Transistor – Second Opinion
This second opinion editorial will contain full spoilers for story and gameplay and is intended as a supplement to Jack Ragasa’s video review.
Transistor, as Supergiant Games’ second foray into video games since their previous title Bastion, is of particular interest to me. I am a sucker for cyberpunk and science fiction after all, and something would be amiss to pass on this offering.
Of course it was impossible to isolate myself from the opinions of the internet and I went into the game knowing of its most discussed shortcomings: length and the controversial ending. Bastion was already a mixed bag that didn’t fully resonate with me but I was prepared to give this game a chance just on the themes alone. Unfortunately it seems that little has changed for Supergiant Games’ since the time spent developing Bastion. Transistor sports the same familiar isometric view, creative art direction and beautiful artwork they are known and lauded for.
Bastion was already what I would categorize as ‘style over substance’ and Transistor follows in its footsteps as the definite example of such. Behind the beautiful environments, the great soundtrack and superb visual design sits a set of mechanics that borders on schizophrenic; but I’m getting ahead of myself.
Story and Themes
The story in Transistor seems to be divided into three plot-threads that are dropped faster than a hot potato. First, and most prominent, is Red’s revenge-plot to avenge the death of her boyfriend, now imprisoned in the Transistor. Second, and more interesting, is the attempt at the deconstruction of democracy and the discussion of censorship through the use of the city of Cloudbank, the Camerata and the loss of Red’s ‘voice’. Third is the origin of the Transistor and saving the known world from The Process, an all-consuming amorphous force with an alien collective consciousness.
All these plotlines do not compliment each other in a way that one would expect, and they also do not get resolved in any way, shape, or form, courtesy to the abrupt and nonsensical ending.
After the player goes through all the challenges both the Camerata and The Process throw at them as well as hunting down and killing the person responsible for killing Red’s boyfriend, the plot grants Red the power of unlimited creation with the Transistor. Rather than resolve the plots, because she is now effectively god, Red then decides to restore her boyfriends corpse and stabs herself with the Transistor, laying down beside him.
If this was supposed to evoke some kind of emotion from me it did so marvelously, but it probably wasn’t the emotion the designers intended me to feel. I did not feel sad or moved by Red’s decision: rather, I was confused and frustrated.
With regard to the metaphor of Red losing her voice, it’s highlighted by the fact that all her comment posts at OCV terminals go straight to moderation, the city of Cloudbank is run and changed by popular opinion and Red regaining her voice at the end by recovering the original Transistor. Instead of resolving the issues presented in Cloudbank Red decides to permanently silence herself by giving up her most basic right in that world.
Another problem is how the plot handles The Process. It forcibly and chaotically changes the environment for its own purpose and despite the Transistor being the only device able to keep it in check, Red also gives up on the hundreds of thousands of helpless people inside Cloudbank. Cloudbank is mostly destroyed or assimilated by The Process at the end of the game and despite the fact that Red’s new power with the Transistor that would allow her to rebuild or at least drive away or contain its remnants, she decides instead to join her boyfriend in death.
Ultimately, only the relationship/revenge-plot gets resolved, as Red kills Grant when he takes the Transistor from her, but does so in the most anticlimactic way possible, with two Camerata members committing suicide before they can be held responsible.
In the end, Cloudbank is still plagued by The Process with the survivors evacuated to uncorrupted parts of town, homeless and awaiting certain death like the tens of thousands who likely already died. Red does not fight The Process because she wants to save Cloudbank; she only wants revenge.
It’s even cruel that Red kills herself with the Transistor, the blade her boyfriend is trapped in, as this essentially forces him to watch as she fatally stabs herself in the chest. Of course it turns out well in the end (at least for Red and her boyfriend) as they are reunited inside the Transistor, but Cloudbank is still on the brink of extinction.
Perhaps Transistor is the antithesis to Dark Souls. While one game is about overcoming obstacles, pushing forward, getting better and making a sacrifice for the greater good, the other seems to be about giving up and ultimate selfishness. Which wouldn’t be in any other medium other than games since other media are not contingent on overcoming challenges. However, in the context of gaming and inside a game’s story they work adversely with the gameplay.
Transistor was praised and highlighted in the press due to its purportedly innovative combat system, which supposedly mixed real-time action combat with turn-based combat. I was skeptical about this mix, as real-time and turn-based combat fulfill different roles in game design and often have diametrically different design-philosophies that have to be applied.
No surprise then that the game is a complete mess mechanically.
If any analogy could be made, playing Red in real time is like playing a character in any given current MOBA. There are four abilities, called Functions, on a quickbar with various cooldown length and potential damage that activate when their corresponding key is pressed. Movement can be either mouse-driven by left-clicking on a spot on the map or by use of the WASD keys. When switching to turn-based mode by using the Turn() function however, there is an action-point pool to be used for the commands, but the ability cooldowns are now lifted, making it possible to use several abilities in a row fluidly. Additionally, whenever you use up your action-point pool and execute your turn-based plan, all your abilities go on cooldown until your turn-charge returns in real time. This design means that there are only two ways Red can attack: in turn-based mode or when her Turn() meter is full (which would allow her to use turn-based mode).
