On Saturday I attended the GameCamp 08 event organised by The Guardian. This was a meeting of a couple of hundred people with an interest and ideas about games of all kinds, modeled after the BarCamp series of gatherings, or in my frame of reference, like a larger-scale, greatly more informal and parallelised version of Dorkbot.

Being over-tired and under-prepared, I gravitated towards presentations by people I’d heard of for the most part, and away from things to do with table-top gaming and ARGs. (I’m not convinced that ARGs - essentially very weakly interactive marketing tools - have any place in an event about games, but as the day had been partly organised by an ARG company, there they were.) With talks going on in seven rooms at once, it was impossible to see everything, but I regret that I didn’t see some of the more esoteric talks (the controller hacking sounded intriguing) or anything focussed on MMOs.

I had originally planned to give a little post-mortem talk about what I’ve been doing in the mobile games sector for the past couple of years. The theme and message was that the mobile games industry had become preoccupied with fighting the ’symptoms’ (complaining about device fragmentation, fiddly controls, failure to address the casual audience, etc.) rather than tackling the underlying ‘disease’ (the fact that mobile games companies do vastly less to gain the trust and participation of their audience than their counterparts on any other format, when the nature of the platform and distribution method demands that they do much more). But it quickly became apparent that what I’d prepared would perhaps be a bit too specialised for the general audience.

Oh alright, I bottled it.

Highlights of the day for me included Ste Curran’s talk on games as a facilitator for shared experiences (which makes it sound a lot dryer than the set of funny and engaging vignettes it actually was), the afternoon’s rather freeform chat touching on Flash development, Ukranians, Peter Molyneux and Twitter, and the closing meeting of the ‘People’s Revolutionary Committee’ where many gaming bugbears (including console exclusives, tutorial levels, and Twitter, again) were condemned to death by firing squad.

All in all, it was a great opportunity to meet some very bright and games-obsessed people, and I hope we don’t have to wait a year for the next one.

My photos are here, and various attendees’ photos are here.

In a break from our usual programming, I made a game:

“Scrumper is a game that allows the player to engage in the ‘victimless’ crime of apple theft from the comfort of their own home. The object of the game is to catch as many falling apples as possible.”

It has already received glowing testimonials from beta testers:

“it actually seems fun now”

“are you sure it’s not virus?”

You can download the game (Windows only at present) from here and further explanation is here.

Judging by the traffic and feedback that I got from the earlier Games for Windows 2000 article (where I collected together workarounds that had been found to allow various supposed “Windows XP only” games to run on Windows 2000) it would seem that there’s still a small but dedicated userbase for this venerable OS among PC gamers.

In the intervening months, two industrious Windows 2000 stalwarts going by the handles of OldBoy2k and OldCigarette have taken up the gauntlet to fix every artificially incompatible game. The results of their efforts so far are catalogued on their Windows 2000 Gaming forum.

Most intriguingly, OldCigarette has developed a collection of API wrappers which goes some way to providing a general purpose solution for current and future games afflicted with these problems. In theory, the Windows 2000-using gamer need no longer rely on the good grace of the community to develop a workaround for a specific game - with this toolkit all they have to do is note the error messages produced when they try to run the game and simply drop relevant DLLs into the game’s working directory.

(There’s slightly more mucking about than that required initially - copying some files, a registry tweak and a reboot - but once it’s set up once configuration for subsequent games is minimal.)

I’ve so far successfully employed this box of tricks to play Stranglehold. Next on the list is Company of Heroes: Opposing Fronts, assuming I can circumvent the utterly deranged battery of booby traps that Relic are passing off as an installer program. But rare exceptions aside, I certainly have more confidence in buying Games for Windows (XP/Vista only) branded games now that I know that I can run them. Take that, inevitably rising tide of progress!

In the last couple of years, the line between PC and console gaming has been (in some respects) almost completely erased. The simultaneous release of high profile titles on PC, Xbox 360 and PS3 is becoming the norm. It’s easy to forget that before the mid-1990s, computer and console gaming were completely different worlds - two hobbies running in parallel with very little crossover between them.

While console gamers gawped at Starfox, PC gamers chuckled to themselves and went back to X-Wing. We’d … seen things you people wouldn’t believe. Of course then the Playstation came along and all the fascinating things people had been doing under DOS got swept aside, but for a few short years PC gamers had a legitimate reason to be smug.

Today, the vast majority of development projects are railroaded into long-established genres, known quantities where schedules and budgets can be (usually over-optimistically) drawn up at the outset. Back in the early 1990s, developers and publishers seemed to have no such play book to work from, resulting in a raft of games that were staggeringly ambitious and seemed to have no obvious precedents - games which would be almost inconceivable as commercial ventures today.

