TOP 2012-08-25

Portal 2: Overrated?

I want to say at the outset that I enjoyed Portal 2, that it’s a very solid, very fun game, and all things considered, I’m glad that it exists. That’s the disclaimer. Now it’s time to lay into it.


Portal did not need a sequel. And before you decide to disintegrate my house with a material emancipation grill, let me clarify: Portal’s story did not need a sequel. Portal’s gameplay needed a sequel in much the same way that the pudding industry needs sultanas. One might say that I wanted more portals, and not more Portal. I wanted to see how the portal mechanic would operate in environments more interesting and varied than the monotonous grey test chambers of the Aperture Science facility; I wanted to see portals in urban landscapes, using them to escape from oppressive forces, or in catacombs or ancient temples or enemy bases. I was interested to see how portals could complement, or even replace, the sort of combat situations we find in Valve’s other games. The portal gun does not need to be an Aperture Science invention for all time: it could be reimagined in a different scenario, so that the gameplay, the fun bit, could translate onto a new story and different characters.

And even if you could argue that the portal mechanic only works in the context of contrived test chambers – something I’d rather put to the scientific method and test out before believing – then a different scenario and a different setting were still called for. Because, after all, the Aperture facility and the delightfully murderous A.I. controlling it were both destroyed at the end of the original Portal – remember? Ending song notwithstanding, GLaDOS was defeated, and surely we all know by now that bringing villains back from the dead is a cheap trick designed to squeeze as much money out of a franchise as possible. Yes, GLaDOS was a popular character and she was popular for a reason, but surely such a sagacious outfit as Valve wouldn’t turn the game into an easy cash cow, but realise that resurrecting a villain may be a feasible profit-making strategy but it’s a literary and artistic no-no.

In order to force a sequel onto the game, Valve needed to contrive reasons for GLaDOS to come back online and for Chell to get back into the facility. Somewhat ironically, Chell is literally dragged back into the building by a robot, fitting for a player being forced back into a story that’s already ended.

When we first see her in Portal 2, GLaDOS is strewn about the room in which we fought her at the end of Portal 1, despite the fact that Portal’s ending clearly suggests that she was lifted out of the room and strewn around the car park. Wheatley comments, “she’s off” – not “dead”, but “off”. This sets us up for a potentially interesting discussion about what it really means for Artificial Intelligence to be ‘dead’ or ‘alive’, but if Valve ever had any plans to explore such philosophical issues, they certainly must have dropped them, because it all just gets too confusing from here.

Everyone knows Portal’s ending song by now and if they don’t know the lyrics then they at least know the title, Still Alive. Yet if there’s one thing she just won’t stop going on about in Portal 2, it’s that Chell “murdered” her. “I’ve been really busy being dead,” she says. So much for “still alive”. But so much for being “off”, as well, because she later reveals that she was somehow forced to watch a video of Chell killing her over and over again while she was “dead”. So much for “doing science”, but it’s hard to know how she could be doing anything when she doesn’t have any power. And that’s if you ignore the fact that we burnt up several of her component parts and then let her get torn asunder in an impressive explosion. All that’s necessary to ‘resurrect’ her is for Wheatley to commit a minor blunder and end up throwing a few hundred switches, and suddenly everything we achieved in Portal 1 is undone. I just couldn’t help feeling a bit put out.

There’s a flagrant disregard for consistency here, and it’s quite probable that we’re supposed to just shrug it off because it’s only here for fun. But if that’s the attitude we were supposed to take then Valve has missed a major opportunity to explore the Artificial Intelligence topic, and to further their storytelling capabilities in the context of this already popular and captivating fictional world. Most gamers have caught on to the current debate about the status of video games as a medium, whether they count as ‘art’, and what possibilities exist for coupling interactivity and engaging storytelling, and Valve is usually cited as being on our side – on the side of games as a literary medium – and many believe that their games are not only fun but possess real literary merit. I’ll return to this theme throughout the review, but for the moment it suffices to say that I don’t believe Portal 2 comes anywhere near the bar that Portal 1 set in terms of developing this balanced craft. It almost seems to throw away the proposition that story is just as important as gameplay for the medium, replacing it almost entirely with values like “fun”, “easy” and “nostalgia”.

