Here is a game that somehow managed to slip under the radar for me. As a snob for strong narrative-based games, I was surprised that a project like Until Dawn managed to escape my attention until a couple weeks prior to its release. Once I heard about it though, I was immediately intrigued. I knew it wasn't going to be a proper survival horror game, but it looked to have a lot of potential to move the horror genre (and gaming in general) in interesting directions. I was doubly surprised when I went to go by the game a couple days after its release so that I could play it over the weekend, only for it to be sold out in the two stores that I went to. It's the first time in about ten years that I've had trouble finding a game on store shelves within a week of its release, but I doubt that I'll have to break my long-standing boycott for pre-orders. So I had to resort to ordering it off of Amazon Prime with 2-day delivery and play it the next weekend.
Suspend your disbelief - and your common sense
Don't be fooled into thinking that Until Dawn is something other than what it is. It is an interactive movie with branching story. It is not an open-ended survival game! Anyone familiar with Heavy Rain or the Telltale Games will have a good idea of how the game will play out. The things you do and the actions and dialogue available are very tightly scripted. You won't be making decisions on how the group splits up, who goes where, or even what any individual character might be doing at any given time. Large chunks of the game are just dialogue and cutscenes, stopping you every now and then to let you make one of two choices, or showing a button prompt on screen to keep the action going (and sometimes keep the character alive). There are even some action sequences that could have been playable, but which are strictly non-interactive cutscenes.
How about lighting some friggin' candles instead of groping around in the dark?
The only time that the game opens up more is when you must explore rooms for clues or evidence. In these cases, you have complete control of character movement and can walk around mostly freely. But interactions are severely limited. You can only interact with the select few objects that the developers intended for you to interact with.
These limitations can be very frustrating because the game doesn't let you do some obvious, common sense things. Upon entering the lodge, I'd like to have been able to light the candles instead of having to wander around in the dark for the next few chapters. Later, when investigating something crashing through a window, I'd like to have been able to take the rifle I just found on the wall. And even later, after the rifle didn't have enough bullets to shoot the murderer, it would be nice to have been able to open up the revolver I just found to make sure that it's loaded and find out how many bullets are in the chamber. And the list of dumb oversights goes on...
Common sense precautions like taking a melee weapon or checking that your new gun is loaded are not possible.
These limitations are further exacerbated by the esoteric nature of some of the decisions. Since all decision are binary (usually consisting of a "helpful" / "safe" option or an "antagonistic" / "risky" option), it's often unclear exactly what the character will do, and the outcome often plays out in a non-interactive cutscene. The character may not say or do exactly what the option described, which might lead the player to think "that's not what I meant to do / say!", and sometimes a decision might railroad you into following through in a way that you don't want to.
Granted, the options need to be somewhat vague, and the consequences shouldn't be obvious. That would make the game too easy and dull. The game has to utilize some of the classic horror movie tropes in order for the narrative to work. After all, the characters don't have the foresight to know that they're in a horror movie game. I accept that there needs to be limitations on the precautions that the player can take, but the player also needs to feel like they have more agency.
Contortion
I also had some frustrations with the game's controls. Movement is sometimes rigid and clumsy, and camera angles can make it really hard to tell if you can reach certain areas of the environment or if there's another path available. I often walked back and forth across an invisible wall right in front of the camera because I couldn't tell if walking past the camera would take me to new area.
The sixaxis sensor is fidgety, which makes aiming the flashlight and weapons difficult.
The motion sensor controls for the flashlight and decisions are also problematic. Camera angle changes regularly locked the motion sensor in some weird, non-level position that made aiming the flashlight or pointing at a decision more difficult than it should have been. I sometimes had to contort my arm in very uncomfortable ways to get the flashlight to turn in certain directions or look at certain things. Some kind of reset button or command would be nice, but instead, I had to resort to pausing the game or going into a menu while holding the controller level in order to reset the motion sensor. This would break immersion whenever it would happen, and it did lead to one of the characters getting killed because I couldn't aim a weapon properly.
