(Amis visiteurs francophones, retrouvez cet article en français ici)
Even though Call of Duty Modern Warfare 2 has many flaws, many agree its multiplayer maps are among the greatest ever made. (Okay, maybe not Estate). Today, we’re gonna look into how Favela was made, and specifically what has been made to balance the map. A little disclaimer here: I’m gonna use some reverse engineering on how I think the map was made and what were the design decisions behind it. So it won’t probably be very accurate in front of the real process Infinity Ward did on this map, but anyway, the exercise is still relevant and above all, quite fun. We will se later the map topology carries a lot of history, and it’s quite do-able to see what has been changed iteration through iteration. So let’s go down there One Forty-One!
What are the 3 words coming to your mind when I say « favela » ? The very first step is to gather relevant photos from actual locations. Based on them, we can extract some keywords that’ll define the overall gameplay of the level. Here are a few reference photos from favelas we can encounter for instance, in Rio, Brazil:
How about Verticality, Confined and Swiss Cheese ?
Let’s start from here. The major gameplays constraint every call of duty map has to face is the ability to handle all the objective-based gametypes without altering the geometry. So every map must be suitable for Domination, Capture The Flag, Search and Destroy and Headquarters. (Demolition and Sabotage have same objective locations than S&D).
Let’s start with the level design rules carried by the Domination and Capture the Flag objectives. They will define the basic topology of the map.
- 3 flags to capture and hold
- 1 flag at each base
- 1 flag at the middle, which is reachable at the same time by both teams when beginning a game
- the flag at the middle is the hardest to capture or defend
Capture the flag Gametype:
- 1 flag to capture, 1 flag to defend
- topology is « team base – route – middle – route – enemy base »
- there are less than three major routes to get to the middle
- these are the fastest way to reach the middle
- these are generally the most exposed to gunfire
- there are a few secondary routes
- they are more protected from enemy fire
- they are longer than the majors’
Okay, now we’ve got ourselves covered. Let’s use the S&D gametype to narrow it down further:
Search and Destroy Gametype:
- We can basically use the Counter Strike defuse gametype level design rules
- 2 objectives to choose from
- one of them very close to the defending team
- the other is equidistant from both Start points
- 2 major routes
- several minor routes
- they both inherit properties we saw for CTF gametype (about length and cover)
Let’s take a look again at the three keywords we got and let’s try to convert them into a level design feature:
Verticality: The map needs to have a single height variation. It must be built on a slope, but we need to be careful to prevent too steep slopes (kills the field of vision), or giving too much an advantage to the team in higher grounds (even if there is a side switch mid-game).
Confined: The map needs to have very narrow routes and tall buildings alongside. We also need to be careful because on one hand it kills proper player navigation (i.e. when two player on the same corridor are colliding because of the lack of space) and on the other hand, it kills interesting combats : a player will have more pain in dodging enemy projectiles.
Swiss cheese: The map needs to have litterally a lot of holes, climbable windows, partially constructed roofs, … Here the trick is to prevent the map from going too complex by adding loads of openings everywhere. It prevents the player from easily memorizing paths and objective locations. So again, we’ll need to be careful.
With all the information we have, it’s time to build the first draft. The goal here is to get a first version somewhat on par with our 3 keywords.
1st step, bases and mid locations extracted from final layout:
Given the final topology, we can deduce the location of those keypoints easily. This is the history of the map that is visible here and let us place these points: The map has a width greater than its height, so naturally, the points A, B and C will spread out along a virtual line crossing the middle of the map. A and C go at each end. B goes where the space is the largest; i.e. the soccer arena.
Based on this, we can now extrapolate on the verticality. One of the base will need to be higher than the other. It’s easy, when we look at the floorplan, to determine which one it is. Actually, I don’t think it was the other way around during the development of the map. Indeed, it may be C that was higher at the start, but it’s harder in a level building point of view to retake the whole map and change the topology to make C higher than A, when you can simply mirror it. As to put the slope north/south or south/north, it may be the same thought process driving this.
Before we start to dig, here are a few conventions. B stands for the Middle. The attacking team is coming from the A area, the defending team is coming from C.
Ok, we clearly see there’s a problem with the routes and the middle. It’s too close to the C spawn. Elsewhere, the north secondary route does not directly meet at the middle. The middle may be misplaced after all. That’s a shame, because it was meeting many constraints : open space, hard to take or defend, and junction point for the routes. Speaking of the routes, the north one does not exactly meet on the middle; which is also an issue we’ll look into later.
So let’s nudge the middle towards the A spawn, right where the secondary routes are meeting. That’s an interesting twist of the original CTF rules, but hey, you can still break the rules if you know what you’re doing…Meaning you’ll have to understand them first. So yeah, the B flag will be hard to defend within those corridors. It’ll probably end up in a grenade fest. We will need to add some extra layer of depth to make all of this interesting. Moreover, those secondary routes become primary means of getting to B.
There’s a strong problem for the defending team with those 2 paths to reach the middle. They are too short sighted, meaning you can’t have a good line of sight of the enemy entrance path.
Opening the bottom part with door and stairs does the trick in giving more control of the lower team in defending or attacking the objective. By the window, a player can handle both guys coming from the narrow street and the side door.
But maybe the tweak we did is a bit too overpowered. Because from the window, we have a clear advantage over our enemies, and since they appear right in front of us, dealing with them is easy (flashbang + mid cover of the window frame).
To balance all of this, let’s add secondary routes and openings, located specifically to force the window bystander to move farther its mouse to aim at his opponents. The first thing I would have made, is placing an entrance here, on the clear advantage of the attacking team:
But IW did it this way:
Why ? Simply because there is no more interest in taking the original path. If I put it differently, the attacking team will have a clear tendency to use the new path, and the opponent won’t know from where the attacking team would show up. It also forces teamplay as the attacking team is bound to throw a flash grenade to quickly reach the side door. But now the defenders know exactly from where the attackers will come from (and that’s bad!). It’s time to add another layer of depth by letting the attacking team reach the upper floor and have a line of sight on the shooter of the window.
Sounds good. The terrace is also exposed to one of the B pathes which is a good thing to prevent the place from getting too much power. And it might get it since the window and the path are roughly in the same aiming angle, which is…a bad thing again. We’ll look now at how to balance all of this.
The easiest way is indeed by letting the players walk on the roof. At first glance it seems very dangerous because having a higher altitude generally means having a tactical superiority (you can see and track the enemy better). By giving two opposed access, we make sure the terrace is not overpowered anymore.
Moreover, it makes more dangerous the B point, because now any attacking team will need to watch at least 6 spots, placed at 360 degrees! (Terrace, 2 roofs, 2 paths and the window). So maybe we will need to add extra cover and some cheap trick to ease up a bit the capture process. That’s where micro tuning comes in place. The next pictures are worth a thousand words:
We’re done at this point.
I thought this article wouldn’t be that long to write! Anyway, in the part two of deconstruting Favela, we’ll look into the process of balancing the flag bases A and C (as well as their CTF counterparts), the main routes and the bomb sites. I hope it’ll be ready by the end of the week, but as usual, I can’t promise.
By the way, I strongly encourage you to do the same on other maps, especially if you’re an hobbyist or a professional. It’s a great learning process. If any IW employee is reading those lines, please share with us what went right and wrong in the makings of Favela!
Update: Part 2 is live!