Jump to content
Sign in to follow this  
  • entries
    10
  • comments
    31
  • views
    2,204

In Defense of Consistency

jblewis

345 views

blog-0359461001379470727.pngOnce upon a time, there was a period in which Chip’s Challenge levels were fairly manageable. As I mentioned in my very first blog post on here, one of the first CC level sets I ever downloaded was just called “LEVELS01.dat” and contained the levels that would eventually grow to become CatatonicP1. There was also a set called “New Levels.dat” that had an unsolvable first level and an open melee level called “Guard Dogs” that involved teeth, which later inspired me to create my own version of the concept. And despite the shortcomings of these levels, I could tell that their designers wanted to create their own version of the CC1 experience that made them fall in love with the game in the first place.

 

But oh, how far we’ve strayed from that since then.

 

The years that followed brought about thousands of levels, and as far as I can tell, the simple life principle of “just because you can do something, doesn’t mean you should” has gone out the window. “Bigger is better” became the mantra of level design for many, with several designers who felt obligated to fill up the entire map just because, well, they could. And if they wanted to make a smaller challenge, the design principle of choice would be to cram in as much stuff as possible in one tiny room because, you know, anything less would just be a waste of space.. And on top of that, the natural tendency of many was to complicate a level and feature as many of the game elements as possible because, hey - easy and homogenous have to equal boring, right?

 

The sad truth is that when examined in light of assembling a set like CCLP1, all of these design principles fall short of - and in some respects, run diametrically opposed to - the design principles that made levels generally considered CC1 favorites what they were. In past blog posts, I’ve discussed why variety is important for any CCLP. However, in this blog post, I’d like to focus on something else, especially with respect to CCLP1: consistency. Now, that may sound rather self-contradictory at first. But let’s take a moment and think about all the different types of levels out there:

  • Feel: open / closed
  • Design: matriced / organic
  • Gameplay: action / melee / puzzle / maze / speed / collection
  • Size: small / medium / large
  • Length: short / medium / long
  • Structure: linear / non-linear
  • Dominant tiles: hodgepodge / element(s)-focused

This may not be an exhaustive list, but I hope we can see each of these types of levels well-represented in CCLP1. Now, I’d like to present a notion to you. It may sound radical, but just hear me out. What if, among all these different types of levels, design principles that transcend their differences existed among many of them in CC1? We toss around the term “CC1-ish” quite a bit, but considering that CC1 has been hailed as a variety-filled set, what does that really mean anyway? We seem all too content to just find “equivalents” to many of its levels for CCLP1, but if CCLP1 is meant to be a CC1 substitute featuring today's levels, shouldn’t we at least take some time to nail down what characteristics defined many of the most-loved levels from the original game and keep them in mind while we vote and at least clarify what we mean when we say “CC1-ish” so we can give each level the consideration it’s due? Or are we too afraid that establishing such standards would eliminate levels that fit our veteran sensibilities and steer us toward those that may feel less interesting to us? It may sound impossible, but I believe I’ve found three characteristics that fit the bill - characteristics that are conducive to beginner-friendliness and (at least so far) seem to have been forgotten by many voters at the ballot box. I’ve mentioned them here and there in passing, but I’d like to examine them more in-depth in a different context. If you examine CC1 closely, you will find that many of its levels exhibit at least two of these three characteristics. Hey, whenever we play any other game, we expect both variety and consistency in quality, so we can't the same be true for CCLP1?

 

Freedom. This is a big one. It’s probably the biggest one that’s often overlooked in today’s level design paradigm. Remember how most of CC1’s levels allowed players to run around without the fear of tripping some button or collecting some item too early? Or just to experience the joy of running around to begin with? Whatever happened to that? This isn’t an issue of “open” vs. “closed” levels. It’s an issue of gameplay rigidity. It’s a tendency that manifests itself in level design because as designers, our natural tendency when we build levels is to think in goal-oriented terms, like this:

 

1. Decide what the player has to do to get to the exit.

2. Build everything such that every tile in the playing field contributes in some way toward that objective.

3. ???