I did like the the parts in which the the game requires that you hold the controller perfectly still while hiding. It's probably one of the best and most creative uses of a motion controller in any game that I've played (but I'm not fond of motion-based games, so I haven't played many). The sensor is fairly sensitive too, so I even found myself trying to hold my breath in order to avoid failing. All the controls (including the holding still function) are very well tutorialized in the first couple chapters with mundane, non-threatening interactions. The gun tutorial probably could have been woven more organically into the story with a snowball fight or something, but it still worked.
Trite character-building
Possibly the most interesting element of the game's design - at least on paper - is the character personalities and relationships, both of which are dynamic based on decisions made by the player. Each of the eight characters has a set of personality traits such as how honest, humorous, charitable, or romantic they are. They also have a relationship meter with each of the other seven characters. Actions and dialogue options made by the player can change these attributes, which can, in turn, change how the character behaves, how they treat other characters, and what options the player is given with that character.
There's a lot of potential here for organic storytelling, particularly in a horror setting. A humorous character might be able to make a joke to lighten the mood and help comfort the other characters before they all panic. Being unable to trust the other characters might lead to stupid or brash choices that can set up drama or conflict without having to rely on contrived horror cliches. Similarly, loyalty (or love) for another character could influence a character to endanger themselves to protect the other, providing further options for exciting challenges and plot branches.
Unfortunately, none of this ever really pans out. As far as I could tell, whether characters like each other, or whether their attributes make them particularly heroic, the player is always given the same set of options for key plot points. Some of the dialogue might change, or the result might be superficially different, but the choices to risk character A's life to save character B is always left up to the player. There aren't any situations that I'm aware of in which a character not being controlled by the player makes an automated decision based on their stats or relations that affects the narrative in any way. At best, they might say something to you in dialogue that might slightly influence your decision, but it's still your decision. And more indirect decisions (such as which path you take at a fork) are almost completely inconsequential.
Characters have dynamic attributes that change and supposedly affect the story, but don't.
The way this probably should have worked is to make every choice, clue, and action significant, and to create more than two options for the explicit plot choices. These options could be tied to the relevant character's personality traits, and the game could lock out or hide certain decisions if the character doesn't meet the personality prerequisites for that choice. That way, there's more of a rippling effect that makes minor choices and dialogue that affect attributes more relevant to the player's decisions and resulting narrative. For example, if character A loves character B beyond some threshold, then the player might not even be given the option for character A to take an action that may endanger character B, even if it might be the best option for the group as a whole. A loyal or heroic character might be willing to risk his or her life to protect another, regardless of their relative relationship status.
On the other end of the spectrum, if character A has a very low trust threshold with character B, then the player might not even be given the option to trust character B. Or if character A is too dishonest, other characters won't trust A, and certain options or paths may be locked out.
The building blocks are all there. The game just never does anything with them.
I'm also not even sure that this approach is particularly appropriate for the narrative and set up that the game provides. I feel like this system would have been more appropriate for a group of characters who didn't know each other and have to build familiarity, trust, and respect for each other (ala Will the Real Martian Please Stand Up?), rather than starting the game as a group of good school chums. There's already a built-in assumption of trust and loyalty to each other, and these characters know each other pretty well, so there isn't much opportunity for betrayal or completely selfish behavior.
I'm also not sure if I like how the game instantly informs you when you make a decision that affects the story via its "Butterfly Effect" feature. Such decisions result in a flash of ghostly butterflies in the corner of the screen and a little prompt telling you that there has been a Butterfly Effect update. You can press the menu button to immediately jump to the Butterfly Effect screen, which shows all the branching decisions that you've made so far and what consequences they have triggered (if applicable).
The Butterfly Effect menu reminds you of just how many decisions have been moot.
