One False Move

by Dan H

Dan muses about Gunpoint, with extensive references to Mark of the Ninja.
I bought Gunpoint in a Steam Midweek Sale when I was on a stealth-high from finishing my New Game+ playthrough of Mark of the Ninja (yes, I know MotN came out a year ago – I play games slowly). It cost me £3.99 and I don't think I wasted my money. Then again, I'm on a reasonably decent income, and anything under a fiver is well within my “impulse purchase” limit. I've paid £3.99 for games and then never played them and not really minded.

I think I probably put Gunpoint in a difficult position, because playing it so soon after MotN meant I inevitably made comparisons (particularly since I'd bought them both in sales, and had actually managed to snag Ninja for £3.50). This was probably a bit unfair, since Gunpoint is an indie game that some guy produced in his free time, while MotN was put together by a full development team with corporate backing. On the other hand, I actually paid more for Gunpoint (and the cover price is actually comparable if you buy one of the special editions), so I'm not convinced it's completely wrong to hold the game to the same standards.

Anyway, in Gunpoint you play a pixelated trenchcoat-wearing freelance spy who accidentally implicates himself in a murder and then gets drawn into a faintly tongue-in-cheek web of intrigue, betrayal and data theft. The core gameplay elements are the Crosslink – a device allowing you to connect items to other items, meaning light switches can open doors and security cameras can fire guns – and your Bullfrog Jumping Trousers, which allow you to perform incredible leaps in order to access secure areas or avoid guards.

A common problem with reviewing anything, but I think particularly with video games, is trying to make sure that you address the thing you're reviewing on its own terms. Tom Francis, designer of Gunpoint made this much easier on me by putting his game design manifesto on the internet. Now as it happens I quite like a lot of the content of Francis' manifesto (although I find his aversion to variable difficulty settings faintly bizarre) – certainly I find it more bearable than the sort of nonsense you get from Tale of Tales and the like. Anyway, the ten subheadings Francis provides give a pretty good idea of the sort of thing he is trying to do. The question is whether he succeeds in doing them.

So that's what I'm going to look at.

1: I want to make games that generate cool experiences

Francis' first point is that he doesn't like games which try to provide a “directed, cinematic experience.” As he puts it “if only there was some other medium where people could direct things cinematically.”

I'm not sure I quite buy all of his rhetoric here (he suggests that in the future we will look at cinematic games with “the same tragicomic pity” with which we look back on obsolete children's toys; this seems needlessly dismissive of both cinematic games and traditional toys) but I appreciate the sentiment. I'd much rather have a game be cool because of the way it responded to something I did, rather than because I reached the right point in the corridor to trigger the cutscene where my character surfs on a shark.

Does Gunpoint “generate cool experiences”?

Umm … sort of.

I think I'd need to play around with the level editor a bit more to be sure, but I think it has the potential to generate plenty of cool experiences, but its elements are too frequently isolated from one another. You have a guard in a room with a lightswitch, and you can jump on the guard's head, or turn the lights off, or electrify the lightswitch. Then you can do the same thing all over again with a different guard in a different room. I kept sort of vaguely expecting massive, unpredictable, wacky chains of events, but they never emerged. If nothing else, the most efficient route was almost always to find a conveniently located light switch, and control everything from there.

I think part of the problem here is that the game is caught awkwardly between puzzle game and – I'm not sure what to call it – freeroaming chaos simulator? If you could rewire the levels however you wanted, then you'd have the freedom to manipulate things however you liked, and that would lead to more emergent gameplay. But that would completely destroy the puzzle aspect of the game, because it would make your objectives too easy to achieve.

I think it doesn't help that there really isn't that much variety to the interactions. Guards basically wander back and forth and shoot you on sight. They can open doors but often choose not to (for reasons I don't understand). Most of the time they stand stock still in one place, turning around only when they hear specifically coded noises (they can hear a lift arriving from across the floor, but don't respond at all to a door being kicked in). This makes a lot of the funky boobytraps you can set up next to useless – sure, you can wire something to a hand-scanner that a guard might use to open a door, but that doesn't change the fact that the guard is just going to stand there doing nothing unless they have to go to switch a light back on.

I did quite enjoy Gunpoint so I really hesitate to use expressions like “the worst of both worlds” but it feels a lot to me like you get, well, the worst of sandbox and puzzle gameplay. For levels to present any kind of challenge, there needs to be a puzzle element to the gameplay. But for the game to be open-ended, puzzles have to afford multiple solutions. This means that most of the puzzles are quite easy (because finding one of several solutions is easier than finding the only solution) but that they still feel like puzzles. I never found myself thinking “hey, I wonder what happens if I do this?” only “hey, I wonder if doing this will work.”

This brings us to the next point on the manifesto.