4. Profit!

 

Okay, scratch those last two items off, but the second point is the clincher here. It’s the major contributing factor to an issue I like to call Designer Disconnect. In the previous blog post in this series, I talked about another tile-based game called Escape, which allowed the player to see the entire map and provided an unlimited undo function - both of which made it much easier to attempt hardcore puzzles. Chip’s Challenge, on the other hand, is not only more than just a puzzle game, it also allows the player to see only a 9x9 grid! This should give a lot of designers pause, but the sad truth is that it does not. Designers have the freedom to see the complete 32x32 map, and all too often, levels are designed as if the player could see the complete 32x32 map - or wants to see it. Is it possible that maybe - just maybe - we’ve forgotten about the thrill of discovering what’s around that next corner?

 

What this all boils down to is that Designer Disconnect often interferes with a player’s natural desire to explore and experiment. When a designer builds a level with a solution in mind and throws down the pieces of that solution, it’s easy to forget about what happens or what is attempted when a player has no idea what that solution even is. Generally speaking, the more complex that solution is, the less freedom the player will have to take a look around and the more the player would have to remember, especially since the game has no checkpoint / undo feature. And the more the player has to remember, the more the game feels like work. I’m just going to speculate, but if I had to guess why CC1’s levels were not exactly all that complex, this is probably at least one reason why.

 

The issue of rigidity may not be the same as distinguishing between “open” and “closed” levels, but our fascination with rigidity has certainly thrown many “open” levels under the bus. I honestly have to wonder how many of CC1’s simple, open levels would do if we had never seen them before and they were in CCLP1 voting today. Would we vote down Nice Day just because it wasn’t hard enough, and it didn’t give us that sweet satisfaction that comes with figuring out something difficult? Would we have a negative view toward Force Square because the wide open rooms around the titular square weren’t jam-packed with puzzles and were just nearly empty spaces? How about Metastable to Chaos and its spacious dirt room to send the bugs into? After seeing how voting has proceeded so far, I honestly have to wonder. CC1 often flew in the face of modern level design by not reducing everything to function and not resorting to irreducible complexity. Instead, we had arbitrary pieces of fire littering some levels, open spaces with absolutely nothing, and dead ends without any items to collect here and there. They may have seemed random, but from a design standpoint, they contributed to the aesthetic of each level and, in many cases, were quite calculated in their placement.

 

At the end of the day, freedom is not just limited to the feeling of openness in a level; it’s a state of mind with which one can approach the game. Let’s take one of the more difficult, confined levels of CC1: FireTrap. This level has a variety of challenges involving monster manipulation and chip collecting, but although these tasks must be completed in a certain order, the player is left to discover what that order is without too much fear of necessitating a restart. It’s relatively simple to roam around without accomplishing anything for a while - and for the most part, there’s no need to worry about cooking the level! There’s even an extra block toward the bottom - how about that! This is a perfect segue to…

 

Difficulty management. In the previous entry in this series, I briefly mentioned this characteristic as a positive trait of difficult levels. However, much like any transcendent quality, it can be applied to just about any level of difficulty when making a level that can be universally enjoyed. The premise behind managing difficulty is this: a game with over 40 unique elements is inherently going to be complex; therefore, a designer must manage a level’s inherent complexity if the level’s goal is to provide a fun experience. I understand that we’ve got levels in our midst whose sole purpose is to stump players everywhere and encourage them to draw diagrams that are meant to be studied for months before a solution reveals itself. And that’s great. But, for the purposes of this blog, I’m talking about levels for CCLPs, which are meant to be played by all types of players.

 

The first step toward difficulty management is to identify exactly what component(s) of a level contribute(s) to its difficulty. Some levels are a challenge because of the concepts featured in them, and for those of us who are veterans, this type of difficulty is easy to spot. However, difficulty can also manifest itself purely on the basis of design, and I believe it’s this form of difficulty that we as designers and veteran players often underestimate. Levels that take a long time to solve or levels with a linear structure are, by their nature, at a disadvantage in the difficulty management arena. This is not to say that these types of levels are bad. It’s just that CC itself is limited because the measures that games often provide to make these types of challenges more bearable, such as checkpoints, aren’t available in this game. And if the level’s goal is to provide a fun experience, then the other elements in the level that contribute to its difficulty must be constructed with caution.