Some such games caught the gaming public’s imagination, allowing their creators (such as Sensible, Origin, Bullfrog, or the Bitmap Brothers) to found dynasties and ensuring their games are still widely remembered. Other games (such as Alone in the Dark) didn’t stay the course, but inspired later, more mainstream games (Resident Evil) to secure their place in history as a footnote.

There were still other, equally fascinating games that haven’t survived into posterity.

One such game was Stunt Island, developed by The Assembly Line and published by Disney Software for the PC in 1992. Stunt Island was a game that had the odds for historical recognition stacked against it from the outset. It shipped on six floppies bundled with a 180-page manual and a poster map, pushing its retail price towards £49.99, not a reasonable price for any game, then or now. As far as I’m aware it never saw a budget or CD-ROM re-release, and its status as Disney’s intellectual property has frozen any chances of reviving the franchise stiffer than Walt himself.

As one of the lucky/spendthrift few who did play it the first time around, I still regard it as one of the most impressive pieces of software I’ve ever seen, as well as the main reason I’m so disappointed when each new GTA game fails to implement a demo recording/editing mode.

Stunt Island was the brainchild of Adrian Stephens, a maths guru who had previously worked on the cult Amiga and ST cyberpunk puzzle-cum-shooter thing Interphase (and subsequently worked on projects as diverse as Comix Zone, Sonic X-treme, Vigilante 8, and most recently the True Crime series of GTA-alikes for Activision).

The game operates on two distinct levels. The core of the game is a flight simulator, which allows the player to fly one of 48 Gouraud-shaded aircraft around the titular Island, and to try their hand at flying 32 predefined stunt missions (storming a barn, flying under bridges, parachuting off a building and other such Hollywood action staples). More intriguingly, the game also provides what is essentially a complete virtual studio for producing machinima (eight years before the term was coined).

Creating an original movie in Stunt Island comprises of three main stages (which can be iterated through multiple times depending on the scope of the film and the number of scenes required).

The first port of call is the extremely flexible Set Design mode, which allows the placement of props, vehicles, cameras and trigger objects anywhere on the island, and even allows for some simple scripting to trigger events, switch cameras and direct the movement of entities during the action phase. All of the game’s pre-canned stunt missions were created using this tool, and while it’s slightly cumbersome by modern UI standards, it’s still possible with a bit of patience to build fairly complex scenes.

Once the scene has been set up, the player/user must then ‘fly’ the stunt. (The game’s terminology is heavily geared towards aircraft-based stunts, but in actuality the player’s controls can be bound to any vehicle or prop. They can even just park out of shot and let the cameras capture the scripted actions that they’ve programmed in the Set Design mode.)

Finally, the footage from each of the (1-8) cameras recording the scene is fed into the Editing Suite, where it can be spliced together into a final cut, and where a soundtrack and visual effects (fades, titles, etc.) can be added. The game shipped with an external player program to allow films to be distributed to people without the full game, and there is evidence that this subsequently happened on a small scale (on AOL and Compuserve mainly, as the game predates the popularisation of the World Wide Web).

Following the movie making process from conception to completion demanded quite a lot of effort on the part of the user, but it was possible to achieve good results, and the process itself taught a lot about movie editing and game scripting. Although most of my own efforts have long since succumbed to hard disk failure, a few Stunt Island epics have latterly made their way onto YouTube. (Unfortunately most of these are rubbish, so you’ll just have to take my word for it that it was possible to make something vaguely watchable.)

Stunt Island was able to realise its extraordinarily ambitious brief thanks to two things. The first was Stephens’ very robust engine technology, which allowed the game to render the (massive) game world as one contiguous map, through the use of a rudimentary Level of Detail system. Interesting scenes in the world were rendered in more detail (for instance zooming in on a city would pop up individual roads and buildings).

The accompanying map listed the coordinates of dozens of places of interest that were tucked away (including the Golden Gate Bridge, Alcatraz island, St. Andrew’s Castle, Stonehenge, farms, towns, army bases, railroads and aqueducts). Of course by modern standards it looks like how Google Earth might look in the world of Dire Straits’ Money for Nothing video, but at the time it was an impressive technical feat.

The second enabler was that the developers very cleverly limited the scope of what would be simulated within the world. Planes, simple landscapes and military hardware had been the stock in trade of 3D computer games for many years; rendering convincing people (to say nothing of trying to animate them) was a non-starter. Basing the game around traditional Hollywood stunt work (going so far as having quite a lot of in-depth background material and an interview with a Stunt Coordinator in the manual) explained away these limitations quite well.

