I've just had the pleasure of receiving a response from the project manager of Subclipse, Mark Phippard, regards my off-the-cuff bitching about an installation issue with the product. This follows my previous bitching about a breakage using Technorati, that was responded to and fixed by a community manager.
This makes me wonder if these responses are because bugs can be closed while blogs are forever. Is publicly griping about a product a better way to get a response than privately going through the intended support channels?
Is blogging going to become the best channel for moaning about stuff? (Is it already?)
Tuesday, 29 July 2008
I've just had the pleasure of receiving a response from the project manager of Subclipse, Mark Phippard, regards my off-the-cuff bitching about an installation issue with the product. This follows my previous bitching about a breakage using Technorati, that was responded to and fixed by a community manager.
Monday, 28 July 2008
In case you missed it, my latest The Amateur column is up on GameSetWatch, suggesting as gamers that we perhaps worry about whether games should be taken seriously too much. And a creepy clown picture...
And because I mentioned the words Artificial Intelligence, the content hoovers of aigamedev.net sucked it up and spat it out as a part of their weekly article summary. Their summary makes me sound very irregular...
Aside from the fact that it thermally shuts down my laptop every time I play the game for more than 10 minutes, Civilisation IV is a great game.
At least, I'd like to think so. My understanding is that Soren Johnson 'single-handedly' redesigned the somewhat clunky Civilisation III interface to stream line the amount of interaction required to achieve world domination, whilst retaining the 'essential' game play. Of course, not streamlined enough for a console, which is why Firaxis has seen fit to release Civilisation Revolution, which at least gets away from having to develop a separate Archery technology from Bronzeworking.
And that is my basic problem. The technology tree. I'm not especially interested in the differences in the very convential technology tree on offer - the order I learn Engineering against Monarchy is not of the slightest interest to me, provided that the game demands I learn both. Sure, I can fail to advance a particular technology beyond the dark ages, but all that guarantees is that every lesser civilisation will offer it as a trade item every turn in return for my continued understanding that we are friends.
And where the technology tree fails to deliver.
I want to break convention, to escape the mundane and predictable advancement across the middle ages from swordsman to war elephant to Renaissance pike. I want to experience a world where no one discovers the Bernoulli effect - which confused even Einstein - restricting modern civilisation to lighter than aircraft: zepplins, dirrigibles, balloons, (even the words sound beautifully steam punk, rolling off the tongue). I want to plot a path through civics where genetic relationships are more signficant than social bonds, democracy is never discovered but sibling hive minds and the social glue of naked mole rats and of termite societies holds sway over the realms of men.
I want to end up in the year 2000 with flying cars and rocket pistols and jetpacks as common technology modes, and the mundane existance of petrochemicals, solar power and hydro turbines just a whisper in high Aztec society.
I want a technology tree with exclusions.
In my continuing quest to comment on the process for trying to get a SVN-based C compilation process up and running on Windows, I see that the Eclipse install process I wrote is horribly broken again.
What is with open source projects arbitrarily changing the user interface required to get them installed and working. Do developers not actually want people using their tools?
This time around, having followed the latest installer documentation as best I can, I'm getting the useful message: Unable to load default SVN client, which given I've just downloaded and installed Subclipse should include an SVN client.
Thursday, 24 July 2008
Thanks to the 49 people who voted. This poll was as much about how computer games and user interface have usurped tradition. It also suggests that about 20% of people would like to see a roguelike based on a fantasy role playing game other than Dungeons & Dragons (RuneQuest anyone?). The faints reference is to the roguelike Pokemon Mysterious Dungeon series, where you never die, just faint away.
Critical injury prone
Next poll: What would help you contribute to the Procedural Generation wiki? I'm always looking for contributors, and I'd like to find out what's holding you back.
Wednesday, 23 July 2008
Feel free to carefully peruse the publicity shot of Halo Wars to the left. Then notice the jarring cliche of laser beams fired with discrete bolts, as opposed to a beam connecting the firer to the target that hits virtually instantaneously.
I blame Star Wars and the rubbish Industrial Light & Magic special effects for starting this - a movement that crescendoed in the original Transformers movie which features a long shot of a metropolis at war where laser bolts curve over the city, like artillery fire.
I want fidelity in my science fiction, preferrably written and designed by someone who knows what little c is. Otherwise you're just playing a fantasy game dressed up with robots.
