Only a Game

Little White Flag

I surrender! I took one day off work, and now I am swamped under more than 200 emails, more comments than will fit on the comments roll on the blog and a record number of trackbacks (two!) Since clearly I must attend to work before I can attend to the blog, many things will have to wait a day or so at least.

In brief, I've read and enjoyed Koster's book and will be posting on that soon enough, and there has been a good response to the request for external level designers for Fireball, and there may be some delays in contacting people because of this.

I'll write more as soon as I find the time!

Murphy's Law of Trackbacks

Although I am not a believer in Murphy's Law, surely its application to blogs would state that your intelligent and well formed posts will be ignored, whilst your rambling stream-of-consciousness posts will not only get a trackback, but will do so at the most inopportune time. And so it is today, with a trackback from Raph Koster that I simply don't have time to properly respond to because I have to travel 400 miles to (hopefully) sell a game to a publisher, and since I'm taking tomorrow off for my wife's birthday, there is little hope of finding the space to write any time soon. Still, I'm awake early this morning - there may be time for a rambling response.

Regarding unifying theories... When Craig says: "It seems to me that all of those different reasons to stop can be summarized as "nothing interesting to do", I am inclined to say "Yes, but..." :) I believe there's a time to unify and a time to subdivide. Animal life could be summarised as entities that feed and reproduce, but that doesn't tell me much about the behaviour of squirrels! Since I'm usually touting diversity, I'm generally less interested in unifying theories of game design - they tend to bleach out the details, and God is in the details, as van der Rohe said.

One quick clarification. Koster suggests:

“Everything known” and “nothing new to experience” certainly seem to have tremendous overlap both with each other

The distinction here (in the context of Rational versus Idealist archetypes)  is between knowledge and unique experience - it's related to the distinction between science and art. Consider a stack of textbooks versus an art exhibition. Everything is known only when all the textbooks have been read, but the experience of reading a textbook has been exhausted within a few pages. Conversely, everything might be known about the art exhibition when everything has been seen once, perhaps briefly - but the experience of the exhibition is not exhausted without taking the time to connect with and explore the pieces on display.

That said, my comments in this regard were part of the ongoing process leading towards our next audience model, and should not be seen as anything short of structured speculation at this time. I post these kinds of thoughts to the blog in the hope to get a broader perspective, and to get them out of my head and into some form suitable for discussion.

Since Koster's A Theory of Fun is already on my reading list, I should probably move this closer to the top, especially if he is going to be kind enough to read our book. I'd like to say, however, that I'm slightly sceptical of reducing games to learning (if indeed this is what Koster's book does!) although I'm certain one can form a plausible model by doing so. When I used to play tabletop role-playing games, much of the play was in the performance, the creativity. I accept that this could be seen as learning to be creative, I just question whether this is a useful approach or merely shaving a square peg into a round hole. Similarly, what exactly does one learn playing a game of pure alea, or from the aleatory elements of most tabletop RPGs? Not to mention games of ilinx, especially those closer to paidia than ludus... I have many questions, but I'm unlikely to get answers without reading the book.

Right, I must be going. Have a good weekend when it comes for you, and wish us luck selling Fireball!

Diplomacy versus Harmony

The trouble with calling the player fitting the Idealist temperament archetype a "Diplomatic Player" (after the Diplomatic skill set) is that in current games, the skills employed are rarely diplomatic. Consequently, I'm debating whether we should rename this to "Harmonising Player" or something similar. (The trouble with 'harmonising' is the z versus s split in US versus UK English). "Empathising Player" (same z/s problem) or "Unifying Player" are alternatives. The downside is drift from the source material, i.e. Temperament theory. The upside could be greater clarity. Thoughts? Opinions?

The End of the Game

When do people stop playing a game? And, critically, are the patterns in which people come to stop playing universal, or distributed by some aspect of personality?

I have been thinking about this topic for sometime, but the motivation for tackling it today comes from Clive's Collision Detection blog where he has posted his most recent Wired article (either link will take you to the content). Clive's opening gambit:

An obsession with a game always ends suddenly. Why?

This is a bold statement to begin with! What does it mean to claim an obsession ends suddenly? Clive's description:

This is one of the abiding mysteries of games: Why do they let us go so suddenly? Every gamer I know describes the same abrupt drop-off, totally unexpected, arriving after hours or even weeks of feverish play. It is like a curious, unintentional form of cold turkey. You wake up one day fully expecting to spend another four hours in an eye-glazed stupor, only to discover that the thrill is gone.