Very few games have followed in Stunt Island’s footsteps. The dependency on permanent storage and a (cursor-driven) editing mode has until recently been a major obstacle to the development of similar games on consoles. In the PC gaming space, the emphasis on ever-increasing graphical realism and diminishing consumer tastes for games that demand extensive learning and time investment have had a similar effect.

The only game in recent memory even slightly similar has been Lionhead’s The Movies, which rather awkwardly married Bullfrog’s tried and tested Theme Park/Hospital business management formula to the popularity of The Sims series (both as a game and a machinima-creation platform) and sank without trace.

In the wake of the Nintendo DS and Wii, and the rise of the various online distribution platforms (e.g. Steam, Xbox Live Arcade, Playstation Network), there seems to be a more positive attitude to experimentation throughout the industry of late. I believe that in this climate, a canny developer could reinvigorate the “virtual movie studio” concept. Nintendo’s Mii system (as used to great effect in Wii Sports, Wii Play and the upcoming Mario Kart Wii) has shown that gamers aren’t obsessed with photorealism. A “Mii Movie Studio”, supported by a channel on the dashboard (where all Wii users could view and give feedback on user-generated movies), could be huge.

Failing that, there’s always the chance that Rockstar will finally do the right thing with GTA4. We can but dream.

Earlier this month, Lindesay Irvine made a post on the Guardian’s books blog bemoaning the news that thriller writer James Patterson was collaborating with Oberon Media on a casual game.

Irvine’s confusion and apprehension at this specific instance of a writer crossing over into a new medium (he seemed, perhaps understandably, to be unaware of the casual PC games market that has been courting Patterson’s middle-aged female demographic for the past few years) quickly segued into a highly defensive rubbishing of games as a whole. He scoffingly suggests games we can expect from other literary figures, including a “Martin Amis first-person shooter”. (Presumably Irvine is unaware that Amis wrote a book on video games over 25 years ago.)

Setting aside the ill-informed bluster, the kernel of Irvine’s argument is mixing imaginative writing with any other form of media serves to dilute it. I have to admit that the bulk of the evidence is on his side. What works on the page may not transfer to the stage, screen, or even audio recording successfully without major rework acknowledging the conventions of the medium. It’s easy to see why someone who has a narrow and outmoded view of games would have difficulty believing that anything of worth could be carried over.

Alastair Harper promptly penned a response which tried to set Irvine straight and make a case for increased involvement of established writers in games. He made the insightful point that storytelling in games allows things that wouldn’t be possible in other media (giving Bioshock’s big reveal as an example). He waxed lyrical about point and click adventures but didn’t touch on the interactive fiction genre, which would have furnished still more examples (from Floyd’s death onward).

Irvine rather sportingly commented back, conceding that maybe there was something to these games after all (and rather quaintly using the term “games arcade”, as well as for some reason feeling the need to resort to a straw man argument about games replacing books in the cultural landscape).

I personally agree with Harper’s position that there need to be more professional writers involved in games development. It’s alarming that games with multi-million dollar budgets (often including a large chunk set aside for professional voice work) still frequently treat writing as an afterthought. As a result, writing for games limps behind the advances made in graphics, animation and music, where it has long been recognised that trained professionals are required.

The embarrassingly witless script of The Elder Scrolls: Oblivion is a good recent example, jarring terribly not only with the obvious effort and attention that has gone into the audiovisual elements of the game, but also with the comparatively high quality of the writing on display in other PC RPGs going back decades.

In my own experience I’ve seen game scripts which the developers (who I’ll spare the embarrassment of naming) have insisted were contracted in from “professional Hollywood screenwriters” which have turned out to be semi-literate gibberish, obviously the work of writers who have been able to dupe the developers (not native English speakers) into thinking they’re the genuine article.

On the other side of the coin are games like Portal and the Mario & Luigi GBA and DS RPGs, which prove that good writing can be a defining characteristic of a game, and that the audience are perfectly capable of appreciating the extra effort and generating good word of mouth for the game as a result.

Assuming that a developer has recognised the value of good writing (both to help communicate a cohesive vision for the game during development, and to do the same for the audience in the finished product) this then poses the question of how writing should be used. It’s increasingly common (especially with the rise of consoles as the lead platform for more traditionally literate games) for writers to work purely on dialogue. Too many developers seem to be afraid of asking gamers read even brief amounts of text to be able to understand the game and the story.