James Cameron, come home.
Sunday, 20 July 2008
I want to point you in the direction of a brilliant article on dynamic narrative in Far Cry 2 (Redefining Game Narrative: Ubisoft's Patrick Redding On Far Cry 2) recently posted on Gamasutra that you should read if you are at all interested in this field. While Far Cry 2 may not interest you, Patrick Redding says he's trying to sneak sophisticated narrative techniques in through the FPS back door, and he suggests a complete separation of the roles of writer and narrative designer, which I hadn't thought of, but seems to make sense in the context of big budget commercial games.
What makes this story so interesting is the comments section, which seems to have brought out commentors on the state of the art in dynamic narrative including Lee Sheldon, author of Character Development and Storytelling in Games, which I hadn't heard of, but I suspect I'll be purchasing soon.
Thursday, 17 July 2008
Wednesday, 16 July 2008
Results for 'Which game featuring procedural content are you the most looking forward to?'; new poll
Should have figured: read a roguelike blog - vote for a roguelike. Thanks to the 50 people who voted. The final results were:
Far Cry 2
Infinity: The Quest for Earth
S.T.A.L.K.E.R.: Clear Sky
The new poll is distinctly hit point themed.
Sunday, 13 July 2008
(You'll probably want to read parts one, two, three, four and five before starting this).
Damage is usually modeled as damage per second, or in a turn based game, damage per turn. DPS can be used to gauge the relative power of two different weapons over time, that is, how much damage on average is delivered by each weapon over a ‘long time period’. DPS is the traditional method of balancing attacks, by ensuring that the DPS of similar threats is approximately the same. I strongly recommend you read Craig Perko’s article on landscapes and level design, which discusses how DPS and environment interact to form game behavior. In this article, I’ll be discussing damage as it applies to attack abilities with infinite range, which are 100% accurate and hit immediately: I’ll then discuss to how to modify attack abilities in part seven.
In a hit point based game, maximum damage can be far more significant than DPS, particular as the threshold of 1 hit point is alive, 0 or negative hit points is dead, is such a significant game component (particularly in games featuring permadeath). A weapon with a high immediate damage but low damage per second may be capable of eliminating an enemy instantaneously, without the enemy being able to return any attack at all. So trying to balancing high immediate damage with a long cool down timer or high mana cost (or any other method discussed in part 3) may not necessarily be effective. And the greater the restrictions on use, the less likely the player will be to remember to use the ability, or to remember that a particular enemy has this ability.
It may be better to model damage as a sliding scale, in which what percentage of enemy attacks will eliminate the player from the game. Ideally at 100% of player hit points, no immediate threat should kill the player – any such attack should allow 1 player action to try to avoid a second attack of the sort. As player hit points decrease, there should be an increasingly level of threat to the player, as the percentage chance of elimination rises. Of course, this ideal of requiring two attacks to kill the player is virtually impossible to prevent in practice, as the player may be hit by two attacks simultaneously, or in a turn based game, two attacks before being given the opportunity to act next.
This suggests that in a single player game, the function of damage against the player is more important than damage delivered by the player.
With a fixed amount of damage delivered by an attack, the chance of the player being killed in a single attack is either 0% or 100%. This does not provide an interesting enough a risk curve for the player. When the damage is a random value between 1 and a particular percentage of the player health, the risk curve becomes much more interesting.
Where the maximum possible damage causes 100% or more of the player’s current health, there is a high probability of death, which is to be expected.
But the percentage chance of death for damage from maximum damage that is a low percentage of the player’s total health behaves as follows. On turn x-1, where x is the player’s health divided by the amount of damage (where x > 1), there is 0% chance of the player dying. On turn x, however, there is a 0.01 ^ x chance of the player being killed – that is a small, and unlikely chance, progressively more unlikely the smaller the percentage of player health that the damage is. However, on turn x+1 and higher, the chance of player death grows dramatically more frequent, as the bell curve defining the total potential damage to the player is shifted to the right.
Damage which is distributed in a statistical probability that models the way that the damage adds up over multiple rounds is the best way to train the player to understand this damage accumulation process. By this I mean, the damage from a single blow should approximate the distribution of damage from multiple blows. A bell curve, such as modeled by summing several random values from 1 to n, intuitively teaches the player this long term damage mechanic. Coincidentally, this can be most easily be represented by rolling several n-sided dice and adding the values together, a method which has been adopted from Dungeons & Dragons.
