Only a Game

Early Playground Worlds

Playground_1 When a child enters a new playground for the first time, they are assaulted by choices – swing sets, slides, see-saws, merry-go-rounds, climbing frames… the playground offers them toys (there is  no goal beyond personal entertainment) and they excitedly begin experimenting with the amusements on offer.  

The term ‘playground world’ was coined (possibly by myself) to describe the kind of games typified by the current Grand Theft Auto franchise in which the player is presented with a wealth of entertainments in a game context. The success of this approach is marked not only in the huge commercial achievements of these games, but also the extent to which companies attempt to copy this formula (often unsuccessfully). These modern playground worlds are expensive systems to build because, according to the obvious logic at least, they require enough different “toys” (vehicles, mini-missions, equipment) to provide the sense of choice and the illusion of freedom.

Recently, it has occurred to me that the kind of play indicated by a ‘playground world’ has been with us for a considerably longer time than perhaps I previously considered – although of course, it was much simpler in the past. 

Let us go back in time to one of the most incredible years in European gaming history… back to the 8-bit era of flimsy plastic home computers. Back to the year 1985.

Bbc_elite Elite (Firebird, 1985) 

David Braben’s Elite is considered by many to be one of the most remarkable achievements of the 8-bit era. (I met him briefly in a bar in London, introduced by one of my bosses who had worked with him at Firebird). I assume everyone is aware of the game, but for those who are not it is a basic space trading game in which you can fly to any system (or even to new galaxies), buy and sell goods, fight pirates, mine asteroids or go renegade and fight the police.

Elite was inspired by the seminal tabletop RPG Traveller (GDW, 1977) – it even included the same sample character, Commander Jameson. This is not a criticism, but it does serve to demonstrate just how influential the tabletop RPG games of the 70’s and 80’s were in the development of modern videogames. 

What makes Elite qualify as an early playground world? Of course, this is a subjective decision. However, the key elements as I see it are as follows:

  • Freedom to travel: you can travel to any system, explore a huge area of space, and different areas feel different largely due to simple changes such as altering the political      stability of a system i.e. changing the frequency and intensity of attacks on the player.
  • Choice of activities: although there are not a great many different activities available, there is some choice – one can ship cargo (the most reliable money spinner), mine      asteroids, earn bounties for defeating criminals, engage in piracy and steal cargo, or even attack the police and become a fugitive. 

Both of these elements were incredible back in 1985! We simply hadn’t seen anything like it, and Elite (for the BBC micro computer) entranced me.

Of course, it’s an old game, offering only wireframe vector graphics, but it remains remarkable. For instance, one can download a version of the original game for the PC and the file is a mere 97k! That’s vastly smaller than the help file for most PC games. 


160pxmercenary_pic_2 Mercenary (Novagen, 1985)

Also out in 1985 was another wireframe graphics classic for the Commodore 64, by Paul Woakes. This one, alas, has not aged as well. The screenshot here was the best I could find for the game. Yes, it’s just about that thrilling graphically. But remember, in 1985 most games expected you to perform a single activity, and also that 3D graphics were still amazing (even as wireframes) – this is only two years after the wireframe graphics of the (second) Star Wars (Atari, 1983) arcade game had wowed people with its “realism”. 

Since you are unlikely to have played this game, allow me to summarise: you have crashed on an unknown planet, and have to find a way to escape. You can move around on foot, or board any of several wireframe vessels to fly your way around. The world is vast, and contains many hidden areas, some of which contain wireframe corridors with more hidden vessels.

The key elements which made this game feel like a playground world were: 

  • Freedom to travel: the planet’s surface is a flat green plain with roads (white lines) and buildings (white frames), with many hidden buildings, and corridors inside some  buildings which can be explored.
  • Variety of Vehicles: the game was compelling at the time because each of the different vehicles had very different properties – and some were amusing, too. The Cheese, for instance, was one of the best vehicles in the game.

The game was a hit, spawning several sequels. 


Paradroid Paradroid (Graftgold, 1985) 

The last game in the class of ’85 I want to discuss is the classic Paradroid, by Graftgold stalward Andrew Braybrook (who I worked with briefly during the last days of Graftgold). This is one of my most fondly remembered games, and for many years was an all time favourite. Once again, graphics are not the strong point – you and every other droid in the game is represented by a circle with a number indicating the code of the droid in question. That said, the bas relief effects used were groundbreaking at the time and seemed really quite impressive compared to other games of the time.