On the one hand, this serves as a nice reminder of the choices that have influenced the story so far, and provides a sense of overall progress (as if the countdown to dawn whenever control switches to a different character weren't enough). But it also makes it overtly apparent when you've made a plot-branching decision, and serves to remind you of just how moot all the other decisions that you've made have been, as well as how inconsequential some of the "major" decisions really are. This sort of thing is great to have as an end-game recap or in replays after completing the game, but I'm not sure that it was the best idea to include it in the in-game menu in the first playthrough. It may encourage (and enable) the player to try to manipulate the outcome of decisions rather than letting them organically evolve.
Teen slasher horror is only the beginning
So the mechanics all feel a bit limited, but something that might be just as important as how the game plays is whether or not the game is scary. It is, after all, a horror game, isn't it? But I didn't find it particularly scary. It certainly has some very tense moments, and the potential consequences of irreversible death certainly forced me to keep a firm grip on the controller to ensure I didn't fail any quicktime events. But I was never really scared.
The first two or three chapters are very slow, as if trying to go for a slow-buildup of tension as you build the relationship dynamics between the characters, explore the decrepit lodge, find the clues scattered about, and come to the realization that you're not alone. In the meantime, the game throws a barrage of silly, cheap, jump scares at you that really do a poor job of setting any kind of tone. The slowness of the game doesn't do an adequate job of building the tension back up before the next jump scare. Things start to kick into gear around the third chapter, the game takes a sharp turn into Saw territory, and people start dying.
Some scenes are blatantly derivative of the Saw movies.
Teen slasher horror isn't really my cup of tea, and so I'm glad that Until Dawn turned out to be much more than just a slasher horror flick. The Saw copy-cat scenes were kind of "meh", since they're mostly just gore fests and lack the perverse (albeit pedantic) morality of the first Saw film. The supernatural twists that show up around the same time generated some mystery and intrigue and made me more apt to want to find clues as to what's going on. This combination manages to create some genuine surprises. I can see how the Saw-inspired choices were intended to fit in with the character attribute and relationship mechanics described above, but since those mechanics aren't well utilized, it kind of falls apart.
I immediately regret my decision!
The game also rewards thorough investigation with details of the game's backstory and clues as to what is really happening on the mountain. Piecing together these clues is engaging, and some of the clues are actually helpful. The desire to explore is one of the ways in which the game subverts the player's better judgement. There's also totems that provide brief glimpses into the future. They can inform you of what characters might die, how they might die, or even give you "fortunes" that show non-intuitive solutions to puzzles that lead to better outcomes. It's all meta-knowledge that the player possesses, but which the characters are oblivious to. It throws a helpful bone to the player and provides some additional incentive to explore.
Fortunately, the game is very good about manipulating the player into making certain decisions - even without the foreshadowing of the totems. These clever misdirections are the greatest strength of the game. As you jump from character to character, the scene progression often gives the player enough information to trick you into making decision that you think might be smart, but which are actually really dumb, or to make you second-guess the obviously-smart choice. There are some decent early examples of this, but the best example (for me) comes very late in the game. If you don't want spoilers, then skip ahead to "End of Spoilers". Otherwise, continue reading, but be warned: SPOILERS AHEAD!
[Show Spoilers] [Hide Spoilers]
So, the best-crafted decision in the game (for me) came after the "safe house" chapter, when you have to decide whether to kill Emily or not. Same, Emily, and Ashley go searching for Mike in order to warn him about killing the wendigo. We had just learned about how the wendigos are created and how they mimic human behavior in order to trick and lure prey. We cut briefly to a scene of Jessica waking up in the mines after having been missing (and presumed dead) for half the game.
This choice was brilliantly set up to confound my better judgement.
Then we cut back to control of Ashley, who falls behind Sam and Emily and begins to hear crying coming from a side tunnel. I had the option to catch up with Sam and Em or explore the noise. Having just been made aware the Jess was alive, the game put me in a mindset in which I felt compelled to look for her and try to rescue her. So I decided to make the monumentally bad decision to split up from Sam and Em and see if the crying was coming from Jess so that I wouldn't accidentally leave her behind.