Damage against enemies should be modeled the same way as damage against the player so that the player is given the opportunity to recognize the underlying behavior. This also addresses what Craig Perko identifies as the predictability of enemy encounters: it should be possible, but unlikely, for an enemy to sustain a much greater number of player attacks than anticipated, or to be eliminated by a much smaller number. This has the unfortunate consequence of adding Skinner box like behavior to the game – where a rule set’s underlying behavior leads to a randomized risk-reward strategy.
Using damage per second, along with random damage fitted to a bell curve, the trade off should be a high risk/reward strategy versus a low risk/reward strategy. Damage per second itself is not a risk, so that all attacks should be balanced against the same DPS. A high immediate damage attack with a DPS balanced against other equivalent attacks, will tend to have a greater probability of eliminating an enemy immediately, which is a high reward, and consequently should entail a high risk. This risk can take a number of forms, which are usually based around putting the player at increased risk of counter attack. Examples of this risk include: having a period of time where the player is unable to attack at all, and therefore no threat to the enemy; limiting the range of attacks to require the player place themselves at more risk; or forcing the player to remain motionless or move at a reduced speed while attacking and are featured in a number of games.
Having multiple attack choices complicates the DPS balance. In combat, a player could switch from an initial use of a high immediate damage attack to an alternate attack method, which increases the overall DPS that the player is able to deliver. To avoid this requires a common cool down timer to all possible player attacks, which may not be appropriate for maintaining suspension of disbelief (Such as sprinting and the flashlight using a common battery in Half-Life 2). It may not be a requirement to avoid this, of course, as the intention may be that the player always attacks with the high immediate damage attack, where possible, or that the counter attack risk discussed above is sufficient counter-balance.
A common alternate method is to tier the player attacks, so that only the first tier has the top DPS, and restricting the player to a single first tier attack. A second tier of attacks can be balanced against a lower DPS rate, without significantly imbalancing the game. The quality of second tier attacks should be dependent on the immediate damage caused by the top tier attack – this can be used to calculate a revised DPS based on a single top tier attack, followed by multiple second tier attacks.
In fact, the player can be given multiple tiers of attacks, from a high DPS to a low DPS, which can be used to trade off time against another game resource. This implies a three way trade off for choice below the first tier attacks: high immediate damage, high DPS, and highly efficient resource utilization. Note that the highest immediate attack can also have the highest DPS in this equation, so it is not a strict three-way trade off, but if this is the case, the next tier down should be much weaker, to ensure the effective DPS remains balanced.
(Note that by balanced, I mean balanced against another theoretical set of abilities in the game. This does not necessarily imply that the player can choose different sets of abilities, but using tiering strongly suggests it. More in parts eight or later).
With the complexities that damage per second vs. immediate damage cause in a game, it is surprising that games seek to complicate damage even more. But inevitably they do. In parts seven and eight, I’ll discuss several complications common to many games: elements, attack shapes and damage over time.
Friday, 11 July 2008
My initial thoughts on Valve's heavy weapon alternatives.
Melee weapon: Heavy on rye
The heavy on rye halves the rate of fire of the heavy weapon's guys punches but the alt-fire allows him to consume a nutritious sandwich; giving him health up to his regular maximum. The heavy carries 4 sandwiches and they have a low rate of fire and a relatively large cool down preventing other weapons be used. Recharged at regular ammunition dispensers and cabinets.
Advantages: Less dependent on the medic for combat recovery. Re-use existing sandwich assets.
Disadvantages: There is little disincentive to use this weapon over the regular heavy melee weapon.
Secondary weapon: Heavy heavy fuel
Heavy heavy fuel loads the shotgun with fuel pellets. Fuel pellets do much less direct damage, but do moderate splash damage, and more importantly, ignite explosives including sticky mines in the vicinity of the impacting shot, causing them to explode immediately, damaging any enemy in the vicinity.
Advantages: A heavy who wants to advance on a sticky minefield has limited choices. This makes the heavy useful for quickly clearing these fields, and keeps them at the front of the action.
Disadvantages: There is little disincentive to use this weapon over the regular heavy secondary weapon.