I recognise this behaviour. But I do not believe it is universal.

I'm going to start by using myself as a case study - not generally considered good science, but not all intelligent investigation needs to be scientific. (We sometimes forget this!) It is only comparatively recently that I have been freely able to stop playing games. As a teenager and even into young adulthood, I finished more or less every game that I started playing, both on home computers, in the arcade and, eventually, on consoles. There are very few exceptions in my life. Sometimes, as with the original Prince of Persia, I continued to complete the game even though I was not even remotely enjoying it. I wanted to beat it. To stop playing (in my mind at the time) was to admit defeat.

However, not all these games produced the kind of obsessive behaviour to which Clive alludes. In fact, as my wife can attest, there is one particular genre of game in which I display obsessive behaviour: CRPGs. Once I reach the state of obsession, I find it difficult to resist the game... my wife (who is a nurse) has described in some accuracy my physiological symptoms of addictive behaviour when I am in the grip of such games. It is a reason that I play them now only very rarely - when my work demands it, when the game can be completed in less than 8 hours, or sometimes to overcome jetlag. Otherwise I avoid playing them.

My problem is that I do not generally experience the sudden termination of obsessive play that Clive alludes to. I have recently experienced the sudden end of interest in a game - but in essentially all cases, the degree to which the game engaged me in the first place was never obsessive. A key change in my behaviour is that I no longer want to beat games - this has ceased to be my primary motivation. But still, there is generally no end to a game obsession for me until the game is completed. On the basis of case studies, I would say that this is a relatively rare state of affairs, and that what Clive describes is a more common experience than unfettered addiction. But not every player shows this behaviour.

Indeed, there are players we have interviewed and observed for whom obsessive play is not an issue. This includes but is not restricted to players for whom extroverted play is a primary factor (Lazarro's The People Factor; Type 4 Participant by DGD1 ; Katherine Ibister's Extroverted Play). For such a player, ludic elements are considerably less important - and indeed, highly ludic games such as Chess and (in extreme cases) card games are of no interest. The content of play for such a player is interpersonal - such obsessive behaviour never emerges.

This immediately changes the shape of the issue - obsessive play of games is apparently not a universal behaviour. Indeed, even among players for whom extroverted play is not the primary factor, there are cases which do not fit the pattern. We have, for instance, a case of a more mature player who plays essentially one game - The New Tetris. She does not play obsessively, and she has not stopped playing it, despite long ago completing all the framing goals. She simply enjoys playing it. There are many similar instances of this pattern of play, especially with games such as solitaire card games on PC.

Clive turns to Koster for a partial explanation:

"You kind of see through the game to its underlying mechanics," Hayes says, "and it suddenly seems no longer worth the investment of time." This is much like what Raph Koster argues in his Theory of Fun: We humans seek constant novelty, so only gameplay that has nearly infinite permutations -- like chess -- can hold our attention forever.

This looks to me to be a ludic fallacy - which is to say, an assertion made by someone with a strong affinity for ludic (structured) play, without taking into account other approaches to play. There is a tendency for people who enjoy agonistic ludic play to forget or overlook players who prefer other styles of play.

Consider my own relationship to Chess, for instance. I have played it perhaps a dozen times, and I have no interest in ever playing it again. Despite the fact that I love and enjoy ludic play, Chess is only barely a game for me. To play a game of Chess, I am required to perform state space searches, an activity I personally consider more suitable for a computer program than a human being. Chess bores me. Chess only has infinite permutations if the different states of the game hold meaning for you. If, like me, the component patterns of a Chess game are not interesting, then the compound game states are equally uninteresting.

However, this is not to say that the capacity for infinite permutation is not a relevant factor in the extent to which a game can hold attention - it is no coincidence that the case study I cited before was a player of The New Tetris, which is a game which offers infinite permutations. Obviously, the more varied the experience, the longer the potential to hold attention - but it takes something quite different to turn this potential into an actual instance of a game holding our intention indefinitely. (If anyone has an instance of someone playing a game over an indefinite period of time which is not based on a mathematically large state space, please let me know!)

It is our belief that what sustains interest in a game is access to flow states (Csikszentmihalyi's model - boring writer, but his model is a valuable addition to psychological thinking). As we write in the book, different players achieve flow states for different reasons. According to this hypothesis, obsession with a game would be expected to correspond with the capacity to produce a flow state - therefore the sudden end of that obsession would presumably correlate with anything which undermines or collapses the flow state.