 

What CC1 accomplished quite brilliantly was limiting its difficulty to one component throughout the vast majority of its levels and dialing down other components to help the player focus more easily on that particular one. Notice, for instance, that almost none of the pure melee levels in CC1 are linearly designed (and if so, they take on a simple A → B structure like the one in Problems rather than a complex A → B → C → D → E structure). Instead, they take place in wide open rooms where the player has the ability to walk wherever he or she wants while avoiding the monsters. The CC1 designers understood that something like monster dodging was going to be no easy task for their audience and tried to implement measures that would keep the player focused on the task at hand - dodging - rather than creating constraints to make things more difficult just because they didn’t look hard enough.

 

So let’s take that principle back to the realm of linear/lengthy levels. Notice how CC1 hardly ever resorts to pulling out increasingly harder challenges as an extremely linear level progresses, nor do most of these levels punish players for making poor choices at a late point. This is because the more opportunities for failure are presented in a long level, the much greater the likelihood for failure and, by extension, the much more frustrating it would be. The hardest sokoban in Mix Up, for instance, was toward the first half of the level, whereas most of the challenges that followed were arguably easier. Even Four Plex, which is probably the most rigidly linear variety level in CC1, had challenges that were fairly self-explanatory. As the level progressed, the challenges became less ambiguous. The opening ice maze contained lots of different possibilities, fake walls to uncover, a few unknowns to discover, and a healthy dash of guesswork. In sharp contrast, the second area’s only big surprise was the bouncing blocks (if one didn’t explore before activating the cloner). From that point onward, there wasn’t really anything unexpected in each room. The northeast room was a monster manipulation puzzle. The next area involved navigating walls. And then, the final challenge was a simple itemswapper. Players could tackle each of these challenges knowing what they were and without any sudden bursts of designer trickery late into the level.

 

Unfortunately, we as voters have tended to approach difficulty management with a stringent “all or nothing” standpoint based almost entirely on our playing experience (“How did this level make me feel?”) when evaluating levels for CCLP1. For many of us, it seems like the general rule of thumb is that if a level can’t be solved on a player’s first try, it must not be beginner-friendly at all and should be automatically discarded. I’m sure we’d have voters giving the proverbial thumbs down to Four Plex purely on the basis of the first room offering a few opportunities for failure. But not only would this voting approach be unfair to many levels, it also fails to recognize that failure is inevitable and to distinguish between different placements of opportunities for failure and how they affect the propensity for an enjoyable experience as they relate to a level’s structure. An isolated tough puzzle in a non-linear level, for example, that allows the player a greater degree of freedom in choosing the order of tasks to complete would be theoretically much less frustrating than a linear level that strictly establishes that same puzzle’s placement in the level at a late spot. The same could be said about the level that involves a bit of guesswork but actually contains multiple solutions or is somewhat open-ended.

 

A similar principle applies to navigation challenges involving recessed walls, especially if there are multiple solutions involved and minimal complicating variables entering the equation. I’m sure even a simple (in design) navigation challenge like Apartment would be slaughtered in CCLP1 voting just because someone didn’t quite get it on their first try. The reality is that Apartment is manageable because it understands just how far it needs to go. Chips, walls, and recessed walls define where one can travel. That’s it. No force floors, no boots, elements, or thieves, no doors or keys. The level works because the design is basic, consistent, and symmetrical, and the end result is a beautiful challenge. But just watch well-designed navigation challenges like Anaconda (Omicron #2) and Cell Swapper (Mushroom #48) or open-ended, multiple-solution levels with a bit of guesswork like Puzzle Box (Jacuzzi #35) fall by the wayside because some people rely solely on feelings and not on discernable principles that helped make CC1 what it is to determine their landscape of CCLP1 ratings...even though both of these levels exhibit a remarkable degree of simplicity and difficulty management. And speaking of simplicity…

 

Identity. This design principle worked in tandem with difficulty management for much of CC1. In fact, you could say that all three of these characteristics worked hand-in-hand throughout many levels. Once again, let’s start with the basics: identity is, in essence, a level’s ability to effectively manage itself. Levels with a strong sense of identity tend to be easily describable, primarily because designers who create such levels actually think through what they want their levels to be before dotting the map with tiles and building away. Levels that don’t have an identity crisis usually contain a discernable objective (or two) that doesn’t require a ridiculous amount of lateral thinking to discover. When we play, say, Digger, we know that the point of the level is to collect chips and find ways around teeth monsters to do so. When we play a level like Mishmesh, we know we're going to be pushing some blue walls.