In theory, this is supposed to leave the player the option to play the game in real time and in turn-based mode. In practice, however, this means there is no incentive to attack in real-time mode, as in Turn() mode the mechanics change and time is frozen, which gives the player incredible burst and combo potential by chaining abilities freely. Not to mention abilities like Crash(), which open enemies up to vulnerabilities, have an incredibly short debuff duration (around 2 seconds) which makes them essentially only usable for combos in turn-based mode, as abilities fire slowly and go on cooldown in real time. There is one exception to this rule though with the Jaunt() function which can be used during Turn() cooldown.
Functions can be installed in the Transistor in three ways: as abilities, upgrades to abilities and passives, each providing a unique behavior. For example, installing Jaunt() as an ability lets you teleport a short distance to get out of harm’s way, but installing it as an upgrade will make the upgraded Function usable during Turn() cooldown and reduce its animation reset during Turn().
However, a function can only be slotted once and there are no doubles, creating a situation where sacrificing Jaunt() as an upgrade on any function will give you one (sic) attack to use during Turn() cooldown. Meanwhile Jaunt() as an ability in itself can be used during Turn() cooldown anyway, and can be upgraded with damage or other effects, like regaining life or creating a shadow copy.
Adding to that the fact that some functions like Help(), which summons a controllable pet, can’t be used at all during Turn(), the design becomes a bit of a mess. On the one hand, it’s clear that playing in real time is clearly inferior to using Turn(), as it limits the player to arbitrary ability cooldowns and slow execution, often getting swarmed and overwhelmed. On the other hand, some functions are restricted to real time mode for no reason whatsoever and the game restricts using Turn() to a charge, while pretending that both play styles are somehow complementary.
This is not the same situation as with games that have an active pause. Taking Baldur’s Gate for example, the rules between real-time play and active-pause do not change. Baldur’s Gate functions entirely turn-based with a surface level presentation of real time combat while the code ticks away in turns. When the player hits the pause key he only stops the ticking clock; they do not become superpowered and remove casting restrictions or ability cooldowns.
I think the only truly valid use of Jaunt() as an upgrade is in the end boss, Grant, which uses the Transistor as the player would to stop time. Since the boss does not use real-time attacks between Turn() cooldowns it is your opening to deal additional damage. But given that this is the very last combat encounter in the game (incidentally also the only time someone else uses the Transistor against the player) it hardly seems justified to even have it.
Besides that, some Functions and their combinations in Transistor seem to be borderline pointless in completing combat encounters. Switch() lets you change enemy allegiance for a brief moment for example, but the damage enemies do to each other is almost entirely non-existent, which means the function only becomes useful when used against support-enemies like Weeds (healing) or Cheerleaders (shield), functions which can be implemented with Tap() or Bounce() by the player themselves anyways.
Another problem is the small enemy variety in general. There are only seven enemy types besides minibosses in the game with progressively stronger attacks and abilities. Upgraded versions of enemies do have additional abilities, but they also behave exactly the same and their composition never becomes more lethal: only the unit itself. Curiously, the strongest and most dangerous enemy type is not either of the miniboss types but a regular enemy called Fetch, a dog-like construct. Fetch is faster than the player, does massive amounts of damage in a cone, has high health and with upgrade 3.0 can cloak, becoming intangible. It is quite possible to become overloaded in a very short time if there are more than one on screen.
After fighting the same enemies dozens of times the gameplay severely overstays its welcome, even with the large amount of Functions to combine, the players tactics never truly change. Ironically, this is also the only mechanical element of the game that is in unison with the game’s theme, and especially Grant’s belief that “If everything always changes, nothing truly changes.”
With Transistor’s extremely short runtime of about four hours, I consider it a feat that the gameplay already becomes stale after the first two.
As mentioned in the beginning, Transistor is the definite example of ‘style over substance’ design, but not for lack of talent or ability in the development team. In Transistor there is a small hub called The Beach that the player can visit through Backdoors scattered across Cloudbank, in which they can participate in specific mini-game challenges. This is the most interesting part of the game mechanically, as the designs of the challenges and the abilities the player gets slotted for their duration actually require tactics and thought for completion.
Of course this doesn’t negate the real-time and turn-based divide of the core gameplay, but it at least shows that the ability and talent was there to put together challenges that require varying tactics and thought. Transistor is hardly a difficult game, especially since you have up to four lives total and the main combat encounters are hardly challenging. Its focus clearly is the story, which is surprising considering how little there is of it.
Transistor hides behind a facade of pretty visuals and moody sound with a lick of Noire storytelling, but it’s a mess in execution and implementation. It aspires to fly higher than it actually can, both in narrative and gameplay, and parts of the game seem to be either unfinished or severely simplified from their intended state.
Perhaps Supergiant Games simply ran out of money. I sure hope they didn’t just run out of ideas.