This is a belated response to Brian’s post (which is a response to Adam’s post).I’m going to speak as a designer rather than a gamer, and list some titles which I think teach us something about game design rather than necessarily being the best gaming experiences available. Yet I accept they are biased towards my favourite games, as you might expect.

In no particular order:

1. Football Manager (2005, 2006, etc). This is an almost entirely abstract sports game. Unlike 99.9% of games, it’s mostly text based, so you’re immersed in the same way as when you read a book. The graphics don’t try and depict reality so there is never an opportunity for the visual failings to jolt you out of the experience. Nor is there an explicit narrative – instead, the game has you tells your own story through your experience of the simulation. Apart from the obvious stark reminder that immersion does not necessarily flow from graphical fidelity, it also shows that not all strategy games need to revolve around geography, whether abstract (Chess) or simulated (eg. Command and Conquer). Most strategy games do rely heavily on some sort of positional play or topology so this is an interesting example of breaking free of that pattern.

2. Ultima VII . This is my favourite RPG of all time, and perhaps my favourite game of all time. There’s a large world, rich with detail, exuding a sense of believability that eludes most games today despite almost 20 years of technological advancement. The plot is interesting, the writing good, the lore engrossing. Oblivion is shallow by comparison. But Ultima VII also teaches us things through its mistakes and idiosyncracies: for example, the unusual viewpoint and control system frustrated many from an early stage, as did the click-and-hope combat system which resulted in the unnecessary death of many a companion – yet the latter was refined (to some success) half a decade later in Baldur’s Gate.

3. Thief: The Dark Project. I think a lot about how games do not ‘need’ story, or simulation: but for me this game has a perfect blend of story and simulation on top of the core gameplay. There are several valuable game design lessons that Thief demonstrates: how fuzziness and uncertainty in the AI changes the way a player has to act, how a game can encourage you to attempt to recover from failure instead of reloading, how audio can be an integral part of a game rather than just polish, etc. I think a lot of players don’t like it because it’s hard to get used to the very different type of influence you have on the world compared to other first person games – treating it like a shooter will get you killed. Similarly, the story aspect in Thief is quite hit or miss for a lot of people, anecdotally because they just want to sneak around, not fight zombies. These issues could explain the sales figures being disappointing for such a critically-acclaimed title, and possibly demonstrate the importants of setting your audience’s expectations when pitching a game, because being the best stealth game on the market is no good if all your players are disappointed by it being heavily story-based, and vice versa.

4. Deus Ex. Again, I love the story in this game – and it wouldn’t work without it – but this is essentially a game about making decisions. Most games with a strong narrative don’t let you make any meaningful choices via the gameplay, but this one does. One mission in particular has quite a different purpose depending on a dilemma you face in a previous mission. And many encounters later in the game depend on actions you took earlier on – not just dialogue choices, but actual game actions. Thus the story vs. gameplay issue no longer seems like a zero-sum problem, as it is often portrayed. Sure, Deus Ex doesn’t deliver on the ability to choose as much as it could, especially later in the game, but that’s a churlish complaint given how much more it does than most games. Perhaps then it’s notable more for showing what could be possible than for what it actually did. Who will take this to the next level?

5. Doom (2). Our home PC at the time wasn’t up to playing Doom but when we finally got a 486 (SX, 33MHz. 4MB Ram I think?) I got hold of this sequel to the seminal first person shooter. Back then, it was as if the sort of game we used to dream would be made was finally possible – a world in 3D that you could move through in real time, and plenty of demon-based action to boot. It’s a shame that I don’t think I’ll ever have such a horizon-expanding moment at a computer ever again. But looking back we can still see that the game was more than just its revolutionary graphics: the maps in particular were amazingly crafted, complex despite never using more than 3 door keys, appearing fully 3D despite being essentially limited to 2D with height values, and making the FPSes of today seem trivial. Sadly, this seems to be what the players want, given the strangely common notion of recoiling at the idea of getting lost during what should have been an exploration game. Doom was also one of the last games to have entirely 2D opponents, which allowed you to face a wide range of different-sized encounters. Future games, needing many polygons for each character, had to cut back on the number of foes that could attack you at once, compromising the experience in the process.