Portal’s (original) ending was, in my opinion, almost perfect. I think it could have done without the ‘cake’ cutscene, if I’m honest, but if we ignore that, the ending was as lean and crisp as the rest of the game (which was as lean and crisp as a frozen home-grown lettuce). GLaDOS’ true purpose is revealed, her plan is thwarted, the facility is destroyed, nobody will ever again have to suffer her twisted testing cycle, and Chell is free to escape into a potentially hostile world. We do not need to know what happens to her. When a story has sufficient closure to make you feel that you’ve achieved something but also not so much closure that you can stop thinking about it seconds after you finish it, then it has done a good job. It’s no good having so many questions left hanging that you tear your hair out in frustration and have to beg the writer repeatedly to make a sequel (c.f. Dreamfall: The Longest Journey), but there do need to be some matters left open for pondering purposes – what will happen to Chell? Where is Rattmann now? Was anyone else in the facility? What exactly is happening in the world at large? We don’t need answers to those questions because the player is quite capable of imagining various possibilities on their own. Like I say, Portal achieved that balance and therefore did not need a sequel.

For me, a good ending is not just one that leaves a few mysteries open but nevertheless achieves closure – it also needs a balance of ‘happy’ and ‘sad’. A totally happy ending is just as unsatisfying and unrealistic, in my opinion, as one where the bad guys win and all the good guys die. There are many games which achieve this sort of balance; Final Fantasy VII springs to mind. Since the game was such a commercial success, it was written in the economic stars that there would be sequels, breaking the usual mould of this franchise, which ordinarily starts afresh with each entry. And so we’re forced to accept that certain characters from VII weren’t quite as dead as we thought they were. But there was always a way out. If I wanted FFVII’s ending to have the same impact as it did when I first played it, then I could simply ignore the sequels and pretend they never existed.

But unlike books and films and every other medium, a digital games distribution platform has the unprecedented ability to release updates. And this has generally been a great thing, especially since Valve has cared enough about its existing products to keep fixing bugs and adding new content and syncing it to us in such a trouble-free manner. But changing the ending of the game? I cannot understate how truly monstrous, how insidiously terrifying this development is for the medium. It means we’re not safe. If this practice becomes widespread, we will never be safe from good stories getting ruined after they become wildly popular, after the invisible hand comes along and snatches out the original game’s soul and fashions a shoddy sequel out of it that only exists to rake in the cash. I would not be surprised if J.K. Rowling gets forced to make a Harry Potter sequel, but at least she cannot change the ending of all existing copies of the seventh book so that it foreshadows Voldemort’s return. (Actually the ending of the seventh book was terrible, but that’s another matter.)

After contriving a reason for Chell to get dragged back into the Aperture Science laboratories, Valve decides to have her put into suspended animation for hundreds of years, thus, perhaps even ironically, distancing itself quite significantly from the events of the original game. What possible danger or mystery could the outside world present, what possible omen, let alone dramatic effect, could GLaDOS’ cryptic “What’s going on out there will make you wish you were back in here ” (from Portal 1) possess, when it’s all hundreds of years ago now, and the whole world has probably changed beyond recognition?

When the game finally gets underway (because the moving container, while technically impressive, is little more than a long and elaborate cutscene), we find ourselves back in the same initial testing track that we experienced in Portal 1. The developer commentary reveals that this was done deliberately to show Portal 1 players how the familiar locations have changed over the centuries. For me, it felt like confirmation of what I’d feared, that Portal 2 was going to be nothing more than a nostalgic trip through the original game, albeit with a destructive face-lift (if such a thing is possible). To be honest, I don’t care what a drab science facility would look like if it were abandoned to nature for hundreds of years; I care about seeing interesting new gameplay and storytelling opportunities.