Then there’s the knotty problem of storytelling. There is clearly a place for games that try to interlard portions of a traditional story into the gameplay, hence the mega-success of the likes of Half-Life, Max Payne and Bioshock. The prospect of having a story structured in this way is familiar proposition for the player. However I think that increasingly the place where writers will be employed most effectively in games development will be in the creation of worlds, characters and back story which can then be used as building blocks for the creation of new stories through the actions of the player(s) and the AI.

More thoughts on this (maybe) to follow.

I played through the single player campaign of Call of Duty 4 a few weeks ago. I’m not sure what I can add to that statement, as judging by the sales figures, most of you will have also played it and formed your own opinions already. Shifting over seven million copies in a few short weeks is nothing short of phenomenal, and the game will surely add a couple of million more to that total in budget and Game of the Year re-release incarnations.

Activision and Infinity Ward judged the market perfectly, managing to recapture the ‘extended audience’ that turned the Deer Hunter franchise into a $100m industry at the turn of the century. Which isn’t to say that the game is aimed squarely at jingoistic Americans, it’s just very idiot-inclusive in its design. It’s the gaming equivalent of a Summer blockbuster movie, and guaranteed to spawn many imitators.

The game is the latest in a long lineage of sanitised Hollywood-indebted war games, harking back to Combat, Ikari Warriors, Operation Wolf and Desert Strike more than its more immediate predecessors (the Medal of Honor, Brothers in Arms, and Call of Duty series et al, with their Band of Brothers/Saving Private Ryan inspired aspirations to historical reverence), and a million miles from the scant few attempts to portray the ugly realities of war such as Operation Flashpoint, Hidden and Dangerous and (oh, alright then) Cannon Fodder.

I’ve never quite been able to reconcile CoD’s subgenre of FPS games, the rigidly linear, AI-less shooting gallery where everything is boiled down to moving between pieces of cover towards checkpoints and repeating the process as increasingly elaborate setpieces are triggered around you. (The game rather lamely pretends to be a simulation during a tutorial which demonstrates some ‘authentic’ military procedure which you subsequently never have to use.)

The reviews I’ve read make it clear that the sheer spectacle is supposed to compensate for the narrow range of interactivity. The annoying thing is that it sort of does. About 80% of the game goes past in a blur of dramatically anesthetised mayhem, leaving no lasting memory but engaging at a superficial level during play, but it sure is pretty.

The game’s standout sequences are ring-fenced away from the waves of AK-47 wielding stereotypical cannon fodder. One level presents a visually extraordinary flashback to early-1990s Chernobyl (which looks rather nicer than S.T.A.L.K.E.R.’s version, albeit at about one millionth of the size and complexity) which makes a commendable leap across the Uncanny Valley through the use of desaturated colour, subtle shading and the masterstroke of obscuring your NPC companion’s face, building the character purely through motion capture and voice work. Elsewhere, the US Marine Corps story thread is terminated with a sequence that can be interpreted as either poignantly heroic or depressingly futile, depending on your viewpoint.

The game’s signature level (”Death from Above”) casts the player as the thermal camera operator in a AC-130 Spectre, the culmination of the thought processes that were set in motion by footage of the first Gulf War, which was often likened to a video game at the time. It’s head and shoulders the most brazenly tasteless level in the game, but at the same time one of the most compelling to play, giving a little taste of the ridiculous godlike power that such weapons afford.

I expected the tone of the game to be crass and jingoistic (in the Tom Clancy mould), but this wasn’t quite the case. Unfortunately the view of modern warfare that the game presents, in its reticence to question or explore the consequences of the process, is still a rather dubious one. After the credit sequence, there are effectively no civilians in the game; everything that moves is a target. Your AI squadmates don’t get angry, scared or confused, they’re all rugged, jargon-barking killer robots. The SAS characters are given a bit of personality, but this mostly manifests itself as sarcastic comments and cheeky cockney swearing. It’s also stupidly easy, allowing the player to heal within seconds, giving them vast amounts of ammo and the enemies no sense of self-preservation. Getting caught on the scenery is the most frequent cause of death, followed by dogs.

It’s a bit worrying that the game has just suggested to seven million people that war is a bit of a laugh, but perhaps that’s straying into hand-wringing, point-missing “if only you could talk to the monsters” EDGE territory. Playing at war remains excellent fun.

Another of the highlights of 2007 that I’ve only just gotten around to playing is Irrational’s Bioshock, a game that has already been ludicrously hyped, critically feted, endlessly discussed, lashed back against and had its backlash lashed back at.