6. Diku MUD. MUDs were the first games I played where the world that kept going when I stopped playing, with people from around the world continuing to adventure in my absence. As with Football Manager the immersion comes from the text rather than the graphics (and there was no sound to speak of anyway), and these games showed the first major use of user generated content in conjunction with the internet as a way to efficiently distribute it. All this meant that the pipe-dream of virtual worlds might come true. And yet, while many of the systems still live on – in particular, games like World of Warcraft are still talked about as having DIKU mechanics – much of the potential of early MUDs has gone sadly untapped, as the potential for meeting new people and adventuring with them got shunted aside in favour of providing something akin to a single-player game with the piracy-averting benefits of providing software as a service. Most MMOs today have a distinctly different feel to early MUDs, with few people interacting in a more than cursory manner with people they didn’t already know before joining the game. There are parallels here with the way that the more open Facepartys and MySpaces of the world gave way to the essentially closed system Facebook, where the mass market, rather than enjoying the new freedoms and opportunities available, sought only to interact with existing social circles and fought against any attempt to expand them.

7. Elite. This is another poster-child for procedural generation – how else could you fit 8 galaxies in under 64 kilobytes of memory? On the surface the actual gameplay is quite simplistic – travel from one planet to another, trading goods, and buying new equipment – but the beauty came from the infinite combinations that the content provided so effortlessly. You could always be on the lookout for another planet with more favourable prices either for buying or selling, or perhaps somewhere a little bit less lucrative but with a stronger government and thus less risk of being attacked by space pirates. But while this might have got boring on its own, the well-paced goals such as being able to upgrade your ship in various ways kept you going back. This is evidence that, with the right systems in place, procedurally generated content can be enough to keep a player entertained.

8. Magic: The Gathering. I only have a small amount of experience with the computer game but I played the card game for several years and so will speak about that. MtG demonstrates the way in which a game can be played on several levels: the game itself, where you pit your cards against an opponent’s cards, or the meta game of constructing a deck of cards with which to play your next match, or even the level above that of trading for cards in order to make one or many decks. Few games can boast such depth for their players before they even start the first turn. MtG is also an interesting example because it was one of the first ongoing games where players continued to get stronger as time went on. Although the power curve levels off due to the nature of the game (eg. with larger decks of cards being less predictable, and thus adding a new card inevitably means giving up an old card), there was still a significant amount of power creep, with various cards making older ones completely obsolete, some expansions being deemed overpowered (which is natural, given that underpowered expansions would fail commercially), and so on. The various game types that limit a player to certain expansions perhaps hint at how MMOs could address the elder game aspect, deprecating parts of the game to keep things fresh for those who choose to keep up.

9. Civilization. Sid Meier’s strategy classic perhaps demonstrates best how a game can thrive when there are many well-balanced but ultimately minor mechanics all contributing towards a whole. Most games fixate on a couple of core features, such as choosing a weapon and pointing it, or optimising a few certain stats, but Civilization gave you many different levers to pull and no right answers. Do I build a Granary or a Temple to make the most of this city? Should we study Mathematics to build catapults or Map-Making to build Triremes? Build on the coast to get a port or inland to improve farming prospects? Attack my militaristic neighbour to avoid a future invasion, or save those resources for now in the hope that war never comes? Sid Meier is claimed to have said that “a good game is a series of interesting choices” and Civilization bombards you with small yet interesting choices at every step, making each playthrough different to the last. As with Football Manager, you’re given many variables to tweak, each with consequences, and as a result you form your own narrative through the unique dilemmas that you face.

10. Pen-and-paper tabletop roleplaying. The typical system here is Dungeons and Dragons, although I’d argue that this is perhaps not the best choice. However, the actual system you use is generally irrelevant. The key here is that the game is partly improvised by a human controller in a way that computers are still unable to adequately emulate. Starting with a small amount of source material – eg. a map, a list of non-player characters, and a rough plot outline – the players can choose how to approach the tasks ahead, and the gamesmaster can extrapolate from his or her written content to accommodate the wishes of the players. From there, the players may present even more creative approaches, and the two sides work half collaboratively, half competitively, to create a shared experience for all, without the need for expensive content. As with some of the text based games above, the players are immersed by their imagination rather than by visuals and this is compelling in its own way. This is in stark contrast to how computer games have evolved, generally relying on very detailed content created before the game, with the result being that the game duration is reduced and the player’s scope for choice is also significantly reduced. Arguably this direction is unsustainable as the costs of content creation grow. Either way, it’s surely not an optimal development because we already have great experiences that are short, linear, and involve few choices, known as movies – games should ideally tap into a different type of fun. Can we ever replicate the benefits of pen-and-paper roleplaying on the computer? I think we can, and the project I’m currently working on is one of several steps in that direction, which I hope to talk about in the future.