Paradroidinfluence The essence of the play is that you are an experimental influence droid who has been sent to a ship full of malfunctioning droids. You must pacify the ship, which is to say, destroy all of the robots. However, what made this game exceptional was that you had two choices of ways to do this: you can destroy them with weaponry (but not all droids have weapons) or you can take control of them by playing an innovative ‘influence game’ (pictured) which remains one of the best integral minigames in any videogame to date. 

The dynamic of play is unlike any other. One can only control the different droids for varying lengths of time, usually related to their power level – which in turn relates to the droids code number. A pathetic 123 cleaning droid lasts a while, but can do little, while an awesome 883 battledroid (with devastating disruptor weapons) would not last as long. In between are many other droids, such as the 476 – a maintenance droid that happens to have a handy welding laser. The higher the code number, the stronger the droid would be in the influence game. The result is the player ‘climbs the ladder’, working from weaker to stronger droids – but ultimately must go back down to a lower droid for stability. One can take control of the top droid – code 999 – but not for long. Strong strategies revolved around the sequence of acquisitions and eliminations, although the game could also be played tactically with no forethought.

The key elements which made this feel like a playground world were: 

  • Paradroidmap Freedom to explore: the entire ship is open to you. There are no artificial barriers to prevent you exploring – although certain decks have many more battledroids on them, and are therefore harder to traverse. Deck transitions are via elevators on the map, but the feeling of exploring a ‘world’ (the ship) is palpable.
  • Choice of activities: admittedly only two – use weapons or control by influence – but still, the player is afforded choices.
  • Variety of droids: each droid feels like a completely different entity, with a different power level, and different weapons in many cases.

Paradroid was remade several times - by Steve Turner as Quazatron (Graftgold, 1986) on the ZX Spectrum, as Paradroid 90 (Graftgold, 1990) for the Amiga and Atari ST. These games discarded the convention of seeing droids as their code numbers, but the play of the games remained much in the spirit of the original. 

The lack of a save game in the original makes modern emulated versions all the more playable.


Other Games 

I’m certain there are other games from this era for other platforms that I am either not aware of or am overlooking that also fit this criteria of an early ‘playground world’.

Ultima IV was released in 1985, but I think does not quite meet the criteria by requiring the player to solve puzzles to progress – freedom of exploration is curtailed structurally. However, I have not actually played this game and I may be doing it a disservice in characterising it in this way. 

Lords of Midnight (Beyond, 1985) is another candidate from the same year – in this one the player has the freedom to explore and has a choice of activity (quest or fight; negotiate or defeat) and a variety of troops – you can recruit a large number of different allies in parallel. However, it’s turn based play feels markedly less like a playground, and the variety is not a choice the player makes but an accumulation of more and more options.

I welcome suggestions for other games which might be considered as early playground worlds, especially those from 1985 or earlier. 



In 1985, it wasn’t possible for the most technically advanced games to support wide scale activities because the high tech games were in the arcades, and had to be designed to encourage players to put coins in slots. On the home computers, pioneers like David Braben, Paul Woakes and Andrew Braybrook considered new ways of structuring play, taking advantage of the fact that in this context players could play for as long as they liked. 

These early playground worlds are remarkable because they created a kind of open play from exceptionally limited resources. They showed that the elements that make a playground world interesting and compelling need not be expensive – provided the game itself is simple and cost effective. These elements appear to be:

  • Freedom of exploration
  • Choice of activities
  • Variety of acquirable avatars (vehicles, droids etc.)

Any game which meets two out of three of these general conditions seems to feel like a playground world. However, freedom of exploration must not be constrained by any kind of overt ‘hoop jumping’ or it would not represent actual freedom (although partitioning of later areas is probably acceptable). Also, variety of avatars should perhaps reflect the choice of a single instance, not an accumulation of units in a strategic sense. 

I played and enjoyed all three of these games in 1985, and all three have exerted a certain influence on my game design work. I consider it incredible that they managed to achieve so much with such limited resources – an example of technical limitations forcing greater creativity. By examining how games with so few developmental resources managed to support a kind of play akin to the freedom and excitement of a child’s playground we can perhaps learn how we can create the same unique experiences in the games we choose to make today.


Feed You can follow this conversation by subscribing to the comment feed for this post.

Personally I've always found it very hard to get into these "sandbox games" as too much choice often means there's little focus to what you're doing and so it lacks the drive and short-term goals that more structured narrative driven games do.