As a game, Bioshock isn’t as good as the best games-of-vaguely-comparable-genre out there (e.g. Deus Ex, the Metroid Prime trilogy - I’ll admit I’ve only ever dabbled with System Shock I and II). As a story, it was probably as good as anything that’s been done in the medium so far, although one that strained awkwardly to stretch a few brilliant high-concept narrative ideas across a standard length PC game.

I think that the astronomically high review scores that the game received can be put down to a mixture of outdated expectations (it shouldn’t be a surprise for a game of this type to have as good or better presentation than Oblivion and Gears of War, at this point), a lack of directly comparable games in recent memory (and a long drought of genuinely creatively interesting games, which affected the PC and 360 worse than most), and a widespread unconscious urge for there to be a ‘landmark’ title to decisively announce that the ‘next generation’ had arrived. In short, it was the right game at the right time. Which isn’t to say that it’s undeserving of praise, but by my reckoning it’s a solid 8/10 rather than the 10/10 that a lot of places gave it.

Bioshock’s setting and central set of characters and concepts are genuinely compelling and original. Rapture is a suitably awe-inspiring monument to Andrew Ryan’s will, and the macabre ecosystem that has developed in the ruins provides fertile ground for many evocative moments and scenes. Everything is bent to the purpose of generating the maximum amount of immersion and ‘ooky-ness’.

Even the more clichéd elements it draws on are given a little twist to make them seem fresh again (the spooky little girls aren’t just dumbly aping Asian horror iconography; the dark, claustrophobic, corridor-based world dodges criticism by having water outside the indestructible windows instead of a Martian landscape). And as the gushing reviews have pointed out, the art style is phenomenal (although I still think that the decision to make everyone - Splicer or not - look like badly melted ventriloquist dummies lets the side down a bit). The rendition of water is the best yet seen in a game, taking the crown from Super Mario Sunshine.

Unfortunately the effort put into selling this unfamiliar imaginative premise seems to have limited the amount of resources left to populate Rapture with fleshed-out characters. There is little in the game that isn’t linked directly into the main plot (the rare exceptions often being among the most affecting parts of the game, perplexingly). Many of the environments do not feel like they have been lived in or served a functional purpose. Unlike Deus Ex and it’s lineage, Bioshock often seems to think that showing without telling is enough to suspend disbelief. The decision to have virtually no text in the game that isn’t presented as audio logs exacerbates the problem.

Holes in the logic of the game world are waved away with ‘funny’ audio logs a few too many times for comfort as well.

On the macro level, the fact that the game leaves so many issues unanswered and open for interpretation works in it’s favour, in the mode of Half-Life, Lost and Twin Peaks.

The Big Moral Choice underpinning the story is nicely done, and presumably only came as a letdown to people expecting some kind of Molyneux-esque implementation of morality as a game system. (I didn’t harvest - by a strange quirk of fate, I was hacking up defenceless kids in Ultima VIII when news of the Dunblane massacre broke on the radio, which took the fun out of virtual child murder for me.)

Because the story is so obviously the point of Bioshock, a lot of reviews seem to have glossed over how (and how well) the game mechanic works.

In a nutshell it works like this: the player traverses a series of large levels (’decks’), each of which is a free-roaming area where several objectives need to be completed to progress. Instead of scripted combat scenarios, puzzles, bosses, exploration, etc., most of the game’s action is based around random spawns of generic enemies (Splicers) whose health and abilities are ratcheted up as the game draws on, and the odd security turret. The game sporadically respawns some enemies into previously visited areas to keep you on your toes (although never in overwhelming numbers).

There are also a few Big Daddy/Little Sister duos wandering each level. Big Daddies don’t attack unless provoked, and can be optionally engaged so that the player can get to the Little Sisters (who give you the resource to buy new plasmids a.k.a. force powers). It’s possible to coerce any of these actors to attack the others. With a few plot-driven exceptions, it’s not necessary for the player to kill everything they encounter.

This system works, and is quite well balanced for most of the game. However it just works. It doesn’t produce moments of exciting emergent gameplay like your Halos and Grand Theft Autos. There aren’t multiple ways to approach situations as there are in Deus Ex or Resident Evil 4, or rather, the different approaches you can take tend to blur into an indistinct mass, delivered in the same way and resulting in the same outcome.

Few reviews have mentioned how pared down the game’s environments are. It wouldn’t be much of an exaggeration to say that Bioshock is the flattest game world since Wolfenstein 3D. The game’s maps illustrate this well - all easily represented in 2D and chock full of right angles, as if Rapture has been designed for a pen and paper RPG scenario rather than a computer game. There’s little concept of cover, few open areas, no platforming and certainly no mutability in the game world. There are still a few minor puzzles, but it’s clear that the graphical gloss comes at the cost of interactivity. I expected the game to make more use of the fact that you’re in a pressurised container at the bottom of the Atlantic (flooding or draining areas, floating platforms, or something), but it’s hard to complain that it’s ‘just’ used for spectacle when it fills that role so well.

