Tales of the Rampant Coyote

Adventures in Indie Gaming!

Frayed Knights: Things to Do in a Dungeon (on a Saturday Night)

Posted by Rampant Coyote on April 3, 2015

FK2_Vault1_ss6One of my design guidelines for Frayed Knights  is to avoid having too much repetition of an encounter “type” along what is the likely a critical path in my levels.

In other words, I don’t want a dungeon that is just full of similar combat encounters. That’s boring. In fact, I try to avoid having more than two planned encounters in a row that feel too much alike. I developed that rule-of-thumb during the development of Frayed Knights: The Skull of S’makh-Daon, although I was basically trying to turn my “gut feel” of design into an actual checklist. Of course, given the player freedom to do whatever and the chance of random encounters or patrols, it’s still possible to encounter three or more vanilla fights in a row, but my guideline helps me minimize the number of times that happens.

Of course, having some kind of lock or trap or dialog counts towards that level of variety, but while I don’t have a specific written rule about that, I don’t want a dungeon to follow a formula of fight-talk-fight-trap-repeat either. Nor do I want a repeat of vanilla-fight, vanilla-fight, somewhat-different-fight either. Variety is the spice of life, and it must be the spice of dungeons, too. As a player, I want dungeons to be full of mystery and wonder, even if it’s just an orc lair.

I guess in that way, I kinda follow Richard Garriott’s design philosophy from back in the 1980s (from Shay Addam’s The Complete Book of Ultima): “Assorted activities and the diversity of activities are what makes a game rich in my mind.” When discussing, further talked about FTL’s Dungeon Master – which was pretty revolutionary at the time and a bit of a harbinger of RPG design to come. “I was interested in seeing all the neat new things I could do, and after it ran out of different activities, I was finished.I enjoyed it till that point in time, and was ecstatic for those few hours.”

Of course, that’s all well and good to start as a philosophical standpoint. It’s right up there with saying, “Make sure your game is polished, well-balanced, and bug-free.” Sure, nice goals, but actually achieving that can be a problem. When sitting at the computer at one in the morning trying to be creative figure out something cool and interesting that would give the player something fun to do and would still belong in an old, until-recently-abandoned underground complex. Preferably one that doesn’t require a bunch of new code or new models.

To come up with ideas, I spend time looking over maps & walkthroughs of old games, browse through old D&D modules, brainstorm, or I can do what I used to do when I was fifteen (and had the time!) and just lay down on my bed and daydream. I think the last is probably the best approach, but the first two are good for getting me thinking along those lines.

So if you want to volunteer any ideas, go ahead. Otherwise, it may look like I’m just napping on my bed with a notebook and pencil next to me (I learned the hard way not to do it with a pen…), but really, I’m working hard! I promise! 🙂


Filed Under: Design, Frayed Knights - Comments: 3 Comments to Read



  • Eldiran said,

    The solution: block puzzles! : D

    Seriously though, I’m working through this issue too, now that I’ve hit the level design phase on my own RPG. You’ve pretty much mirrored my thoughts exactly.

  • Maklak said,

    Off topic rant:

    I’m wondering why computer RPGs are pretty much stuck with simplistic systems made for Pen and Paper RPGs, where most calculations might as well be done in the head with a dice roll or a few. Examples:
    *) HP (usually healable in combat, (even in modern settings like Wasteland 2) for a ludicrous effect of killing dozens of people and shrugging off what should be moratl wounds) rather than something more realistic, like hit locations, lasting critical wounds effects (including neve damage and limb loss), blood loss, infections, etc. Combat should generally be dangerous and something you want to avoid, not XP grinder and mass murder simulator.
    *) Grenades / Fireballs are most often modeled as a circular area, such that if you are inside you roll for evasion and damage (which are the same, regardless of distance) and if you are outside, you’re completely safe. A more reasonable system would be something like concentric circles (the father you are, the easier it is to evade and the less damage you potentially take) plus sharpnell as a random number of “rays” from the center. If any of them touches you, you get a chance to be hit. I’ve seen a Neuroshima (Polish post-apocalyptic RPG) do a decent model of Claymore mine with cone-shaped area of effect and about 5 zones based on distance with diffrent damage and number of sharpnell. Pretty good for a Pen and Paper RPG.
    *) Ranged weapons tend to have a very limited range. With high level of skill and short-ranged weapon you can have almost 100% chance to hit in one spot and 0% one square further. Even if you want to stick to % chance and not do the whole weapon spread / area of target / hit locations thing, there are more gracious ways to do it. For example Eve Online has a formula where each weapon has an optimal range and a range where accurracy starts to drop rapidly. WH 40k allows to shoot at up to 4 times maximum range, but at increasing penalties to hit. In short: dropping accurracy is the way to go here.
    I mean you have computers with ludicrous memory and processing power and are already using complex graphics, pathfinding, etc, so why not do full frontal Dwarf Fortress? My theories:
    1) Programmer Laziness – this stuff needs to be programmed by hand, while libraries / frameworks do a lot of heavy lifting in areas like graphics. Combat systems from Pen and Paper RPGs are “good enough” and widely accepted, so why change?
    2) Perceived player laziness – players like to optimize and therefore like systems that they can understand and mimic in something like Excell. They don’t want to read a PhD on game mechanics just to know how to build their characters. So familiar, simple, good enough systems are all-around win.
    3) Balancing – there is a strong notiona that games should be like theme parks, with everything balanced between classes, levels, enemies and so on. It is nigh-impossible to do it in a complex system, but something like DnD is basically built around this concept. To me an especially repulsive concept is that the number and strength of the enemy group should varry depending on power level of the murderous hobos. Simply put, the game universe should not revolve around them.

