Basic game design suggestions Thread - Claustrophobia: The Downward Struggle

Hello there dev(s)!

Roguelikes are (thankfully) experiencing a revival right now, which grabbed my interest in the "genre" once again and so today I purchased your game and have been playing it for hours on end up until now - I love the art design, the humor and the overall feel of the game. Being a psychology student however, I've also been paying close attention to how the game mechanics translate into what I as the gamer feel and think while engrossed in your roguelike. That said, I would like to offer up some suggestions to consider about the most basic game mechanics by taking a step back and analyzing what distinguishes average from really great roguelikes with the hope to contribute a tiny bit to make Claustrophobia the latter type of game.

First let us consider two very different roguelikes, FTL (faster than light) and Desktop Dungeons - both truly great games in their own very unique ways.
Also, both are probably very close to the opposite ends of a spectrum that might describe the genre of roguelikes (if it can even be said that such a genre exists).

FTL plays out in real time, there are no turns and you order individuals or groups of your crew around your spaceship in a manner usually reserved for real-time-strategy games.
However, a pause button lets you interrupt the action at any time to let you think and give your orders while everything is paused. So you fly around with your crew in one of ~18 spaceships and have to flee the "rebel forces" by navigating from left to right by visiting "beacons" along randomly generated star maps, while (nearly) each beacon features some kind of interesting, random encounter - more often than not fighting other spaceships in various game-changing backdrops (ion storms, sun storms) or visiting space stations to sell and purchase gear. You can't visit all of the beacons however, because you are being chased relentlessly by the rebels. The game is really hard, I've logged close to 80 hours and depending on the starting ship type I choose I only win and defeat the end boss somewhere between 10-40% of the time. Permadeath is a given. The game is great and truly addicting, give it a try if you haven't yet!

Desktop Dungeons on the other hand looks much more like Claustrophobia - its turn based, tile based, randomly generated and has permadeath. It's so absurdly turn based in fact that enemies don't even move, only your character moves - so you decide when you engage which enemy. Enemies have levels and wait for you to come around and slay them. Generally with enough backup from spells and potions you can kill a monster 2 levels above you (for important extra xp), but anything above that is usually suicide - if you don't keep whacking them they also regenerate with each turn. So how can this possibly be fun? Simple: the limited resource here is "undiscovered space". It's a completely absurd sounding game mechanic but once you try it you'll get it: every new tile that you discover restores your life and mana. Once a tile is discovered/visible however, you don't get that bonus the second time you pass along and "see" the tile again. The goal ist to slay the level 10 Überboss - and to that end you must reach level 10 or at least 9 as well with nearly full health and some mana to batter him with spells. The key to winning then is to navigate the fog of war in a manner where you pick your fights very carefully - because if you're at full health you don't get benefits from discovering new tiles, so almost every single turn you play becomes an important descision that has big risk/reward calculations going on.
If you have never played it, I recommend to check it out the free alpha version for inspiration: Desktopdungeons.net

Another roguelike is Spelunky, which is more of an action platformer and maybe even further away than FTL from the standard roguelike formula, but I would also happily classify it as a very successful roguelike, because it executes the core of what a quality roguelike is really about marvelously...

Okay so where the hell am I going with this - The point I am making is that the unique "roguelike experience" can be captured by fundamentally very different games. Games that on the surface have very little in common like FTL and Desktop Dungeons. What they have in common however is this: Permadeath, randomization and,above all, important decisions. And I will argue that last one is really the most important "feature" of roguelikes, because permadeath and randomization are really just common tools to give the decisions of a player more importance and weight, which through adequate types of feedback ultimately results in a more gratifying and satisfying game experience.

The decisions feel important in roguelikes because the consequences are huge... namely permadeath, you lose a unique character/spaceship and have to start over from zero. The decisions also feel interesting because of the randomization aspect of those games, since if you fail and start over you can't "automate" them and just brainlessly reiterate your decisions in a way you can automate your shooting or jumping skills in action games. In roguelikes you can only automate your decisions up to a point because you can't just make the same decisions you did in your last playthrough because everything is different via randomization - you don't know what's in the next room, you don't know what the potion does (yet). Maybe this time you find an epic bow early on so it pays to put points and skills into agility instead of strength like you initially may have planned to.

Fundamentally every good roguelike is not really about permadeath or randomization, it's really all about making thoughtful and interesting decisions and handling risk, to see the results and rewards of your decisions. And what's more important is this - the consequences of your decisions should become apparent immediately or very soon after you took your decisions, in other words the uncertain yet to some point predictable feedback of your decisions is what is so addicting and drives you as a player forward.

Playing Claustrophobia today however, something became apparent to me: The decisions I took usually weren't very interesting to me and I felt like I was very rarely forced to make a decision. Because of the extremely labyrinthian design I felt like most of my time I was actually "backtracking" through already discovered and looted corridors and rooms (being aggravated by traps, which are a great idea but horrible when you want to quickly backtrack through rooms) to loot the (fewer and fewer) places I hadn't discovered yet. Since this isn't an action game but a turn based game I'm interested in making interesting and risky/rewarding decisions instead of running back and forth through empty corridors, which just feels like I'm wasting my time. However, in order to build a great and strong character before descending further, I feel like I have to mop up the entire level before actually descending to the next level. My gaming sessions simply felt too long and drawn out, which hadn't as much to do with how much time I actually spent playing, but with how my subjective feeling about the time progressed - namely slow which is a bad sign and pointing to the fact that the decisions I made weren't interesting in many cases.

