The Wii U has promised to be all that and a bag of potato chips. Games for everyone, and games for you, or so they say. It’ll have 1080p-resolution graphics. It’ll have high-resolution Blu-Ray-comparable discs. It’ll have a controller with a hi-res touchscreen in the middle. And we’re promised it’ll have all the online you could hope for. And, as if to prove their point, the first thing Nintendo did was present to the hardcore masses, depending upon your viewpoint, either proof in the pudding… or an ultimate peace offering: the Legend of Zelda in crisp, clear 1080p. All in all, it’s a buffet of promises presented to us to excite and amaze. Who could say that there’s anything bad in this? Who would be foolish enough to dare say that this shouldn’t be what we get?
Certainly not I.
Yet despite knowing that this move to HD was one that Nintendo desperately needed to make, I cannot help but wonder just how brilliant the first ever HD offering from the bounty of Zelda will shine.
Now I’m not saying that Nintendo can’t actually do it. That would be rather naïve; I mean, the Wii U Zelda tech demo was simply beautiful. You can see ambient occlusion, real-time lighting, dynamic shadows, high-res textures, shader technology, and other good graphics buzzwords, all of which end up leaving nothing to imagination. And if that’s not proof enough, you should stroll on over to YouTube where they have managed to emulate existing Wii games at 720p resolution, and the result is absolutely brilliant. Despite the fact that actual Wii hardware technically taps out at 480p and doesn’t give you a pixel beyond, the fact that the textures still look good upscaled is actually fairly surprising since that is the first thing you’ll notice the moment you make that leap.
So clearly, they’ve got some degree of knowhow over in Nintendo secret code labs on how to get this done. What’s the problem then? Why should I be worried?
The problem though is that being able to render a single room in high-res is a very difficult task than running an entire full-length game in HD. The programmers for both the Xbox 360 and the PS3 have had their work cut out for them this generation, and though this isn’t uncommon when crossing hardware generational gaps, the transition to HD not been without its growing pains whatsoever. Though the precise details of the Wii U’s processor and graphics chip aren’t entirely public yet, there are a good many things that we know need to be happening—both at the bit level and at the artistry level—to really make a game shine in HD.
First, any game making the jump into HD has to have its code be both concurrent and parallelized. For those of you who don’t know programmer speak, that’s just geek talk that means that the Wii U will need to run multiple tasks at the same time. Way back in the day, games only had single threads of execution, and these were fairly easy to make. (Eventually they have gone to two threads, with one of them simply being a streaming thread to pull data from the disc into memory.) While this seems easy enough at first, it really isn’t as easy as you might think. You see, the human brain performs is reasoning logic with just one train of thought; sure, you may multi-task a bit, but you’re usually juggling very different problems or situations, and like all those people on cell phones on the freeway, you tend to not be able to do one of those tasks well at any given point.
But that’s not really enough when it comes to gaming in the HD era. What we’re looking at now is that the software for any future game is going to have needs to be able to solve related and interconnected problems at the same time, and each of the individual problems need to be broken down into smaller subtasks that can be solved independently from one another (and therefore occur all at the same time) so that they can be combined and packaged together at the end to produce a single frame of video and audio, only to be run all over again.
If you’re confused by all this technobabble, my apologies. Let’s just say that stuff has to get done really REALLY fast, and simply doing all of the steps one at a time in sequential order is not even close to fast enough.
And this has to be done because, simply put, there’s MUCH more stuff that has to get done on the graphics end just to keep up at 30 frames per second (much less the games that endeavor to go 60fps!). What that means is, every 33 milliseconds, the game has to figure out what buttons on the controller that you’re holding (and remember, this has to travel through the ether, wirelessly to the console!), figure out what that means for the characters in the scene, figure out the physics of the world to see if the player can actually do what he or she wants, actually do the thing that can be done, and then figure out the color of over 2 million pixels before finally shoving that onto the screen. And that doesn’t account for other possible things like getting data from over the Internet, pulling data from the disc and storing it in memory, moving the NPCs and AI enemies in the world, activating a bunch of sounds, and maintaining a consistent idea of what state the game is in. 0.033 seconds. You can’t press the start and stop button on a stopwatch in that span of time.