Though I know there are millions of people who'll disagree with me because if I think if you take any modern MMORPG like World of Warcraft or EVE: Online, you've got all the open ending exploration and possibility you could ever really need.

So although I'm not really a fan of the genre, I think the kind of experience you're talking about has evolved really well into the current stable of MMO gaming.

I think the argument you're making here is that playground content is possible to do "cheaply" if the basic system is robustely interactive to the point where such content can be largely procedural. Hopefully, this is just what I'll find building characters with the Rocket Hearts engine. Exploration will be constrained in the resulting game (Magic Circle, if we get that far) but there will be a choice of activities and a variety of avatars, the recombinance of which will add much to replay value.

Mr. Perrin's comments about an intrinsic duality between sandbox and narrative is interesting, because it represents a common attitude that I'm excited to be trying to disprove.

Speaking of sandbox games, I thought this was a pretty good sandbox game idea:

Snow Day (game idea #9): An open-ended sandbox game from the perspective of a kid in a small town covered in a blanket of snow. You can make snow men, have snow fights, and (if you want to quest) help adults do various tasks for rewards. May we also recommend an online MMO portion alongside a crafted single-player adventure. And, if you decide to use more contemporary music (bad idea), consider Minus the Bear's "Hooray" (good idea, and the lyrics are related to snow fights).

Hi all; thanks for the comments!

I'd like to start by suggesting that 'playground world' and 'sandbox game' are not necessarily synonyms, although they do share in common the capacity to support open -ended play.

The typical definition of a sandbox game focusses on three elements: neverending gameplay, freedom to experiment and non-existent (or at least non-linear) narrative.

I think also sandbox games imply the capacity to create one's own goals and play (as with Sim City and the Sims), which may be present in a playground world but need not be.

While the freedom to experiment might be seen as a common factor with what I mean when I say 'playground world', 'neverending gameplay' is less certainly suggested.

Lastly, I personally use 'sandbox' to talk about situations where the risk to the player is non-existent - i.e. Sim City and The Sims. This is not the case in the games mentioned here! Death is a very real component of these games.

Elite has neverending gameplay, but it still has an overarching goal: to reach the ranking of 'Elite'. (In practice, though, this is not why or indeed how most people play this game).

Mercenary has a plot and a spine - it does not have neverending gameplay of any kind. In fact, it has a semi-linear plot which sends the player from location to location - much akin with GTA. In fact, I would say it is a spiritual precursor to GTA.

Paradroid is a very goal oriented game in some respects - like many shooting games, you have to clear levels. In Paradroid, a level is a deck of the ship, and when it is clear, the lights go down - creating a satisfying sense of closure. The unusual feature is the capacity to clear all the levels (decks) in any sequence - which provides tactical and strategic options not normally present thanks to the influence game. I do not believe anyone would consider Paradroid a sandbox game - but I suggest it can be considered a playground world.

My interpretation of Richard's comment is that he does not like sandbox games because he prefers to be part of structured play i.e. a narrative or goal-oriented structure (not an uncommon attitude). These structures can and are built into playground worlds. Witness GTA: there is a spine to be followed; one can ignore the opportunities of open play if one wishes. By all accounts, many players enjoyed these games in this way.

A playground world can support a linear or goal-oriented structure (Paradroid, as mentioned, is goal-oriented around clearing decks; Mercenary is goal oriented around the narrative goal of escaping the planet). This, perhaps, is the key divergence between the notion of a sandbox game and a playground world: a playground world might be used as a sandbox, but it may also be used as a basis for goal-oriented play.

A playground world, in my view, supports diverse approaches to play. Ideally, it should support a goal-oriented and/or narrative structure as well as an open-ended 'experimental' aspect to its play. Catering for different play styles in this way enlarges the potential audience such games can entertain.

(Richard - I would be interested to hear more from you as to how you relate to, say, Morrowind or GTA, and examples of the sort of games with structures you enjoy).

As to Patrick's suggestion as to the argument I'm making, well, I didn't really have a purpose when writing this other than to explore these games which struck me as early examples of playground worlds and see where it took me. :) But it must be said, looking at how older games managed to create more diverse play can be of great benefit to indie developers - the older games were limited by technology, and indie developers inherit these limitations by having limited budget.

It is definitely the case that I am suggesting that it is possible to create playground worlds cheaply under certain conditions - although not just procedural, as with Elite, as Paradroid and Mercenary have no procedural content at all.