But things aren’t exactly the same as they were in the original game. There are the retcons. The elevators are different (forgiveable) and a ‘furnace room’ has been added that was clearly never meant to exist. And in general, the Aperture Science facility is obviously so much larger than it was originally intended to be in Portal 1. The enormous test subject ‘accommodation’ unit is there partly to illustrate the gigantic scale of the place. But where did this test subject suspended animation facility come from anyway? My understanding of Portal 1 was that GLaDOS forcibly enlisted Aperture employees to be part of the testing cycle, and anyone who did not comply was left to die in the neurotoxin. Portal 2’s own exposition makes it clear that test subjects in the 50’s were transported to and from the facility in a limousine; in the 70’s, homeless people were enlisted and then immediately discharged; in the 80’s, again the employees themselves were exploited. It seems incredible that Aperture would then have the money to build an enormous facility capable of keeping large numbers of test subjects alive for very long periods of time. But even if they did, I find that it removes a lot of the drama, the horror, from Portal 1, if Aperture already has a history of forcing people through testing tracks. And one final question I had to ask myself: where exactly is the equipment that keeps Chell alive for hundreds of years? The game lets us believe that Chell just went to sleep in an ordinary bed, without food or revitalisation technology anywhere to be found.

Discussion of plot holes is often perilous, since die-hard fans tend to be able to come up with explanations for them of a more or less contrived nature – but I’ll mention a few more anyway. I’m of the conviction that a story should not force people to hunt for explanations, at least not for technical aspects of the story that have no reason to be mysterious. So let’s continue reviewing the story and see what happens. After resurrecting GLaDOS and finding the dual portal gun, lying inexplicably under a pile of rubble near the furnace room, we’re put straight back in a GLaDOS-led testing cycle. One has to wonder, I think, why GLaDOS doesn’t simply kill Chell immediately and spend her time rebuilding the facility and getting the cooperative testing initiative up together. One might suggest it was the influence of Caroline in GLaDOS’ brain, but that didn’t stop her back in Portal 1, did it?


Rattmann’s dens make an appearance at the start of Portal 2, which I can only interpret as another nostalgia factor, because as far as I’m concerned, Valve messed up an opportunity here. Rattmann was kept mysterious and enigmatic throughout the original game, but the Lab Rat comic indicates that Valve actually came up with a very interesting backstory and personality for this character. But curiously, they chose to waste this backstory and kill him off at the end of the comic. If you have a mysterious character that you then develop further for the purposes of a sequel, it seems strange to have him killed off before the sequel even starts, and then just throw in a few token references to his existence at the start of the sequel. And then even these token references end up thinning out and the character is forgotten about completely.

A little later in Portal 2, GLaDOS taunts us with the idea that she might reunite Chell with her birth parents, but this comes to nothing, suggesting that it was an idle subplot designed to fill up the time and drag out this nostalgia trip a bit longer. And if they really needed an idle subplot just to drag out the gameplay a bit longer, I think they could have done better. Valve admits that they purposefully leave Chell’s backstory mysterious because players justifiably feel as though they are the protagonist, and don’t feel as though they’re role-playing Chell. It seems uncharacteristic, then, that they should threaten to break this feeling by raising questions about Chell’s past. But given that they decided to do this, it’s then all the more disappointing that they don’t in fact answer any of these questions: the subplot simply stops abruptly, without achieving anything, without any character development and without even being particularly funny. (“Your birth parents that you’ve never met are here in the facility! … Oh no they’re not! Joke!”)

Now GLaDOS finally decides she’ll kill Chell after the next few chambers but Wheatley rescues her. We troop along walkway after walkway, being told where to put each portal before entering the bowels of the facility to shut down the turret production line and the neurotoxin, this being the primary goal of chapter 5. Later, Wheatley manages to restore both of these things in preparation for the final boss fight, despite being an incompetent idiot. So this is not only inconsistent with his character, but also renders yet more of our hard work completely redundant. What is the point of doing anything in this game? We kill GLaDOS, she comes back; we destroy the facility, she repairs it; we euthanise our Companion Cube, it comes back; we cut off the neurotoxin, it gets restored. The restoration of the neurotoxin and turret production is all the more incredible given a) that GLaDOS couldn’t do it and b) that she supposedly couldn’t access those parts of the facility (“she can’t touch us back here”), raising the question of how Wheatley was able to do so despite being connected to the same mainframe as GLaDOS was. The whole “she can’t touch us back here” is quite contrived as it is: consider that GLaDOS was perfectly capable of going “to the surface” where she “saw a deer”.

