The Last of Us: Why next-gen is overdue
It’s no secret that I’m a big fan of Sony’s first party studio Naughty Dog. Over the past generation the studio has become the paragon for interactive storytelling in the minds of many, consistently raising the bar for narrative-driven games. While Sucker Punch, Quantic Dream and Kojima productions were demoing their next-gen facial animation technology at E3, a game with comparably realistic performances was already arriving in stores. This week, Naughty Dog released their latest game, The Last of Us, to both critical and commercial acclaim. In the same amount of time, The Last of Us has grossed more than Hollywood’s latest summer blockbuster Man of Steel has in the UK.
Many reviewers reason that the game’s peerless performances help convey a unique, provocative message that makes The Last of Us stand out as one of the most interesting games of the year. I wholeheartedly agree. Although the facial animation showcased at E3 was impressive, Naughty Dog has been using a system with even more potential since the PS3’s release. Even with state-of-the-art motion capture technology; many studios are still confined to the limitations of live-action filmmaking in terms of acting.
Naughty Dog isn’t. Instead, experienced animators use an actor’s best facial performance as a guide for handmade facial animation. The team then chooses the best vocal performance and full-body performance separately for each character. This means that Naughty Dog doesn’t just use the best take, they use the best of every take. On current-gen consoles, where character models have limited complexity, it’s hard to imagine any other process producing the kind of subtlety integral to The Last of Us’ story.
This is one of the reasons that I feel The Last of Us is one of the best story-driven experiences of this console cycle. However, it’s also one of the reasons that I believe the game had more potential, and that the next generation of consoles are long overdue. To illustrate my point, let’s look back at Naughty Dog’s rise to fame with the release of 2009’s Uncharted 2: Among Thieves. Uncharted 2 happens to be my favorite game of all time. There are plenty of subjective reasons for this, but technically it’s also one of the most polished console games ever made.
Over the course of the one hundred and eighty hours I’ve spent with the singleplayer campaign of Uncharted 2, I have noticed no visible frame dropping or judding, only two instances of noticeable texture loading, and one sound effect that is occasionally out of sync. Even if you hate this game, it’s undeniably the most stable AAA experience on a 7th gen console. A remarkable achievement for its time, considering the game has no loading screens after the initial startup, includes complex set pieces, and was one of the most visually impressive games on any platform at E3 2009. It was hard to argue that any other console game was technically more impressive, and I would argue that no console game has made any real improvement since.
Fast-forward to November 2011, and the release of Uncharted 3: Drake’s Deception. Characters have a higher poly-count, lighting is more complex, and set pieces are more impressive than ever. There are more animations per character than in any other game released to date. However, with every technical step forward, the game also takes a step back. Frame dropping, texture popping, aliasing, and out of sync sounds are present and noticeable throughout the campaign.
Uncharted 3’s AI was a step down from its predecessor, relying more on difficult-to-shoot animations than taking cover effectively and avoiding open areas. Controls also felt more sluggish. Aiming lag was eventually fixed with a patch, but the multiplayer felt adequately responsive from day one. Was it a lack of communication within the studio that caused the initial difference, or was it simply that the aging hardware of the PS3 struggled to deal with the visual fidelity that was expected of Uncharted 3’s campaign? It’s no secret that the PS3’s hardware is outdated.
While most console cycles last four to six years, this generation has lasted seven (assuming next-gen consoles are bound for a November 2013 release). One could argue that since Naughty Dog had been split into two teams, sticking to the two-year dev cycle could have been what led to these problems, rather than the limitations of the PS3. However, their recently released The Last of Us had a much longer development cycle. As much as I appreciated The Last of Us, technically the game once again took as many steps backward as it did forward.
The game is more responsive than Uncharted 3, nearly twice the length, and features far more gameplay latitude. It has less frame dropping, less noticeable texture loading and, once again, more complex character models. Yet, the game never seems to reach thirty frames-per-second, the game’s initial loading time is extremely long by any standards, many players have run in to numerous bugs throughout the campaign, and aliasing is some of the most extreme I’ve ever seen, making shadows blockier than any Uncharted title.
While gameplay demos that show the game’s AI companion, Elle, throwing bricks to aid the player might seem impressive, the AI in the final game is actually far more inconsistent. While sneaking, Elle will pass directly in front of enemies searching for you, and will sometimes dart back and forth aimlessly, unable to decide where to take cover. Though the enemy AI doesn’t become alerted, it’s still an immersion-breaking compromise that stems from the fact that the AI couldn’t be as intelligent as the game’s story and mechanics required it to be. Despite being fairly linear, there simply weren’t enough AI processes to adapt to every contingency.
So what does all this have to do with the facial animation I mentioned at the beginning, and the relevancy of current-gen consoles? Well, everything really. You see Naughty Dog’s facial animation process is one that other developers are still catching up with. It’s essentially a next-gen idea. It requires a level of character fidelity that takes up a lot of power. The PS3’s limitations forced both Uncharted 3 and The Last of Us to make a choice: better character performance, or better game performance.
Since the characters have been the driving force of every Naughty Dog game this generation, the answer was a no-brainer. However, it meant neglecting what developers conventionally prioritize above all else. I don’t want prospective players or Last of Us fans to think I disliked the game. It’s actually one of my favorite games this gen. As an avid gamer and thorough critic, I know the limitations of hardware, and can recognize when a talented team has taken the right steps to achieve the best possible version of their vision. However, I feel that The Last of Us is a next-gen game trapped in a current-gen game’s body.
The game managed to achieve what was most important for the story to be effective. Still, its attempts to make soft ambient lighting beautiful are negated by its distracting aliasing, and the limited number of processes that could be assigned to each character made the idea of a realistic AI companion… well… unrealistic. Uncharted 2 was the peak of the PS3’s technological relevancy. By the time Uncharted 3 was in development, many sacrifices had to be made in order to achieve incremental improvements, and ideally that’s when Sony’s in-house developers should have been looking to migrate to new hardware.
If it were a PS4 game, perhaps The Last of Us would have had fewer limitations restraining it from fully realizing Naughty Dog’s apparent goals. Having said that, if next-gen consoles are indeed shipping this holiday, this is a good way for gaming’s longest generation to come to a close. The aptly named The Last of Us left me with two messages – one which the story intended, and another, more uplifting message, which is this:
When it comes to Naughty Dog games with 7th gen limitations, this is the last of them.