And lastly, thanks for the link, Erik; I find it amusing that someone is trying to do a game concept a week. It's a perfectly achievable goal - but what does one do with a catalogue of 52 game concepts? (I have notebooks and notebooks of game concepts gathering dust, as do most game designers). I'm not sure what he hopes to gain from doing this, but if it gets people thinking about different kinds of play then I doff my hat to him!

Best wishes!

I enjoyed GTA to a degree, and Morrowind probably less so. The problem with putting a linear narrative to follow in an open world is that the two things don't fit very well.

I've been playing a bit of Oblivion recently and when you're following an "end of the world" story and go off to chase after deer for a few hours before returning to the fight you've lost any control of pacing in what could otherwise be a quite thrilling story. In Morrorwind it felt even worse because if you forgot what you were doing you'd end up wandering around incredibly generic villagers full of people repearing common phrases while you're trying to get back on the progression path. All too much stuff to remind me that I'm in an open ended game and so I have to work harder to enjoy myself.

When I play GTA I generally follow the linear story path, because when I try doing other missions it just feels like I'm doing a handful of irrelevant mini games until I'm ready to get back to the "real" game.

The thing is despite my earlier comment I'm not fundamentally agaist these kind of games but in my experience of playing open ended games you've just got so much choice and it's hard to know if you're going to run off and find something exciting or just spend hours going through incredibly generic or procedural missions so you might as well just follow the original linear path if you want a more interesting experience.

As I said before I think the only place where this has been dealt with well is in MMORPG games. Eve Online is the real successor to Elite, you've got a wealth of choice what you want to be in a wide galactic economy. Also the fact that wherever you go you're running into real players means it doesn't suffer from the generic conent problem. Of course in an MMORPG you can never really be the hero, the saviour of the world.

In terms of single player games though I'd love more choice, more freedom but every approach to this I've seen to this either has one "ideal" way to play the game where you get more out of the game, or the game is ultimately just unstructured and nothing you do every really advances or changes much.

So maybe I want something like GTA where you can join the police, or any random gang and the story and experiece you follow will be equeally rewarding and not just a generalised series of missions. But that just means the creation of masses of content for the player's every whim.


I certainly didn't take any of your comments to mean you were "against" anything, but you did strike me as someone typifying a goal-oriented focus to play and as such I was very interested in your views on this subject. I'm always seeking new case studies, new data... :)

Regarding GTA, the beauty of the set up is that there is a tight linear spine for a goal-oriented player to follow, but there's an open world for a process-oriented player to have the freedom to explore. I think this structure worked fantastically in isolation (despite minor flaws), and undoubtedly increases the audience size for these games. Expensive, though, as we're all well aware...

I agree with your supposition that creating a rewarding experience out of any element in an open world requires masses of content creation, which in turn is expensive. For Reluctant Hero (which may yet live!) I was exploring the idea of making the different elements of the world effectively procedural, rather than mission based, to see what effect that had on the feel of the open world. We may yet find out. :)

May I ask, would you characterise yourself as being tolerant of repitition in games? Do you find yourself, for instance, tackling a tough challenge several times in succession before earning victory? Do you ever become obsessive about beating a tough challenge? What about recurring play elements in, say, cRPG games - such as powering up or purchasing equipment - that occur often? How do you feel about this kind of repitition?

Please forgive my public inspection of your play style, but alas, I cannot resist exploring! :)

Best wishes!

I noticed that most (all?) titles being discussed here imply physical / geographical exploration of some some sort. I do feel, at the same time, that Playground experiences are not only about that, but also about material manipulation of some sort, like Erik's idea illustrates.

That's why I would like (as usual) to add early Jeff Minter's titles to the list (like Psychedelia and Colourspace).

I remember describing these titles to folks back in the very early 90s as examples of the next generation of game design. When people rightfully discuss the joy of procedural content, it is fitting to bring up David Braben's early and essential contributions. Spore has one galaxy? Elite had several if my fading memory serves me.

One additional game I'd add to the list is Nethack (1986) and it's early varients such as Rogue (1980). To me these are fore fathers of games such Oblivion.

What is interesting is how the specific genres of Paradroid and Elite never really took off. There have been sporadic open space sim games throughout the years, but only as a minor genre. The whole science fiction parasitic robot genre never really got a foothold though the mechanism has been used in other titles from time to time.

