If you get a bunch of game designers together to talk about any sort of game system, it won’t be long before one of them asks the primordial question – “Yes, but is it elegant?”. Elegant is design short hand for effectiveness and simplicity, combined with a certain grace of application. Grace is a nice hand-wavey word that means that group of designers can argue until the cows come home about which is the most “graceful” of the options, but at the end of the day everyone agrees that elegance is desirable, and the most elegant solution should be the goal.
I wonder if that’s true, however. One of the things we’re good at is unravelling complexity, and there’s a satisfaction that comes with working things out that is deeply embedded in the human psyche. It seems to me the more I look at complexity (even wanton, needless complexity) the more it can offer a sense of learning and collective experience that in fact enhances the experience of play. Perhaps elegance isn’t always the optimum goal?
Sports offer a great example of games with complex and esoteric rule sets. The Australian Open is on at the moment, and offers a great reminder of the inherent zaniness of the rules of tennis. If you’ve got no points, that’s called love. Scores go love-15-30-40 and then game. Unless both people have 40 points, then that’s deuce. First to six games wins, unless they both get to five, then you keep playing. Which is the same mechanic as deuce, but with a different name (not very elegant, that). You play three sets, unless you play five. That’s not even getting into more esoteric elements, like challenging decisions, the role of the umpire, foot faults and so forth. Once you have all of those “basics” clearly understood, then you can start looking at the actual gameplay that takes place underneath them – positioning, choice of shots, and ability to make the shots one wants. This is for tennis, which is one of the simpler of the mass audience sports. Lets not even start on baseball stats or american football plays.
The interesting thing about all of the inelegance that hangs around the sporting world is that, although it does raise the barrier to entry for people new to the sport, it also gives people a shared language and experience. Once you’ve crossed the threshold of knowledge, you have meaningful minutiae of detail to discuss with other fans. In this case, complexity works for you, not against you.
Over in video games, we can see complexity working in the favor of everything from RPG’s, to the ever escalating systems present in FPS’s. In fact, genre’s of games regularly add complexity to previous games in the series or genre, on the assumption that a large element of their audience is familiar with what came before. This can work well (adding RPG progression to FPS multiplayer) or can eventually reduce your audience to a small number of experts, who continue to demand harder and more complex scenarios that alienate newcomers (like the various bullet-hell shooters).
Gameplay goes from this ...
To this crazed example of why they call them bullet hell games!
Let’s look at one of the triumphs of complexity, the Pokemon series. Not only are the game systems composed of a bunch of relatively esoteric relationships (check out the grid below to see how different types of Pokemon interact) the games grew more complex as they proceeded. Coming into Pokemon late in the series required a huge amount of learning, because it was assumed that you’d played and understood the predecessors. Nonetheless, once you’ve climbed that hill, there’s an immense satisfaction and sense of community – and it gives you something to discuss with other Poke-heads the following day.
If you think this is complex, try understanding optimal use of effort values. It makes the Spiderman Clone saga look simple.
So, while in any discussion about game design you’re always going to get someone who wants to find the absolute simplest solution to any problem, effectively devolving every game into a Go-like Platonic perfection, there’s lots of reasons to embrace complexity of systems and interactions. While complexity alone isn’t going to make your game interesting, feel free to explore the spaces where it can add to what you’re doing. This was certainly one of the strong points when we developed the character creator for Freedom Force. It was the initial goal (let people create their own superheroes, and make sure they can build just about anybody) that opened the flood-gates in terms of complexity. While we tried to keep the systems as simple as possible (with varied success) too clean a solution would have stopped people building heros that played and felt like the heros they were familiar with. In the end, we jury rigged and special cased around the existing systems to try and be as complete as possible – and ended up with a system that was one of the best parts of the game.
Fundamentally, I agree with that guy who said “games are a series of interesting decisions,” – complexity allows for additional axes of interesting decisions to be made, and therefore offers opportunities for people to learn, engage, and discuss the systems you’ve built. Elegant systems are a great virtue, but they’re only part of the picture.
There’s an excellent article by Malcolm Gladwell in the New Yorker discussing what he calls the Quarterback Problem. In short, there are certain jobs where almost nothing you can learn about a candidate will accurately predict how they’ll do once they’re hired. That, combined with reading Joel Spolsky talking about hiring good coders, led me to give some thought to the issues in the games industry. Years of hiring game and level designers, and being involved with the hiring of programmers, artists and producers has me thinking that there are some definite parallels with finding good people in games – and while there’s a lot of ink spilt talking about hiring good coders or artists, there’s not many people talking about how to hire great designers.
