The Stack

More office Team Fortress 2 yesterday — this is becoming a regular thing. I’m skipping it today — I’m facing a deadline — but made up for it with a lengthy session after work last night, in a big chaotic battle on a public server. Fortunately, it seems that whatever it was that caused my machine to shut off is limited to the Developer Commentary mode. I experienced no problems in the game proper.

By now, I’ve played enough to get three special special class-specific items. There are two ways you get items in TF2. First, you apparently just get items handed to you for playing a lot — I’m not sure if this is bound to specific milestones of total time played or score or if it’s just completely random. One thing that’s definitely random is which item you get — it can be for any class, and not necessarily one you’ve been playing, or have ever played. The other way is to achieve a certain number of class-specific Achievements, which always gives you an item appropriate to the achieving class. Two of my items are of the first type, and the third was for reaching the first Achievement milestone as a Sniper. Which is strange, because I had been playing mostly as a Pyro lately, and actually had more Pyro Achievements than Sniper Achievements. But the Sniper class has fewer Achievements in total than the Pyro, so its milestones come quicker.

The reason I was playing the Pyro so much is that my first Item was a Pyro one: the Backburner, a special flamethrower that always does critical hits when you attack someone from behind. It’s a nice bonus, and more importantly, it alters your tactics by giving you an incentive to be sneaky. Its drawback is that, unlike the standard flamethrower, it can’t do a “compression blast”. I had never used that anyway, so it’s not much of a loss. On the other hand, I kind of want to try it out, now that I can’t do it. So I’ll probably swap the Backburner out the next time I play.

For the Sniper, I got the Huntsman, a bow that replaces the sniper rifle. The guy who started the office session specializes in Sniper himself, and really likes this bow, because it gives the sniper a better chance in close quarters. This may be the right choice for small-team King of the Hill mode, where all the players converge into the same room, but in a larger game, I found I missed the rifle’s scope. Without the ability to zoom in and pick people off precisely from halfway across the world, the Sniper loses its main appeal. So I don’t care for the Hunstman.

The third item is a watch that lets the Spy feign death. I haven’t even used it yet, because playing the Spy is difficult enough to get a grasp on without extra features. I hadn’t played Spy before getting this item, and even now only played it briefly and completely wrong. The difficulty here is that the Spy’s powers mostly revolve around changing his appearance or turning invisible, but there’s no obvious feedback about this: as in most first-person games, you can’t actually see yourself. Most classes play pretty much like any FPS, but to play the spy, you need special instructions.

And I don’t know where to find those. There’s no obvious in-game tutorial or documentation for the classes — there’s some intro videos for the different map modes, but that’s all I’ve seen. The official TF2 website just has a blog and a link to Steam, and the “View the manual” link on Steam just shows a bit of promotional literature about the sentry guns that the Engineer class can build. Everything I’ve learned about gameplay, I’ve learned by word of mouth. I suppose it’s possible that this is how everyone learned the game — starting with trade shows and interviews with the developers, spreading through web forums. If that’s so, and there really is a complete lack of official documentation, that would mean that the game is even more dependent on its community than I had imagined.

Anyway, that’s three items, and it looks like I’m not using any of them at for the time being. Maybe the next one will be better.

Everyday Shooter makes one really big departure from standard practice in shooters: in most cases, you don’t get any points for shooting stuff. In fact, you never get points just for shooting stuff. At most, shooting stuff releases square chips that you can then collect for points, possibly wasting valuable time or putting yourself at risk in the process — and these chips are the only source of points in the game. But most things don’t even release chips, unless they’re destroyed as part of some kind of chain or combo.

You might wonder why I care. I’m seldom concerned with score, unless it affects game mechanics somehow. Which it does. Certain milestones yield extra lives, of course, and while they don’t come as fast as in Robotron, I always seem to manage to get a few. (The first extra life is at 200 points. I’m not sure how it goes after that. I typically seem to wind up with something over 2000 points per game, but I don’t think I get anywhere near 10 extra lives in the process.)

