Review: Ico

SCEA (creators)

Review written by Andrew Plotkin


Graphics
Stupendous
Atmosphere
Stupendous
Plot
Good
Writing and dialogue
Almost nonexistent
Gameplay
Very good
Puzzles
Mostly jumping/climbing/exploring, but excellent
Difficulty
Not too hard

A few days ago, writing about Silent Hill 2, I said: "In the past, designers struggled to make 3D games less crude; from now on, they can concentrate on art."

I was cheating, somewhat. I wasn't just thinking of SH2 when I wrote that; I was also recalling Ico.

While playing SH2, I kept saying "Aargh." Or "This is farkin' insane." While playing Ico, I kept saying "The light. My god, the light."

You can tell I'm impressed, because I'm writing in short paragraphs.

You are a boy, maybe twelve or thirteen years old. You have horns. (That's not a helmet.) You are being carried through the forest, willya-nillya, by three large men on horses. (One wears a grey cowl; two have horns.) You are taken to your destiny, which is an immense castle on an island, and you are thrown into a rune-covered sarcophagus to die.

That's the setup. That's what you know. The instructions booklet has a bit more detail, but you might as well not read it, because it removes some of the sense of mystery while adding nothing of value. Go with what you experience.

A timely earth-tremor lets you out of the sarcophagus, but that's the last help you get. Everything else is up to you. (True for both the player and the protagonist, eh?) So you explore.

You might easily compare Ico, in its basic game genre, to Tomb Raider. You explore. You have certain abilities -- jumping, climbing, swinging on chains, pushing crates -- and you use those abilities to navigate through, over, around, and within rooms of the castle. The rooms are designed to be tricky to navigate. Occasionally evil things come after you, which you must beat off with a stick.

But that doesn't even begin to convey what the designers of Ico have done.

First of all, the castle is huge. No, bigger. Really big. This is the biggest place I have ever seen on the screen of a computer game. Starship Titanic wasn't this big. I have no idea how they managed this. It's just polygons on a television screen, right? The sense of scale is entirely arbitrary.

But the scale's not arbitrary; it's immense. Maybe the camera angles do it, the close-ups and pull-backs as you move through these spaces. You are small, the castle is larger than you, and the horizon vastly larger than that. If you have any tendency towards acrophobia, you might want to play Ico in small bursts. I mean this.

(The rooms fit together, too. Towers to basements, you could build a model of this place. It's not passages tunnelled arbitrarily through living stone, nor irregular rooms stuck together with gaps. You go back and forth and up and down and you see everything.)

And there's the light. I don't know how they managed that, either. Some rooms are shadowy; sunlight filters in, turning the dimness to a dusty haze. Some rooms are open to the sky, and the light burns on the stone. The grass in the courtyard glows like a summer afternoon. Sunlight shimmers across the leaves of the trees, and when you look up into the sun, you are dazzled blind. I am not making any of this up, that's what happens on the screen.

I had no idea this was possible. So many modern games, you know, are in love with darkness. "Horror" covers more action-adventure games than every other category put together -- shadows, slime, flesh, decay, mist, biomechanical Giger rip-offs. And I love that stuff. But why has no-one before imagined what can be done with sunlight? Not magical radiance or plasma-flare; just sunlight. Sunlight and summer hazy heat. I swear, more often than not, I wanted to squint and shield my eyes.

All of that is just detail. You meet this princess. You rescue her from a cage. She's taller than you (thirteen, remember?) She doesn't speak your language. But, very obviously, you have to protect her.

It's all in the body language -- I'll leave it at that. No I won't. You are quick, nervous, aggressive. The girl is graceful, easily startled, curious. She'll come over to you if you wave. You can take her hand, and she'll follow you; or run after you, if you run.

-- and you're staggering along the corridor and she's right behind you, laughing, tugging on your hand, you're pulling one way and she's yanking the other and you're both laughing and running and there's an archway and you stumble out into the sunlight together and you see --

I did that with words. Ico does it with motion. I'll leave it at that.

Now, the plot revolves around the girl, and so does the gameplay. She can climb ladders, and you can pull her up onto ledges, but she won't climb chains or scootch along ledges. Contrariwise, there are doors that she can open that you can't. So the challenge is not just to navigate through the castle, but find (or create) a path that you can both navigate. Because the evil things are after her, too. If you leave her alone too long, they get her.

Since you spend so much time thinking about the girl, the designers put in a lot of work on her. She is not a totally passive character. She wanders around a bit; watches you; goes off to look at interesting things. Sometimes she runs ahead when you open a door, excited to see what's beyond. ...Truly, I think there could have been even more of this. She does get fairly predictable by the time you've played most of the way through the game. A few more behavioral quirks, to sustain astonishment, would have been nice. But this is a minor, minor kvetch; overall they've done a terrific job with her.