2: I want to make games that let the player be creative

To an extent, I think that this one is on a hiding to nothing. In the absolute sense, a player can never be truly creative in a video game, because everything that happens in the game must have been coded into that game by the developers. You can do things the developers didn't anticipate, but that isn't really the same as being creative. You're always working within somebody else's framework.

The best you can usually get out of a game is being made to feel creative, and Gunpoint didn't really do that for me. Again, I think that this comes down to the puzzle elements. Because every mission has a clear objective and shows you everything upfront your thought process goes immediately to “how do I achieve my goals with the resources in front of me.” Perhaps I just played in a funny way, but I felt that I was being asked to solve a problem, not to explore a scenario, and as a result I usually went for the first solution I hit upon. This had the unfortunate consequence that even if my ways around problems were utterly unique in the history of Gunpoint players (I don't believe for a second that they were) I still felt like all I'd done was solve a slightly arbitrary puzzle. It also meant that the game didn't invite replay to me, any more than I'd want to replay a puzzle game to see if the puzzles had different solutions.

3: I want to make games with clear rules but surprising results

I think Gunpoint sort of semi-succeeds on this one. The rules are definitely clear insofar as the game is pretty clear about what you can and can't do, but some things are quite annoyingly vague, in particular the way the game handles noise. Again, I've probably been spoiled by Mark of the Ninja on this one, because its noise handling is utterly crystal clear. In Gunpoint, however, most of the time a guard will stand where he is doing nothing, no matter what you do on other floors, but sometimes they will run up several flights of stairs to investigate breaking glass. And sometimes they won't.

Similarly I'm not really sure how well the game succeeds in producing surprising results. I suppose the slightly unpredictable way the guards behave in response to noise is surprising, but otherwise the game is rigidly deterministic. Every item in the game is a switch that can be in one of two states, and flicking between those states is entirely predictable (unless you lose track of what you're doing). The levels just don't seem complicated enough to give rise to the same kinds of domino-effect chains of events that characterise emergent gameplay.

I should stress that a lot of this is going to be down to personal preference and, to some extent, buy-in. Whether something feels like it emerged organically from your gameplay or not is ultimately a judgement call everybody has to make, but to me it always felt like I was just unpicking the pre-prescribed solutions to each level. Once I'd seen how something worked, I saw how it was inevitable that it would work that way, so in that sense the rules were clear, but the results weren't surprising.

4: I want to make games that are a bit different

I liked point four of the manifesto a lot. It basically says “I don't want to make yet another game in a genre that's already tapped out, but at the same time I don't want to be all … ooh games are not games because something something chess something” (sorry, that's my wording – the Tale of Tales manifesto annoyed the crap out of me).

Gunpoint does succeed in this area. It's a puzzle/platformer/stealth game that is kind of like other puzzle/platformer/stealth games while still being its own thing.

5: I want to make games that are fun to learn

I'm torn on this one.

I really appreciate the fact that Francis includes this in his manifesto, and I like the way he expresses it (“The mindset that interactivity has to be stopped in order to teach something is fucking insane”).

I think I'd call this one a qualified success. Certainly the process of learning to play Gunpoint is pretty much indistinguishable from the process of playing it, so you very much do learn by doing. On the other hand, this means that a lot of both the learning and the gameplay comes down to trial-and-error. I've also seen other reviews suggest that you don't really feel you've learned the game properly until you've finished it, and I think I'd agree with that. It's only in the final mission that you really feel you have to put everything into practice, so it almost feels like the whole game is a giant tutorial.

Still, it's certainly better than stopping you every thirty seconds to bombard you with popup windows.

6: I want to make games that feel good but still use your brain

This is another qualified success. It does work as a puzzle/strategy game, and so in that sense it does indeed feel good but still use your brain, but I don't think this is particularly unique. This point in the manifesto talks a lot about the artificial distinction between “dumb but fun” and “interesting but frustrating” games. The thing is, the manifesto acts like these are the only two types of game that exist, and they aren't.

A great many games are fun. A great many games are interesting. Some games only manage to be fun, and that's fine. Some only manage to be interesting, and that's fine too as long as they aren't crap and frustrating with it. But there are also a great many games that manage to be both interesting and fun. And indeed for a lot of gamers (particularly the sort who are likely to play and enjoy Gunpoint) “interesting” is part of the fun.

To put it another way, Gunpoint does indeed “feel good but still use your brain”, but so does Plants versus Zombies.

7: I want to make games that value the player’s time

Again, I really appreciate the sentiment, and I while Gunpoint has caught a bit of criticism for its 2-3 hour playing time, I actually really like the fact that it's a short, accessible game that you can pick up and put down as you like. And I would far rather have a two hour game that is fun throughout than a ten hour game that is chock full of filler.

