Linear VS Open World

Over the years we've experienced many contrasts related to the play area of our bandicoot pal. Throughout his adventures, Crash has followed linear paths and explored vast fields, often in more ways than one. It's a result of the evolution of video games and an attempt at keeping the series relevant. Let's try and figure out which method has worked best for Crash.

The Simple Approach

The classic Crash Bandicoot games, as well as Crash of the Titans, are well known for their linear levels. In fact, they're not too dissimilar to traditional 2D platformers when you think about it. Though you can move around freely in any direction, the paths you run through are hallway-shaped, and the choice of taking an alternate route is practically non-existent besides some sporadic road forks. This is why the occasional 2D segments never feel out of place.

Linear design is by no means a bad thing on its own, and it's always worked really well for Crash. The various different crates and hazards you can interact with are there to keep your mind off the thought that you're essentially running along straight lines most of the time. This variety helps making the games such a joy to play.

One advantage to this approach is that you don't have to worry about getting lost since you'll always end up where you need to go. The camera, on the other hand, is both a blessing and a curse. If all you're doing is trying to reach the end of the level then you don't need to worry about a thing, but in the few occasions you need to go backwards, it can be a bit impractical to know where you're going and what's coming.

The series stuck with the linear design philosophy even after just about every other 3D platformer had adopted open areas to explore. In fact, 3D platformers with large areas were already around when this franchise began, so Crash had a more limited level design than some of the competition from the get-go. Again, nothing really wrong with that, but by the time the series left the original PlayStation, fans and reviewers alike wondered when the bandicoot was finally going to visit less straightforward areas, hoping the next games wouldn't just be more of the same.

Expanding the Horizon

Unless you count the jet-ski levels in Crash 3 (or the empty hub world of CTR), it wasn't until 2004 that we finally got a taste of what it's like to play a Crash game in wider environments. Crash Twinsanity introduced open, interconnected hub worlds with non-linear design, and though the actual levels were still as linear as before, they were no longer just long hallways.

I'll admit I was among the players who wanted a free-roaming Crash game with plenty of areas to explore and secrets to find, so when I first heard about Twinsanity, it felt more or less like a wish come true. Ever since I'd played Naughty Dog's Jak & Daxter I craved a game where you could guide Crash across large playgrounds and take on different objectives.

My first impressions upon playing the Twinsanity demo were highly positive, as it was pretty much what I was hoping to find. N. Sanity Island was big and filled with a lot of different stuff that added diversity to what would otherwise feel like a barren beach. There was a geyser that spewed out fish for no reason, Wumpa trees that grew upon planting seeds, a cannon to shoot statues with, and more. The place felt lovingly crafted, and it was fun to explore because of the diversity.

When I finally got to play the final version of the game, I was disappointed to see that the remaining hubs were all emptier and less interesting than the first one. None of what I liked about N. Sanity Island so much was present in them, including the pleasant environment full of details. The lesson here is that territory means nothing without something interesting to do or look at, at least in a platformer of this kind.

As for the levels themselves, the same thing happened to a lesser extent. The very first level, Jungle Bungle, has a surprising amount of hidden nooks and crannies for you to uncover (and even more were present during development), but the other levels don't do much to let you explore what's beside the main path, if there's even anything there to begin with. Crash may benefit from having clearly delineated levels, but this doesn't mean they should be just a single path to the goal. Road forks, wide areas, and optional secrets are all things that can spice up a level and make it feel less restrictive.

The Player VS the World

One of the first things I thought upon playing Twinsanity was: "I hope I don't have to break all the crates as usual..." I was relieved when I realized this wasn't the case, because finding a lot of objects in a large area or a long level can be quite stressful, especially if you need to do it in one go. The developers made the right choice when they changed the way gems work, hiding them for you to find instead of waiting for you to break all the crates.

As much as I personally love Twinsanity in spite of all of its flaws, I was disappointed by its lack of any real objectives or missions. You might argue that this keeps the gameplay simple, and I agree with that, but additional objectives add an extra layer of complexity that can really turn a game into something even more enjoyable. You wouldn't even need to change all that much either — N. Sanity Island already has that going for it. Driving chickens into nitros, visiting the small island across the sea, and all those other things I mentioned are essentially optional missions, and you do get rewards for doing them. It's a shame the game is quick to drop this in favor of simple puzzles with explosive crates. Creativity just makes things so much nicer.