So it goes. The background of events becomes clearer. You explore more. You fight off more evil things. (The fights did get a bit tedious. Not very tedious, because the action is not difficult and quite viscerally satisfying, but it is mostly all the same. You eventually get to beat them off with a sword instead of a stick, but that's about the extent of the variety. I do understand why they included the fighting element -- the game, medium-sized at best, would become very short without the pacing effects of the periodic battles. Nonetheless, they needed more variation.)

The graphics, believe me, keep getting better. The early parts of the game are all stone and brick. Oh, you think, stone is trivial -- do huge flat rectangles and map a texture onto it. But look closer. The stone is old, damaged -- sections have fallen away, edges have worn off. Here and there. Just little details.

Then grass and trees. Later, water (amazing)... views out over the ocean (better)... a few caves and caverns... and that was when I said, "This looks as good as Riven did."

Okay. It's still true that pre-rendered scenes are better at some things. But Ico does natural caves, and jury-rigged wooden walkways -- twisted and irregular, not simple and foursquare. They look just fine. Not like wood mapped into a Quake level, but like wood. Riven had huge machines; Ico has huge machines. They look like machinery. The views are, okay, you get the idea, the scenic views are art.

(And yes, the camera conspires in all this. This game has dramatic pull-backs and vista shots -- even though the camera is completely dynamic and follows you around. No, I don't know how they did that either.)

So it goes, to the endgame. Now, one quirk I will note right away; there are no opportunities to save anywhere in the endgame. You are stuck in front of your PS2 for a solid hour. (Mind you, by that time, you won't want to drag yourself away.) This doesn't mean that if you die, you have to start the entire hour-long sequence over. The designers aren't that stupid. I'm pretty sure that there are implicit save points; you only lose a little bit of progress. (Furthermore, I say "pretty sure" because in fact I didn't die in the endgame. It's challenging, but not something you have to try over and over. Which is another clue that the designers are smart.)

The endgame is one final bout of exploring, climbing, scootching, ...(seeing even more amazing scenery)... and then there's the final battle, and the final puzzle.

The final battle is long. (Not infinite, but long.) And the final puzzle, well.

I believe this is the first adventure game puzzle I have ever seen that builds the challenge of a classic, thinking-person text adventure puzzle -- insight into a system of laws, experimentation, feedback -- out of the full range of possibility of a 3D action interface.

Sorry, that's a long and complex thought. I shall simplify.

The most trivial kind of puzzle, in an action game, is the key-and-lock. The key and lock are locations in the 3D world, but the essence of the puzzle is binary; you've reached the spot with the key, you've reached the spot with the lock. Two changes of state. The puzzle is really unrelated to the kinds of interactions that the rest of the game is about; it's not about how you walk or run or jump. The puzzle could be moved identically to a text adventure or a tile-based RPG.

A riddle is more complicated, and may be a good puzzle in its own right, but obviously it's just glued onto the 3D game. A dialogue box pops up and interrupts the action. Similarly, the legions of peg-jumping and tile-sliding locks that we all hate.

A switch-and-elevator is more integrated with the underlying nature of the action game. Sure, the switch is binary; but the elevator moves you through the 3D world, just as climbing or jumping does. It's less arbitrary, you see, less an alien intrusion into the kind of game you're playing. There are many possibilities for this kind of physical reaction in the game world: pushing over a pillar to form a bridge, for example, or rolling a boulder down a hill to smash through a door.

The Tomb-Raider sorts of puzzles (which comprise most of Ico) are very well integrated with the 3D world; the challenge is entirely a matter of physical positions of things. But, on the other hand, you usually know the rules. The truest form of an adventure puzzle involves insight; the deeper, the more satisfying. Realizing that you can push crate A over to ledge B to climb to pillar C is an insight, but not in a very deep way. You know, overall, how the game physics works; you're just figuring out how to apply the details.

(I am simplifying my case, of course. At the beginning of Ico you don't know all your physical abilities; you encounter them one at a time. You have to realize that, for example, a hanging chain can be swung on as well as climbed. These moments of insight are deeper puzzles, and they turn up regularly throughout the game. But that's the rarer type, you see. More often you say "Ah, a chain, I wonder if I have to swing this time.")

Now imagine a puzzle that has both strengths. It's entirely about position, the 3D room you are in. But it also is about strange new laws, magical effects. And it involves your inventory -- objects in the room, what you're holding. And all communicated through the action-game interface; no verbal explanation at all.

I'm not going to give away any more than that. I was impressed. I want to see more of that sort of thing. Lots more.

(Plus, really nice magical effects.)

Conclusion: Ico is certainly the most well-designed console adventure game I've played. It's a simple game -- it doesn't do very many things -- but everything in the game is fresh; interesting ideas and experiments everywhere. Details all attended to, and done right. A clear vision, implemented beautifully.

Stick around to the end of the credits. I'd never have thought of that, either.

I'll stop babbling now.


Game Reviews

Zarfhome