What's wrong with Heavy Rain's controls?

Above: Heavy Rain characters contemplate internet criticism.

On Friday, I mentioned that some of the criticisms of Heavy Rain seem off-base to me. The biggest one has to do with the controls. There seem to be two essential complaints about them. One is that you're not really controlling your character. The other is that they are arbitrary quick-time events. The result: You are providing inputs at certain times, but you're just watching the characters perform the actions.

This is factually true, and experientially insignificant. I agree with Tom Cross when he says: "I’m not sure what everyone is complaining about when it comes to Heavy Rain’s controls. They’re too abstract, or something? They’re not like 'real' game controls? It’s really unclear."

Start with the question of abstraction. Sure, buttons don't map to specific actions. That's because the actions your character can perform are so different. Using the right analog stick to perform the bulk of the character's actions does seem to me to do the job, especially since Quantic Dream made an effort to have the control input track to the onscreen action (like pressing it to the right in order to open a sliding door). I suppose you could have a context-sensitive action button instead, but, really, what's the difference? Well, I can think of one: pressing a button feels less like opening a door than moving a control stick does.

Heavy Rain does have its share of action scenes that wouldn't look out of place in other games, like fistfights, and, yes, they do control differently than you'd expect. You don't have specific actions like "punch" and "guard." Instead, you have to quickly respond to onscreen prompts within a time limit. This, again, seems to bother people who are used to more traditional playing styles. I've found it to be rather exciting. I never know what's coming next, and I tend to mess up just enough to leave the outcome of the scene in real doubt.

"But you're not really playing it!" some complain. Well, I'm pressing buttons to create onscreen action, aren't I? Maybe I'm being obtuse, but I don't see a dramatic difference between these two scenarios:

-Press X when prompted to dodge an attack (Heavy Rain)
-Press the left trigger when the opponent's weapon flashes to dodge an attack (Bayonetta)

Bayonetta has been praised as pure gameplay, while Heavy Rain has been criticized for being all cutscene. Yet in each case, it's a matter of split-second reaction to visual cues onscreen. I'm not saying they're exactly the same, but I am saying that it's a difference of degree and not of kind.

A control scheme can only be judged by how well it allows the player to experience the game. Other than the first 10 minutes or so that it took me to get used to making the characters walk around, I've found the controls of Heavy Rain to be up to the task. No, they wouldn't work in many other games. They don't need to.

Big Red Potion

I mentioned it in passing yesterday, and now I'm happy to post the link to Big Red Potion Episode 20, "Out of Our Hands."

I joined hosts Sinan Kubba and Joe DeLia, along with guest and friend of Insult Swordfighting, Michael Abbott, to discuss game controls. As N'Gai Croal has said, we see games with our hands. The controller is the nexus of intention and action. Yet most discussions of games skip right over this topic, or approach it with no more insight than "the control was good or bad." Or maybe we make a snarky comment about waggle.

With that in mind, we tried to take a broad view of play control. We talked about notable successes and failures in the history of peripherals (somehow without mentioning either the U-Force or the Sega Activator in the latter category), what effect the Wiimote has had on game design, and what the future holds for Project Natal and Sony Motion. And we wondered whether standard control inputs are limiting gameplay possibilities, or whether new controller types will lead to a bunch of shallow gimmickry.

It's always hard to judge one's own performance on these things (plus, the sound of my own voice gives me hives), but I can say with confidence that Sinan, Joe, and Michael all said fascinating things about this most important of videogame topics. Have a listen!

In defense of the gamepad

Above: The humble, hard-working gamepad.

Pity the poor gamepad. He's the least popular kid on the block. The ever-evolving controller has been the primary input for almost every game that's come out over the past 30 years. From the single-button Atari joystick to the precision-engineered machine you see above, the gamepad has been an integral part of the video game experience, even has the industry has grown into the multi-billion-dollar behemoth it is today.

And the console manufacturers can't kill it fast enough.

Last night, I was a guest for the recording of the Big Red Potion podcast (I'll post the link when the show is available). The topic was play control: the state of it, how it's evolved, where it's going. We spent some time talking about specialty controllers, like the Guitar Hero guitar, and the future of motion control, especially with regards to Wii MotionPlus, Project Natal, and the PlayStation motion controller. To look at the direction the hardware is going, you'd think that the gamepad has become obsolete.

Anyone would agree that game controls are more complex than they used to be. Game designers can go overboard with button combinations and functions, sometimes with hilarious results. It's not immediately clear, from looking at a gamepad, what each button and stick does. Non-gamers can be frustrated or intimidated by these bulky, multi-functional devices. This is all true.

But look at what else the gamepad does. It stands in for a steering wheel in a driving game. It replaces a stick and pedals in a flight sim. It eliminates the need for a keyboard and mouse in a first-person shooter. It supports the split-second reflexes you need to play Street Fighter or Devil May Cry, and it credibly stands in for all the sporting equipment you can think of. This thing does it all -- no bold new paradigms needed.

