Home Artists Posts Import Register

Content

Review code provided by Ubisoft

I was excited about Watch Dogs: Legion for two main reasons. One: the game is set in London, which makes a change from the usual American settings of open world games. And two: the game was directed by Clint Hocking.

Clint is a game designer who I respect a lot. He directed the best Splinter Cell game, Chaos Theory, and then made one of my all-time favourite games: Far Cry 2. I've also watched all his super smart GDC talks, and regularly quote him in GMTK videos on topics like intentionality and improvisation. (He also coined the term "ludonarrative dissonance").

After Far Cry 2, Clint bounced around studios like LucasArts, Valve, and Amazon before finally returning to Ubisoft to direct the new Watch Dogs game. So even though I was never a massive fan of the series (didn't play the first, thought the sequel was okay), I was totally on board for Legion, as the first Clint Hocking game for over a decade.

Unfortunately… I was left disappointed. So, what went wrong?

Now I don't really have too much to say about the majority of the game's action. It is, by all accounts, a perfectly fine open world game. You've got driving and shooting, story and side missions, and - unique to Watch Dogs - a clever hacking system. This lets you watch through CCTV cameras, remotely detonate explosives, distract guards by sending smut to their phones, and make drones betray their users.

It's nothing special. And its actually alarmingly similar to Watch Dogs 2 - more so than most Ubisoft sequels, which is saying something. But it's mindless fun. A comfort food. Easy going and just engaging enough to keep you occupied.

So the saving grace would be the game's much-touted and heavily-advertised "Play as Anyone" feature.

Basically, the city of London is populated by millions of randomly generated citizens - and you can play as any of them. After picking your first character from a pool of randos, you can then walk up to anyone on the street and recruit them. They could be a plumber with a wrench, a banker with a sports car, a granny with a machine gun, or a podcaster with a gambling problem.

When Ubisoft showed this system at E3, I thought it looked very interesting. It was certainly innovative, and looked to bring something new to the tired open world genre. In the right hands, it could be on par with Shadow of Mordor's Nemesis system, which used a procedurally generated cast of characters to great effect.

Sadly, the Play as Anyone feature is half baked and falls flat in many areas.

Okay, so from a technology stand point, the system is truly impressive. Every NPC in London has a unique face and voice. They have a job and whole backstory. They have family and friends - who also exist in London. And they even have a daily schedule that might see them going to work, before heading to the pub, and then going on a date.

You might even find them mourning at the grave of their sister… who was killed when you mounted the curb while rushing to a mission six real-world hours ago. It's pretty bonkers.

Perhaps most impressive of all is the voice acting. As it turns out, the entire game script has been rewritten and re-recorded in dozens of different voices (posh, cockney, Irish, old, young, Scottish, etc etc). So no matter who you are playing as, your character will play an active, speaking role in cutscenes. It would have been easier and cheaper to make your current hero a silent protagonist, but that's not what happens.

The devs also used a bit of subtle sound modulation tech to create even more voices, like deep and high pitched variants on the recorded voices.

But technology alone isn't enough. The most important thing is how it impacts the rest of the game: the narrative and the gameplay.

So to quickly touch on its story impact: it's a little odd. Your randomly-generated character doesn't feel like a true protagonist, and so always feels out of place in main story cutscenes. It also frequently undercuts the tone of the game: sometimes its a serious sci-fi story with allusions to real-world events like Brexit. But then you're playing as a living statue with a goofy Eastern European accent.

But there are advantages. You can play as a non-traditional hero like an old black woman with a thick Afro-Caribbean accent. And your team of heroes feels uniquely yours - which I think is an advantage, even if they're not as well rounded as, say, Sitara, Josh, and Wrench from Watch Dogs 2.

But it's gameplay where this system really falls down. And it's ultimately because, if you ask me, Watch Dogs Legion fails to adequately answer a key question: why should you engage with this system at all?

Because after the gimmick of playing as a hiccuping granny wears off, I think the game needs to actively persuade you to use this system - simply because there exists two pretty significant points of friction.

One is when recruiting new characters. You'll need to take on a multi-part mission which are often repetitive (in gameplay and location) and involve travelling halfway across London. The second is when changing character, which involves a loading screen. Both added just enough friction to make me think twice about engaging with the system at all.

So one way for the game to encourage recruitment and swapping would be to make different characters useful for different missions.

And in some instances, this is true. For example, if a mission is in a gang-occupied hospital, a nurse can just walk straight through the door - Hitman style. And a construction worker can summon a cargo drone at any moment, which you can ride up to the rooftop for a better entry point.

But in reality, every character is perfectly capable of taking on any mission. For one, every single person in London is a typical open world hero: they can drive a car, hack computers, shoot with perfect accuracy, perform takedowns and executions, and use any of the game's gadgets. And for two, many missions have multiple ways to succeed: you could shoot your way in or just send in a spider bot to do the hacking for you.

Another way to encourage recruitment would be to give characters really interesting traits. And this is where Legion utterly falls down.