Primary weapon: Really heavy weapon
While firing this weapon, the heavy weapon guy cannot move - but equally they cannot be pushed around by turrets or enemy explosives. The weapon incorporates a shield which protects the heavy weapon guy by halving damage to the front 45 degrees, excluding sniper head shots while moving or spun up.
Advantages: The really heavy weapon is a variant of the 'give the heavy main gun a shield' - intended primarily to allow the heavy weapon guy to wade through multiple turrets, and present a front line of defense without requiring a medic.
Disadvantages: Fixing the heavy weapon guy in place makes him more vulnerable to grenades and well-aimed rockets. If necessary, make having this weapon prevent the heavy from being overcharged by the medic.
Also see Tom Francis' Tantyana or Deep Plaid's Vacuum Gun.
I read a lot of game design blogs. I also happen to be in the middle of writing an article based around designing special abilities for games. And I also saw that Team Fortress 2 had started up a blog.
No - I turned up at the TF2 blog, saw that they hadn't got an RSS feed, and walked away.
As a result, I missed Robin Walker publicly soliciting advice for the Heavy's new power ups. No other game designer blog I read mentioned a thing. And now, a week plus change later, they've had so many responses, they're overwhelmed.
The good news is that they've had such a good response, Valve will keep doing this in the future. But shame on my fellow game designer bloggers for not mentioning this great challenge and/or shame on me for missing out this time.
[Edit: looks like Valve is still taking suggestions. This is really a challenge about how to pace a fight. Ideas and discussion in the comments thread. As Robin puts it:
What other goals should we have for the Heavy update? Half the battle of good design is choosing the right goals and the right constraints. In A Heavy Problem, we specified a goal for you, and the constraints that had to be kept in mind. Try taking a shot at defining the overall set of goals that the Heavy unlockables should be trying to achieve. Watch out for the common circular logic trap of finding an idea you like and then trying to extract goals from it. We find it's best to not think about ideas at all at this point, and to focus entirely on more abstract goals. What are the biggest problems in the Heavy class? In what situations is it the least fun? Like idea evaluation, goal generation is a big topic in itself, and one we'll go into in a later post.]
[Edit 2: Fifty pages of web fora discussing the problem for you to read here.]
Tuesday, 8 July 2008
For those of you who contributed to the comments thread from my Minesweeper vs. Solitaire discussion, you may be interested to find out that Minesweeper cheats to make the game easier. [Edit: Nothing to see here. Someone else has pointed out that Minesweeper only cheats in the way I originally pointed out - by ensuring that you never click on a bomb as your first move].
Monday, 7 July 2008
2008 continues to take people who have profoundly affected my thinking. Camp Concentration was and continues to be my favourite novel. And the world has lost another original and challenging voice...
We're half way through 2008, and Grand Theft Auto IV and Metal Gear Solid 4 have achieved huge commercial success. But despite the initial critical acclaim these games have received, more recent analysis of both franchises has started to sound a little... hollow. Insult Sword Fighting's Mitch Krpata reviews MGS4 with 'Old soldiers never die, they just talk a whole lot'. Junot Díaz writes for the Wall Street Journal that Grand Theft Auto IV is 'no Scarface', just more of the same. And Joystick Division's Gary Hodges sums the problem up best when he says of Metal Gear:
Playing through MGS4 in all its extravagant glory, I can't help but think of it as something like a Tyrannosaurus rex: the biggest, most extreme, most fully realized example of something that's ultimately an evolutionary dead end.
Part of the problem, of course, is that these games are part of established franchises which come laden with the baggage of having to meet the expectations of an pre-existing player base without straying to far from the fold. Part of the problem, as the Brainy Gamer points out, is that Hideo Kojima needs an editor.
On the gripping hand, it looks like that the modern game criticism process is fundamentally flawed. At the moment, the pattern for a blockbuster release seems to be overwhelming positive reviews prior to release, then a slowly building groundswell of disatisfaction followed by a wave of outright backlash, following the patterns laid down by Bioshock, Metroid Prime: Corruption, Super Smash Bros and others in the so-called golden year of gaming of 2007. But dare a game critic ride ahead of this swell: they'll either raise the ire of the fanboy masses or, as in Gerstmanngate, a publisher's wrath.