    Slighly more on topic:
    It is probably a bad idea for FK, but I’d like to see an RPG that doesn’t “think” in terms of encounters. Rather, there are some places with creatures who try to defend against wandering bands of murderous hobos (player characters), but they can do it intelligently, rahter than divide into level-appropriate groups, stand around, then fight to the death and drop only a small fraction of their inventory on death. So no encounters, just competent defenders with a mastermind. Examples:
    1) A dragon rules over a band of Kobolds. She has alarm spells and enough telepathy to command her minions. They will attack with ranged weapons from murder holes over trapped choke points, use cover, fall back through small tunnels to the next choke points and so on. The dragon will be reluctant to fight the hobos and try to escape when wounded. Or circle-strafe and breath / magic a party that’s too stupid to have ranged weapons for everyone.
    2) Hobgoblins inhabit a small fortification. Rather than aimlessly wander their corridors in small groups, they sound an alarm when intruders enter, fall back, some of them use a secret tunnel to go behind the party, then do a massed assault where their enemies are surrounded.
    3) Forest creatures (faries, elves or whatever) spy on a party of high level intruders, while massing a few groups of militia some distance away. They use their native terrain advantage for sniping and hit and run, plant control and illusions to lure them into a trap, then massive assault to finish them off.
    4) Counter tactics: A high level party uses illusions to make the enemy try massed assault with most of “encounters” in an area. Then cluster-bomb it all for t3h l00lz.
    But then player stupidity indiced TPKs (for example from charging into encounters on the assumption that they must be balanced) are mostly great fun for the GM in Pen and Paper RPGs and it wouldn’t be trivial to do in a computer game, so it’s probably a moot point. However, it might be feasible to have several wounded enemies fall back and join the next encounter or an alarm that makes every encounter in the area go towards the alarm.

    You want a deeper world with more things to do? Give me more realistic economy and logistics, like in this book: http://suptg.thisisnotatrueending.com/archive/27934657/images/1382726888015.pdf Sure, it is an infinite money machine and most today’s RPGs abhor that, but I like the concept that economy makes sense and trading and crafting can be profitable.

    Any thoughts on another one of my long off-topic rants?

  • Cuthalion said,

    Maklak, I agree with you that computers don’t need to stick with aping tabletop rules. When they do that too closely, it can actually be frustrating. KotOR or Baldur’s Gate I don’t think showed you the die rolls or really explained what different things meant; they just assumed you knew what a fortitude save was and whether a +1 to that was as important as a +5% to hit, and that “to hit” did not mean “damage”. If you’re not going to explain your tabletop rules or show the dice rolling, you’re just making a poorer version of an existing system.

    However, some games suffer from being a little too pleased with the computer’s ability to do complex math, to the point where it’s impossible to strategize. Don’t tell me I can build my character if the numbers relate to each other in obscure ways that you don’t explain. I’m playing through Tactics Ogre: Let Us Cling Together on PSTV right now, and it’s awful at this. I can’t tell whether 2 of one thing is better or worse than 3 of another, how much spending precious skill points will help if I choose this skill versus that, or how the different numbers I have translate to giving or taking damage. It’s all just guesswork and going with whatever characters seem to be effective. I have a whole spreadsheet’s worth of stats that are useless to me, because how they work is obscured. It’s so bad that there’s at least one heated thread on GameFAQs arguing over whether one of the skills does anything at all or if it’s completely broken due to developer error! They’re arguing over this because it’s that hard to test conclusively.

    So, I say let the computer’s capabilities expand the horizon, but don’t forget that if you’re making a strategy game or an RPG (which is part strategy game), you’re asking players to make decisions. Make sure they can tell what the difference is between their decisions. Otherwise, you’re just making an action-adventure game that might be turn-based.

top