But what really makes parallelization so difficult? It’s because parallelizing a very complex task (such as a game) into small, independent subproblems is simply not how the human brain really works, at least, at a conscious level. Most programmers simply end up being unable to disassemble a game into so many individual component pieces or incapable of actually implementing it, much less figure out how to optimally schedule those tasks so they get done as fast as possible. That’s not to say it isn’t possible; developers have been doing this since 2006 and before with the advent of the Xbox 360.
But even with the 360 and the PS3, it really did take about a full two years or so for people on both sides of the fence really figured out how to master the hardware and to bend it to their wills. The launch titles for both consoles were, admittedly, graphically weaker than their eventual predecessors. And it’s not because there’s new technology or new languages helping the process; consoles really don’t change over time. It’s because developers had to learn from their own and others’ mistakes, figure out the proper methodology for pumping graphics through the system via experience.
A second performance complication Nintendo might run into is the lack of a dedicated hard drive inside the console. While the Wii U is allegedly going to have some amount of internal flash memory, it’s not certain whether or not the full capacity of the flash memory will be going towards gave saves or if some of it will be reserved for caching data on the optical discs. Accessing data directly DVDs in the 360 and Blu-Rays in the PS3 is extremely slow. However, accessing data from a hard disk cache is significantly quicker, which is why many games require gratuitous install processes before you can play the game. If you can’t use the internal flash memory as a cache for game data streaming from the disk, it simply adds to the complexity of the problem in delivering content at such a blazing speed. Games have managed to get around this, but again, it’s all part of an uphill learning curve.
When Wii U launches in 2012, Microsoft will have had a six-year head start, and Sony will have had five, and knowing that it took developers roughly two years to really find their stride means that they’re now beginning to fully master the sheer power of their mighty consoles. While admittedly it is a little odd to say this, thankfully, Zelda will likely be in the boat of being one of the later new titles to be developed for the new console due to Skyward Sword coming out so late in the Wii’s lifetime. It means that Nintendo will have learned from mistakes and will hopefully by then have a much more established process for parallelizing their game engine, which will (or so we hope) deliver on the promise of a rich Zelda experience.
Second, the next potential problem area that Nintendo will be facing is being able to create high-fidelity environments. The tech demo looks pretty sharp and slick, but again, this is just one room, a single room polished and perfected to incite your imagination to go wild. Making an entire game of the same graphical quality takes a lot more time, effort, dedication, programmer coordination, and—of course—money.
Looking backwards through time, it’s actually pretty tempting to hold up Twilight Princess as a pretty decent standard for how graphically advanced Zelda games have become. After all, TP’s scenes actually look fairly well oiled and polished by comparison to the likes of Ocarina of Time and Wind Waker. But a more detailed inspection of TP—for better or worse—reveal that, just like its predecessors before, the game still heavily relies upon abstraction of a world to sneak by with less detail than a totally “realistic” world would have.
For example, Link’s house and Rusl and Uli’s house in Ordon Village don’t have any beds in them. Hyrule Castle doesn’t have a legitimate way to reach the second floor from the first (and neither are there impassable door façades along the walls to mask the fact that there are more rooms elsewhere). The world relies upon placing you within a deep valley so that you cannot possibly escape out the sides of it. Eldin Bridge passes over a seemingly bottomless pit. And somehow Peak Province is covered in snow while the similar or higher altitude Zora’s Domain is not. And of course, if there’s ever a treasure chest or a clay pot inside a house, you can open or break it without anyone getting upset.
Nintendo does a lot of hand waving over many incongruous elements of the game in order to reduce the sheer amount of detail that would be necessary to portray true realism in games. When compared to games such as Assassin’s Creed with its accurate-to-history environments and crowd simulations and Grand Theft Auto IV with its fake radio broadcasts and real-time traffic, the detail of Zelda pales in comparison. Perhaps this is merely a thing, and Zelda has the power to skate by with its abstractions just based upon the power of its name alone, but I really do think that Nintendo will need to up the ante just a little bit if they’re saying that HD is now the future.
