Learning From Crysis: The Making of Crysis Warhead

Tomorrow, Crytek and EA Partners will ship Crysis Warhead, a PC-only follow-up to last year's Crysis that falls somewhere in between an expansion pack and a traditional sequel.

In addition to further optimizing the game code (Crytek hopes to combat the belief that high-end PC gaming demands thousands of dollars), the Crysis team learned a number of lessons about design and pacing, taking player criticism of Crysis to heart.

Though recently-formed Crytek Budapest took the lead on the project, there has been overlap from the first game -- such as with Bernd Diemer, a designer on that title who now serves as producer for the franchise.

Diemer recently sat down with Gamasutra for an in-depth postmortem-like discussion about Crysis Warhead's development, during which he addressed the reasoning behind the game's unusual format, the tricky balance between player guidance and player freedom, the maddening intricacies of NPC AI, and the problems with the often-maligned drastic shift in gameplay towards the end of the original Crysis.

Here, in a discussion shaped by Gamasutra editor Chris Remo, we present Diemer's own words about the development process of Crysis Warhead.

On Warhead's pseudo-sequel format: What we tried to do is the best of both worlds. Back in the day, when we first started thinking about Warhead, it started out as a vanilla expansion pack. That's the kneejerk reflex of developers: "We're doing a game? We're almost done? Let's do an expansion pack next."

The initial design was more along these lines. You needed the original game to play it, it was single-player only, and a lot shorter.

Then we brought in the Budapest team. Around that time, when we were finalizing Crysis, we were building up the Budapest team with their own project -- giving them full studio status within our small Crytek family. We brought them over to Frankfurt to meet everybody and get them trained on our tools and our engine. So we suddenly had twenty guys with weird names in our office, speaking a language nobody could understand -- although they speak English as well.

It was a very intense time for us, so we decided, "Let's have these guys take a look at the expansion pack we wanted to do." We told them, "Here's our box of toys. Please think about how we can improve it." Back then, we knew we should have done certain things in differently given more time or a less hectic development schedule.

They went and did prototypes in the editor, like what a mod team would do. They used whatever was available in Crysis to prototype new gameplay, such as the new vehicle rides, aliens breaking through the ice -- those made it from the first prototype video into the final game.

After we looked at all the ideas and prototypes they had for the then-expansion pack, we said, "No way is this going to be used in an expansion. We have to make it standalone." We then decided to develop a longer campaign -- a full single-player campaign -- but we kept the $29.99 price point, which leans more towards the expansion pack.

On structural comparisons to Valve's Half-Life 2 episodes: The main difference is that Valve continues the story. What we did is tell a parallel story on the other side of the island. It gave us the opportunity to evaluate our own franchise a bit more. Crysis Warhead is the first time we, as Crytek, explored a franchise we had ourselves created, past the original game. It was an opportunity to look at what we could do to keep it fresh.

We decided to have the first game's character of Sykes be the main hero, and tailor the game a bit more towards his personality and his love of blowing things up -- wherever he is, things tend to blow up around him. It's not always his fault, but very often it is. That drove the whole design.

We wanted to keep the original Crysis vision, which is the sandbox freedom and tactical freedom that allows you to play the game however you want -- stealth, action, whatever. From the beginning, there had to be a lot of things that blow up on the main path -- if you stick to the main path, it's a lot more action-driven than Crysis used to be.

On concerns that the action focus could detract from the sandbox feel: From the pacing side, we decided not to have gameplay-specific levels like we did in Crysis, where we had, for example, an entire tank level and we tailored the entire level to have the most preferred experience being in a tank.

In Crysis, we had a bit of an old-school mentality to level design: "Now, it's the vehicle level." In Warhead, we have shorter sections within levels -- you don't have one level with driving all the time. There would be a short vehicle section, an infantry section, maybe another vehicle section, then some aliens, then some North Koreans.

It's a lot more entertaining if you mix it up more, so players can't identify, "Now I'm in the sniper level," or, "Now I'm in the VTOL level." There's a possibility that something else is coming.