As Mitch Wahlberg of the Game Anthropologist highlights, many game critics are aware that something is rotten in the state of game criticism. Game criticism is tied up too much with commercial reality because too many game critics, particular those in the 'enthuisiastic press' as N'Gai Croal labels the gaming media, have a vested interest in creating buzz around a new property and are dependent on high review scores in order to get pre-release access to a publishers upcoming line up.
Equally, reviewers and writers in a position to not to have to depend on the outflow of this commercial sewerline are acutely aware of the infancy of the medium. Game criticism doesn't yet have the academic credentials of film and literary criticism, and the still-warm corpse of the comic book industry is a vivid lesson to what happens when a new medium fails to bridge the divide between niche and popular success. This leads to what the title of the article suggests, an endless search for the Citizen Kane of gaming, to validate and justify the medium as critically, as well as commercially significant. Games critics periodically engage the Roger Eberts of this world to justify whether games are art, challenge themselves to stand the greatest games of the medium alongside the literary or filmic greats and explore the personal and political significance of gaming to try to validate the time that they see 'wasted' playing games.
It may be enough to plot a path from the current state of affairs to one where game criticism matures as an industry, or perhaps fractures, into high and low criticism, academic review and popular opinion, just as criticism has survived in the worlds of film, music, art and literature. But that would be a failure of vision. Because games are unique amongst all those mediums. No critic would dare suggest that Monet should have used different brush strokes, Orwell different film stock, Mozart a different key or Shakespeare different lines - although they may discuss the choices made and how those choices impact the final work. But every game critic will at some point consider the rules of a game and how different rules could change the game play experience. The 'what-if' approach is fundamental to the play and by extension to game criticism. And the game critic therefore is as much a peer of the game designer, in fact, is the game designer in a way that they can never be a film maker, musician, artist or writer.
(In fact, not quite true. A critic is as much a writer as a novelist, and the theatre director has to have the eye of a critic, but they are not film makers, musicians or artists unless their critical response is equally a film, music or work of art).
What do I mean by this what-if approach making the game critic a peer of the game designer? Fundamentally, the process of being a game critic is the same as being a game designer (is the same as being a game player). That is, it involves the exploration of a possible game space, and trying to validate whether that game space is interesting. The difference is the output: the critic is after the fact, and their output is criticism of the game, whereas the game designer is in a position to prune and reshape the game space to their satisfaction.
But the game critic is doing the same pruning process: because they are searching the space of all possible game released at a point in time, and trying to select which games their audience should play. The critic also influences the audience's reception of these games. I played Half Life 2: Episode 2 with the full knowledge that the in-game gnome caused one player-reviewer endless grief in trying to get it through the episode to launch it into space. This not only changed part of the game for me, so that I spent five minutes looking for the gnome at the start of the game, but I could have chosen to completely change my game experience by turning it into a game of 'get the gnome to the rocket' instead of the narrative that Valve intended for me.
This is not just a case of spoiling the Unusual Suspects by guessing a third of the way through the movie what the outcome is. What it is, is more fundamental to defining what a game is. Game Intestine gets close to the answer, with its charts of time vs fun that help you figure out when is the optimal point to quit Final Fantasy XII (or in Stephen Totilo's case, asking Chris Zukowski directly). Game Criticism should be asking 'is there fun to be had here, what sort of fun is it, and how can I maximise fun in the experience?'. It is as much practical as it is theoretical.
For a game with a fixed narrative, Game Criticism will resemble traditional criticism of film and literature. But for a game with an endless possibility space, such as we will see in Spore and Little Big Planet, my games of 2008, and arguably those that will be proclaimed as gaming's Citizen Kanes come the end of the year, Game Criticism will come to resemble an ongoing dialog. It is as much about suggesting ways to play and places to travel to. This is why the blog is a much better medium than traditional magazine reviews, and suggests that the travelogue, like Jim Rossignol's This Gaming Life: Travels in Three Cities, will be as valid a form for game criticism as academic papers. It is as much why game designers and would-be game designers, and even game players are as much Game Critics as traditional big media writers.
For those game critics who wish to continue writing in the magazine review style, if not in print media, there is still hope. The fact that the so-called evolutionary dead ends of modern gaming have made so much money will assure us of blockbuster summer release style gaming for a few years to come. And the publishing pipeline will ensure that the magazine format and enthusiastic press will survive for some time yet. But, like the dinosaurs, the magazine review is already dead, and the hind-brain between its legs is the only thing still keeping it standing. And the agile, mammalian blogs scurrying around its feet will have their day in the sun.