How can this (in my view shortcoming) be addressed?

I think the major design philosophy should not be to superficially "recapture the roguelike genre", but to constantly ask oneself: Am I making interesting decisions in this game as I'm playing it? Why do or don't those decisions feel interesting to me? Can we improve the frequency of important decisions? How and when does the player get feedback about the quality of his decisions? And finally how can the game mechanics be tweaked to facilitate the frequency, feedback and felt importance of the player's decisions?

Desktop Dungeons would be much more boring without the limited resource of "discoverable space" forcing you to make tough (and thus interesting) decisions. FTL wouldn't be as awesome if there was no rebel fleet pursuing you, limiting the number of beacons you can visit... if you could visit all beacons as often as you wanted it would not be as awesome, because it would make your decisions about when and which beacons you visit essentially pointless and uninteresting, detracting from the experience.

Claustrophobia is missing a similar game element entirely as of now... I can take my time to explore the entire floor before I descend further, reducing the overall time I make interesting decisions drastically and exponentially (because the more of the level I discover, the more backtracking is going on to the point where I'm almost bored and where I'm rapidly losing interest in entering the next level).

How about we do some brainstorming about possible game mechanics which facilitate this kind of interesting decision making I was talking about - the details aren't really as important, as long as they fulfill their psychological function.

We can imagine a game mode (let's call it "Inferno") where the level isn't a square but a long, drawn out rectangle where the player is running from left to right (or from bottom to top) while escaping from flowing lava or fire. The lava could swallow everything that is not a wall or a rock (doors, bookshelves, etc.) and spread out - one tile every 10 turns or so forcing the player to make tough decisions all game long and limiting the time you can spend on any given level. Right now if I see a tiny pile of gold (or any other kind of useful item) there is no interesting decision going on in my head, I just get there and get it - very marginally happy and slightly bored. But if it's lying way across the room while the door next to me is starting to melt it suddenly becomes a very interesting decision to me.

Remember this is just an example, the particulars don't matter as much, as many other imaginable game mechanics could force you to do similarly interesting decisions - very mighty but slower moving enemies could start to swarm the level after a while and slowly push you to the next stairs for example.

And limiting the players time in an area is just one of many potential game mechanics that can increase the frequency and quality of a players decisions. I bet there are many, many more that we can think of together if we put our heads to it.

As of now I definitely feel there is some game mechanics missing to push this game over the edge to make it the roguelike greatness it could become. Something that cannot simply be rectified and glossed over by 90 bugfixes, adding 100 new monsters and 1000 items for variety's sake. I think some fundamental game mechanic(s) are still missing to make nearly every turn feel like an interesting decision to me, because what great roguelikes are about is really about thoughtful and interesting decisionmaking, not the superficial characteristics like permadeath and randomizaton (as I hope I could convince you in the course of this epic essay). If you can think of a way to offer more interesting decisions more frequently to the player, I truly believe this game will become a true classic.

What are your thoughts on the matter?

Edited by: 24Immortal

Hello! First off, I found that very interesting to read. You make very structured, well thought out points. I have played Desktop Dungeons (and have friends who have played a hell of a lot of FTL), and I agree that the core mechanics of those games are what defines them within the genre. I don't think Claustrophobia is missing it's core mechanics, rather that they are spread too thin to be having the impact on the gameplay that they are meant to. This has been on my mind for a while now, so let me explain what I think the problems are, and what I'm planning on changing.

First of all, floors are way too large, and because of that, the terrain generation goes a little over the top, spawning the labyrinthine corridors you mentioned, and causing the player to back track through areas they have already explored. This defeats a lot of the point of considering your plan of attack when starting out exploring, as dead ends and poor design cause you to end up back where you started.

Secondly, the enemies you encounter are simply spread randomly over the map, meaning that the enemy scenarios you come across as a player are normally only a couple of enemies of the same type, which require little thought to defeat. On top of that, enemies do not really incorporate many interesting tactics into their combat, meaning each enemy can be beaten in a very similar way.

Finally, the skills which you spend talent points on do not require much thinking ahead. You can put points into anything you want at any time, and, should you put a point into a skill and then change your mind, you need only put your next point, a level on, elsewhere. This means that, while I want the player to be able to customise their class, you do not need to think about your class design past "what skill do I want this level".

These are the things that should be shaping the core mechanics, that give the player the need to consider their next move, and that make interesting decisions worth while. At the moment, however, they are not working as they should. Based on these points, the changes that I think would really tighten the core gameplay are:

• Decrease the size of each floor, and increase the number of rooms on each floor, reducing the amount of winding corridors which go nowhere.
• Redesign how enemies spawn, and give enemies different combat mechanics which must be considered when fighting them. This is discussed more here.
• Change the skill engine so that you must commit to certain skills trees before spending points elsewhere. This is discussed more here.
• Change how health regeneration works. Although I have not discussed this here, this is another point that has been brought up a few times before.
  

I hope that goes some way to explaining my current thoughts on the mechanics matter. If you have any thoughts, questions, or suggestions, then let me know!

Thanks,
~TheIndieForge

Edited by: TheIndieForge