Hunting the Big Daddies is a bit of a let down, considering their focal role in the game. They’re basically bullet sponges with limited AI. For the first few levels of the game there isn’t any effective strategic way to fight them beyond emptying clips into them and running like hell (and more often than not respawning repeatedly). The level design doesn’t help here either - there’s no way to have a cat and mouse running battle, because there’s nowhere to hide from the Bouncer’s nearly impossible to dodge drill charge attack. Once the player has the chemical thrower and a few other gizmos Big Daddies represent no threat at all.

I would have preferred the Big Daddy battles to give the impression of fighting a massively stronger enemy, which could be cut down to size using indirect means (think Ewoks verses AT-STs). There are hindrances that the player can use such as oil puddles, proximity mines and trap bolts, but they’re fiddly and underpowered. I wanted to be able to lure them out onto thin ice, or weld them to the floor in the path of a train, or drop a load of scrap metal on them, or anything other than just having a shootout. When games like Resident Evil 4 essentially offer designers a bumper illustrated book of fun and interesting boss ideas, there’s no excuse for just sticking a diving helmet on a pinky demon from Doom.

Most of the combat in the game is uninspiring. The fiddlyness of switching weapons and ammo types and constantly reloading (magazine sizes for your weapons and plasmids are generally tiny) is an unwelcome distraction when fighting multiple enemies. I found that I was using the wrench, chemical thrower and grenades almost exclusively. The pistol and machine gun tend to burn through ammo and the shotgun is useless even when fully upgraded. The dustbin lid sized ‘crosshair’ (necessitated by the console version’s autoaim) makes the ranged weapons feel imprecise.

We then come to two ill-thought-out secondary interfaces which both serve as good examples of lazily following Irrational/Looking Glass conventions without stopping to consider if they’re suitable: hacking and the research camera. Both of these serve their purpose, but both are hopelessly outclassed by their equivalent solutions in Metroid Prime 3 Corruption: welding and the scan visor. The research camera gets an additional black mark for only being able to ’see’ what the game considers to be standard enemies. You can’t photograph boss characters, or incidental appearances of enemies (such as the Big Daddy walking around on the seabed).

My final gripe is more of a plea to the industry at large - do we really need to have security cameras and automated turrets in every game? They’re lazy and cheap to a degree that’s practically insulting to the player. And don’t get me started on the security bots. Let’s punish the player for walking in the wrong place by annoying them with noisy flashing Daikatana-esque flying enemies for a whole minute! Furthermore, in Bioshock none of these systems make a shred of sense in the context of the game setting.

While the narrative aspects of the game are unimpeachable, it’s a shame that the rest of the game falls between the action and adventure stools. It never really fully commits to either camp and feels a bit muted as a result. You don’t really do anything much more ambitious than opening doors and turning valves. By way of contrast, in Metroid Prime 3 (which, in a nice bit of symmetry, has great mechanics but an execrable narrative component) the player at one point visits a fantastically baroque steampunk version of Cloud City, in which they have to construct a giant bomb, which they then have to fly down to the surface of the planet, fighting off boarding space pirates and repairing the damaged escape capsule, against a time limit. (There are similarly memorable sequences scattered throughout the game.) Bioshock doesn’t have anything that works as well as a game as that, but then no game has ever had anything as good storywise as the pivotal scene at the Hephaestus.

For all it’s shortcomings, I still found myself completely immersed in Rapture for five hours at a time, so it must’ve been doing something right. It’s not really a rehash or a dumbing down of System Shock so much as an adaptation of System Shock’s qualities to the demands of a modern audience - not idiots, but not willing to put up with the clunkiness inherent in 1990s PC games. It’s not entirely successful in that aim, but it’s the right sort of thing for developers to be doing, rather than concentrating solely on games as vapid and generic as Halo 3, Crysis and their ilk.

2007 has been a banner year for games. Not the best year ever, but at least the best year of this century, with each format seeing their share of classics. Having gotten out of the habit of playing regularly, I’m currently scrambling to assimilate as many of this year’s ‘must play’ games as possible, so it’s likely that I’ll only get around to writing about most of them retrospectively.

In the meantime here’s some thoughts on what will surely come to be seen as the game of the year, Super Mario Galaxy.