I'm not sure it values your time as much as it thinks it does, though. Its autosave system is cute, but because it only tends to wind you back a few seconds it can, paradoxically, wind up costing you more time than a checkpoint save system – I recently had to restart a longish mission from scratch because I wanted to try going for a “kill every last motherfucker in the building” playthrough. I knocked a guard out by accident (once somebody's KOed you can't go back and kill them later), and the game went past the 10s autosave limit before I hit load.

The game also makes the extraordinarily peculiar design decision of including exactly one depleteable/grindable resource. Several of your special gadgets (the Longshot that lets you rewire enemy guns, the Hushcracker that allows you to break glass silently, and the Prankspasm that allows you to electrify objects) use batteries. These batteries are not replenished between missions, but batteries you pick up on missions are carried over. Since there are some missions which contain batteries but are otherwise short, it means that grinding for batteries is actually quite a sensible plan a lot of the time – certainly it's the only way to make heavy use of the Longshot (which takes four batteries at a time – you'll normally only have four in total unless you upgrade your capacity, and most battery-containing levels only drop one or two). For a game that seems intent on letting you play however you want, this is an inexplicably poor choice.

8: I want to make games that let you choose how much challenge to take on

This is the peculiar section where Francis complains about games with variable difficulty settings. Even though, actually, manually adjustable difficulty settings are the perfect way to allow a player to choose how much challenge to take on. I mean, it's what they're for

The problem with the alternatives that Francis suggests (“optional objectives, perfecting performance metrics, taking on a late-game challenge early, or adhering to a personal play style”) is much the same as the problem with Shamus Young's self-balancing gameplay. It presupposes that a particular set of player behaviours indicate a desire for an easier or more difficult gameplay experience. They often don't. If I try to take on late-game challenges early, it might be because I'm deliberately pushing myself, or it might be because I'm the sort of impatient player who follows the plot without worrying about sidequests. If I try to stick to a personal playstyle, it might be because I am trying to make the game more challenging for myself, or it might be because I just like the idea of playing the sort of game where I do those sorts of things.

To take a concrete example, in most stealth games, the 100% nonlethal, no-fingerprints-no-witnesses runthrough is the hardest (or, in some cases, second hardest after “don't use the stealth mechanics at all” or “kill absolutely everybody for no good reason”) while the easiest is usually the path of greatest expediency, killing where it's helpful, sneaking where it isn't. And to an extent this does allow a certain amount of player challenge selection, particularly since – within the stealth genre – it's usually understood that “no fingerprints” is “supposed” to be the most difficult way of doing things. But this is just a genre convention – there is no intrinsic reason that deciding “I am going to play this game without killing anybody” has to be the same as deciding “I am going to play this game in the most difficult way possible.”

Even more problematically, Gunpoint actually reverses this genre convention (not necessarily intentionally). I've gone back and replayed a couple of levels and, generally speaking, the hardest way to play the game is to play as a total psychopath. The way the game is set up, you can incapacitate most guards in a variety of nonlethal ways, the most straightforward of which is to jump on them and punch them. Having punched a guard to knock them out, you can carry on punching them as long as you like, and ten punches will result in a kill. This means that the main lethal takedown is simply the main non-lethal takedown repeated ten times, making it strictly less practical. Not only that, but there are only two other lethal attacks in the game, one of which is to throw somebody out of a high window (I've yet to work out what counts as “high”, sometimes a fall of a single storey does it, sometimes it doesn't) or to shoot them with a gun you can purchase for $1000. The gun has limited ammo, and if you shoot a single guard, a countdown is immediately triggered after which a police sniper appears at the exit. The police sniper will instantly shoot you dead the moment you step into the light, meaning you have to make sure you take them out with a single shot from your gun (which has very limited ammunition), delivered from a dark room, or else the level becomes unbeatable.

All of this adds up to make the most challenging way to play Gunpoint the completely psychopathic way. But not everybody wants to play a complete psychopath. Particularly because playing a complete psychopath differs from playing a pacifist only by the addition of a few red pixels. It doesn't change the way the rest of the level responds to your actions (it isn't, for example, like Mark of the Ninja where the bodies of dead guards will trigger reactions in other guards, potentially even causing them to panic and shoot each other). So varying your playstyle is ultimately quite unrewarding, and while it can make the game more challenging, I (and I suspect many players) would rather have the game be challenging by default.

It doesn't help that several of the challenges you might choose to set yourself on a particular level are dependent on the level design allowing those challenges in the first place. Non-lethal is always an option, but if you wanted to try a “perfect gentleman” run (no witnesses, no violence of any sort, no loud noises) there are some levels where it is simply impossible, because you are required to break glass to progress through the building (you could use the Hushcracker, but that takes batteries – see above). I haven't played around enough to tell if every level is open to a no-violence-no-witnesses strategy, but I'm fairly sure that some of the early ones at least involve guards who have to be punched out in order for you to progress.