Character ‘Development’

But let’s talk about the villain-swapping itself, where GLaDOS is removed from control of the facility and Wheatley installed in her place. This sequence at the end of chapter 5 is something I wouldn’t have expected to be so bad, given Wolpaw’s remark in an interview that he feels guilty about telling the story explicitly (“here comes the exposition buddy”). Valve also generally refrains from taking control away from the player and rarely uses cutscenes. In this sequence, the player is trapped in the elevator and is forced to listen to the argument between GLaDOS and Wheatley. We learn that Wheatley was one of the cores that Aperture tried to use to restrain GLaDOS’ power and that he was therefore designed to be a moron, and these facts are simply shouted at us – curiously, they never become relevant again. And then we have to sit through Wheatley’s character transformation, which seemed to me to be a good example of how character development absolutely shouldn’t work. Characters are supposed to go through gradual transformations, not stay constant for ages and then suddenly shift (unless they have a particular psychological condition that makes them like this, I suppose).

From my perspective, Wheatley was a bumbling comic idiot who meant well, despite a slight intolerance towards humans, and in a few moments he became an evil megalomaniac. For the rest of the game he acts as the primary antagonist, suggesting that we’re supposed to believe that power genuinely corrupted him into someone as evil as GLaDOS. If that’s true, the transition was unrealistically fast and contrived; there were no hints beforehand that Wheatley might have a latent resentment towards us or a lust for power. There is an alternative view, however, whereby the mainframe itself can instantly corrupt any core inserted into it. This would explain the sudden shift of personality, but it makes Wheatley’s treatment at the end of the game definitively unjust (if it wasn’t already). And besides, there is no explicit indication that this explanation is the one they intended. I think Valve intended for us to believe that Wheatley had become a real villain, and clearly didn’t care if the shift was sudden. I guess it was more important to fill up all that gameplay time with useless adoption subplots rather than worrying about real character development.

GLaDOS also goes through a transformation, and this time it isn’t so sudden. She discovers the personality of Caroline within her, and it makes her somewhat more human, and culminates in her deciding to save Chell’s life. This was reasonably well executed, but has the unfortunate consequence of removing everything that we liked about GLaDOS in the first place. GLaDOS’ remarks in the second half of the game are hardly ever entertaining and are delivered in a much more human tone. If you compare any of this voice acting with GLaDOS from the original game you will notice how stark the difference is, and how much better the original is, because part of the fun of GLaDOS was that she said funny things in a computerised voice and in a digital style. Acquiring more human traits, it’s almost hard to believe that GLaDOS is still a computer in the latter part of Portal 2. It’s hard enough as it is to understand exactly what it means for Caroline’s personality to be part of GLaDOS, or why it was conveniently inactive for such a long time, or why she didn’t delete it earlier.

In the developer commentary it becomes quite clear that this “GLaDOS origin story” was opportunistically grafted into the game when they decided to use Ellen McLain for Caroline’s voice actress for economy reasons. Frankly, it shows. GLaDOS already had an origin story. She was Aperture Science’s backfired project of creating true Artificial Intelligence. Did Caroline’s personality get connected to GLaDOS to preserve her for eternity in accordance with Cave Johnson’s dying wish, or was it to try to tame her murderous urges, as with all the other cores? There is clearly a conflict of plot points here, a natural consequence of shunting new ideas into a story when you’re already halfway through the sequel. And Cave’s remark, “hell, put her in my computer” suggests that GLaDOS is somehow the evolution of Cave’s personal computer, which just makes the whole thing more suspect.

What Are We Testing?