One of the main points made in Gladwell’s article is that the same is true of teachers – and the impact that a good teacher can make on their students is exceptional. I won’t try and suggest that good designers are as important as teachers – but they can and will make a substantial impact on the quality of what can be a multimillion dollar project. It’s worth trying to find great ones.
Yet, having sat through a lot of interviews (on both sides of the table) I’ve been largely surprised at the lack of precision involved in hiring designers. In fact, the higher the position being hired for, the less substantial the process seems – ultimately boiling down to hiring creative directors without requiring a design test of any kind.
Even so, is the traditional design test a great tool for finding good designers? For people who haven’t had to do one, most companies ask prospective designers to fill out a design test. It’s basically a document that asks the candidate to write up a spec for an example level (usually a level in the game or sort of game they’ll be working on). At Pandemic, we used the first Destroy All Humans! as our example for the design test, and asked people to design a mechanic, ask how they would bind together the saucer and on foot mechanics, then ask them to develop a weapon and scenario for the game, along with asset lists.
Now, this is a good representation of one of the tasks a designer will regularly need to do – documentation and nice logical spec designs are crucially important to ensuring that tasks begin nicely. As I’ve worked in software development over the years however, it’s become clear that this is really only the beginning of the job of a designer – and in many ways, it’s the easiest part.
Where a great designer really shows their chops is in their ability to simplify and convey the goals of the game, to implement exciting play, to respond to user testing and build things better, and to work with everyone else needed to get those elements into the game. This ability to collaborate and respond to the experiences players have while engaging with the things you’ve built (and to put yourself into the players shoes in advance) are much harder elements to test.
So what’s the best response? One of the suggestions is to throw the gates wide open, and see who pans out. That seems like a good idea (and one I’m at least partially in favor of) but it can be very difficult to put into practice under the normal time pressure of game development. Generally speaking, you’ll need to find designers and ship a game with them over the course of two years – and it’s hard to simply hire 20 people with the intent of cutting them down to 5 after a few months. That approach may fly in financial services, but it’s not so easy in game development – especially as the cost (in terms of the resources of the rest of the team training up new staff) of each new hire is highest during those first few months.
My first suggestion is that it’s absolutely key to know what you’re looking for. This sounds simple, but my experiences tend to suggest it’s almost unheard of with designers. People are looking for someone who comes across well during the interview, or who can talk about games, or who has nice hair … it’s rare that a studio has a codified set of requirements that outline exactly what they’re looking for.
So first, determine what makes a good designer for your studio. Every studio works slightly (or extremely) differently, so the requirements you have will differ. At Pandemic, I was hiring designers on the following criteria :
1. Collaboration : We had an extremely high focus on working with the team to achieve goals, and our designers were the ones ultimately responsible for putting assets into the game. So a key part of their job would be working with the artists to make sure it looked right, and the programmers to make sure it worked right. For our team, collaboration was a must.
2. Responsibility : Related to the above, we needed to find people who would take responsibility for the things they were working on. On our team, designers were both the first port of call and the last responsible for anything in the game – and that meant making sure that we hired people who would take pride in what they implemented. In a lot of ways, this is equivalent to the “Get Things Done” requirement in Joel’s article on hiring.
3. Insight : Understanding how game mechanics work, and what changes have which impacts on the actual game. From insight comes craft – the ability to create elegant structures, to make changes which simplify and improve the game you’re working on.
Now, with those three criteria in hand (while I’m not proposing these as universals, they’re a good start) I set out to try my best to find ways to test for them. It’s not as easy as you might think, and I’m not suggesting there are any magic questions that will give you a positive test as to whether someone has the requirements.
Nonetheless, armed with explicit knowledge of what I was looking for, it became much easier to ask questions and have a conversation around those areas. I can ask which games the candidate has played, and then (providing it’s a game I’m familiar with) we can start to talk about the themes and mechanics. Not only is this one of the things I enjoy, it generally gives me a sense of the way the person thinks about mechanics. In terms of collaboration, I’d ask about past working experiences they’d enjoyed – both to get a sense of what they’d done in the past, and also to see if it involved other people. If so, I’d dig deeper and ask how they worked with those people.
So generally, setting criteria and conversing around them can give a pretty good impression as to whether the people you’re hiring have the basic chops or not. The key to building a system that works, however, is to add some feedback to it. As a result, I’d keep notes as we went through each hire, and then return to those six months or a year later and reflect on what had worked, and what hadn’t. Over time, this allows me to refine and reflect on how the criteria I’ve selected have turned out – and whether it needs any tweaking.
