So:

in the centre of your screen you have a crosshair. This is where your shots go. Now, on a PC, where you can aim with a mouse, you don't need any autoaim, because you can be very precise. But with a pair of thumbsticks, it's unfair to ask such accuracy of the player - especially when the sticks themselves may not be entirely accurate.

So you have a degree of autoaim; if it's pretty near the middle, it'll probably hit what you want it to.

Now, the other thing that modern semi-realistic games feature is a penalty for accuracy when you move and shoot. Counter-Strike managed to express this to the player by having a four-line crosshair. As the player moved, or sustained automatic fire, the bars widened. They returned more central when firing ceased and motion stopped. The widening target indicates the inaccuracy of the shot - the potential area where aim could go.

Rainbow Six 3 does things a bit differently. The target is a red dot, surrounded by a circle. The diameter of the circle indicates the area affected by auto-aim. As the player zooms in with his weapon, the red circle gets larger. The dot, incidentally, is always central. As the player moves, though, the red circle shrinks, until at a run, all that is left is the dot. So: as you move, the computer will compensate less for you, and your accuracy also decreases. It's quite effective; as long as you can get the target into your sight - which at long range can still be quite tricky - you can probably hit them. Shorter-range weapons have smaller diameter circles, and smaller zooms. It's a surprisingly effective system - it may only be effective on a console, where autoaim is permanently on, but it also removes the need for hyper-accurate input device, and insane thumbskills on the part of the player. It also, I guess, takes into account that the player is meant to be a highly trained soldier, not a half-blind rookie. It takes a while to get used to, but once you do, it's a particularly effective system, that allows for rapid target acquisition, and accurate-enough aiming.

There was a brief discussion piece on the Today programme today, about the popularity of online gaming and... well, I can't really tell what it was about. I think it was about whether or not it's a good idea for people to spend so much time in virtual worlds, or whether it was the decline of society... but it was all very poorly worded.

It started off well, because they had Susan Greenfield and Richard Bartle on to discuss it - big names, really significant. But referring to Bartle as "professor of games design at Essex University" essentially omits what he's done, why he's so important to the field, and why he got that title in the first place. There was a lot of confusion - we clearly weren't talking about computer games, we were talking about massively multiplayer online stuff, but that wasn't clearly defined. Greenfield made an interesting point - that one should consider being on the mobile phone or listening to music as pretty much the same to being online, because we are being distracted from the real input being fed to our senses. Despite John Humphreys' protestations.

The piece was four minutes long, almost went somewhere interesting, but could have done with being half an hour. Two great people to discuss a very interesting topic they both have great perspective on - and it was thrown away as a novelty piece, pretty much; the tone of the programme towards the two doctors appeared to be one of sneering disbelief; "we don't care what you're saying but we have to run this article anyway because the editor said so, so here we go...". Which was a shame. Close, then, but no great discussion, and not much I hadn't heard before.

When I read Tycho's curt dismissal of Vice City, it seemed a good time to put my thoughts about the game down on paper.

Tycho is dismissive of its popular appeal; ie, the fact that it is hugely popular with all gamers, especially casual ones, and probably in part because its structure lends itself to casual play. I recently read an article wherein casual gamers discussed new PS2 releases (it was, if you must know, on a mountain biking forum). Someone was rating XIII, but pointed out he'd nearly finished it in a couple of weeks, and Vice City had taken him a couple of months and - as in implicit assumption - Vice City must therefore be better.

Vice City is an interesting beast. It has a well constructed plot, that often has several strands running at once (depending on the order you chat to bosses). It has a more cinematic feel than GTA3, possibly best signalled by giving the player character a voice and role, and its plot features several heavy pop-culture references (to films and TV shows such as Taxi, Miami Vice, and Scarface). At the same time, though, it immediately encourages experimentation. Almost any car is free to jack. The bridges to the mainland (which accounts for about 50-60% of the physical area the game covers) are blocked until halfway through, but even so, there's enough stuff to play with. It's quite possible to load it up, play for an hour with no discernable goal, and the thrill of random violence might well appeal to many casual gamers.