Currently, I am looking at how people play games in terms of Temperament theory. The research is still ongoing and hypothetical at best. For the sake of discussion, here are my immediate impressions of how I would relate Temperaments to the experience of coming to the 'end of the game':

• Rational (Strategic player): since a key theme is knowledge and competence, loss of interest is likely to correspond to achieving a state of total knowledge, or a state which approximates total knowledge - "seeing through the game to the underlying mechanics", quoting Hayes above. The end of the game may be when everything is known or can be anticipated.
• Idealist (Diplomatic player): a key theme is unique identity, loss of interest may correspond with a loss of uniqueness. If a game tries to sustain interest through permutation, this is unlikely to work - the game is probably up when there is nothing new to experience. Narrative games, whose identity is sustained for the duration of the story, seem to be preferred to strictly ludic play.
• Guardian (Logistical player): I would hazard a guess that task completion is a factor here. When the game has no tasks to complete, or the player has become unable to carry out their assigned tasks, the game will no longer sustain.
• Artisan (Tactical player): freedom to act is a key theme. I would suggest the game is up for someone fitting this archetype when they no longer have the capacity to personally affect the game world. Like the Idealist, permutation isn't necessarily going to be sufficient to sustain interest.

Of course, any one individual could express more than one of the above traits - and furthermore, the assertions above are at best speculative.

I'm not working towards a conclusion, merely thinking aloud, so I apologise to anyone who has read this far in the hope that I have answers rather than fresh questions. However, it strikes me that there does appear to be issues of personality to take into account when considering how and why people stop playing games.

Clive ends by asserting that no other media shows the same sudden drop off in interest as games - and I broadly agree. I believe the distinction here is the extent to which the media can diverge from the enjoyable. We continue watching TV shows sometimes out of a sense of loyalty - even if it ceases to be as engaging as it once was, it may become so again, so it takes a while to break the habit; also, some people feel obligated to continue watching shows for which they have seen every episode, if only for the sense of completeness.

Disenfranchisement with games can occur more suddenly because the game only needs to cough up a challenge fractionally beyond our abilities to break us out of flow. (Although for players fitting the Type 1 Conqueror template this may actually engage them further in an ongoing quest for fiero). We keep playing as long as the hope and promise of more flow states remains. The player remains engaged as long as the play remains engaging. This seems like a simple truism; it's only when we start to explore how different players become engaged that it becomes something considerably more interesting.

Cold Light of Day

Reviews for the book are flooding in at the rate of, oh, maybe one a month. Still, the reviews have nice things to say and unsurprisingly this is more enjoyable than being torn apart by wild tortoises. Our first review is Jack's at the Amazon sales page (five stars - thanks Jack!), and the second (as far as I know) is over at the consistently excellent Lost Garden. Danc's review not only praises the book but it raises all sorts of sensible issues for discussion. I can think of no better claim to make about one's work than it provokes intelligent debate among the bright and talented.

Danc makes the claim that the likelihood of the DGD1 audience model being correct in all its details is approximately nil. Personally, I think the word 'approximately' isn't needed in that sentence! We did the best we could with the data we had, but the reason for the '1' in DGD1 is that by the time we were finished, we already had new directions to explore. My hope is to encourage the creation of many audience models - what a terrible place the games industry would be if there was just a single lens to look through!

The review breaks into an engaging aside, looking at the dangers if the audience model approach becomes dogma instead of thought food. It will come as no surprise to my readers that I detest rigidly dogmatic thinking (although, as my wife is keen to point out, I do dogmatically believe in supporting diverse belief systems!). One of my chief concerns with Noah's 400 Project, which I discussed with him at length when he first set it up, was the danger of people taking it as canonical dogma... the trumping schema  seems to feed into this problem. Noah seemed confident there was little risk  - I'm not so sure myself.

With regards to our own work, the idea that anyone could rigidly adhere to a single audience model and use it to dictate cookie cutter game design is chilling. Danc is right to raise a warning flag, but I remain optimistic that the game design community can take DGD1 in the spirit in which it was offered - as a way of thinking differently about the play needs of a diverse audience. (Perhaps I am showing the same naivity as Noah in this regard, though). Ultimately, future audience models may provide a template for wide scale economic success - my hope and belief is that if and when this happens, the niche audiences will not be left out in the cold as smaller companies will be able to meet their play needs on lower risk budgets. In fact, the direction we've gone after finishing this first round of research into audience models was to say: lets see if we can make it economically viable to support the play needs of minorities. I know we're not alone in thinking this way.