Where to start? Well, firstly, it looks like this:

So presumably Robbie Bach is feeling a bit silly right now.

I had expected Super Mario Galaxy to be good, but I had also expected to have to apply the Nintendo blinkers which came free with Sunshine, Wind Waker and Twilight Princess to be able to overlook the inevitable convention-bound shortcomings, rough edges and annoyances.

I was surprised and relieved therefore to find that Galaxy is a proper flagship game for a console. In terms of levels and content it feels several times bigger than Super Mario Sunshine, and in terms of production values it’s the most assured game Nintendo have produced since Legend of Zelda: Ocarina of Time. The initial impression that the game gives is one of quality (even luxury), with a classy orchestrated soundtrack and an art style so clean and shiny you could eat your dinner off it.

Galaxy is essentially a massive toybox containing several hundred small, finely crafted gewgaws and puzzles. This description could kind-of-sort-of be applied to Super Marios 64 and Sunshine, but with Galaxy the little bursts of fun aren’t padded out with repetition, filler and experiments that with hindsight didn’t quite work. For the first half of the game, the rate at which new treats are thrown at the player is perhaps a little excessive, hardly leaving a moment to process the information or to pick out single elements for closer inspection.

The last 3D Mario game (Super Mario Sunshine) was heavily influenced by the structure of Grand Theft Auto III. At the time, making huge sprawling levels was considered to be the ideal that all 3D games should be striving for, even though in the context of a platform game this resulted in largely empty areas which involved much trekking around to get to the next place of interest. (In Sunshine’s case they were frequently broken or unfinished to boot.) Galaxy does away with this dependence on monolithic, thematically consistent areas, instead allowing the player to be bounced around self-contained plateaus and planetoids which the designers were able to develop independently and then compose together into complementary arrangements.

There’s so much stuff in Super Mario Galaxy that it would take me many thousands of words to catalogue it all in any depth. As well as transmuting the idea of a 3D platform game environment, the chaps in Tokyo haven’t hesitated to realise that Mario is essentially a distinctively coloured marker in 3D space, presenting the player with a plethora of new interfaces that have nothing to do with platforming, from slinging Mario around like a billiard ball, to surfing, swimming, skating, performing a variation of Super Monkey Ball (except motion controlled, and fun), and at least four distinct methods of flight. (There are of course many ‘trad’ 2D and 3D platforming areas to get stuck into as well.) The game doesn’t wear out any of these new mechanics, usually featuring at most two levels based purely around each one.

For a long time I’d felt that I’d grown out of platform games, lacking the patience and manual dexterity to plug away at them for hours for little reward. Galaxy makes me realise that this was more the fault of the games available in the genre up until now than of a lack of enthusiasm on my part. For the average player it’s a challenging game, but one that never feels unfair. Towards the end of the game there are levels that will require many attempts to beat, but in nearly all cases each attempt didn’t feel like a setback, but rather another chance to improve my skills at the task at hand. If I’m being completely honest there are perhaps half a dozen stars in the game that just aren’t fun to endlessly fail at - most of which uncharacteristically focus on the shortcomings of the controls, or feature needlessly protracted platforming sequences where a moment’s unwariness results in instant death and a return to the start.

Anyway, you’ve probably read the reviews and don’t need to be told just how well Nintendo can put together a platform game. There is one other extremely important part of the game which doesn’t have a precedent anywhere in the Mario franchise, and which many (boorish, American) reviewers have overlooked or even expressed disdain for. I am referring to the Storybook, a trail of small, unassuming scraps of narrative that the game casually leaves for the player as they progress. It tells the fairy tale story of new character Rosalina (okay, bit of a spoiler but you surely figured that out), a space-faring girl who has become the ‘mother’ of a race of cutesy star people and gives Mario his star-hopping ‘powers’ relevant to the game’s premise. It’s brilliantly done, and even a bit subversive in a weird, anodyne, Nintendo sort of way.

The Storybook is striking as it introduces human interest into a game which has unswervingly presented its main cast as simplistic cyphers for twenty years. It is brave in its form of representation, being told in simple text overlaid on a series of static watercolour images. It is affecting in that it tackles themes of death, displacement, responsibility and emotional growth. It’s not particularly subtle or original, but if you don’t get a bit of a lump in the throat at one or two points, you’re probably some kind of heartless monster who laughed at Watership Down. It adds a little bit of resonance to a game that’s mainly about a funny little man jumping on baddies, and I hope that the game’s director gets some positive feedback about it among the braying of jackasses who have been conditioned to see such a thing as a ‘lame bonus gallery’ that doesn’t ‘win’ you anything in the game.