More important are the unlockables. See, your points build up from game to game, going into a pool that you can spend on stuff. There seem to be three categories of things you can buy: graphical effects that make no difference to game mechanics, additional starting lives, and the ability to play specific levels in isolation without going through all the previous levels to get to it. (For that last one, you have to actually reach the level normally before you can buy it.) Obviously the first few lives are must-haves, but the marginal price seems to go up exponentially. (And I do mean exponentially. That word gets misused a lot, but not on my blog.) I can foresee a point when the single-mode levels become a more attractive expenditure: when I’m having difficulty mastering the most-recently-reached level and want to practice it without wasting time on levels I’ve already mastered.

Because, in the tradition of the arcades of old, there’s no permanent progress within a game. When you run out of lives, all you can do is start over from the very beginning, which becomes tiresome. I suppose the buying of unlockables is an attempt at finding a compromise between this uncompromising design and the more modern approach, where progress is regarded as your right just for playing a lot. And really, it works pretty well, but I think I’d be happier if it worked a little faster.

In the last 24 hours, I’ve been involved in two more office Team Fortress 2 sessions. The first was apparently on the game’s anniversary of release or something: all the characters wore little party hats (on top of any other hats they normally wear), and, when killed, exploded into balloons and confetti.

As a result, I’ve given a serious try to two more character types: the Demoman and the Sniper. The Demoman, master of the grenade launcher, actually seems pretty bad for this small-team stuff. When there’s only three to a side, you spend a lot of time alone, waiting for your teammates to respawn, and the Demoman is essentially helpless when alone: the delay before his grenades go off means that he can’t really kill any but the most oblivious of victims. His value, it seems to me, is more in the threat of damage than in the damage itself — to limit the opponents’ options by placing obvious threats in front of them. I’m told that the use of grenades in real-life combat is similar — that the point of them isn’t so much to kill the enemy as to make them take cover or flee. Anyway, I found the Sniper much more satisfying, even though I’m rubbish at it. Although classified as Support rather than Offense, killing is all the Sniper does.

The first session left me wanting more, so after I got home, I tried running it on my home PC for the first time. Trying out the Developer Commentary tracks, I was alarmed to find that my machine spontaneously switched off, multiple times. This is an unprecedented problem. I’ve seen games exit to the desktop, freeze up Windows, and BSOD, but never just make the machine power off without so much as a beep of warning. Maybe the graphics card is drawing too much power or something?

Everyday Shooter is a title I’d heard before, but didn’t know much about: I knew it was abstract, and it had received some attention around the same time as Portal, but that’s about all I could tell you. Somehow I had got the idea that it had a great many levels, each with its own rules. That’s half-right — there are only eight levels, each more elaborate than I had been led to believe. And yes, each level does work a different way. The controls stay consistently Robotron-like, but they vary in enemy mechanics, and in particular in how you create the chain reactions that clear the most enemies and potentially net the most points.

I also wasn’t expecting it to turn out to be a member of that severely underpopulated genre, the Music Shooter. Instead of zaps and explosions, your shots produce notes, or even entire riffs, played on an electric guitar. These sounds become part of the music playing though the level, always an unaccompanied guitar piece, as abstract as the shooter itself. The underlying songs are linked to the levels’ structure, and in a way that suggests that the song, rather than the level, is the dominant element. Each level lasts exactly as long as it takes to play the song, and changes in what’s going on in the game are governed by shifts in the music more than by anything the player can do. The game’s creator even refers to it on his website as an “album”. I’d almost say that it turns shooter mechanics into a kind of dance, but really, that’s something that’s always been inherent in the genre — particularly in scrolling shmups, which share Everyday Shooter‘s unstoppably flowing nature. All too often, however, those games interrupt the flow by stopping the music and the action when the player gets hit. Everyday Shooter understands what it’s doing too well to make that mistake.