(Also of note: GTA3 begins in the industrial district, with largely rubbish vehicles to steal. Vice City begins in the beach and leisure areas, with the quickest, neatest, most stylish vehicles - many sports cars and motorbikes. Later in the game, the "duller" areas open. In terms of balance, this is clever - because evading the cops in a family sedan is harder than in a Countach. Also, though, it plays up the "fun" factor more. In GTA3, you earn the fun stuff. In Vice City, the keys to the toybox are laid out in front of you).

Then, when you're bored of that, you could take one of the many plot missions. There are a certain set of missions that comprise the "main story", and it's possible to advance very far in these with very little extra experimentation and exploration. To get the final two, though, I believe the player is required to have a "60% complete" rating according to the stats screen.

Now, the completion percentage does not just relate to the essential plot missions; it also relates to all the bonus missions, to finding the 100 hidden packages around the city, to completing the 36 unique jumps and 20 rampages that lie scattered. The hidden packages are a prime example of the need for experimentation: some require some investigative footwork to find; others demand inch-perfect superbike jumps.

Oh, and you have to acquire all the property in the game to contribute to that 100% rating. Now, some property will give you income, and some just functions as extra safe houses, helipads and garages. The acquisition of property is quite a neat feature, because (as well as giving you more hideouts from the police, so you don't have to backtrack so far to save or evade the law), it also gives the player a feel of rising in status; I began in a beachside hotel, but now I've got a Condo and a mansion, two helicopters, and four garages to my name. I feel like a "bigger" criminal.

And that's before I mention the subgames - stealing a taxi will allow one to play taxi driver and earn money for as long as you want; it's like Crazy Taxi, basically, built into the game.

And so it's quite easy to get diverted from the main plot. All these little bonus features also reward "pick up and play"; through experimentation or simple errors, the player advances towards completion - not through persistence, skill or logic, the two attributes that feature strongly in more "hardcore" games.

Also, what is "complete?" I'd argue the game is complete essentially when the story is finished to the player's satisfaction. You can still drive around, mess around when you've got that real 100%; does it make one feel any better? It's unfair to deride other games for their shorter length when such an open-ended nature is not their goal. XIII is more stylish, more cinematic, and has a far better executed plot than Vice City every could have; it achieves its goals admirably. Length is not a guarantee of quality - Final Fantasy VII was returned by the dozen because it was just too turged for the 90s Playstation generation, hooked by Wipeout, to cope with. I feel a similar fate will befall Knights of the Old Republic. What Vice City has is a subtle balance: it is epic in scale but not in plot; it has variety on its side. It's also a long game full of short sequences, rather than a long game full of long, drawn-out missions where one slip spells game over. I can't think of a mission yet that's lasted more than about 15-20 minutes at best. So rather than being properly "long", it just consists of more short-things stuck together than most games.

Don't get me wrong; I love the open-endedness; I love the fact that the game rewards having fun; I also love the fact that, depending on the time I have, I can either have a serious go at plot missions, or just try and nail one more jump. Or maybe find something new I never knew about. At the same time, it's almost too pick-up-and-play. It gets the casual-gamer kudos because of this, and also because of its trademarks - recognizable licensed soundtrack; mature language/plot and graphic (if cartoonish) violence; glamorized, senseless violence. None of these make the game better - how many times do games have M- or 18-ratings that turn out to be their only unique feature? And, conversely, how many wonderful games are passed over because snobbish gamers complain they look too childish?

Tycho's point is fair; there is more to gaming than GTA. Indeed, there is more to the pinnacle of gaming than GTA. No question, Vice City is one of the best games I've played, but it comes into my top ten or twenty, not my top three. It just happens to be one of the few times brilliance has slapped the casual gamer - for whom the PS2 or Xbox is something to do before going out or in idle moments, not something to invest time into - in the face. No casual gamer would pick up Ico, Rez, Knights of the Old Republic and their ilk on a whim; too complicated, too surreal, where's the guns? And yet it's here that the real treats of gaming lie.