Danc wants to see various things that we want to see too - such as additional studies done with more statistical rigour. Oh to have the money to fund such an endeavour! I dream of a partner organisation with the resources to prove us wrong and take us to a new and better model. In the meantime, we plod remorselessly towards DGD2 on our own, and with a budget more suitable to buying tinned spaghetti hoops than carrying out research.  Still, we have to do what we can.

In terms of the applicability of audience models to game design, we have been using DGD1 (and Lazarro's Four Keys) to inform our game design for two years now... I guess I should really write about how this has worked for us at some point - but how does one get around the non-disclosure agreements? I know we need to do more with this.

As for exploring the business consequences, I thought the book already did this! But Danc is right that there are potentially negative issues which are glossed over in our enthusiasm to suggest that there was a better way to do business than copying the most successful genres ad nauseum.

I loved reading the Lost Garden review. It was that encouraging moment when you go from lingering self doubt to the confidence that your once fanciful ideas actually make sufficient sense that a fellow intellectual is willing to spend the time looking at both the advantages and the potential pitfalls of your approach. I wouldn't want DGD1, which is effectively our prototype audience model, to be  anything other than a first step towards an industry which is more in touch with the reasons why people play games, and a lesson in how we can better meet the play needs of our audience. But this journey has really only just begun.

Mental Pressure Valve

Time to take a break from coherence and indulge in a touch of broomwork in my crowded mental space.

Am I posting too much? I'm finding it a rewarding way to start my day, but I worry that it all comes too rapidly. Posting every day makes sense for someone like Alice, whose posts are short, but it takes time to read all the nonsense/ideas I pump out. I don't really know what my relationship to the people who come to this blog is or should be. Maybe this is not something that one knows, but something one must simply take on trust.

The blogs I read stopped talking about games this week and turned to other subjects. In the case of Man Bytes Blog, this led to a number of posts about myth which I have really enjoyed, and which have prompted modestly verbose comments from me. I've really enjoyed it - mythology is a subject close to my heart. For anyone interested, the four posts are archived under Myth Things (in reverse order).

Corvus looks completely different with the shorter hair! Almost respectible. :) It makes me want to host a 'Years Apart' exhibition, and show sets of photos of people taken five or so years apart. Seven years ago, I looked like the leader of a religious cult! :)

Sanjit, lead programmer on Fireball, delivered a build yesterday which fixed all the problems in the burning mechanics that were giving me some anxiety. Now it's certain the gameplay works, and it definitely has the capacity to support both the  Hard Fun of puzzles and environmental challenges and the Easy Fun of simply watching things burn. A colleague suggested yesterday that we might like to show the title and author of levels in the game - and this is quite tempting. Authors names would by like the names of crossword makers - handles, if you will. It would allow the player to build a relationship with the different level makers, potentially. We need the people to make the levels, though. We could even change the framing structure to be more like a gallery than a game, potentially. See here for details if you fancy being involved.

Last night, in the "Not Work" evening, we had an impulse to play Wave Race 64. I never realised before how much the lower resolution graphics lend this game an impressionistic feel - it's like playing a Monet painting. No wonder Blue Storm was a disappointment - it's graphics are shiny and "realistic". All the beauty was lost. Wave Race 64 (along with Blast Corps) was the game that persuaded me to buy my very first console, although I'd played on many consoles before this point, of course. It still retains an elegance of design and environment that I admire.

Went to see the local squirrels last Monday. They're doing well - but too busy burying supplies for the winter to hang out and play with me. Grateful for my free handouts, but easily distracted. My wife took some pictures and films - I haven't seen them yet, but I might post a few if there are some good ones.

Right, my mind is swept clean. Enjoy your weekend!

The opening image, Broom Brain, is by Ryan Arthurs. No copyright infringement is implied, and I will take the image down if I am asked.

Quantum Items

When is it appropriate to have a vastly diverse collection of items to find as rewards in a game? When is it appropriate to have a more rigidly defined compact schema of reward objects? Is there an audience need to be reflected, or is it merely an issue of identity that each game should resolve in its own way?

The pattern for treasure objects in RPGs was set back in 1974 with Dungeons & Dragons (the white box version). It was expansive - more treasure tables and magical items than you could ever need. They must have felt it was the right way to go, because when Advanced Dungeons & Dragons came along in 1978, they added even more items and even more tables. RPG games have stuck with this tradition... Diablo and other hack games have very little to offer but the diversity of their objects - a fact neatly lampooned by ProgressQuest.