It’s definitely what I’d call an art game, which is a little ironic, given its origin. According to the author’s notes, it was created to get away from the mistakes of a previous project that he describes as “a ridiculous concoction of self-indulgent, games-are-art-theory-innovation wankery” by getting back to basics. But of course the basics are art. Like those Grecian urns that Keats liked so much, it’s an art born of human requirements. Theory is all very well, but its importance can be overstated.

Well, I’ve managed a least-effort pass through every level in Crayon Physics Deluxe. Overall, I find the balance a bit strange: towards the end, most levels were solvable using the same few techniques. For example, I got a lot of mileage out of creating a large weight attached to a stationary structure (that is, one attached to permanent scenery by two pins), then draping a loop of rope, with both ends attached to the weight, over an obstacle and around the ball, then erasing the stationary structure and letting the weight pull the ball upward and to its goal. My first few attempts at doing things similar to this were more complicated and involved more elements, so if nothing else, repetition helped me refine the technique. Still, the number of places where this was applicable makes me wonder if I was expected to come up with easier but less general solutions first.

In my first session, I had used my usual gaming trackball for input, but the second time around, I came to my senses and hooked up the Wacom tablet. It took a bit to get used to the feel of it — I don’t use it very often — but it’s definitely worth the switch (assuming you already have a tablet, of course). Not only does it make it far easier to draw straight diagonal lines, it’s also more mimetic in this context: a stylus, obviously, handles a lot more like a crayon than a mouse or trackball does. I can imagine a world in which tablets of this sort are included with Crayon Physics as a Rock Band-esque custom controller.

Anyway, I’m putting this aside for the moment, because I’m eager to try other new things, but I intend to get back to it before long. There’s still that final spot on the map to be unlocked before I can really feel like I’m done, but that looks like it’ll be a fairly long haul: unlocking it requires Elegant, Old-School, and Awesome solutions for fully half of the game’s levels, and on a lot of the levels that doesn’t even look theoretically possible. My next post on this game will probably describe the physical mechanisms in detail.

The recent announcement of Zuma‘s sequel has also got me thinking about what a peculiar notion that is. To my generation’s grandparents, talking about a sequel to a game makes no more sense than talking about a sequel to a cheese or a hatstand. But this is because the word “game” has shifted in meaning. Modern videogames, unlike traditional games, are a narrative medium. A game can have a plot, and that’s all a sequel needs.

But does this really apply to Zuma? It has a plot of sorts, but it’s paper-thin, and about as important to the experience of the game as the inter-level animations in Pac-Man (you know, the ones where the ghosts chase Pac-Man across the screen and then he chases them back). There isn’t a trace of a story until you get past the third chamber, at which point you get some nonsense about an ancient prophecy and some sort of spirit guardian. (For some reason, prophecies seem to figure into videogame plots a lot, much like abducted princesses. Perhaps because inescapable fate is an easy excuse for constraining the player’s actions.) My point is, no one can seriously be expecting that anyone will buy Zuma’s Revenge to find out what happens next.

Rather, people will buy it for whatever new gameplay and graphics it provides. But even those can’t be too new. It looks like they’re keeping the fundamentals intact, which means whatever innovations the new one provides will be the sort that, under a different business model, would be provided as an upgrade to the original. The fact that PopCap is releasing them as something they call a sequel, rather than as an expansion pack, or even just an incremented version number (as a freeware game like Nethack would), is clearly a business decision, not an artistic one. PopCap knows how to sell new titles, not updates to old ones.

It’s been over a month since my last workplace Team Fortress 2 session, but we finally managed another one. We ran a private server with only six players, three on each team. Over the weeks since the last session, there’s been some consideration among this group of what game mode to use for small-team play. In the last update, Valve gave us the answer: King of the Hill.

