Why this "Perfect" Idea Didn't Work
Deadlight is an old school game with an inventory, but it isn't a graphic adventure. Graphic adventures rely on players revisiting areas in order to look for missing items and to try new item combinations. And at that moment, we couldn't support this backtracking.
Firstly, we had already built a lot of the scenarios of the game (around 40 percent of the content was already there), and they simply hadn't been designed for it. They had been conceived as a completely linear, cinematic experience.
Zombies and traps were triggered assuming that the player will arrive from a specific direction; cutscenes were triggered on the same basis; even the navigation and the layout were meaningful and interesting only if the player was moving in a predefined direction.
That means that if the player decided to go back, he would find completely dead and empty areas without any interest. Backtracking would have been painful and boring.
Secondly, as Deadlight is a game that relies heavily on graphics, we had plenty of no-turning-back zones in order to deal with the streaming of Unreal Engine. If the player misses an object in one scene, we simply won't be able to go back for it.
In the end, we had already created an extremely linear, but appealing, game experience. And then it was supposed we should simply "add puzzles" to it.
We tried to include a few inventory-based puzzles, but we simply didn't have time to rebuild the levels. We had half a year for the alpha submission and we still had to finish the ranged combat system, the AI, the melee...
We desperately needed to close the layout of all the levels. We didn't have time to keep looking for solutions. So we decided to cut the feature. Our puzzles couldn't rely on an inventory system.
We changed our focus, starting to look in more detail at games of reference. At the end of the day, Limbo was an extremely successful game full of puzzles and no inventory.
What We Had to Change
Removing the inventory meant that the player would not need to look for and to carry items to solve the puzzles. But we still believed that making the player to explore and investigate was something really important to establish the mood of the game.
Moreover, our artists had already created dozens of evocative, highly detailed game areas, some of them off the main path, and it would had been painful to not guide the players to them.
They only thing we could do was to take the collectibles and the ammo picks and make them play this role. In this way, collectibles turned out to be something much more important than just an achievement or a trophy. Apart from their narrative role, they let us to populate every corner of our maps with something to reward the players and to foster that survivor-like exploration we were seeking.
The other big consequence of removing the inventory was that we needed to simplify the puzzles we had already built. From that moment, all the puzzles had to be solved using just the navigation system, simple combat actions, plus the standard interaction button: "Press A to do X," where X could be "open a door," "turn on the lights," "push the crate," etc.
In other words, we started to morph our graphic adventure puzzles to Limbo-like puzzles.
|
Printer-Friendly Version
Absolutely. Super Metroid was one of our references, but to create a game like this, you need to make this core not only to the level, but to the game design.
Just to mention two key features: new skills you are learning now and then, and easy to spawn, varied, enemies that fight with nice patterns and that create interesting combinations.
At some point of the development we haven't implemented yet these features, but we needed to keep building levels if we wanted to meet the deadlines.
So we built them with the "tools" we had at that moment. It's a nonsense to build levels assuming that you will have core features "in the future".
With the tools we have, we only could create a linear-one way experience, so we focused on that.
I'd argue that Braid and Monkey Island are just as immersive as more cinematic games. If the player is expecting a cinematic experience, and runs into challenging puzzles (as you wrote about), that immersion is lost. Similarly, if you were playing Braid and ran into a bunch of cut scenes and heavily scripted events, the immersion would be lost, too. Perhaps priming in the early stages of the game is essential to communicate to the players what to expect from the rest of the game.
I guess that achieving "immersion" is always one of the goals of any game designer, in any case.
With Deadlight, however, we also wanted to craft a "cinematic" experience, and for that, variety seemed to be more important that depth.
I applaud the team not only on the design decisions, which I agree made the game a better one, but also on the production decisions that allowed the game to ship on time. Scraping a feature that isn't core to the experience, and instead embracing the available systems to create content - this should be mantra among game devs.
How important and sadly sometimes undervalued is playtesting!