I can understand the casual praise for Vice City; I can understand the hardcore praise too. It's not a be all and end all, and it makes sacrifices (frame rate, graphics) to achieve its goals. To dismiss it from on high, when one has experienced its peers, its betters, is unfair; I've played many games few have and adored them, but that's to my benefit; I need not harp on about it being their loss. It's a stylish, brilliant crossover, and it takes what the original game tried on the PC so long ago to a logical and excellent conclusion.

I'm mainly playing three games at the moment: Vice City, Panzer Dragoon: Orta, and The Two Towers. Vice City is fun, there's new stuff to try whenever I get bored, and I don't feel too bad if I fail a mission, because I can probably take a bike out and find some more jumps. Orta is fantastic - visually sumptuous, and yet unique, and seriously challenging; I've dropped it down onto Easy and it feels perhaps slightly too easy, though still a good challenge; Normal requires nothing less than real skill to survive. It's challenging in a way rail-shooters seem they shouldn't be. The real surprise of this triptych is The Two Towers, which despite being quite brief and containing physically small levels, is a real challenge - a rarity for a film license. To survive the later levels as Legolas, or to stand any chance on the hardcore challenge that is the Tower of Orthanc requires real skill - the parry button is far more vital than it might initially seem. It's hugely satisfying to survive some of the tougher levels - in Helm's Deep, you really do get the sensation of being entirely overwhelmed.

Alexandra wrote below on games, mainly RPGs, that let you create a character and thus let you win "as a specific character". What's interesting is that, even in this late stage of videogaming as an aftform, such a freedom as to the role you play is very limited.

The most basic choice as to the character you play is one of gender - male or female. Most of the time, this is cosmetic; bar a few personal pronouns, dialogue, plot, character abilities barely change. Alone In The Dark tried this with a little twist - the male character, tougher and better with weapons, was a PI investigating the house Decerto, wheras the female character, Emily, was a relative of Decerto's owner and so knew the house better. But I don't think the camera angles changed, which is a shame - she wouldn't be as surprised by some things as Edward (the PI) would be.

One of the earliest games, I guess, where gender made an impact was Alter Ego. This was a life-sim for the C64, where the player made personality choices for a character from the moment they're born (beginning with the marvellous line "You are in a warm, dark, friendly place.") One side of the disk contained the male version, the other the female, and each was radically different - as the development of men and women is.

It seems that in videogames, the choice of gender is often arbitrary. Black Isle Bioware's role playing games - Baldur's Gate, Knights of the Old Republic, Fallout, etc, with their very open-ended structures did make some good use of gender - gender specific romances, including one for female players in Baldur's Gate II for a start, plus gender-specific conversation options. It's also worth noting that a lot of the time, the option to choose gender allows players to mirror the real world in the game world, and not to role play; surely the point of the escapist fantasy videogames provide is not to mirror the real world? Other, more interesting questions, such as player's apperance are never questioned. This is a choice that would be quite interesting in RPGs, and especially in Massively Multiplayer games, where the software company need not write routines to respond to appearance; players would go by what they'd trust in the real world. (For instance: you might choose to have a scar on your face. Will it scare people? Will it make people think you're a fearsome warrior? Will it just make you look "cool" and not effect the game world at all? I can't think of any game that models this, though perhaps some MMORPGs do, and there I can see it working).

Another common choice in gaming is the chance to name, or rename your characters. Though the hero of the Zelda games is always referred to as "Link" (and, indeed, that's the default name the game offers for him) it is possible for the player to rename the character to anything he or she chooses. It's the same with characters in Japanese RPGs such as Final Fantasy; there are default names, but personal names can be chosen.

Why is this? Well, to go back to Alexandra's idea of "winning as a certain character", this allows me to rename Link to "Tom" and then win as me. Except I'm not me; I'm a teenage elf with pointy ears. This isn't so much roleplaying as wearing a mask. The player in the real world overlaps into the game world, and yet to what end? Not to much immersion, that's for sure. It's the character the game is designed for, but wearing my name badge.