In the next part of the game we get a good look at the history of Aperture Science, and this was my favourite section. Still, awkward questions arise. It transpires that Aperture had the portal gun (or “Portable Quantum Tunnelling Device” as it was known then) back in the 50’s. It’s highly implausible that they would decide to use this device only as an unnecessary prerequisite for their other product testing initiatives, but even more implausible that they didn’t come up with a way, in all that time, to cash in on the device and solve their funding problems. But this isn’t the half of the problem. There is a more fundamental question, which also applies to the original Portal but becomes more acute in the sequel, and it’s something I’d be interested to see some discussion on. The question is this: what exactly are we testing? This part of the game suggests that the products themselves are being tested, albeit in very strange ways, and that implies that the testing tracks of the original Portal are designed to test the portal gun itself. Of course, you’d have thought that it would have had enough testing by now if it’s been used in over five decades of testing cycles for other products, and was presumably tested on its own before that too.

But there is another possible and in many ways more plausible answer to the question of “what are we testing?” Arguably, it’s not the gun that’s being tested, it’s the person. The test chambers are quite obviously designed to test the intellect of the person going through them, an idea reinforced by GLaDOS’ quips about the results of the test (“you are a horrible person… we weren’t even testing for that”). So which is right? Is Aperture testing its products or is it testing people? And to what extent can you call that “science”? Isn’t science about formulating and testing hypotheses? I don’t think I’m overthinking this: I think this is a question that deserves to be asked because it cuts to the heart of the entire story concept and is related to the very purpose of Aperture Science.

The Ending

So now we come to the ending. Wheatley undoes our earlier hard work, we defeat him, GLaDOS saves Chell but not Wheatley, deletes Caroline and lets us go. Notice in this final speech the way that the tone shifts from the emotional, flowers-and-sunshine vomit fodder of Caroline’s influence to GLaDOS’ customary cynicism, and notice how relieving it is. Of course they couldn’t have let the game end on such a sweet and pathetic note. But it proves that nobody really wanted Caroline, and we’d much rather have GLaDOS being mean and murderous.

That is a minor point, however – minor in comparison to how Valve genuinely mucked up the ending. Shooting a portal onto the surface of the moon was fantastic, yes: it’s what happens immediately afterwards that’s the problem. It’s called a cutscene. Now, if you’re trying to avoid using cutscenes, it’s generally more forgivable to save it until the very end rather than sprinkling them throughout the game, but I honestly think that this cutscene was unnecessary. Valve knows the deal: the player needs to be in a reasonably confined space without many options in order for story information to be delivered to them. Good examples are rooms with no exits and elevators: both are used to great effect in Half-Life 2 to ambush the player with exposition, without removing control from them for a moment. And guess what, Portal 2’s ending cutscene takes place in a room with no exits where GLaDOS delivers her speech, and an elevator (ignoring the short section on the moon). Since we’d previously endured story exposition in this environment, back when Wheatley took over the facility, there’s really no excuse. And then we simply take an elevator ride and meet the turret orchestra – it would not have been hard to make this scene interactive. The player wouldn’t have had much freedom in the elevator, but the illusion of control is very important in video games. The worst thing, I think, is the very final scene, where Chell steps out into the field outside the facility, confronted with an uncertain future. How much more powerful would this scene have been if we’d been able to walk out of that door ourselves and venture out into the world? The scene could have faded out enigmatically after the player had traversed a certain distance, and this would have been a far more satisfying conclusion.

If Valve’s philosophy is to keep control in the hands of the player unless there’s absolutely no alternative, then it’s all the more disappointing that they missed these opportunities and wrenched control away from us in such a sudden and altogether unnecessary fashion. I imagine they realised that they’re actually quite good at making FMVs and wanted to do one for an important part of the game. FMVs do entail a higher quality, of course, but the cost is not just the absence of interactivity, but also inconsistency, since the prerendered version of a model always looks slightly different from the realtime equivalent. And in this case, I think this higher quality was wasted. The clean, white, simple shapes of GLaDOS and the turrets hardly look much different in the realtime engine. Maybe it would have been nice to see a fully prerendered video fly-through of the Aperture facility, or an aerial view of it; but I don’t think it was worth rendering turrets and a dark elevator shaft. And that final movie was probably a sizeable portion of the download time that I found so excruciating on release day.