Again, I should stress that a lot of this is about personal reaction and buy-in, but having played through the game once, I just didn't get any real urge to go back and try it again in a different style.

9: I want to reward anyone who supports me, instead of pointlessly fucking them over

This is about Francis' stance against DRM, which doesn't really reflect on the gameplay very much. I will say that I liked the way he tiered the pricing of the game (you can basically get a standard version for £8, a special edition for £16 or an exclusive edition for £24) but otherwise I don't have much to say on this one.

10: I want to make exciting games

Again, I think this was a qualified success. Although I think sadly I was most excited about Gunpoint before I started playing it.

I think the idea of the game – a tongue-in-cheek noir stealth-em-up where you can rewire the level to manipulate guards and defences – is potentially exciting, but the implementation of it here is just slightly lacking. Everything you can do just feels a little bit samey: open a door into a guard, they fall over. Electrocute a guard, they fall over. Punch a guard, they fall over. Shoot a guard, they fall over and put you on a timer. I know that the game was made on a tiny budget, and complaining about animations in an indie pixel game makes me a gigantic asshole, but the problem I had was that the different approaches to a situation just didn't feel meaningfully different. I mean I could electrocute a guard with a plug socket, or knock him out with a door, or zap him with a rigged lightswitch (I never got that to work though) but it all amounted to the same thing. The guard walked along, I pressed a button, the guard fell over.

I feel kind of bad, because I've spent nearly three and a half thousand words nitpicking the hell out of this game, and I actually did quite enjoy it. Apparently Tom Francis made enough off of Gunpoint that he can now develop games full time, and I genuinely am interested in seeing what he does next, because Gunpoint is a very good start. But I'm not sure it totally lives up to the high standards he sets for himself.

And I'm not sure I'd have paid eight quid for it.

bookmark this with - facebook - delicious - digg - stumbleupon - reddit

Comments (go to latest)
Guy at 06:05 on 2013-09-26
To an extent, I think that this one is on a hiding to nothing. In the absolute sense, a player can never be truly creative in a video game, because everything that happens in the game must have been coded into that game by the developers. You can do things the developers didn't anticipate, but that isn't really the same as being creative. You're always working within somebody else's framework.

It's rare, but I think it is possible for player creativity to exist in games. This depends of course on your definition of "creativity", which is a seriously contentious term, but I'd say it's got something to do with putting together elements in a way that lets them do something new or unexpected. "Working within someone else's framework" can still be highly creative if that framework is sufficiently broad/flexible. Otherwise we'd say there's no creativity in music anymore since people are just re-using the same notes and scales that someone else created.
As for an example of "real player creativity in games", I'd say the strongest example that comes to mind is SpaceChem. You could make a pretty good argument for MineCraft, too... but again, it depends on your definitions of "creativity", "game", &c... at 20:32 on 2013-09-26
"putting together elements in a way that lets them do something new or unexpected" -- yes! So this point is closely tied to point 3 in the manifesto, about having clear rules with surprising results.

Maybe a good example is chess. Even once I'd clearly understood the simple ruleset, it would never have occurred to me that it might be a viable strategy to, for instance, deliberately throw away some pieces, in order to buy yourself time to bring other pieces into play. What a delight to discover that this bizarre strategy could actually work! In other words, the ruleset of e.g. chess is rich enough that you can discover or invent strategies that are genuinely surprising.
Arthur B at 21:18 on 2013-09-26
I think the distinction I would make is that playing a single-player computer game with pre-designed levels and linear narratives isn't so much like playing chess as it is like doing a series of chess puzzles: rather than having the sheer variety a human opponent provides, the game presents you with more or less the same challenge each time - and here the "chess rules" seem interesting in principle but the actual "chess puzzles" that are presented aren't especially rich.

In other words, the rules of a game are a map that defines a territory, and playing that game is an exploration of that territory, and playing a computer game with pre-designed levels and a linear plot forces your exploration to stick to a certain route, within the bounds of a certain degree of freedom. A good linear game can spool out a fairly substantial degree of freedom, but even then it's not going to leave you free to just roam around exploring the territory as you please because by setting a particular scenario the designer has implicitly limited your freedom to explore the potential of the game rules to the context of that scenario.
Dan H at 21:32 on 2013-09-26
I think this, again, comes down to different definitions of creativity. I don't think I'd use the word "creativity" to describe - for example - coming up with new strategies in chess. But this is mostly quibbling over terminology.

Either way I'm not sure the way elements interact in Gunpoint allows the player to be creative, simply because there aren't that many interactions, and you're working against a static AI.

You'd never need to work out how to sacrifice pieces in Chess if your opponent always made the same moves every time.
In order to post comments, you need to log in to Ferretbrain or authenticate with OpenID. Don't have an account? See the About Us page for more details.

Show / Hide Comments -- More in September 2013