Even with all this in hand, hiring great designers remains somewhat hit and miss – I know some criteria that make it easier to work well in the studio setup we had, and have some ideas as to how to search for people who meet those criteria. Nonetheless, I was often surprised (and generally pleasantly!) by the new approaches and insight the designers I was fortunate enough to work for brought to the table. I’m interested in what sorts of approaches other people use – and how that’s working out for them!
I spoke to Yug and Matt at the 48 hour game competition, which can be heard on their podcast http://www.australiangamer.com/podcast/176_live_from_48hr_game_making_challenge_2009.html
I also did an interview for Amanda Dell on ABC radio as to why exactly a pro-team would be interested in the 48 hour game competition – MP3 here.
So, the 48 hour game competition is now done with, and the level of entries was fantastic. I was especially impressed with the efforts of some of the amateur teams – amazing work from people outside the games industry, in crazy timelines, and with a surprising amount of polish. They also tended to sleep way less than us, who for the most part actually got some sleep in the middle there – we’re old men (well, some of us are) we can’t stay up for 48 hours straight making games!
On the nice side, our game “My Mechanical Romance” won the pro team competition. We used the three seed words (Giant, Storm, and Love) to tell a tale of Giant Robots who, after fighting oppression for many years, have finally been allowed to be wed. You control the forces of the wind in order to blow the clouds around Central Park, keeping them away from the Giant Robot Weddings and (once you get to Level 2) into the various groups of protesters who arrive.
It’s a simple piece, but I find the mechanic really interesting, and I’m fascinated with where we can take it. There’s a build below for people who’d like to play it on PC – I’d provide the Mac and Webplay builds, but for some reason they’re huge – I’m going to take a look at that tomorrow and see if I can’t get them smaller.
Edit : Large Intel Mac build now available for your delight!
So, enjoy our little piece of 48 hour pro-Giant-Robot-Wedding touch controlled fun : My Mechanical Romance!
My Mechanical Romance, a literal screenshot
Some quick instructions – hold down the mouse and drag the cursor to blow a breeze across the clouds. Keep the mouse held down for longer breezes. Keep clouds away from weddings, and over protesters. Enjoy having a play!
PC Build : My Mechanical Romance
Intel OSX Build : My Mechanical Romance (40 odd MB)
48 Hour Games Competition
I’ll be at the http://www.48hrgamecomp.com/ along with the gang from Curious Bear. Will be really interesting to see what we can come up with in two short days, as well as how everyone else does. In addition, this will be our first project using Unity and will act as a great test bed to see how that goes. I’ll provide updates over the course of the weekend on Twitter, and then a summary here once done.
Cluck It!
I worked with Curious Bear to help develop Cluck It! – a casual iPhone game of chickens crossing roads. It was accepted into submission last night, so we’re now live and avaliable! Go grab a copy (it’s only a dollar!) and let us know what you think. We’re pretty pleased with what we’ve accomplished, and we’ve had some great feedback so far. Some of it even from people who aren’t our mates!
Shipping an iPhone game is an eye opening process, really. While there are a reasonably large number of hoops to cross, it’s simply amazing how much easier it is than passing any console TRC process, in terms of requirements and amount of work. We probably went to submit about a week later than anticipated, due to finalling issues, but that’s still pretty good.
Cluck It! has some really neat features – I’m looking forward to discussing in detail some of the thinking behind the level design (my primary input into the game, along with the suggestion of keeping it to a one button interface). I was especially looking to build a game that, at the highest levels, captured the old school arcade experience of flow – but that was at the same time super casual. I think we achieved that nicely – especially once you get into the star rounds towards the end.
One of our early stages
There are a couple of major trends we’re seeing in the commercial side of game development at the moment (and for the last year or more), both of which have a pretty dramatic impact on the sorts of games that are being funded at the moment. Given the usual development schedule of between 2-4 years, this means we’ll really start to see the impacts of these strategies from the major players show up on shelves in the sorts of titles we have avaliable in the next 12-18 months.
So, what two elements am I talking about?
The first is the splitting of the market in terms of profits. While there used to be a decrease in sales between the number 1 selling game and the number 10 selling game, they tended to (historically) be in the same ballpark. No more. These days, a really enormous blockbuster (think a GTA, or a Halo, or a Call of Duty) absolutely dominates the charts. There’s a huge cliff between number one and number three at any point.
The second trend we’re seeing is the rental market. For every copy of a game that gets sold, it may get resold three or four times – usually at only 5-10% off the cover price! All of the profit from those sales goes to the Electronics Boutiques and Gamestops that are responsible for the second hand market. None of it goes back to the developers and publishers. Again this is something that’s been present for a while, but is rapidly becoming endemic. Often you’ll see second hand copies on sale the day after release! In fact, game stores offer serious incentives for you to purchase and return your copies as soon as possible.