Most of the traits are actually really, really boring. For example, a character might be a pickpocket who can steal cash when doing takedowns - but money is only used for pointless cosmetic items. A character might have a personal vehicle, but the streets are lined with unlocked cars. A character might have slightly increased defence against punches, but that's not very handy when everyone in London is carrying a gun.

There are a few interesting ones. I found a guy leading a protest, who carries a megaphone and can recruit nearby NPCs to rise up and fight. So I recruited him, shouted some inspirational words into my megaphone, and stormed a gang hideout. The gang members instantly shot and killed all the NPCs - and then me. Funny, but not very effective!

You are more likely to find a character with no traits whatsoever. Or even negative traits! An old man might have a bad back and so can't crouch or sprint. Why would I even bother to recruit him? I spent a lot of my time in the game reading each character's bio, and realising that I didn't want to recruit 99% of the people in London.

There's only a few characters with legitimately good traits, but they aren't part of the Play as Anyone system. If you liberate a borough (by, say, doing three side missions in Westminster) you'll get given an elite operative such as a John Wick-style hitman or a drone specialist who can summon a shock drone. These are better than all the NPCs in London and I used them for the majority of the game.

The third way to encourage recruitment is to simply kill your operatives.

When you start Watch Dogs Legion the game asks you if you want to play with permadeath on or not. I assumed that permadeath was some bonus hard mode challenge for a second playthrough. A few hours in, I realised that it was fundamental to making the game work, and so had to start the whole game from scratch with permadeath turned on.

Because if you can just use the same character over and over again, there really is no reason to recruit more team members or change your hero. But if your character can die… well that changes things. This really encourages you to pick the right person for the job, really encourages you to keep a large team of characters on stand by, and literally forces you to switch character.

(It also has other advantages like ramping up the stakes, and creating great moments of drama. When my spy lady finally bit the dust, I was as distraught as when I lost key units in XCOM and Fire Emblem.)

Again, however, the system doesn't really work. Even playing on hard difficulty, the game is very forgiving. And worse still, there are so many ways to play that don't put your operative in danger at all: I finished most missions while hiding in a bush, using drones, robots, and CCTV cameras to do the dirty work for me.

The only time I lost operatives was during a couple story missions where the game just spawned loads of goons on my location and forced me to survive the unfair ambush. Otherwise, my characters just kept on living.

So only a few hours in, I found myself realising that I didn't want to bother recruiting new characters because they're not interesting enough to warrant doing a dull mission. And while I changed character sometimes, it was only between a small cast of special operatives. And while my operatives could die, that wasn't likely to happen.

Ultimately, this system must have required so much work - it is a technical and logistical feat - and yet, by not propping it up with good design, it loses all purpose.

I mean, just to give you an example of what you could do in Watch Dogs Legion.

You need to hack a server inside a hospital. So you find a paramedic who works there, but learn that he's not a fan of your hactivist group, DedSec, and doesn't want to help. So you follow him after work, and he walks to a pub to go on a date with his girlfriend. Later, you talk to her, and she says she needs help: she's in debt with a gang. So you break into the gang headquarters and erase her debt from the computer. Her boyfriend is now happy with DedSec and agrees to be recruited. You switch to him, and then walk into the hospital and hack the server.

This is literally something you could do in Watch Dogs. All the systems are in there - the relationship mechanics, the schedule system, etc. And yet, you could also just send a spider bot into the hospital, disable the security with a single button press, and hack the server. Or you could recruit a different paramedic who does like DedSec already. Or finish the mission in many other extremely simple and straightforward ways.

Good systems fall apart if they are not propped up with good design that encourages the player to actually engage with them. There's something about optimising the fun out of games…?

I don't want to play armchair game designer, but my mind raced with possibilities while playing Legion.

I think the main thing would be to make every character completely different.

There's no guarantee they'd be able to drive a car, hack a computer, or fire a gun (or be willing - they might be a pacifist). They could have lots of really cool benefits, but also some serious disadvantages to balance them out. For each mission you'd receive a brief about what it might entail (some hacking, some driving, and a few unarmed guards) and then recruit or swap to the operative you think might be best.

You could really invest in characters - removing their negative traits (like in Darkest Dungeon) and adding positive ones. This will make their permadeath even more powerful.

You could even send multiple characters, and swap between them GTA 5 style. Perhaps have a hacker and a getaway driver sat in a van outside, and a podcaster with an uzi sneaking in through the basement.

As always, it's easy to look at a game in retrospect and think what they could have done (especially when you have an infinite financial and performance budget). But boy, was this game crying out for something else.

So yeah, I was left really disappointed by Watch Dogs: Legion. What could have been a really inventive and disruptive game (just like Far Cry 2 was), ended up being a typical open world title with a fun and marketing-friendly gimmick tacked on.

Hopefully I don't need to wait another decade to see what Clint Hocking tries next.

Files

Comments

Anonymous

Such a shame. I am so bored of generic open world games, they all feel exactly the same. I was hoping this game could shake it up a little. I understand these multi million dollar games are very risk adverse, but they really need to start taking some. Loved Far Cry 2 for the same reasons as you.