The desire to save the weak, lost, and vulnerable Jess overwhelmed my better judgement and the recently-acquired knowledge that the wendigo will mimic human speech in order to trap victims. I suspected that Jess might be turning into a wendigo, but I decided to take that risk if it meant I'd have a chance to save her, especially since the game hadn't provided any opportunity to do so previously. I violated Cardinal Horror Movie Rule #1, and I wandered off by myself to look for a friend. What I found instead was - of course - a monster waiting to tear my head off.
The irony is that Ashley had no clue that Jess might still be alive. My own meta-knowledge contributed to making this dangerous decision. You might say that it was bad design, since Ash probably wouldn't have gone off on her own. But one of her character traits was also that she was charitable, so it makes sense that she would be the one to try to help someone in need, even if she doesn't know who that someone might be. In this particular case, all the different components of the game came together to create a vexing conundrum that subconsciously forced me into making a very bad strategic decision.
At least talk to the person behind the door
before blindly opening it!
The only blemish in this design was the lack of common sense that plagues the rest of the game. Yes, I knowingly made a dangerous decision, but the game also didn't give me any opportunity to be safe about it. I would have preferred that Ash try to talk to the person behind the trap door instead of just opening it without hesitation. If I could have been given some reason to believe that it was Jess, then I'd have gladly opened the door. Instead, I do feel a little bit cheated that I wasn't allowed to take a reasonable precaution.
End of Spoilers
Psycho analysis
There are also brief psycho analysis sections in between chapters. This feature seems pulled straight from Silent Hill Shattered Memories, and it doesn't work much better than in that game. Your answers in the therapy sessions don't seem to impact the actual plot; though it does influence the appearance of the analyst and his office. The idea is to psyche out the player (assuming that you answer the questions honestly), but it's all very superficial and severely lacks subtlety. I say I'm afraid of snakes, and so next chapter, a snake appears on the analyst's desk. A snake doesn't show up in game or ever become relevant to the plot, it's just there, sitting on the desk because you said you're afraid of it. The only time that your responses to the therapist result in any changes to the game is if you say you're afraid of needles, then the psycho knocks out a character with an injection rather than with knock-out gas. Oh, and you might find the psycho's hit-list of character photos arranged in the order in which you said you most dislike the characters. That's about it. The analyst himself is actually a very interesting character that was weird and creepy in his own way, even without the gimmicky props. So I still looked forward to the end-of-chapter therapy sessions, just to see what the analyst would do or ask me next.
The opening chapter cannot be changed and, as far as I can tell, those choices have no impact on the rest of the game.
At first, it seems as though the psychiatrist is breaking the fourth wall and talking directly to the player, as if to find out what scares you and then use that against you in the actual game. Finding a clever way to work this into the game's plot and create some genuine scares tailored specifically to you would have put Shattered Memories to shame, but it isn't the case. It seemed apparent fairly early to me that the person receiving the therapy was intended to be the psycho, and these sessions serve to project elements of the player's personality onto that of the killer. I think the idea here is to make the player feel more responsible for the pain and death suffered by the characters by creating an implied connection between the player and the psycho, as if the player's choices are a manifestation of the psycho's will. In that regard, the therapy sessions work well and are pretty clever. It's just too bad that they don't have more direct effects on the actual narrative events.
The psycho, however, is one of the game's biggest problems, and the source of what I can only describe as a motivational plot hole. I can't decide whether the psycho is just poorly written, or if the game is trying to imply that the psycho is too deranged to operate even on his or her own standards of justice and morality.
[Show Spoilers] [Hide Spoilers]
So the whole point of the game (at least initially) is that Josh wants revenge on his friends for essentially causing Hannah and Beth's deaths. But he enacts this revenge on the wrong people! Is he mentally unstable enough to just not know the difference anymore? Maybe by the end of the game, he's that unstable. He seems very cognizant of what he's doing for the first half. Or is it just a case of the wrong people being in the wrong place at the wrong time?