What does this mean for publishers? It means they’re changing the sorts of games they’re making in response. The first point means they’re focusing on hits and putting all their energy behind guarenteed blockbusters. Which basically means we’re seeing huge campaigns promoting sequels and followups, as they’re safe bets. We’re also seeing a lot of studios following the EA Sports model of yearly updates. If you want to put your big cash behind the safe IP’s, then you want to make sure they’re coming out nice and regular.
Of course, it takes new IP’s to build franchises which can be exploited in this fashion – but that’s a different topic for a different time. It is worth noting that Ubisoft seem to have found an excellent balance however – there were a couple of years there where they were leaning very heavily on Prince and Splinter Cell to drive their sales, but that’s relaxed to a much broader strategy where they try to bring something new to the table each time a major franchise sees a new release.
Secondly, we’re seeing a focus on strategies that either reduce resale or get some cash coming in even if there is a copy is resold. In other words downloadable content – either free or paid for, and multiplayer content. The DLC helps to keep you playing (or grabs some extra cash from you even if you’re playing a secondhand copy) and multiplayer content is notoriously sticky.
However, the rush to multiplayer has it’s own perils – because multiplayer content is only sticky IF it’s got a huge community behind it. Which moves us back to the problem of following hits yet again – players flock to best in show multiplayer, but tend to ignore anything other than the best.
I don’t necessarily have an easy answer to the current hard questions facing the pointy end of the business. That said, I do think this makes clear a set of opportunities developers have these days.
First and foremost – take risks and lots of them. The potential reward is huge if you succeed – a new francise is a licence to print money, if it’s a hit maker.
However, take *cheap* risks. There are a lot of places to prove out a concept other than a 50 million dollar headline title. You can make a lot of smaller XBLA or PSN games for that sort of cash, and in the process find the ones that catch fire for you. It’s not as simple as that alone, of course – but we’re definitely seeing the payoffs for the studios that are focused on building IP rather than simply resting on their headliners.
Exercising the Creative Spark
Had a great chat with Janine Cahill of Future Journeys today in which we covered a great range of interesting topics. The one that stood out to me was creativity exercises and the various approaches to creating a workplace with a sense of fun and spontenaity.
I then brought up Brian Eno and Peter Schmidt’s Oblique Strategies, and the more recent example of similar thinking, Jesse Schell’s Deck of Lenses. Both of these represent a great way to integrate some new thinking and perspectives into your routines, especially in places where you might get stuck – either in a rut, or just stalled completely. There’s a difference in approach between them – but they’re more similar than they are different.
What I find most interesting about these sorts of approaches (apart from the ease with which they conceptually adapt to technological solutions) is the potential benefit to our thinking of adding a random element, especially when you’re working in a creative field. Of course, I tend to think all fields are creative fields, but that’s my bias showing through. Enforced randomness is something that’s hard to find in the whirl of day to day business, but there’s nothing wrong with planning fortuity. One of the things I do is add appointments to my calendar that say things like “Take a step backwards and look at things differently” – which is a great opportunity to utilise a tool like the Oblique Strategies, or to look over Noah Falsteins 400 project which even without an update for the last 3 years or so, still remains a great list of things to think about.
So, I’ve been playing a bunch of both of these open world super power games recently, and they make for an excellent comparison of choices that you can make in developing an open world game.
It’s a little sad that they’ve both been released at the same time, bringing the inevitable comparison and competition. Add platform exclusivity into the mix, and you’re guarenteed to get heated discussion on both sides. Nothing I write here is to outline ways in which one performs better than the other, merely the ways in which different choices take you to different places.
Prototype and inFamous are both excellent examples of the new breed of open world games, and the fact that they’ve both metacriticed in the mid eighties is a clear indication of how high the bar is when it comes to open world development. The amount of work involved in making an open world at this level of polish and amount of content is simply boggling – many times the amount of work involved in a classic FPS, or even a linear third person platformer.
Pacing and powers:
It’s worth noting that the classic open world game (GTA) doesn’t unlock new abilities in the fashion most games do. In fact, it’s avowedly open world, in that even those abilities granted by gaining new items (planes, missile launchers, etc) are available if you know where to go to get them. Both Prototype and inFamous take the more normal game approach, teaching the player new abilties as they progress both to add variety to gameplay and to allow them to learn as they progress.
inFamous utilises pretty classic platform game pacing. New abilites grant new power, which often grant access to new areas of play or movement through the neighborhood. With relatively few new powers (there are about 17 that actually provide new functionality, with many of those having three levels of upgrade) the overall game progress feels relatively gently paced.