The last big thing that I feel is currently lacking in the Zelda franchise is to make the player emotionally invested in the story. In my personal opinion, the two best games that Nintendo has achieved when making rich characters and a story that touched players were Link’s Awakening and Ocarina of Time, both of which are old hats when it comes down to it. Now I’ll be fast to admit that all of 3D main console Zelda games since—Majora’s Mask, Wind Waker, and Twilight Princess—have come close to the mark. MM was phenomenal in actually creating a living, breathing city of Clock Town, WW gave the Sage of Earth and Wind live by having them troll through dungeons with Link, and TP allowed the player to have a deep attachment to Midna.
But outside of each of those concentrations, it doesn’t take much to see that Nintendo started skimping a bit on the detail. None of the other Majora’s Mask cities had anywhere close to the same attention to detail as Clock Town. In Wind Waker, the game stiff-arms the player once they’ve fully restored the Master Sword by forcing them to flail about otherwise insignificant islands to find the Triforce pieces. And in Twilight Princess, even the Resistance fails to really gain the screen time they deserve to properly make them as truly as significant as Midna (or even Ilia) were. And let’s face it, it would have taken extremely little effort at all to make a proper love story out of TP; Nintendo just didn’t dare chance the wrath of their fans.
Certainly, not every character on the stage gets the opportunity to have the spotlight, and similarly not every NPC gets to have his special interaction with the character. (I mean, how many “NPCs” do you pass every day in your life?) But with heavily cinematic games like Heavy Rain, Uncharted, Mass Effect 2, and Halo: ODST telling vivid stories that make the player frequently pause and reflect upon the situations they’re experiencing, the bar for storytelling and character development in games has been significantly raised in this past console generation.
The problem is that the Legend of Zelda has never really claimed to be the pinnacle of storytelling or immersion. The actual plot of each Zelda game is just a thin veneer of eye candy overlaying a thick helping of solid gameplay mechanics. Ocarina of Time really doesn’t tell all that deep of a story; it’s much more fairy tale in nature, full of tropes and MacGuffins to help drive the player to complete all of the objectives before him or her. Yet it had just enough spice in several of the main characters—Zelda, Malon, and the Sages—to solidify an emotional appeal within the game.
Provided Nintendo wants to compete head-to-head with the likes of the 360 and PS3, Nintendo is going to be forced with the tough decision of not skimping when it comes to golden opportunities to elicit emotion in players. While I’m not convinced that Link himself should get VO treatment, I do believe that it’s beyond time that Nintendo finally pay to hire voice actors for all of the major NPCs in the story; they really need to do away with Midna’s babblespeech and the occasional “hmm,” “ahhh,” and “hehe” sound effects and give proper voice and intonation to the characters we’re supposed to care about.
An added bonus would be for Wii U’s future Zelda game to take a bold risk with a character or two by perhaps either not letting things resolve “happily ever after” or by allowing the various characters to actually show deep, heartfelt emotions towards Link (or vice versa). After having played Tales of Symphonia, I am still an ardent proponent of a Harvest Moon-like system where you, the player, can eventually choose the relationship Link ends up with at the game’s end. I’d also be in favor of the total surprise of Nintendo actually killing someone that Link was close to (and not just at the beginning of the game when it has no real meaning yet), especially if there’s actually a way for the player to prevent it.
Now none of these obstacles and pratfalls are impossible to work around. In fact, so far this generation has brought about some very interesting developments both in the art of the story and in the engineering finesse it takes to achieve 1080p/30fps. So in all good theory, it’s very likely that Wii U’s future Zelda game could be the shining pinnacle of what the franchise has ever delivered.
But so many of the decisions as to whether or not this will happen rest squarely on the shoulders of the Zelda Team’s producers. I can’t fault Shigeru Miyamoto for being a very creative individual and for being able to design some truly amazing labyrinths and game concepts, but the more I hear about what it’s like to work with him in the interviews with other Nintendo employees, it often sounds like Miyamoto will frequently—as the story is told—upend the tea table and force everyone down a different path. Were the ideas that Miyamoto rejected ones that could have really helped Zelda?
But with Nintendo is going down the road of HD finally, it’s fully acknowledging that the “blue ocean” strategy Nintendo has been on the past five years is over; now they’re sailing full speed ahead into the shark-infested waters they so sought to avoid with the Wii. The question is whether Nintendo is fully prepared to lay arms upon the enemies just off the port bow or if they’ll end up retreating to the kiddie pool, thereby furthering the conception—whether true or not—that Nintendo focuses exclusively on gaming for the younger generations.