Planning for Planning
I’ve been playing a lot of Assassins Creed Odyssey lately so my head is filled with forts, outposts and regimental camp layouts.
I get like this sometimes when I am get completely absorbed by a video game.
I was trying to articulate a layout on this nature lately and I realized something rather important that I used to be aware of last year, but I sort of failed to articulate it properly.
A good game-play loop in a video game like Assassins Creed or Ghost Recon is based on 3 game-play pillars:
- Scout
- Plan
- Execute
The Scouting Phase:
The Scouting phase implies allowing the players to have a good overview of the layout he is about to attack.
Much like in real military operations this phase is about collecting information about what’s going on in the layout, so clear positioning of affordances and realistic functions assigned to key layout spaces help the player come up with a plan.
I am not necessarily talking about about actual functions, like where is the barracks, where do they eat, where to they get their orders, or where do they store their vehicles or weapons, although these are also very important, but I am talking about clearly showcasing game-play opportunities like:
- Where are the entrances?
- Where can I go up?
- Where are the vantage points?
- Where are the enemy patrolling?
The Planning Phase:
This helps the player articulate a mental plan:
- I can take this path here, then climb on that, then hide in the bush.
- I can sneak under the fence, hide in the bush, then climb in the tower and secure a vantage position.
This implies to a large extent that the game-play is driven by these game-play opportunities and more exactly by how many of them are available in the layout:
- Few is bad — Players will feel restricted
- A moderate amount is preferable
- A lot is bad — Players will get overwhelmed
It also important to make sure that the information that the player gets is also incomplete and the player needs to actually explore the map in order to be discover more.
The Execution Phase:
Once the plan is set the players can then engage the layout and attempt to execute it.
Since the information is complete this leads to a more then often flawed plan (We don’t want the players to have flawless plans). This will lead to situations where the plan will need to be readjusted on the fly and even the game-play style that the player has chosen to change.
The trick is to ensure that all game-play styles are valid, so the player won’t feel frustrated for failing the stealth (for example). MGS 5 does this really well, by allowing the player to engage and disengage in a variety of ways. AC Odyssey does the same thing.
I often find that in AC Odyssey this becomes really obvious when then player is fighting Mercenaries on high difficulty options, where direct one on one engagements are not an option.
The Contrast
This allows for a very structured but flexible level design formula that permits the designer to focus on “planning for planning” assuming the player will:
- Have total freedom of choice
- The game will support systemic interactions
- Game-play Style shifts are possible and there is enough space for emergent game-play from the said systemic interactions.
This goes in stark contrast with the traditional way of doing level design where the designer does not focus on “planning for planning” rather on “one beat at the time” experiences.
I took the liberty of drafting a crude table of what I feel the differences between the two types of level design approaches are. Feel free to contradict me if you disagree:
On the left we have the “Planning for planning” approach where the player:
- Knows what the information is because he can scout it.
- The action in the game happens either because the player wants to interfere with how the systems interact with each other or as an interaction between two systems.
- The player has the ability to throw a wrench in the gears and destabilize the systems so it generates interesting results.
- The player can plan ahead and the solutions might be complex.
In this kind of “sandbox “ environment” the focus is on interesting systemic interactions that can lead to interesting player stories.
On the right we have the “One beat at a time” approach where the player:
- Does not have the ability to scout ahead
- He is faced with dealing with each bit individually, in a precise order and can’t circumvent the script.
- The action happens to the Player, as a scripted event. The player just triggers the scripted moment.
- The player has no knowledge of what’s ahead unless he restarts the checkpoint.
In this kind of situation the focus is placed on created awesome set pieces, scripted moments and delivering an already defined narrative that does not refer to the player but to the player character.
Designers are usually trained in one of these approaches as a result of their experience and it’s often hard to switch from one mode to the other.
This might feel like a design direction thing, but in fact it’s driven by the technology that permits this kind of design.
These things however are not mutually exclusive and do overlap sometimes.
However in the context of very large video games, like AC Odyssey, Ghost Recon, Watch Dogs, Far Cry or any other open world game that you might think about planning is important, not just for planning but also because in most cases the muscle that drives the “scouting” brain also drives the review brain, so it’s an important tool to bridge communications between designers and directors that might not have too much time to play the game, but can interpret the affordances you are showcasing in the context of the game vision.
So please plan your maps before you build them.