So with the new DP, internally, we've been talking about something I call "the Inversion," which is a potential solution to a design problem we've had since the start. But before we get into the Inversion (which is cooler than it sounds), let's start with this:

Oh my. New players suck at breaches. Basically a lot of firefights break down into the above. Here's why this is bad:

  1. Doorways are uninteresting; they're always the same space. It’s just not the actual level geometry that's designed to be interesting to fight in.
  2. The strategy is always the same: pie off the room and engage at distance.
  3. Randomized maps matter less when firefights take place in the same area every time.
  4. It does not fulfill the fantasy of room clearing, and fails to capture the violence of action.

That last one is most important. As game designers, one of our primary jobs is to fulfill a fantasy. In this case, one of the core fantasies of room clearing and CQB is Violence of Action, especially when it comes to the moment of the breach. Players are meant to breach, bang, and clear (both the room and the fatal funnel) with as swiftly as possible. MoH: Warfighter demonstrates it well:

He doesn't even clear the fatal funnel. Noob.


So what's happening in that video is an execution of the fantasy of room clearing: a breach is made, the flashbang thrown in, and then the attackers rush into the room, and fucking murder everyone while they're stunned. This is the power fantasy of playing the cops in our game. Except our game is a multiplayer game so we can't resort to slowing down time to briefly make the game into a target range. And we don't have tomahawks.

Right now we’re not fulfilling that fantasy. Instead we get this:

Well, I guess this is my fault.


So why are our new players playing like this? Because it's somewhat effective; it's not a dominated strategy. For example: if the police only got shotguns, they'd approach the game very differently. They'd move into the room as quickly as possible in an attempt to get within effective range. Pieing off the room at standoff range would never work, and players would adapt.

The Cop's assault rifles are designed to be deadly-accurate at range. While aiming, recoil is very controllable, and you can score one taps or double-taps to kill bad guys at a distance fairly easily.

Crackshot BlueDrake

Meanwhile, on Team Terrorist:

See, the bad-guy weapons are designed to be... bad. This is in part a fulfilment of the fantasy--generally criminals don't have access to state-of-the-art weapons, and don't have the training to use them well. But that's not the operative justification for that design decision, it's this:

Since our truck always spawned in the same position, if any door faced westward, defenders had a new strategy open to them: interdiction.

It's not a bad strategy. It's actually quite an interesting one, since the only way it works is if the cops aren’t expecting it. If the defenders had good long-range capability, it would be a dominant strategy.

But why not simply move the truck spawn?

BECAUSE I WAS LAZY, that's why. It was going to be a bunch of work to spawn the truck in a safe position, not in line of sight to the doors. So this was the quick fix. That quick fix is responsible for exacerbating one of our biggest design challenges, which is getting new players to properly breach and clear a room.




The Inversion. Make the police ineffective at range and make the defenders effective at range. As in Counter-Strike, both sides will be deadly-accurate while still, but since the police must “push” or move into position to fire, they’re at a disadvantage. If cops try to pie off a door instead of violently banging and clearing, they will die. That dude with the Kalashnikov is going to double tap them and then they’re going to fall down and bleed to death.

In order to tie this all together we need a weapon that performs best while in motion. Fortunately, one little game happens to have the handling we’re after:

Sun Tzu said: Defeated warriors go to war first and then seek to win, Victorious warriors win PRO90 first and then go to war.


And yes, the truck will move into a safe position. Finally.

It fits!


So how will you play Due Process as the police? You'll make the breach. You'll throw the flashbang, you'll clear the room. You'll do that or you'll die, not because we said so, but because we've designed it so.

Now watch all this get proven wrong when we do our first test.




You're still here? Nerd.

So how can this be avoided in the future? I'm not really sure. The obvious answer is "don't be lazy," but there's a defense of laziness: it gets shit done quicker than properly implementing things.

There's an adage in this business: "fail faster.” In the context of game development, that phrase suggests you should get your game to a state in which you have succeeded or failed as quickly as possible. Basically, favor the quick and dirty implementation and see if your concept is sound before you sink a bunch of resources into it. From that perspective, it was still the right move.

Really the only problem here was us committing one of the cardinal sins of game design: blaming the player. Our opinion for the longest time was that players simply didn't know better. After all, we (the designers) knew better and all of our breaches went hunky-dory. The fact of the matter was that we hadn't implemented the mechanics would enforce the dynamic we desired. If we had viewed the problem from that lense, maybe we would have arrived at this conclusion earlier.

Whatever, at least we didn't design 10 door breaches that play out in the exact same way.