Who wants myriad treasures? In terms of Temperament Theory/"DGD1.5" (since I am apparently using Temperament Theory as a provisional audience model), it is presumably the Strategic players (Rational temperament). With a desire for knowledge and completeness, players with Strategic bias are more likely to be willing to examine the data for many different objects and make decisions - this process lacks the immediacy that a Tactical player would presumably prefer. There could be a Logistical aspect, perhaps; the Guardian temperament (the parent temperament for the Logistical skill set) is associated with a capacity to handle supply mechanics. Perhaps a wealth of items is appealing Logistical players as a sorting exercise, but I feel such players must still have access to Strategic skills if they are truly to enjoy relating to equipment in this way.

Few other games show the diversity of items that we see in CRPGs. Even car games with considerable customisation options such as the Gran Turismo series don't hit the same volume of items. During CRPG case studies, a common element was that many players (principally in the Type 1 Conqueror and Type 2  Manager archetypes in DGD1)  enjoyed CRPGs  in part because of the mechanics. They liked learning the mechanics (surely a Strategic-type trait) - indeed, one female CRPG player went into some detail as to which Final Fantasy games she'd enjoyed in terms of how the mechanics were rendered.

Smaller collections of items give clearer rewards to the player that don't require much learning, and must presumably appeal to a wider audience. In a Legend of Zelda game, for instance, the rewards are extremely rigidly defined: there are a set of different colour rupee pieces (money), certain resource items (chiefly in the later Zeldas i.e. deku nuts, arrows), the map, compass and boss key for each dungeon,  the tool key items that form the basis of Zelda structure, and the ultimate Zelda item - an empty bottle. This is a more abstract item system than the RPG approach of "more items than you could ever imagine!"

It is likely that this kind of approach is better for a wider audience because it requires less learning, and in the mass market the closer the game is to 'pick up and play' the better its chances. One can argue that Zelda's system is slightly too abstract for a mass market audience, however. It's hard for a new player to appreciate why a bottle should be of such value - especially since its range of uses only emerges through play and cannot be derived intuitively.

In reworking the Resident Evil game abstractions for the fourth game, Capcom included some RPG elements. This is not wholly surprising, as there is scarcely an upper market game out there which does not include RPG elements now - in particular, gradiated advancement schemes. Since these systems provide a satisfying sense of progress often without recourse to expensive game resources. The "RPG-lite" system in San Andreas is a great example, as the skills just affect parameters such as time the player can hold their breath under water, threshold of loss of control in a collision etc. Cheap additions to an expensive game. The arrival of an "RPG-lite" system in Resident Evil is a big departure from the usual mechanics of this franchise, but it is not wholly surprising given the prevailing trends.

For Resident Evil 4, the RPG mechanics rest in the weapon statistics - a rather neat system in many ways, as the numbers are tangible and immediately understandable (firing speed in seconds per shot, reload time in seconds, number of shots per clip) - only firepower is abstract, and the game fixes this by expressly informing the player that 1 standard handgun shot is worth Firepower 1. I will not be surprised to see this system crop up in other games, if it hasn't already.

RPG progress mechanics are driven by a resource. In San Andreas it's time spent doing an activity - an excellent choice for a mass market game in many respects, as the player automatically benefits from whatever they are doing. Most CRPGs use "experience points" - an artefact from Dungeons & Dragons and before which is showing its age. In Resident Evil 4 it's a straightforward currency system. Find treasure, sell treasure, buy weapon upgrades. Not an unfamiliar formula, but made more accessible by choosing readily understandable statistics for the weapons instead of largely arbitrary mechanics (as in most CRPGs).

The interesting part about the items in Resident Evil 4, and the reason for the title of this post, is that they are arranged in a scheme not unlike electron shells in an atom - they are arranged in an almost formal pattern.

The basic 'quanta' of treasure is a Spinel (for reference, spinels are gemstones related to rubys - although in the past they had a bad reputation for being passed off as 'fake rubys', modern gemstone dealers have developed more love for the stones). In fact, finding spinels is such a basic part of the gameplay that when my wife and I find one we cannot resist crying out "Spinel!", which in itself has added to our enjoyment.