TF2 has several modes based on capturing “control points”, which you do by standing near them for a period of time. (As I understand it, each player within range exerts influence on the control point, pushing it towards ownership by one team or the other. Once it’s pushed all the way to being owned by one team, it remains owned until the opposing team pushes it all the way back.) King of the Hill mode is such a mode, but with only one control point in the center — a variant simple enough that it’s surprising that it took them this long to add it. It’s good for small groups because it concentrates everyone’s attention on a small part of the map. Not necessarily their physical presence, mind you — a Sniper can still stand a long distance away and affect the battle, as one player proved.

I started this session playing a Soldier, the class armed with a rocket launcher, on the basis that the blasts, even when nonfatal, could push people off the hill, as it were. This turned out to not work: the control point’s range is large enough for people to dodge rockets without leaving it. The Soldier was still pretty effective, mind you — I’m told that picking it is never a mistake, regardless of the map. Still, I switched to the flamethrower-armed Pyro after a while, deeming its hard-to-avoid spread of flame a good way to clear the hill of interlopers. It seems to me that a team composed entirely of Pyros and Snipers could do pretty well on these maps. But what to I know? I’m still a beginner at this game.

Continuing into the Steam indie pack, I first made some failed attempts at getting The Path to work on my machine — I’ll have to return to that one later, perhaps after my next system upgrade. Giving that up, I proceeded to Crayon Physics Deluxe, a physics-puzzler based on the novel idea of letting the player add arbitrary objects to the scene by drawing them with the mouse pointer, and deducing their physical attributes from how and where they were drawn. (I described Blueberry Garden as having a hand-drawn look, but allowing the player to draw things really takes that idea to a new level.) I remember trying the prototype of this, Crayon Physics sans-deluxe, when it was new. It wasn’t clear to me then how the simple mechanics could be extended to interesting puzzles. It seems to mainly manage it by adding more types of things you can draw. For example, at one pivotal point, your repertoire expands to include pivot points — drawing a small circle inside another closed object provides a nub that forms the basis for pendulums or levers.

As is fairly common in level-based puzzle games, CPD doesn’t require the player to complete every level to advance. Levels come in batches (depicted as islands on a map), and each batch is unlocked by obtaining a certain number of stars. Each level can yield two stars. The first is granted for simply beating the level — the level’s goal is marked with a star to make this clear. ¹Actually, I’ve seen a few levels that have two stars in them, marking two goals. Beating such a level requires hitting both stars. It seems like you still get only one star for beating it, though. The second star requires the player to meet three challenges: an Elegant solution (meaning you only draw one thing), an Old-School solution (no pivot points — that is, a solution that would work in the original Crayon Physics), and an Awesome solution. The Awesome solution is of particular interest, because it’s done on the honor system: players are expected to decide for themselves whether or not a particular solution is awesome.

The Awesome criterion is a peculiar, and somewhat controversial, choice for a puzzle game. Generally speaking, puzzle-solving involves a mode of mind in which all you’re trying to do is satisfy the puzzle’s constraints. Critical evaluation of your own work on the basis of its aesthetic merits isn’t part of the puzzle-solving mindset; it’s more part of the Artist or Designer mindset. So the author of CPD is essentially trying to break us out of puzzle-solving mode and put us in a more creative frame of mind. Some people don’t see the point of this, or perhaps see the point but resent it, preferring to stay in puzzle-solving mode, do the minimum, and just arbitrarily check the Awesome checkbox whenever they need to. But such people miss the point of the game. I say that with some confidence: the author has made statements about the game’s point, and why he added the Awesome criterion in the first place. It’s basically because he saw people doing the bare minimum to complete the puzzles and wanted a way to encourage them to do otherwise.

Personally, I’ve never met an honor system I didn’t want to honor. And that applies even moreso in games, where cheating just means cheating yourself out of the satisfaction of winning honestly. Completing this game requires more stars than there are levels, so if I want to get it off the Stack, I’ll have to perpetrate some Awesomeness. I haven’t marked anything as Awesome yet, but even so, I’m finding that my approach to solutions is colored by the mere anticipation of it. Cheap answers just don’t satisfy.