 

Once again, this type of level design feels so foreign to us today. I’m just going to speculate here, but if I had to make an educated guess as to why, it’s probably because to us veterans, such levels feel all too predictable. And predictable must equal dull, right? That’s probably why many of today’s levels feature room after room filled with completely different challenges that require the player to switch gears and think differently every time the gameplay shifts. It’s certainly a far cry from the low-key days of Tossed Salad or even Mixed Nuts. We’ve gotten to the point now where we recognize that trying to be too clever by injecting CCLP3-esque obfuscation and deception won’t win people over, but we have yet another quantum leap to make: recognizing that building a jumble of stuff just to say that our level is amazing and has more tricks than the previous one also isn’t really all that clever in an age where practically more than half of the levels out there aspire to do the same.

 

Maybe I’m just old-fashioned, but personally, I find it much more impressive when a designer knows how to exercise restraint. It’s easy to see all 1,024 tiles in the editor and feel inclined to fill them all up. It’s much harder to know when to stop. Predictability to a person who has command over what to build on a 32x32 board is certainly not the same as predictability to a player who’s exploring that 32x32 board with only a 9x9 square viewport to look at while exploring.

 

Thankfully, for the most part, CC1 knew just how far to go. Do you know, for instance, that adding a border around your level reduces the size to 30x30, eliminating 124 (over 10%!) of the 1,024 tiles? And if you stripped the level down to a more manageable 24x24, you’d be eliminating 448 of those tiles? CC1 certainly isn’t perfect at knowing where to stop, as there are definitely some exceptions - long block-pushing levels like On the Rocks come to mind - but even they avoided the pitfall of “breaking up” the homogeneity by forcing players to get their minds and fingers off block-pushing and onto something completely different in the name of “being more interesting.” From a difficulty management standpoint, it’s much easier to temper the inherent difficulty that accompanies length or linearity with simplicity and straightforwardness in the challenge.

 

What I have to wonder is - why do we carry around a stigma toward homogeneity in levels? CC1’s levels managed themselves well because they understood that the game was interesting enough to warrant that focus on individual concepts more often than not instead of flitting around from concept to concept and - SQUIRREL!

 

You get the point. So, to wrap this series of blogs up, I’m going to pose a challenge to you. As you rate the levels in voting, ask yourself the following questions:

 

1. If I were to take all the levels to which I’ve awarded my highest ratings, how consistent would these levels be with the design principles that made CC1 what it was?

 

2. From that list of most highly-rated levels, could I build a set of 149 levels that included a proper difficulty curve, levels that explored each of the game’s elements, and plenty of variety throughout?

 

Be honest with yourself. Then play the levels. Examine them in detail. Think outside the fact that you failed or solved a level on your first attempt. Then get back to the ballot box. You’ve got an opportunity to help craft CCLP1 into a suitable CC1 replacement that is one of the greatest, most diverse, beginner-friendly level sets ever made by examining what made CC1 work and applying those principles here. Or, you can undermine the purpose of the voting process and throw around ratings solely based on how the levels make you feel.

 

The choice is up to you.



4 Comments


Recommended Comments

Thanks J.B. for your thought out posts. They have been extremely nice to read. I agree with many of your points and it's good that you spell them out and write them down. I was already planning to go back and look at my top votes (once I'm done) and see what type of a set they would make and adjust accordingly, so thanks for the encouragement there. However, we can't expect all, or even many, players to be able to dedicate that amount of time (going through all the levels) so maybe it's enough for them to re-evaluate only the levels they have voted on. Perhaps for someone may have 30 levels that they have voted worth five stars, but they can still try to see what type of a difficulty curve and variety a set with those levels would have; and on the other hand there probably are those that have voted 500 levels with five stars and maybe they should tone it down a bit. Overall, I get the feeling that CCLP1 is in good hands and you guys will do what ever it takes to make this a decent set for beginners!

 

-Miika

Share this comment


Link to comment

What do you mean by "realism"?

 

Levels that uses concepts in a way someone could use them in real life. For example, think about Investment from CCLP3. Part of the concept is to evaluate what bombs need to be exploded to get more blocks to complete the level, but in reality, it could be interpreted as making a bank account, depositing money into it and letting it build year round. However, realism does not have to be a concept, it could also be the environment, textures, or anything that does not appear supernatural. Sorry for the confusion.

 

~lookatthis

  • Upvote 1

Share this comment


Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×