Good Stories

It’s often said that Valve’s strength is in writing good stories. I disagree. Portal’s story is practically non-existent as it is and Portal 2’s, to be quite frank about it, is crap. It’s unoriginal, fails to say anything about the potentially interesting questions of Artificial Intelligence, is riddled with plot holes and inconsistencies and was quite obviously cobbled together without any clear idea of where it should be going. Character development is sudden and contrived; subplots are introduced that don’t go anywhere; lots of things are done only to be undone at some later point, having achieved apparently nothing. And this isn’t an isolated case. I don’t believe the Half-Life series has a good story either. When it comes down to it, it’s little more than “aliens invade, fight them”, which shouldn’t really be winning any awards nowadays. So no, Valve doesn’t write good stories. What Valve does is to construct believable, captivating and incredibly detailed worlds around stories which range from mediocre to trite. That is their success. Since video games are necessarily about a player’s interaction with a fictional world, making convincing worlds is arguably more important in this medium than in any other, and that explains why Valve’s games are still fundamentally good. But ultimately, a crap story that’s told really convincingly well… is still a crap story.


Because I’m already stretching the word limit acceptable for a game review, I’ll make my gameplay points in list form, in the hope it’ll constrain me.

  1. Certain objects can’t be picked up. Compared to the beginning of Half-Life 2, where you can pick up chairs, televisions and paint pots alike and throw them out of the window just for the hell of it, Portal 2 feels unnervingly ‘static’. I suspect the restriction in Portal 2 has something to do with the fact that weighty objects could potentially be used as substitutes for cubes on the pressure buttons, but this doesn’t change the fact that it felt somewhat immersion-breaking. Allowing the player to pick up anything they should reasonably be able to pick up makes the world feel more real and less like a painted setpiece (“you can look but you can’t touch!”). On the same point, there’s the foliage growing out of the test chambers which is completely static: it doesn’t rustle or move; you just walk right through it as though it isn’t there. There are also – unforgivably – some invisible walls.

  2. Some features of the original Portal were inexplicably removed. In Portal, the cross-hairs would only light up when you hovered them over a portalable surface; this doesn’t happen in Portal 2, despite the fact that it could have been potentially useful in those big spaces where you had to search for ages for the right shade of white that permits portal placement. Similarly, the original portal gun has a circular light on it that tells you which colour portal you last fired, again missing in Portal 2 for no apparent reason.

  3. Conversely, a feature was added to Portal 2 which I hardly ever found useful and quite often found annoying. I’m referring to the fact that you can always ‘see’ the portals that you’ve placed, even if there are walls separating you from them. Theoretically it’s to help you work out where you’ve left your portals if you’ve gone quite far away from them. But it has the curious effect that if you are looking directly at the portal, the indicators are also still there, overlapping the actual portal, and, weirdly, everything else (‘always on top’). That means if you’re looking at a portal and there’s an object between you and the portal, the portal-finder will overlap the object despite the portal being behind it and in plain view. At first I thought this was a bug. No, it’s a feature.

  4. Dumbing down. With few exceptions, Portal 2’s puzzles are easy. I can’t prove this, and I know some people disagree, but I was very disappointed to find that I never felt genuinely challenged at any point in the game. I kept expecting a ramp-up in difficulty after we’d learned new mechanics, but it never came. I appreciated that there were fewer timing-based puzzles, or puzzles that depended on quick-reflex acrobatics, but I think the difficulty level overall was much lower than the original game.

  5. Related to the last point is the design of the puzzles themselves. Look at almost any chamber from the original Portal and you’ll be pleased to see that characteristic shade of grey that says “portal me!” covering most of the walls, floor and ceiling. The puzzles were still hard despite the fact that you could put portals almost anywhere. But look at most chambers in Portal 2 and you’ll find that massive chunks of the environment are unportalable; in some cases, the only surfaces that are portalable are precisely the ones that it’s necessary to put portals on in order to complete the puzzle. This was almost certainly a way of guiding the player towards the correct solution, but unfortunately this not only dumbs the game down but also takes away a lot of the ‘playground’ feel from the levels. If I want to stop and randomly do an infinite fall, I often can’t.