So, no sooner do I make a move away from Zuma than PopCap unexpectedly announces a sequel. I take this as a sign from the gods of gaming and synchronicity that they don’t want me to abandon it just yet, and who am I to argue with gods?

If there’s one thing that really separates this game from PopCap’s more recent titles, it’s that its Adventure Mode is actually hard. Of the other PopCap games that I’ve purchased since starting this blog — Peggle, Bookworm Adventures, PvZ — I completed two of their campaigns in a single day, and the other in three. I don’t know if I’ll ever manage to finish Zuma. I’ve been stuck at the same point for several posts now.

And yet, the point where I’m stuck is so very close to the end! This is the point where one wants to just charge through and finish the thing, but that’s exactly the wrong approach to take to a game like this. It asks the player to be patient, to keep practicing without any measure of progress — no virtual cash or experience points building up from session to session. I wonder if the sequel will add such things? It could be a significant way to extend the game’s appeal to magpies like me, but at the cost of changing the game’s character in ways that fans of the original might not like.

So, last weekend Steam had a sale on a big package of indie games. Some of them I had already played, but enough of them were of interest to me that I had to snatch it up. And since it looks like I’m not finishing my last game any time soon, I might as well dig into them now.

First on the docket: Erik Svedang’s Blueberry Garden, a short ¹Steam tells me that it took me “0.9 hrs” to play it to completion. 2D platformer in a charming hand-drawn style. I’m not the first to make this comparison, but: its closest relative is probably Knytt, in that it’s a quiet game, a platformer based on exploration rather than combat, where a large part of its appeal is simply observing the art of the highly open landscape and the weird creatures that inhabit it. But more than that, Blueberry Garden is about figuring out how a world works, without aid of instruction or exposition. Although the world model and controls are platformer material, it’s got the heart of an adventure game. One of the weirder ones, like Myst or For a Change.

Or, for that matter, like Windosill, which it also resembles in its initial price point. It’s listed on Steam for under $5 US, and not because it’s old. Again like Windosill, it’s short and arty, which puts it in the same niche as a hundred free browser-based Flash games. It’s priced accordingly, but some consider even that pittance too much.

And now that the vague generalities are out of the way, it’s time for spoilers.

Given the game’s tranquil, exploratory atmosphere, it comes as a bit of a surprise that there’s a time limit. The main goal throughout most of the game, it turns out, is to save the garden from flooding by turning off a large faucet. Not that I understood this at first — on my first go, I squandered precious time just noodling around. The thing is, the faucet is the first thing that the game shows you, but without context, it’s hard to know what to make of it, and consequently easy to forget about it. On my first sally, I eventually noticed that the water level had risen to the point where I was wading whenever I revisited my starting point, and I wondered why. My best guess was that it was something I had done, perhaps the combined weight of the large items I had stacked up putting pressure on the wrong spot. (The stack of large objects is essential for reaching otherwise-inaccessible locations.) Only on my second try, after failing the first, did I see that faucet with enough information to grasp its importance.

The game’s victory screen contains a URL where you can leave comments. There, I discovered that my experience of the game was actually pretty common. Is it what the author intended? Probably not; I imagine that the introductory scene of the faucet was intended to convey information on first viewing. Not necessarily to make the player immediately say “Aha, that faucet must be a flood threat”, but to make the player say, on discovering the flooding later, “Aha, this must be because of the faucet”. If I’m wrong about this, and my experience was the intended one, I have to say it’s a masterful touch. It all but guarantees that the player will see both the good and bad endings — and, giving the generosity of the time limit, the player will likely see the bad ending only once. It also breaks the game into two pieces, one before you come to understand the threat and one after. These acts are very different in tone, despite the fact that the game doesn’t actually change at all between them.