From a creative standpoint, it makes it much easier to keep players engaged if you can surprise them from time to time, forcing them to experiment. Then, we encourage the player to realize they don't always have to do it in a particular way.

We didn't want to force players into mandatory vehicle sections we had in the original Crysis. In a hovercraft level, for example, you can stop at any time, and there are other weapons and vehicles lying around. If you prefer the classical sandbox gameplay, where you come up on a challenge, observe, and grab the appropriate weapons, we designed for that; or, you can go the Skyes way with dual SMGs blazing and blow the hell out of everything.

On criticism of the very linear last third of Crysis: Vehicle rides, even if they are on rails, can be a lot of fun, just shooting lots of targets, but in Crysis, the switch in gameplay was just too great. We spent eight hours teaching the players, "Do whatever you want," then suddenly you jump into a vehicle and we just turned off the exit key. Of course, the first things players did was say, "I want to get out."

We taught them before, "Do what you want! Freedom!" Then we broke the design rule that we ourselves created and spent a lot of time teaching players. Suddenly, that rule wasn't valid anymore, and that is an abrupt switch.

From a pacing standpoint, we switched almost entirely to alien combat at a certain point in the game, and from then onwards the expectations were set. The players knew, "From now on, it's aliens." Forum posters talked about the first part of the game, and the second part of the game. The public perception was really driven by these design choices -- there was "pre-alien Crysis" and "post-alien Crysis."

In Warhead, we tried to stay away from that. As a team, and as a company, we're now a lot more familiar with the IP we created. Crysis' nanosuit actually came into development rather late, while we were building the game. As a game designer on Crysis before I moved into the producer rule, I had to convince level designers into changing levels to be more suitable to the nanosuit gameplay. If somebody had spent three months working on a level, it can be tough to accept a different mindset for that level. Now, as a team, we are much more comfortable with how levels have to be built. The nanosuit makes sense.

If you give complete freedom, you have to make sure that players at least get an indication of what they could do next -- not just drop them in and say, "Do whatever you want!" So in the level design, we brought everything closer together. The levels aren't smaller, but the action elements are much more condensed.

On pacing and "freedom that isn't really useful": In Crysis, we had a lot of pacing issues, which we sometimes didn't even realize because we had a lot of other crucial things to do. It sounds silly, but we really did get to that point, where we were saying, "We'll fix the pacing later, fix it later, fix it later -- oh my God. Now we have to do it."

We then had to make some harsh design calls. There's a level with the player riding in a Jeep driven by Prophet. That level had decisions made with the best intentions, but also driven by the reality of our situation, with people clamoring, "Get it out the door!"

That level was designed around an experience where you can drive wherever you want, you can get out, or you can stop. In that kind of level, you have to make sure that there is actually fun to be had throughout, and there is something to do when you get out of the vehicle. At the minimum, players need weapons to pick up throughout the level, or you just stand there with no weapons, and then you die. Where's the fun in that? That's an example of freedom that isn't really useful.

In Warhead, we designed those vehicle sections with two approaches in mind. We started with the main vehicle path along a road, then we expanded it outward. We would say, here we want to have an action bubble where the player can stop, and we would tell the designer to be sure there are weapons around, that AI is properly set up and can call in reinforcements -- all the basic rules we set up in Crysis for infantry combat had to be present throughout vehicle levels as well.

Once we learned that language, those design guidelines and design rules, we filtered it through the vision of the Skyes character -- things have to blow up, and the player needs the right tools and the right opportunity for that to happen, if the player chooses. Meanwhile, we had to keep the rules from Crysis, allowing the player to play more stealthily from a distance. That was our challenge as a studio; hopefully, we succeeded.

Even with vehicle levels, you can do the whole thing on foot -- just forget the vehicle, and go action bubble to action bubble as the Predator. Everybody should play the game however he or she wants to -- we tried to cover all the potential approaches somebody would want to take.