Railing against philistines aside, we come to the question of whether Super Mario Galaxy is the best game ever. Personally, I don’t think it is. Underneath all the variations on the central theme, the game is specialised (meaning, it’s a platform game rather than a game which transcends genres), and playing in its world demands skill from the player almost continually, even though it’s typically fun skill rather than robotic learning. You can indulge in some freeform play in the game, but not to the extent of something like GTA or Crackdown. (It’s better than either, clearly, I’m just trying to convey the point that it’s not the only game you’ll ever need.) Also, the hub level, while suitably pretty, is a bit inconvenient in the later stages of the game when you want to channel-hop between galaxies, and unlike Sunshine, it doesn’t contain any goals/gameplay in itself. If we’re really nitpicking, the camera occasionally misbehaves as well (especially underwater).

A while ago I linked in passing an article by Stuart Campbell which argued that the term “videogame” (his contraction) was now being applied to such a broad range of genres and specialisations so as to be almost meaningless, and that the original focus of games - the joy of skillfully and creatively controlling an agent in an abstract environment that couldn’t be done in any other medium - was getting lost in the scrum as developers tried to ape cinema and strive for ever greater realism.

I think we need a term for this which isn’t as pejorative as ‘retro’ and doesn’t reference the useless and misleading ‘casual/hardcore’ distinction. I’d suggest Classical Gaming, and Super Mario Galaxy is as good an example as any of it being done right. Everyone should play it.

At some point I will have to try writing about a game I don’t think is amazing. There are some, honestly.

The whole situation is profoundly regrettable. I hope that there will be a way to salvage Gamespot’s reputation and working environment before it’s too late, but it will require everyone involved to act quickly and cohesively to restore order before the damage can be addressed. Unfortunately I suspect that the person(s) at CNet who found themselves so embattled and out of their depth that they thought it would be a good idea to fire Jeff would be likely to continue to act in the same cowardly fashion. I hope I’m wrong.

There is a handy round-up of past and present Gamespot editorial staff reactions to the news here. Regardless of your opinion of Gamespot (I personally think that they are still better than pretty much any other site trying to cover the same remit, but recognise that they are not infallible), the people caught in the crossfire deserve our support.

UPDATE (06/12/2007):

A second official statement from Gamespot offers some reassurance. I guess we’ll see over the coming days and weeks what the impact if any will be.

A couple of years ago I had the good fortune to meet Ron Gilbert at an informal “meet the fans” event in London. I asked if he’d ever thought about making the production notes and materials for the Monkey Island games available to the public. His reply (paraphrasing from memory) was:

“Arrr! But why? Who would be interested? So anyway, the secret-”

- which I have to say surprised me somewhat.

I’ve since found that this mindset is quite pervasive in the games industry. Attentions are always focussed on the next big thing, and once a game has shipped it’s assumed to be of no interest except as a historical curiosity. Any documentation made during the development process ends up getting trashed or mouldering away in a programmer’s bottom drawer somewhere. (Occasionally fragments do emerge - such as this excellent ‘Vision Statement’ by the designers of Planescape Torment.)

In many cases access to these materials would answer a lot of players’ questions, ranging from geeky points of trivia about characters and plot up to the rationale for major design decisions. Canny developers would also stand to benefit from swotting up - attempts to cash in on the success recapture the magic of well-loved games and genres would be improved if they sought only to retain the features and conventions that enrich the game, rather than the ones that were imposed by technical limitations or a lack or time.

It would also help to explode a lot of myths, especially among obsessive fans who assume that because a game has the name of a popular franchise on the cover, everyone involved in it’s development must have encyclopedic knowledge of (and unquestioning respect for) previous titles made years ago by people they may not have even met.

Part of the reason that the industry has developed such a cavalier attitude to historical preservation is a lack or resources to devote to a process which has no obvious commercial benefit. What’s needed is a repository for these materials maintained by parties who can dedicate the necessary time and effort to preserving and presenting them.

It’s quite common for universities to keep libraries of the notes and artwork created by authors, illustrators and film-makers, but until recently there didn’t seem to be any equivalent initiative for games developers. So I was excited to discover that earlier this year, the University of Texas announced that it was starting a Videogames Archive, with the involvement of Richard Garriott, Warren Spector and a raft of big-name developers and publishers.

Hopefully this archive will lead to similar efforts being undertaken in Europe and elsewhere, and will help to convince developers that there is some value in preserving and sharing these materials. In the mean time, you can donate (money or items - assuming you agree with the concept and how they’re going about implementing it, of course) to the University of Texas’s archive here.