Spinels are valued at 2,000 pesetas (ptas) and are placed in slightly hidden locations. Obvious locations contain contain currency valued at or around 1,000 pesetas (800, 900, 1,000, 1,100 etc) - representing the lowest quanta of treasure in the scheme. That the spinel's value of 2,000 is the basic quanta of the scheme is underlined by the fact that a 'full heal' (three green herbs or one green and red herb mix) is also worth 2,000 ptas. First Aid sprays are valued at 2,500 - encouraging the player to sell these and keep their other heals.

All the treasure exists in a broad quantum hierarchy related  to the spinel/full heal. Here's the basic pattern:

• Spinel/full heal: 2,000 ptas
• Velvet Blue/First Aid spray:  2,500 ptas
• Emerald/Treasure with Slots (empty): 3,000 ptas
• Significant Treasure (e.g. Amber Ring, Antique Pipe, Elegant Headdress): 10,000 ptas
• Treasure with Slots (complete): 20,000 ptas

There are a few variations, but not so many that this general pattern does not stand out. The positive elements of the system lie in the rapidity with which it becomes familiar (hence our cries of "Spinel!" - we have not had such gleeful game catchphrases since Sega Bass Fishing's, "Small one", "Okay, average size - good job" and "This one's HUGE!") and the extent to which one feels the step up to a higher quanta. 10,000 ptas equals five spinels which feels like a big reward. 20,000 ptas is party time. The game hands out treasure like it's going out of style - it is practically the archetypal "Monty Haul" dungeon crawl. Just as well, as frequent supplies of treasure softens the blow of the somewhat pernickity challenges it sometimes forces the player to endure.

Are such quantum item schemes a good approach? In many ways, yes. They define clear patterns of emotional response, are easy to learn and add to the identity of the game (years from now, 'spinel' will still be in our lexicon) - not to mention they are cheap to implement since they don't require much in the way of art resources. That they are "gamey" (which is to say, favouring ludus over mimicry) is not actually a significant drawback as a more "realistic" system would struggle to deliver the cashflow required to power up the weapons without drawing attention to itself. Because the artificial collection of these rewards rapidly becomes familiar,  there is perhaps less temptation to  notice the surreality of spinel ubiquity.

On the other hand, I wouldn't expect CRPGs to give up their cornucopian approach any time soon, especially since these games place a priority on character expression - and there is no easier way to achieve this than giving the player a vast variety of items to choose their equipment from. The Sims is another game with the same premium on player expressivity - and therefore it too must inundate the player with choices (albeit largely cosmetic choices).

Perhaps, then, this is the boundary condition for a quantum item scheme. If the player's focus is intended to be chiefly in the active play of the game, it makes sense to have a more precisely abstracted scheme. If the player's focus is intended to be partly or wholly involved in personalisation and expression, a more expansive system should be preferred.

In many ways, quantum item schemes are a big step forward from the "every idea in my head" approach of Dungeons & Dragons, but there is little doubt that there are still times when a dazzling diversity of items still has its place.

Lessons from Hollywood

It's not a great secret that I have followed the writing of Michael Moorcock for many years, and that I learned a lot about writing from his advice and techniques. One of his literally hundreds of books is Letters from Hollywood which, logically enough, is a collection of letters Moorcock wrote to his friend JG Ballard while he was working in Hollywood. Moorcock's experiences became truly grim and depressing while working for a particularly tyrannical director (unnamed, although a friend has told me who it is most likely to have been) - an incident which makes Barton Fink seem not so far from real life. Moorcock's problem was that he wanted to tell good stories. But Hollywood's not about telling good stories, it's about telling popular stories, stories that many people will pay to come and see.

Moorcock's experience soured him to the Hollywood film industry. He says: "I admire all the good films that are made in spite of the system!" Surprisingly, Moorcock has decided to allow a film version of his Elric saga to be made, a fact which suggests to me that he is aware of his own mortality. Many sci fi and fantasy authors only allow their films to be made when they are in their twilight years and most, like Frank Herbert, do not live to see the theatrical release. (Moorcock, incidentally, already had one of his books turned into a film, although he is not overly fond of the results). Even when you don't like the Hollywood system, there's still a certain majesty to the big screen that is hard to resist.

The problem with Hollywood is that it's very good at what it does, which is make popular movies with the purpose of making money. Thankfully, part of making popular movies is occasionally trying new things, and another part of making profitable movies is capitalising all your niches - which includes avant garde art house weirdness (which does occasionally siphon funding out of the machine) and all sorts of interesting and rewarding films. I won't cite examples - I'm sure you can find your own.