Of course, it's difficult to design games for varyingly named characters, simply because nowadays most games feature speech and speech synthesis isn't good enough to cater for every possible name. Knights of the Old Republic caters for this by never having other characters refer to the player by name, and by never having the player speak (bar odd affirmatory "unhuh?"s during play) his/her lines; instead, the player chooses a line, which passes unspoken, and the other characters respond. This is a very successful method - if you doubt me, try playing KoTOR, because it's one of the best scripted and most immersive games I've played; it also features superb voice acting.

Alexandra raises the point of emotional immersion, and I'd argue that it's impossible to be immersed in a game if you don't care about the characters in it. In a first person game, it's easy to assume that you are the central character - but the moment you break that frame of reference (such as in the third person driving of Halo, or an external cut-scene, you have to ensure that the player still cares about their avatar; if the character looks like a dork, it's impossible to care. Half-Life went as far as to never break from the first-person perspective at all; the entire characterisation of Gordon Freeman was done by external characters' comments. At the beginning of the game, fellow scientists chat to him as a peer; later, one of them looks at the shotgun (or whatever weapon is in your hands) and says "you look like you know how to handle that thing". Voila: instant characterisation, demonstrating Gordon's development from nerdy scientist to soldier of fortune).

A common feature of games these days is having to protect other characters, or to fight alongside AI buddies. Many games successfully pull this off. Halo does it by giving the marines many, many lines of banter and dialogue; their radio chatter is so realistic (and, importantly, so is their AI behaviour) that you can't help but believe they're human and want them to stay alive. The US military commissioned the game the public will know as Full Spectrum Warrior for training purposes. A recent feature in Edge (E131) on FSW remarked:

"The army wanted to teach the players to weigh the lives of their men very heavily. Giving them voices and personalities made it hit much harder when one is killed"

And you don't even need hyper-realistic graphics to acheive this. Sensible Software's Cannon Fodder was a very basic, cartoony sprite-based action game. Over a vareity of missions, you guide 1-4 soldiers, shooting enemy troops and completing objectives. Each of the soldiers has a name - initially, the nicknames of the Sensible stuff (for no-one can forget those first soldiers, Jools, Jops and Stu). At the end of every set of levels, 15 more recruits are added to your pool of soldiers. These are essentially the "lives" of early games. The new recruits come over a hill to join a queue of troops; not just any hill, though, but Boot Hill, with a cross or gravestone for every soldier you lost. Also at the end of the missions, all survivors are promoted a rank, and there is a rollcall of the dead. Just the simple addition of names increases the emotional connection with characters; the galling image of Boot Hill before every mission only makes the impact worse. But there's no digitised voice, no famous voice-actor or face in front of
This filling-in of personality when there is none there bar a name or briefest of animations is one way in which games could make some claim towards being something akin to an artform - the fullest understanding of characterisation is not merely in the explicit depiction the game provides, but also in the characteristics the player gives to his characters, through his/her own imagination.

Of course, games are not stories (as Greg Costikyan argues, and as I shall explore more fully later); they are as much abstract puzzles and toys as adventures. Adding characterisation to games does not always lift a game above the level of an abstract problem. The ghosts in Pac-Man, after all, had not only names (Shadow, Speedy, Bashful, and Pokey) but nicknames (Blinky, Pinky, Inky and Clyde respectively). Because the game is still entirely abstract, there is no need for anything in it to have names, and so the personification of the blob-like ghosts makes very little impact - until, of course, the player blames his own failure on the fact that "that red ghost had it in for me", and suddenly the ability to blame "Blinky" is at least some compensation. (Still, it's curious as to why they get nicknames AND names; and note that the column with the name is not titled 'Name', but 'Character').

Probably the greatest potential for character creation is in massively multiplayer games, mainly role-playing games. One of the key marketing points of such games is the ability to live a fantasy life, as an entirely player-created character, interacting with other real people (who are also living behind personas). Non-Player Characters are far less important than in traditional single player games, as players create their own dialogue, encounters, parties, and even sometimes stories. In The Sims Online, the online version of the hugely popular game, the entire game is about pretending to be another person, having another life. The Sims seems one of the strangest games to take online - for surely it would become little more than a glorified chat room, like Habbo Hotel? - and indeed, take-up figures have been far less than for games such as Everquest and Star Wars Galaxies.