I always come away with the thought that playground games are hard to replicate. They tend to involve large amounts of emergent gameplay with lots of little systems interacting in intriquing ways to create a broad and appealing gamespace for players. These emergent systems are easy to screw up and hard to build in the first place.

Why aren't there more successful GTA-style titles on the market? Some of it has to do with GTA being a genre king, but a lot of it has to do with the quality of the competition. In a playground title, it is hard to identify and duplicate the exact formula that makes the game fun so you end up with crippled copies competing against an actually enjoyable title.

take care

Chico: I think the term 'playground world' implies the presence of a world... Minter's "light synths" have no world, so I wouldn't use this term. But one could stretch the term, of course, to cover other games (playground games, perhaps). Nice to mention Minter, though, while we're in the right temporal neighbourhood (Psychedelia was 1984, Colourspace 1985). Minter's 'nongames' were especially remarkable because no-one else had even considered taking 'games' in this direction at this point!

Danc: Elite did indeed have 8 galaxies. There wasn't much difference between them, though. :)

I'm not sure that Rogue/Nethack qualify as playground worlds, although there's always room for discussion. Taking the three criteria mentioned above, I'm not sure any of them apply to these games - there's no freedom to explore (because the levels are encountered effectively in sequence, I think), no variety of avatars and no choice of activity (it's fighting all the way!)

Paradroid and Elite had a distinct disadvantage in reaching a wider audience: they're both narrowly positioned in esoteric sci fi contexts. Elite has the additional disadvantage of 3D controls, which is a big audience narrower. Of the two, the Elite style setting has a better chance of surviving than Paradroid because robots have an exceptionally narrow audience, whereas space adventures have a curiously romantic appeal to some people.

Emergent systems can be hard to build (more commonly: hard to tweak) but these early games show how to get around this problem by being *simple*. It's something most game designers don't seem interested in (for reasons we might even be able to predict).

Your assessment of GTA's market advantages strike me as sound, and it certainly is the case that many clones fail because of a failure to identify the strengths of the original(s). But the other reason they tend to fail is that the original (a) built it's playground world format originally in a cheaper form (2D) then refined it into a more expensive (3D) version, giving time for the format to evolve organically and (b) has acquired a development budget the size of a small African nation! :-D

Anyone seen a statement of the development cost of any of the last three GTA games? I've looked, but never found.

Thanks for your comments!

Ah, Nethack. That is a game that permanently has its hooks in my soul. At one point, I had to rifle through the hundreds of pages of spoilers in order to beat the darned game, just once.

Nethack is all about exploration. Every corner is new territory and as you advance, you end up uncovering a large open world whose resources must be managed carefully in order to advance.

In terms of characters, you do play a single character class, but how you play can vary greatly. The ability to swap bodies only exists through polymorph and it isn't usually the most recommended option. Instead, you swap out different armor, which customize abilities and such on nearly a per verb scale.

I've always made the distinction between open/playground games and scripted games in terms of the scope of designer intent. The designer, in effect, gets to play god with the game player. How often does the player feel the icy hand of designer intent?

In most early games such as Space Invaders or Pacman, the player is given a relatively short term goal with highly constrained resources. "Eat all the dots." As soon as they've accomplished the goal, the designer swoops in and moves them onto a new challenge. "Now eat *these* dots."

In a more open game the designer interferes with the player less often. Arguably, they show the player trust and respect. Paradroid, Elite and NetHack all fall into this mold. They say "Here is a world. Here are a mess of tools. There's a goal in there someplace too if you really want one. Now go at it!" And then they go away, often for dozens or even hundreds of hours.

To me, that is the essense of a playground game. It gives the player the freedom to do what they wish. To fail, to succeed, to come up with exceedingly clever solutions to unexpected, unpredictable problems in ways that are untainted by Big Designer looking over your shoulders.

Admittedly, such open games are not for everyone. Final Fantasy (and oddly enough many religions and governments) share a scripted cradle to coffin philosophy that is highly satisfying for many. Free will is a great and powerful design idea, but I'm not sure that exercising free will is widely practiced or even enjoyed by many modern players. :-)

take care

Thanks for that description of NetHack; although I've had it demonstrated to me, I haven't played it myself, so my experience is decidedly second hand.

Best wishes!

You should check out the latest Space Rangers 2 demo ...

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Your comment could not be posted. Error type:
Your comment has been posted. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.


Post a comment

Your Information

(Name is required. Email address will not be displayed with the comment.)