On the balance between guidance and freedom: From a creative standpoint, it's okay to guide players. It's not okay to force them, but you can make strong suggestions. For example, when we started working on the hovercraft level, we had a gun on the vehicle. What we didn't want players to do was to stop and shoot too often.

Because we had aliens, of course people stopped and tried to shoot the aliens with the mounted weapons. The result was that the challenging driving section turned into a challenging shooting section, and players were crashing all the time.

We decided to just take away the gun from the hovercraft. At first, everybody said, "No! You can't do that!" But then, we noticed that driving, not killing the aliens, became the challenge, and that's what we wanted. It suggests strongly that you should keep driving, because you don't have a gun. If you insist, you can get out, but the main path is driving the hovercraft.

We discussed player guidance versus player freedom endlessly -- back and forth, up and down. We debated every decision. When we did Crysis, we had very set rules at the beginning: Never force the player to do anything, and never take control away from the player.

In principle, those are good rules, but in some cases it's okay to bend the rules a bit, or to provide a viable alternative to the player.

For example, it's okay to tell people, "I would suggest you drive a vehicle" -- but don't kill them when they get out of the vehicle, or provide no other alternative to driving a vehicle. Otherwise, you're teaching by dying, which is a bad design mechanic.

On finnicky AI decisions: In an escort level we've been showing off, the player has to accompany a pilot character in a vehicle. We originally had the pilot sit in the same vehicle as the player, and would automatically switch seats: If you were on the gun turret, he would take the driver's seat, and vice versa.

We ran into a problem there: If you decided to leave the vehicle, what should he do? Should he get out of the vehicle as well, and follow you on foot? Should he keep driving? We tried having him mimic what the player was doing, but players do the weirdest things.

Some players decided to swim across the bay, and then the pilot would follow and be attacked by a shark and die, and then the level would end because he's a story character, and the mission requires him not to die. He's hard to kill, but he can be killed -- you just have to work at it.

There, we were telling players, "Do whatever you want," but the moment they got out of the vehicle and into combat, enemies started shooting at the player, and very often the player would get a failure message without knowing why. That was the result of situations like an enemy throwing a grenade at the player, the player dodging it, and the pilot attempting to follow the player and step directly onto the grenade. Bam, he's dead.

We eventually decided to add another vehicle for the pilot to control on his own. From a design and programming standpoint perspective, we could make that vehicle a bit more invincible. It's not impossible to kill, but it's harder.

It also allowed the player the alternative options we originally wanted, without the confusion: If the player wants on-rails action, he can just jump onto the pilot's vehicle turret and have fun.

That experimentation actually took about two months of design iteration, but it was very valuable. By making that simple decision of plopping the pilot into his own vehicle, we allowed the player more choice. The player can be in his own vehicle, on foot, or in an on-rails experience. To make sure the player knows that, if you leave your vehicle and stay outside, he eventually calls you over the radio and invites you to jump into his car.

On the importance of design iteration: The implementation we achieved in Warhead for these sections is not perfect, but I believe it's very important to try and explore that kind of "implied freedom," as we call it. The player can do whatever he wants, but the art of game design is to subtly suggest to the player what he should do.

The curse of AI design is achieving, "Do as I want, not as I say." AI has to try and anticipate what the player really wants: Does he want to get on the gun? Does he want to drive? Did he just get in by mistake, and then exit the vehicle? Should I wait? Should I keep driving? That's a nightmare for AI programmers and level designers.

Our original implementation wasn't smart enough to recognize when the player wanted to catch up to the pilot, so we ended up in a situation where the player was running after him, and as soon as you got close, he would drive a bit more. You ran after him again, and he drove a bit more, and so on.

We eventually needed to create a rule for the pilot that allowed him to find a safe spot where no enemies were shooting, wait for the player, wait long enough that even if the player is clumsy he has long enough to get into the other vehicle, and then keep on driving.

It's important to explore these possible things a player would want to do, and try to design them to their logical conclusion, even if it's causing developers a lot of pain and effort to do so. To just stop and say, "That it, just lock the player in the vehicle!" is the easy way out, and you should avoid that if you can.