’t be dismissed out-of-hand.
Jeff said,

I love character generators, I’ll spend a lot of time in them, but I’ve found, especially with more intricate and complicated games that have lots of different things to choose, I often get bogged down in the choosing of state and skills and whatnot, sometimes to the point where I have trouble making a decision and never even get into the game.

A robust character generator with lots of tool-tips and suggestions helps this, but honestly, sometimes I can’t get past char-gen to get to playing.
Andy_Panthro said,

Yes! I do love in-depth character creation.

Even Ultima had it’s system of picking questions based on the Virtues.

Some of the best character creations though have been with Daggerfall, Megatraveller and Darklands.

The key to all of them is that you are allowed a rather large amount of scope to craft a character. You can make them very focussed on one area, or a jack-of-all-trades.

In particular Megatraveller and Darklands both are based on your party being people who have had previous experience in life, and are now turning to adventuring.

I think I spent more time in the latter two making characters than I ever did playing the actual games.
Adamantyr said,

I’ve debated doing something similar in my own CRPG design. I eliminated ability scores, but that wouldn’t stop me from having random generation in the start.

Still, the problem such generation systems for me is, I’m always tempted to just keep re-rolling until I get the stats I want. Or, if I really don’t want to wait for that 18/00 to come up naturally, hack the character data. When does it end, really? Once you start down that path… 🙂
Rampant Coyote said,

@Adamantyr – The ironic but self-correcting idea here is that as you have to generate stats for a larger party – or for a game in which you’ve got permadeath – you grow less concerned about getting that magical 18/00. In a group, the fact your fighter has an extra +1 to hit and +2 damage makes a lot less difference than if you are running a single character. And in a game with permadeath, well, his chance of becoming worm-food before level 4 is decreased by only 3%, anyway, so do you really want to bother to keep rolling?

The trick is, naturally, that you want to make it so that people with less patience aren’t unduly penalized. If someone wants to spend five hours making a perfect party… well, then they should enjoy the fruits of their dedication, and have an easier time of it, IMO. The player who only took a half-hour is four-and-a-half hours deeper into the game by then.

So I don’t know that it’s a huge problem. But I’m definitely one of those guys who will go through DOZENS of stats, at least, before committing …
Adamantyr said,

@Rampant – True, but the whole “random generation” also includes the idea of having a complex system to determine that “+1 to hit”.

That’s really what makes the ability score system fun… it’s playing with the numbers, trying to find an advantage in a complex system. It’s scratching the “Munchkin” itch… if you’ve done up Champions characters, you know EXACTLY what I mean. 🙂

Different CRPG players have different standards. I like having plenty of character space for names; I was SO annoyed when Avernum 4 suddenly restricted me to 12 characters for no good reason. Others want gender choices, or races, or any other number of customizable features. I just don’t see ability scores as an automatic “must-have”.

Apologies if I sound a bit rantish; I’m feeling a bit self-conscious about the “low” rating I would get in this area on GIMLET.
CRPG Addict said,

That’s it, Adamantyr. Keep promoting it! I OWE you a GIMLET.
sascha/hdrs said,

I enjoy the character generation very much! The more the merrier. I actually believe that a throughout character generation process adds to the personal relation that you have with your character(s).

In most modern day games that try to simplify the character generation I feel that some important part is missing and it makes the game feel less precious for me. Take for example the “lite RPG” Mass Effect … There’s some cheap char gen before you start but it’s really just watered down to pick one of 3-4 classes, change a bit of appearance and that it then.

RPG’s are something not to be played during a coffee break but for long, cozy autumn evenings. The game devs of yesterday knew that so they made their games very elaborate.
Today’s mainstream developers want to carter to the occasional player as well. But such players quickly loose their patience if something takes too long in a game so the devs water down anything that requires a tad more thinking and instead make their games look like Hollywood movies (in terms of visuals and story!).
Calibrator said,

“Today’s mainstream developers want to carter to the occasional player as well. But such players quickly loose their patience if something takes too long in a game so the devs water down anything that requires a tad more thinking and instead make their games look like Hollywood movies (in terms of visuals and story!).”