ZoidbergForPresident

So, were you disappointed in your hero designer? :P

Anonymous

I want to focus on the paramedic example you gave, and how it is systemically more interesting to change someone's opinion of DedSec and recruit them, but can also pick someone who already likes DedSec. This can be addressed with characters that don't like you having more abilities and make them more valuable, human resources. So this paramedic can also treat other members' injuries quicker and also take reduced damage overall. I have not finished (I've actually barely started story missions, instead I'vebeen out recruiting people), but I FEEL like that way of engaging in recruiting people already exists. Many characters with multiple, good abilities inherently dislike DedSec; in contrast to those that do like you, they possess only one or perhaps two abilities. These people will be more valuable to the team throughout the story of your game. Handy that I'm still in the early game, because the story is me assembling my team of hackers AND THEN taking down the regime. That's the exact narrative in Far Cry 3, except the game didn't decide this, I decided it. Currently, I think the Recruitment system lacks an element that I didn't appreciate in Darkest Dungeon (and now I want to replay): you have to "fire" people. There exists a limit on how many people you can recruit, but you can't explicitly fire them. In WDL have to send them on suicide missions to get rid of them (and holy fuck, that's a messed-up emergent narrative you've put into your game). And finally I think the number of abilities per character makes it too simplistic and restrictive. In Shadow of Mordor, some Uruk:s can have a few good and bad perks as an even enemy; some can have loads of bad perks to represent them as inferior captains; have loads of good perks and a weakness that makes them stubborn to defeat; or have a few perks and be a mostly basic enemy. The number of perks didn't have a hard cap, and it has made the Uruks overall more interesting to engage with. It is awesome to see this system reworked for controllable characters, but I think this system "making it feel restrictive to engage with", highlights to me how restrictive just doing anything in WDL actually is.

Anonymous

Aside the "guns for everyone" issue (which I hope is back-story related), I am indeed most surprised by "everyone can do everything" statement: why have many characters when they are fungible? I like the multiple solutions, but as you mention it might just auto-optimise the fun out of it; similarly goes for the "hard to kill": being lenient on unskilled (or disabled!) players is good... unless it ends up making the game not fun. I know full well making a difficulty auto-balancing system is hard and very prone to failure, but I wonder if they even tried or considered one

Anonymous

It's not just "guns-for-everyone" in WDL, it's also "hacking-for-everyone" too. I see the appeal in it, systemically: you assemble a proper bunch of shlubs (the game developers did not design shlubs, the game created them) to take down the regime. And an appeal I have is how it's MY bunch of shlubs, just like it's MY Pokemon team that defeated the Champion. I chose which ones will help me reach the end-goal. Personally this game might be a bit lacking because there isn't enough moving parts in moment-to-moment gameplay. Multiplayer will inject a lot of fun into the system, but on PS4 that's gonna take a very long time.

Mikhail Aristov

You know, Mark, "optimising the fun out of games" sounds like an great idea for a whole video. I know that Koster already wrote about it, but surely there have been new ideas and approaches to tackle the issue in the past 15 yeras?

Mark M

Interesting. Sounds like they added the play as anyone system to give the game something new, but didn't want to force players to engage with it if they didn't want to. So they removed any excessive punishments or anything that would require you to go out of your way to strategize. That way it appeals to the broadest audience, but doesn't really excel with anyone.

Anonymous

That hospital server example really stung: there's such great potential in this system, if only the design was more committed to it! I wonder what the game would be like if it went the opposite direction: everyone you recruit is *underpowered*, no single recruit can do much on their own because they're up against tremendous odds, but building a team where each person makes up for the other's weaknesses and having them all perform the mission both increases your odds and makes for a thrilling gameplay experience, switching between them, making sure no one dies, etc

Anonymous

This is so disappointing. In the best world, this game would be something like Mooncrash--an innovative approach to persistent problems in immersive sim game design. It's sad to hear that it's gotten the Ubisoft treatment through and through. I really hope that the team that worked on this gets to iterate on it, now that the systems are in place. Hell, use the same London map! This mechanic has too much promise, and too much hard work went into it, to let it die quietly.

Anonymous

I now have this feeling that this is how the game is meant to be played, given how weak each member already is BUT can be rescued with multiplayer (a shame that it's going to be delayed until after new consoles).

Anonymous

Really interesting, and what I worried would be the case from the previews. You're totally right: you need clear, strong pros+cons to different characters/tools to get players mixing it up and making strategic choices. The "one size fits all" approach just shifts the focus to a brute force stats contest. I come back to "given the chance, players will optimize the fun out of your game" a LOT. When confronted with a challenge, most players will ultimately choose safety and efficiency over their own enjoyment - designers need to give players a reason to play the fun way - if possible they need to make the fun way to play the most optimal way. The Hugo Martin DOOM Eternal interviews touch on this a lot, and it resonated with me a bunch

Anonymous

I'll have to check them DOOM interviews as your synopsis resonates with me a lot too, haha! Yet to play Eternal but 2016 is by far my fav shooter for exactly the reason you said (and never been able to articulate until now - thanks!!).