Something that is essential for a hub-based world is that every level needs to be easily accessible from its hub. Twinsanity has a couple of levels where you have to replay other levels just to get to them, and no one likes that. Even some of the levels that can be accessed from the hub are a chore to get to sometimes, especially since the game doesn't have a map. Let's not forget that you can't get out of the last hub without finishing the game (though at least this is a glitch and not a design choice). A different approach, and one that hasn't been done in the series yet, would be having a single hub (or even a simple warp room) while relegating most of the exploration to open levels.

Crash: Mind Over Mutant arguably has it worse because of the way it handles its world. There are no hubs to speak of, and instead of a classic levels system, the game is just one huge world where every area is connected to one another. This is commonly known as a "metroidvania" game, thanks to the franchises that popularized the genre, Metroid and Castlevania. There is nothing inherently bad about this design type (it's actually one of my favorites), but Mind Over Mutant does not do it right.

Backtracking is a staple of any "metroidvania" game, but when done right, it's not the chore you might be thinking of. When you're required to backtrack to a previous location, your character usually gets stronger, so even though all the obstacles you've overcome will still be there, they're no longer as much of a nuisance. In Mind Over Mutant, this is done to a minimal degree, so backtracking tends to get annoying. I should also mention that the fixed camera doesn't help matters when you need to go against the screen.

There are two other things that make backtracking frustrating in Mind Over Mutant. The first one is that progression in the game is very strict, and you're always given a single objective at a time. You can't take on a different mission on your way there and leave the rest for later, which is more upsetting when your current objective lies on the other end of the map.

Secondly, despite areas being interconnected, most of them only have one entrance and one exit. Some don't even have an exit besides the way you come from: the Evil Public School, for example, can only be accessed from the Ratcicle Kingdom, and once you're there, you can't take a different way out to visit something else. Ideally, each area should have had multiple exits leading to different places. Better yet, all those save statues should have doubled as warp points that unlocked once you found them. The few teleporters you do get in the game appear near the ending and don't cover a lot of territory anyway.

Let's Have It Our Way

We've talked about linear design and explorable areas, but what about being able to tackle things the way we want? Twinsanity and Mind Over Mutant both feature open areas to different extents, but to get through the story, you still need to do things in order. This is a different kind of linearity not tied to the areas themselves, but rather, progression.

That's where Crash Tag Team Racing comes in. This game introduced actual missions to the series, but you can pretty much do whatever you want, whenever you're up for it. It's quite simple in practice: Crystals are your reward for completing tasks successfully, from doing missions to winning races and other events, and they also unlock more stuff to play.

Almost every mission, race, or event in Tag Team Racing is completely optional, and the only thing you absolutely must do is unlocking the next area (which is, sadly, set in a fixed order). You do that by (yep, you've guessed it) collecting more crystals. In other words, you can ignore anything you don't like and choose your own event order. You can have a completely different experience every time you play, and that's one of the signs of a truly functional open world.

The only problem with Tag Team Racing's environment is that all of the missions are identical to each other. It's usually just a matter of finding and giving the right trinkets or money to random people. It's easy to conclude that after a certain point, the larger a game's world is, the harder it is to make it interesting (Tag Team Racing might not be that huge, but it does try to combine two completely different genres into one game, and that probably doesn't help).

Additionally, it's more difficult to create a story using non-linear progression, since you'll need to predict all the possible outcomes of the player's choices. The easy way around this, and the one Tag Team Racing went with, is to have as few story events as possible, but this also means the story won't be that interesting or relevant to the player's experience.

Explore the Potential

A balanced compromise between world size, variety, and practicality has yet to be truly reached in a Crash Bandicoot game. In that sense, the safer approach of a linear experience has been done better, but that doesn't mean the developers should stop trying. Just as there's a place for linear experiences in Crash, I believe there's a place for exploration and a more open-ended nature, but any game trying to adopt this philosophy needs to adhere to the principles we've discussed throughout this analysis:

• The larger an area, the more diversity it needs to remain interesting;
• Collecting a lot of items in a large area can be stressful if you need to do it in one go;
• If the game's world is too large, a map makes everything much easier;
• Even levels with just one exit should still contain secrets and/or multiple paths;
• Extra objectives help spicing things up;
• In a hub-based game, every level should be easily accessed from its respective hub;
• In a hubless world, the more access points in an area, the better;
• Unlockable warp points are definitely appreciated;
• Non-linear progression adds replay value and gives players a bigger sense of freedom;
• The less linear the progression, the harder it is to have an interesting plot.

And now I'm off to see what secrets I can find around the block.

Hopefully nothing gross this time.

Comments