The evolution of game accessibility really has a bell-shape (technical advances in graphics aside):
1) (low) big printed manuals, separate map & elaborate character generation, not too much interaction possible, often cumbersome (Ultima 4 has been criticized a lot recently but it would fit into this category, IMHO)
2) (higher) smaller manuals, separate map & built-in overview map & less complicated char. generation, more interaction possible, less cumbersome (Gold box games, for example)
3) (highest) smallest possible manuals, no printed map but full automapping but optional user remarks & pre-defined characters, highest level of interaction, comfy playing (Morrowind)
4) (downward a bit) no printed manuals but instructions built into the games, complete automapping without user remarks & pre-defined characters with limited stats, good level of interaction but casual crowd can’t be overburdened by too many options (new Ubisoft titles)
5) (even lower) game only available as download/flash, no manual – only simple instructions at the game start, simple map, less interaction (stuff like Pixeljunk Shooter for example)

Step 6 could be fully automatic experiences by downloading a “game” which essentially plays itself. These things exist, I know, but are visually unappealing…

Personally, I think that we had the best time at the end of the 90ies with games like Deux Ex, the Thief series etc. There are still good games today for “real gamers” but the majority has been dumbed down. Plain and simple.

A dumbed down can of course still be fun to play but you and I know that we want more “complicated games”: A strong YES to comfortable but complex gameplay.
Jack said,

The problem is that most of the old RPG’s force you to make UNINFORMED decisions about your character. When you’re just booting up Fallout 2, for example, it’s impossible to know that you won’t see an Energy Weapon until you’re ages into the game, or that there’s hardly anything that you can Throw at people. Someone just starting a game could blunder into choosing those or a bunch of other useless skills and only realise their mistake hours down the track, forcing them to restart or push through the rest of the game with a broken character. It doesn’t help that the actual pre-made characters they offer are fairly useless.

Ideally, of course, every skill and stat would be balanced to the point that the player will end up with an equally useful character no matter what they choose. Unfortunately, that’s pretty much impossible to do while still offering a decent range of choices.

Even if they did, though, character selection screens like these still force you to choose how you’re going to spend the entire game before you even see one second of gameplay. You’re going in blind. Choices don’t work if you don’t know how they’re going to affect things.
Tom H. said,

I can’t help feeling that you sidestepped half of the criticism of up-front character building, which is the opacity of the system and hard it is to make sound decisions before play begins.

I use Eschalon for two examples: first, throughout the game, there’s a fairly ready supply of gold, and a very limited supply of skill points. *Some* skills can be trained for gold, allowing you to spend your skill points elsewhere. You can only learn this by repeated playing or by reading the forums for walkthroughs.

Second, there’s the problem of hidden mechanics. You assign a lot of attribute points to a beginning character; something like 1/4th of all the attribute points you’ll get during the game. The forums have detailed deconstructions of exactly how wrong the tooltips are, how the optimal builds aren’t the builds suggested during character generation, and how there are hidden breakpoints that make over-investment in particular attributes silently worthless.

From my experience, playing Eschalon without the reverse engineering work others have done is an exercise in frustration. It’s easy to follow the designers’ advice and end up with significantly underpowered characters who can’t progress beyond the halfway point of the game. 10-15 hours in is way too late to discover that.

(On the other hand, using a large number of the optimizing tricks on the forums makes it “too easy” to power through Eschalon I.)

I’m not entirely sold on the argument that during play you can discover what skills are useful to you and direct your investment there; too often it seems you have to have invested in a skill already to get past some chokepoint in the game, or that there’s a temporary condition in play that encourages investment in some skill which becomes useless in further play. But there is a real problem in most of the skill-based games I’ve played, building complex characters up-front without some way to assess whether they’ll be viable in play.
Rampant Coyote said,

At one point, I had a longer paragraph describing those very issues, but I edited it down to just the example about a mage not having enough intelligence to cast higher-level spells. But other examples could include a lockpicking skill with very few locks, or the aforementioned issues with Fallout 2. Or how badly “acrophobia” would affect you or even prevent you from winning the game. Or if the game was unbeatable if you didn’t have a wizard in the party.

But it sounded like overkill to me. I hadn’t broken it down in quite as much detail as you folks just did. Those are definitely points of frustration. Thanks for giving the issues more attention here, because that’s definitely something that needs to be overcome in order to make it work.
Yoel said,

I really like character customization. A lot.
Yoel said,

I usually find it significantly more fun than the game itself.