Gamepad design has progressed in a steady, upward fashion. Thanks to years of iterations, dozens of good ideas have become standard, while bad ideas have been phased out. Gamepads today are ergonomic, with sleek, comfortable curves (not like the pointy-cornered NES controller at all), and triggers positioned just so. Rumble is standard. Pressure-sensitive triggers have made vehicle control miles better than it used to be. For all the talk of Wii/Natal/PSMotion being a revolutionary way to play, I'd argue that nothing has fundamentally altered the way we play as much as the move from digital to analog thumbsticks. At the time, it seemed like such a small thing, but playing in three dimensions would be impossible without it.

I'm not saying the gamepad is perfect, and I'm not saying it's a good thing when overly complex control schemes put people off. Simplicity is a virtue in game design, and hardware design. What I am saying, though, is that the gamepad sometimes goes unappreciated. It's not enough for any of these brand-new control methods to have one killer app with which they work perfectly. Natal and Sony Motion are going to have to prove that they can be as versatile as traditional methods when playing traditional games. Otherwise, I'll stick with the gamepad.

Standards and practices

Ryan downloaded Bionic Commando Rearmed and seemed to enjoy it, although he had a complaint about the button mapping. On the Xbox 360 control pad, you press the B button to grapple and the A button to fire your weapon. Something about that seems off. For the sake of ergonomics and usefulness, you'd think playing with the A and X buttons would make more sense. Look at the picture of the control pad atop this post and tell me that wouldn't fit the human thumb better.**

A similar phenomenon has afflicted an awful lot of 8-bit remakes, for reasons I don't understand. They can't -- or won't -- properly replicate the NES experience. A few years ago, I was playing the Mega Man Anniversary Collection for the GameCube, where the developers had swapped the functions of the two buttons. They mapped firing to the A button, and jumping to the B button. This is exactly backwards.

When I remember about playing those simpler NES platformers is that all of them worked the same way. You hammered your knuckle on the rightmost button to jump, and fired your weapon with the tip of your thumb. This was an ideal arrangement. The leverage of your thumb allowed rapid-fire button presses for the main action without sacrificing the more deliberate pace of most jumping mechanics.

Why do so many remakes reverse the button positions? Partly, I think it's a rudimentary mistake: on the original NES pad, the B button was on the left and the A button was on the right. On the Xbox 360 pad, the A button is on the bottom, and the B button is on the right. Keeping the same button assignments results in a reversal of their positions.

This is a bigger deal than it might seem like, and not just because it made such an awkward experience out of Mega Man 2, which I was playing for an audience. It comes down to user experience and expectation. There is a temptation among tech people sometimes to blame their users when the users have trouble with a product. Although this is sometimes true, most of the time it's because the designers didn't do their homework with regards to usability.

The Web usability expert Jakob Nielsen likes to say that website designers ought to do the same thing as 90% of their competitors, because that's what users have come to expect. They have learned to navigate web pages in a certain way. It's the same with games. An Xbox 360 user playing a platformer will expect the A button to make his character jump. If you're going to deviate from the norm, you'd better have a damn good reason.

This isn't to suggest that progress isn't desirable. When a change in control is worthwhile, then it's going to happen. Ten years ago, any racing game you played assigned the gas pedal to one button and the brake pedal to another (mapped to a modern day 360 pad, you'd probably go with A and stop with X). Today, these functions are assigned to the analog triggers in nearly every game, because that's a better way to do it.* User expectation has kept pace. Nobody picked up Burnout Paradise and pressed the A button to accelerate.

Too Human shows some of the pitfalls of going your own way. Users expect to be able to manipulate the game camera in a 3D platformer by using the right analog stick. But Silicon Knights had a different approach in mind for their dungeon crawler, and instead assigned the right stick to melee attacks. It's an interesting choice. I don't want to knock it simply because I wasn't used to it. In fact, it wasn't even that hard to get used to, and works pretty well in some ways.

The question is: Is what the developers gave up with this new approach equal to what they gained? I'd have to say no, and not just because poor game cameras are one of my top 3 videogame pet peeves. While the melee action does what it's supposed to, using firearms is practically impossible. Cycling between enemies doesn't work too well, especially when the boss you're trying to shoot at is standing behind grunts. That's a pretty big part of the combat.

Standards become standards for a reason. Nothing is beyond revisiting, of course, but it's not helpful for 90% designers to eschew decades of user expectation because they think they've found a marginal improvement. Maybe Too Human wouldn't have been the same game if it copied the control scheme from God of War, say, and had you press the X button to attack and use the right analog stick to dodge. But one of those games relied much more on the user's built-in expectations -- maybe it's only coincidental that it was the much better game.

*Another excellent example is EA's Skate. They took a completely different approach to videogame skateboarding than did the Tony Hawk series, and it was a stroke of genius.

**John Barleycorn points out in comments that you can remap Bionic Commando's buttons, if you want to. This is always the best solution. Even so, the default configuration ought to be the most familiar one.