Prototype, on the other hand, accelerates unlock pacing to absurd lengths. There are no commercial games I can think of that provide this many game changing power-ups this quickly. I haven’t had so many things unlock since Burnout Legends (and there, most of the unlocks were cosmetic and not functional). Prototype contains about EIGHTY different abilities (and that’s only counting ones that add a new function – not those that boost stats or damage) and hands them out like candy.
Combat and movement :
Both games feature great moment to moment gameplay, with really differing intents. Prototype has a hugely extensible “mayhem” focused combat system. In general the goals seem to be ultraviolence and freedom of movement, and it achieves this in spades. That said, there’s not so much in the way of tactical decisions throughout the combat, nor is this the goal.
In contrast, inFamous is a study in small scale combat. While you have a decent range of abilities that cause destruction, they are heavily limited in terms of usage (they deplete your powerbar as you progress). This, along with the cover system and fragility of the main character, all lead towards short term tactical usages of power, and pop and shoot gameplay much like many other current gen shooters.
As a result, inFamous has much better thought out combat puzzles. Environment is key to inFamous, both for combat and movement – in combat, it plays more like a Gears of War, and in movement more like a classic platformer (not surprising, given the developers pedigree). In contrast, combat in Prototype leans towards Devil May Cry combined with Spiderman. These choices define the sorts of tools the designers have to create interesting content – and as a result, each game plays distinctly differently, despite surface similarities.
Side Missions :
Prototype lost me very quickly with it’s side missions (called events) – each of the ones I picked up felt unmotivated (there’s no real dialog or cutscenes to set them up) and on top of that, they were unrewarding. This is as much down to my personal play style as anything else – I like to see variety, and I like rewards to move me forward. I tried out a few events, and in each case completed them without scoring high enough to be rewarded. This turned me quickly off them – there was a lot more instant gratification on the mainline game path, and thanks to the huge whack of powers you get anyways, I never felt I needed to grind the side missions to get more.
That said, I returned to the side missions after a few hours of play (and after upgrading my abilities) and found it much easier to achieve success and medals. They still weren’t as compelling or varied as I’d like, but nonetheless the difficulty felt much better placed once my powers were at the appropriate level. While new events get unlocked all the time (and I presume that the events unlocked increase in difficulty as the game progresses) there’s no way to tell which events are appropriate for your current skill / upgrade level.
In effect, the way events have been established against the framework of (the already discussed) power unlocks in Prototype makes balancing them delicately almost impossible. Each one has its own difficulty and power requirements, and may shift from too difficult to too easy depending on the load out the player currently possesses. In this situation, the player really needs tools to understand which are appropriate for their current skill level. That said, I suspect most players will follow my course, and simply ignore them – which isn’t actually a bad thing. Prototype is a better game in terms of content engineered to the players skill level without them. The difficulty ramp feels much more reasonable and the main plot line is put together very well.
I do get the sense that some elements of the main plot line are intentionally made overly difficult (for example the sequence where you have no powers, which represented a difficulty spike for me) in order to encourage you to go back and grind side missions for more XP. I’ve yet to meet anything that requires it, however.
Getting around to inFamous – largely a study in contrast. Side missions are set up with a bit of character work, and have a meaningful reward at each point – they actually change the properties of the environment you’re in (Hive and Military base sidemissions do this for Prototype – and as a result are the only ones I played). On completion, you’ll get a bit of feedback (again, from characters or watching the TV’s in the world) as to how people are starting to view you differently. While it seems like a small thing, this combined with the length and accessibilty of the side missions makes them appealing. It’s always worth five minutes to me to see some progress in the world – and the missions are so well checkpointed that I can quickly progress through to conclusion.
UI :
A quick note on UI. The generic “hi tech” interface used by Protoype I find generally unappealing, especially the small text in dialog boxes. The whole thing comes across as functional, and being functional I want to get past it ASAP. I find inFamous generally more appealing and interesting. A small thing, but I do think it has an impact on how willing I am to browse menus and sidemission screens.
Wrap up :
Mechanic decisions in an open world game determine (to a large degree) how exactly you can build content, and by so doing, what sort of play experience the gamer gets from your game. Prototype and inFamous are both excellent examples of this – with very similar (in theme and surface detail) set ups, they go to very different places in terms of moment to moment play experience.
Radio National
You can hear myself and Lee Sheldon talking about Serious Games on Life Matters from Radio National. Check the program for 11th June 2009 to hear us wax lyrical on games in general, and serious games in particular.