Such gameds allow a greater freedom in terms of "who the player can be" and "who the player is to interact with", but of course, such greater freedoms make structured plot much harder - far more so than in relatively-free-but-slightly-limited games such as KoTOR, Deus Ex, or Fallout.

This leads into the discussion of plot within games, and of the important divide between games and stories - but that is for another essay.

Art is a form which is protected as free expression. Because of this, the extent to which videogames can be considered 'art' (and, by implication, protected as such) is something which has been commented on a great deal. In 1999, Edge stated that they could not be seen as art because of the element of choice involved on playing a game - that "The primary interaction when playing a game is not between the emotional and spiritual themes of the game, but with manipulation of onscreen objects." If this argument is taken as true - that videogames are not art because they concern choice - what does this choice mean about games as a form?

This is concerned with the fact that playing a game is necessarily a choice in itself. It is possible to consume art passively - to take a trite example, many paintings are so familiar we can look at them without thinking about, or reacting to them. By contrast, any game (video or otherwise) must actively engage the user mentally and physically. The player has to respond in order to experience the game; they are involved in the form itself far more directly than with any other medium. Furthermore, in the case of many videogames as well as RPGs, the player will have to create a character. When this happens, the involvement is taken to yet another level - the aim of the game is not just to win, but to win as a certain character. The engagement is not just with an abstract problem, but with something within that abstract problem.

It could be argued on these grounds that this "manipulation of onscreen objects" does have the capacity to involve the player emotionally: whether or not it si to be considered art on these grounds is not the issue at stake. The fact that this involvement happens in the context of a game is interesting. In board games, the pieces on a board are not personified. If, in chess, another player takes one of your pieces, the response is not for the piece itself, but for the game as a whole: what does this mean for the strategy I had planned? In a game in which the player takes a role, all that happens, happens to a character - meaning that as well as strategy, and the player's response through the character they have developed is considered. This degree of (if not emotional involvement) immersion raises the game above the level of an abstract problem, meaning that the game is a space in which players can make meaningful choices about that space. It could be posited that this should afford videogames the protection which art enjoys.

(Obviously this raises other issues, particularly how we are to consider RPGs within this schema. It also (again obviously) requires more argument.)

I know it's been commented upon a lot, but try this for size: the weapon hierarchies of Doom (and successors) versus Halo.

• Halo: You can only carry two weapons at once. This is a pain, because every weapon is useful. The Assault Rifle, so irksomely useless at early stages, becomes a very useful Flood-slayer. The Pistol pulls off headshots and is surprisingly powerful. Even when more powerful weapons are available - the shotgun, the rocket launcher, the plasma rifle - the basic pistols are most useful, because they are plenty, and have unique features (a zoom on the human pistol, the overcharge on the Covenant pistol). Every weapon is an option - you could waste rifle ammo useful versus Grunts on an Elite, or choose more appropriately but risk running out of plasma charge. Or, just rocket launcher the lot of them, but prepare to be low on ammo later. Every cog in the machine is useful, and every weapon is useful until the very end. It's a very socialist/communist perspective, and the limit to two weapons only serves to highlight how useful each weapon can be.
• Doom: You can carry all weapons at once. In general, each one supercedes the previous; the shotgun outclasses the pistol, the chaingun (generally) outclasses the shotgun, and so on until the BFG. In general, you use the best weapon available to you. If that's out of ammo, you use the next; so after a BFG spree in a big room, you're reduced to rocketing a pile of enemies until you have BFG charge again. The shotgun is probably the best overall, being quite powerful but slow; the chaingun, though next up the ladder, is powerful because of rate of fire, not strength of round. That said, the pistol is never used once the shotgun and chaingun are acquired - the shotgun does more damage, and the chaingun uses the same ammo more effectively. Might is right; the higher numbers succeed. Also, because you always have everything with you, there's no penalty for not using it. It's more capitalist in outlook - the higher up the food chain, the more powerful, but you're not penalised for still having the low-end stuff. And just above the very bottom is the surprisingly powerful workhorse.