  6. The portal placement and entry/exit rules were simplified, at the cost of making them less realistic. When you put a portal on a Portal 2 wall, it has exactly one possible position – you can’t put it slightly further up or further down, and if you shoot at an angled surface, the portal will always appear with the correct orientation. When both portals are on the floor, you are given an inexplicable ‘nudge’ upon exit that allows you to escape reentering the exit portal and getting into an endless loop. And entering a portal has been made much more liberal: you can be carrying a box and only just manage to enter the top part of the portal, the box looks as though it should hit the floor and not go through, but it magically comes through the portal with you anyway. Of course these changes were made with good reason – to remove the often frustrating timing/accuracy aspect from the puzzles and focus on the cerebral element. But sometimes it comes across as excessively unrealistic, and it breaks the immersion.

  7. Usually you can’t place a portal on a moving surface, but that’s exactly what you have to do to cut off the neurotoxin supply in chapter 5.

  8. The behaviour of the gels is somewhat inconsistent. This is difficult to explain in words, so I might post a video of it at some point. Basically, if you walk onto some blue gel, you don’t bounce until you specifically press the jump button. But there’s a bit in chapter 6 where you slide from orange to blue gel, and you get automatically bounced at the right moment. Obviously they did this to help you complete the acrobatics successfully, but in my case it backfired: I was convinced I had to press the jump button, and this messed up the system, sending me hurtling into the gunge repeatedly.

  9. There are too many gameplay mechanics. It seems to me that they just had too many good ideas! But it ends up being excessive. In addition to the customary cubes, buttons, moving platforms, turrets and emancipation grills, we have faith plates, thermal discouragement beams, light bridges, excursion funnels, repulsion gel, propulsion gel and conversion gel, all requiring their own introductory sessions to explain them. The result is that there isn’t really enough time to explore any one of them; I was always expecting the game to settle down with the mechanics already introduced and just throw some good, hard, well-designed puzzles at me to test what I’d learned, but it felt as though you’d learn a new mechanic and then it would get pushed to one side and yet another one would come along. The light bridges were criminally underused. Looking on the bright side, at least there are plenty of creative opportunities for custom map-makers who can finally realise the full potential of all these mechanics.

  10. What’s interesting about Portal is that you don’t just go through a series of test chambers in a regimented fashion – you also go ‘behind the scenes’ and see the guts of the facility that you’re trapped in. There are several parts in Portal 2 where you’ve finished one set of testing chambers and you’re on your way to another set, but these sections were often unnecessarily long and trivially easy. In many cases they simply reduced to putting a portal on the other side of a wall or on the opposite side of a gap – repeatedly. These sections of the game were effectively just ‘padding’, making the game arbitrarily longer, and contrast starkly with Portal’s minimalism, where practically nothing was superfluous.

  11. In Half-Life 2, if you go past a map transition barrier backwards or aiming at the floor, this rotation is carried through to the next map once it’s loaded. But in Portal 2, you’re always facing in the same direction at the beginning of each map. This is nitpicking a bit, but it actually made me feel less as though the game world was continuous and more as though it was a sequence of distinct levels (which it is). The fact that it uses full size loading screens with animated progress indicators rather than a simple and unobtrusive “loading…” sign also doesn’t help.

  12. The music and sound design is generally praiseworthy. Many of the gameplay mechanics have their own musical theme, a sonic signature, so to speak, and often interact with each other or with the background music for dramatic effect, which is highly creative and commendable. Sometimes, however, I thought there was a bit too much music. This is an abandoned and lonely facility, after all, and it was sometimes annoying that the music just repeats over and over and doesn’t stop until you’ve finished the chamber. And I’d also like to complain about the new portal placement sound – I’m not the only one who finds it annoying am I?


Portal 2 is fun, clever and entertaining; I’m not going to deny that. But the game has been subject to such a lot of hype, showering it with praise that often goes uncriticised. From my perspective, Portal 2 has provided us with a toolbox of clever and creative gameplay mechanics from which both Valve, and the players, can create fun and interesting puzzles (which is what makes the future DLC so exciting). The single-player campaign is then effectively just a showcase for these puzzle elements, and the plot is only there to string together the various demonstrations. I don’t think it deserves any more praise than that, and that’s a shame, because it could have been so much more.