Greg Costikyan and others have complained that Hollywood tie-in games hurt the games industry. They do. Because most of them aren't very good games, and high profile rubbish is bad for any industry. But they are an enormous help to the game industry in terms of keeping studios afloat. I reckon that at least half of game developers working today owe a debt of gratitude to Hollywood for providing funding that has seen them through tough times. I know my company would not have survived if we hadn't had some license adaptations to work on in the rough years.

One can argue that we shouldn't have to rely on money from Hollywood - sure, if the games industry was capable of running itself as a profitable business. But many developers (and several publishers!) have the business acumen of a block of cheese and at times its a wonder that the industry makes any money at all. In this regard, Hollywood has been very neighbourly, providing capital to keep us afloat at the cost of forcing us to supply some merchandising. It's really not that bad a deal.

That we are receiving financial support from Hollywood is reasonably good news for many people who work in the games industry. But as the expression goes, give a games company a fish and it will eat for a day, teach a games company to fish and it will eat for a lifetime. The games industry could learn a lot from the way Hollywood has identified the key profitable niches (genres in the case of films) and then invested variably in all these genres on a year to year basis, thus providing the audience with a wealth of choice - and meeting the entertainment needs of millions of cinema goers. Complaints that most of these films are utter tosh are fair but misguided. Worthwhile, interesting and inventive films are still made, and the fact that they are in the minority can hardly be considered a surprise - I challenge anyone to name a media where this is not the case.

What have we copied from Hollywood? Well, mostly their franchise ideas. That Lara Croft was inspired by the Indiana Jones movies is a debatable but inevitable criticism; that Halo owes a vast debt to Aliens seems unavoidable. And in that particular case, the story elements of the game are much cruder than the story elements of its inspiration - and Aliens is hardly the deepest story around.  We have a lot of nerve when we bitch at Hollywood when we haven't got our own house in order when it comes to stories, although the situation is certainly improving and game writing is starting to achieve proper recognition as a craft.

Hollywood is an ally to the games industry - a dangerous ally, but any powerful ally has this element. I do not believe that Hollywood has done anything to make the games industry a worse place, but the games industry has made itself a worse place by trying to copy Hollywood directly instead of understanding what works in the Hollywood model, and then adapting it to games. The narrative structures that work in a film are (in general terms) inappropriate to export to a game, the funding model for films is currently inappropriate for games - but it is the games industry itself which is making the mistakes, the film industry is largely an innocent bystander in our collective incompetence.

How many games are made which are basically agon plus fiero (competition with triumph over adversity)? 95% of our output is doing the same thing - targeting a relatively narrow demographic. That's not how Hollywood works - they find every viable demographic and exploit them fully. That's what commercial game development should be doing. (Don't worry - nothing that happens on the commercial front will take away indie games, as these are made by enthusiasts, just as indie films are made by enthusiasts - and a more stable games industry should mean more money to fund indie development - that's what happened in Hollywood eventually, although there's doubtless room for improvement).

Moorcock's line "I admire all the good films that are made in spite of the system!" applies equally well to games. I admire all the good games that are made in spite of the system! And our system is a mess compared to the Hollywood system. Hollywood is proficient at making popular movies for every corner of the audience with the goal of making money. We are terrible at making popular games for every corner of the audience with the goal of making money. And until we get good at this most basic economic challenge, the industry will be too unstable for us to comfortably support auteurship, expansive invention and advances in dynamic storytelling. Let's learn from Hollywood in a manner that doesn't equate to copying their ideas and rendering them more blandly. Let's learn to feed ourselves.

First Look: Fireball

This is Fireball (or Hidama)  a game in which you play a ball of fire and set fire to objects made entirely out of cubes of various different materials. It's abstract, it's strange, but it's fun and different too.  The Japanese version of the title has the implication of a shooting star, which is apposite to the game, as you can drift around in the air before slamming satisfyingly into the ground to start raging fires.

Right now, Fireball is in pre-dev, and we'll be showing it to publishers for the first time next week. Designed as a budget market game for PS2 (although a PC version may also be made), Fireball is the first of our "verb games" which emphasise tight and easy-to-learn game mechanics with simple control mechanisms. The PS2 version is intended to be played with one stick and one button. Press once to jump, press again to slam into the ground and start fires - although you can also just push your fireball into things to start fires.

The first picture (above) is from a basic training level. You start in a stone chamber, with a gap in the wall low enough for you to jump out through.