Most of his torment is focused on people who were the least involved in Hannah and Beth's death. He seems to pick on Chris the hardest. Josh gives Chris the most painful and traumatic decisions, despite Chris being asleep when the prank against Hannah was done, and despite him apparently having no involvement in the prank at all. Making Chris have to chose to kill his crush or his best friend is cruel beyond measure, especially for someone who isn't guilty in any way. It could be that Josh was planning this choice for any couple that he could catch, but it seems deliberately planned for Chris, since he and Josh are best friends. Making Mike have to chose between Josh and Jessica, or making Ashley have to chose between Josh and Chris (for example) wouldn't have had the same effect.
Mike, Jess, and Em are the most guilty in the prank against Hannah, yet they receive the least punishment from Josh. Plot hole? Or just inconvenient coincidence?
In fact, the two characters who seem most responsible, Mike and Jessica, are sent away by Josh and not made a part of his prank at all. You'd think that Mike and Jessica would have been the focus of his hatred, since it was Hannah's crush on Mike that enabled the prank to begin with, and the prank itself seemed to be Jessica's idea. So why does Josh just let them wander off to the cabin by themselves? Why doesn't he send Chris and Ashley (or even Em and Matt) to stay safe and sound in the cabin, and force Mike and Jess to stay in the lodge? Was that the plan, but Em and Jess's cat fight upset the plan?
The only way that I can reconcile this is that he must have had some plan for dealing with Mike and Jess in the cabin, but just never got around to implementing it. This seems unlikely though, since Josh isn't aware of any of the events that happen in the cabin, which implies that it isn't monitored or booby-trapped, and therefore not part of his prank. The only other explanation is that Josh honestly doesn't blame Mike or Jess, or that he lets Mike off the hook as some kind of way of respecting Hannah's feelings for him. These all feel like flimsy excuses though, and I honestly feel like this was just sloppy writing.
Perhaps a better design (from a gameplay standpoint) would have been for Josh to have offered to retrieve Em's bag. This would give him an excuse to disappear, and make Em and Matt available for his "experiments". Forcing Em to have to chose between Matt and Josh (or even Mike and Josh) in the saw "experiment" seems like it would have been a much more effective set up (from a gameplay standpoint). Em's relationships with the other friends are the most volatile, so it would have been easy enough to design the game so that she swings her decision one way or another based on player choices. She could have started with high relationship with both Mike and Josh, and have that relationship deteriorate or grow based on how Mike reacts to her argument with Jess. Giving the gun to Matt and asking him to shoot Em would then put Matt's suspicion of Em cheating to the test. It seems that Matt and Em could have been easily transposed for Chris and Ashley in the story, and the story might have worked better, and provided more opportunity for it to be influenced by the relationship mechanic.
Who built these secret tunnels, and why?
As an aside, I'm also not aware of the game ever providing an explanation for why there are secret tunnels between the sanitorium and the lodge to begin with. Who would have built these, and why? Is the game trying to imply some relationship between the Washington estate and the sanitorium? Josh clearly knew about the secret basement, and so probably would have been aware of the passage to the sanitorium.
So there's no shortage of weirdness and contrivances in the writing.
End of Spoilers
The interactive drama renaissance will have to wait another day
Until Dawn has some genuinely good ideas when it comes to character interaction and the way that it manipulates the player's decision-making process - ideas that other narrative-based games should look at and learn from going forward. There's a lot of promise here that the game just doesn't deliver on due to limitations of the design and irrelevance of some of its features. I liked it a lot better than I liked Heavy Rain, and I certainly felt like more of my individual decisions were significant (even though some were crap shoots).
Despite my disappointments, I'd definitely recommend a playthrough for Until Dawn for fans of horror or highly-narrative game design, but I don't think it's worth more than a rental.
Until Dawn is tense and offers plenty of excitement, but it isn't exactly the revolution that it was hyped to be.