A short way ahead is a wall of green blocks. These are the most combustible blocks in the game, and you can always burn these.

The wall is too high to jump, and inevitably the player will set fire to it, and the fire will spread to engulf the entire wall.

Once the wall has burned down a little, the player can get out and have a look further afield.

What can be seen is that the entire area is boxed in with wooden blocks.

Pushing up against these doesn't start a fire... there must be something else that can be done.

Sure enough, if the player presses jump while already jumping, they'll initiate a slam - crashing into the ground and starting fires hot enough to burn wood.

Burning the wood also increases the temperature of the fireball - meaning it can always burn wood from now on. (Also, because the fireball became hotter it can now jump higher).

Well, might as well burn down the whole wall while we're here!

In a matter of a minute or so, the wooden wall is completely collapsed.

What is left?

Just a wooden tower that was outside of the wall. The exit is on top - and there's only one way to get it down.

Burn it down!

That's the basic gameplay of Fireball, all centred around moving your ball of fire around an abstract landscape and setting fire to blocks of different materials.

There are numerous ways that individual Fireball levels can be designed.

The  screenshot to the right shows another area which consists of a series of bridges made mostly of wood.

The problem here is that when you first get onto the bridges, you set fire to a green 'fuse' which in turn sets alight to the bridges.

You then have only a short amount of time to dash across the bridges and make it to the exit.

If you're too slow, all the bridges burn to the ground and there's no way to get across to the exit. Not to worry - you can restart each level as many times as you like.

This particular level isn't that hard once you know what you're doing.

In fact, completing the individual levels of the main game is not intended to be hard - anyone should be able to play with a little time and application, and progress is automatically saved after each completed level, so you can switch off and stop playing whenever you like.

The levels are short - taking just a few minutes to play. They are designed to be easy to complete - but hard to complete well. The number shown in the top right corner is a count of the number of blocks burning at any given time - the advanced player has the secondary task of working out how and when to start fires in order to get the highest burning block combo.

This is a game designed that anyone can pick it up and have fun with it - just watching the world burn is entertaining and fun. Although in early stages, we're really pleased with the way it's shaping up.

Fireball Needs You!

We are hoping to incorporate a number of levels designed by people outside of the games industry, so this is an open call for modders, amateur level designers, students and artists of all kinds to get involved in the level design process. Experience is not a requirement, and there is no minimum level of contribution. People currently employed by video game companies should not apply - there are legal issues. Your levels can be challenging, fun, pretty, strange... whatever you like!

Everyone in the external level design pool who contributes at least one level which makes it into the final game will recieve a level design credit. In addition, we planning to  set aside 10% of the royalities of the game to be divided between all of the level designers according to their contribution (so if you contributed 20 levels and the final game has 100 levels, you'd be in line for 2% of the royalties). Don't expect to get rich, though - this is a small project, and most video games never actually see any royalties. Still, if the game is a surprise hit, you would get your fair share.

Interested? Click here to email us [note: link now disabled]. Send us your name, age (no under 18's - sorry!) and no more than 100 words of persuasive text about your prior experience or anything else that might make us pick you to join the external level design team. We'd love to hear from you, but we can't guarantee a reply immediately so please be patient!

What happened to this game? Read the Play with Fire post-mortem from March 2008!

ISO Standard Game Design?

Are there any elements of game design which are sufficiently universal that something akin to an ISO standard could be developed for  game design? Perhaps an ISO standard for mass market game design - that is, a description of necessary or explicitely undesirable elements of game design when targeting a genuinally arbitrary audience.

If this were possible, it would be more in the style of a Technical Report i.e. an ISO TR, although I don't propose actually doing it - life is too short. Rather, I am interested if there are any elements the game design process which transcend issues of audience. We'd be talking about limiting factors and useability, primarily, and not about what people like or dislike.

My instinct is that there probably are a few elements that could be specified. For instance, targeting an arbitrary audience implies the need for a low dimensionality of control  (i.e. simple controls). But even this might be insufficient - we may find, for instance, than the direct feedback of the Revolution controller is an easy to operate data entry system with an inherently higher dimensionality of control than joypads et al. Eyetoy is pretty definitively mass market and I couldn't begin to estimate the dimensionality of that control method.

The highly subjective nature of the play experience inherently makes the game design process hard to standardise. Although every different approach to game design can be informative, rules-based methods are markedly less useful than holistic approaches to game design, because only when the game and its audience are considered together do we get the full picture.