Sunday, 6 July 2008
Shortly after I argued for hitpoints as the best way to represent damage because they makes the game shorter, Craig Perko of ProjectPerko has argued strongly against them because they make the game longer.
Hit points are one of those 'trendy whipping boys' of game concepts, like character classes and cut scenes that periodically are trotted out as an example of bad game design. But, like these other examples, hit points are the best mechanism that we currently have to solve a difficult design problem: in this instance to represent the process of damage. The alternatives suffer from over complexity and design weaknesses such as negative feedback loops (See my original article for further discussion.) [Edit: As Craig points out in the comments, I'm actually talking about positive feedback loops with negative consequences]
I don't see anything compelling in Craig's argument against hit points. I guess as a roguelike designer, I'm used to random damage, which seems to be the crux of his problem with them. Having hit points as a resource to manage is a strength, not a weakness, and even a card based game system such as Magic: The Gathering uses them to model damage.
Instead, Craig is using hit points as a straw man to address more fundamental problems around combat design and grind. I have no problem with trying something different with combat design but trying to call out hit points as the fundamental problem with current game systems is the wrong way to go about it.
I've got a fairly large list of RSS feeds in Google Reader now, which means I come across all sorts of interesting articles, links and trivia. If it's interesting enough, I'll 'star' it for later and turn it into either a link article here or at Promised Poems, inspiration for a longer piece or a news item or link on the PCG Wiki.
But that's only addresses about half of what I've starred - there's a large number of links that are interesting, but off-topic enough or which I don't know or care enough to comment on. At the moment, they just stay sitting in my Reader queue and never go anywhere. But that's changing.
I'm now sharing anything I find interesting (and burning it here). If you want some high-quality reading, without comment, feel free to have a browse through what I've linked to. There maybe the occasional roguelike entry there: such as the announcement for voting for the Roguelike Soundtrack contest. If it ends up there, it means that another roguelike blog that you should be tracking either directly or via Planet Roguelike has mentioned it, and I don't have anything to add. And, as always, Temple of the Roguelike is your best source of purely roguelike news.
At the moment, I'm just working through the backlog, so entries will be out of order. But let me know whether you find it useful. I've also added this under a My Stuff link bar on the right hand side of the page, where you'll find links to various 'things I do publicly on the Internet' that are not specific to this blog.
I've argued previously that one way of procedurally generating stories is to 'go deep', that is spend time creating a detailed enough a world that the stories prove interesting. The downside with going deep is that it requires a large time investment. With Dwarf Fortress, it looks like that this is starting to pay off. From a recent development log entry [Edit: you can read more here]:
I was testing some adventure mode conversation responses in a human town. The high priest of the god of revenge came down the stairs, and he was an elf with a human name. I talked to him and learned that both of his parents had goblin names, and his grandparents had elven names. So I guessed that his grandparents had been abducted and that the humans had then liberated the tower, placing the goblin-named parents under human control before they had their child. This was more or less correct, but a bit more happened. So, way back when, the demon was causing all sorts of trouble, fighting both the humans and the elves and destroying their cities. This went on for 40 years -- an elf managed to tear off the demon's nose about halfway through this rampage, but he couldn't finish the job because he had lost a limb in the fight before the nose-removal and lost the rest of his limbs afterward, prior to being burned to death. Eventually the demon was shot in the year 45 and the humans took over the goblin tower. Many of the goblins fled into the mountains, and others were enslaved, including some of their abductees. Ngoso and Bax were a couple of elven children born to abducted parents that were left alone to live in the goblin tower under their new human rulers. When they grew up, they both started shops -- The Enjoyable Paddle and the Lessened Healer -- in the year 45, and in the year 54 they both moved to the human town of Beersreined. The next year they were married and they both decided to wander the wilds looking for monsters to kill (they never found any, as the giants were off on the other side of the world and the dragon was dead). Ngoso had ten children in Beersreined, the third of which became the high priest of the god of revenge after both he and his father were converted by the new temple in 146. Ngoso never converted as she had become a dragon worshipper during one of the attacks prior to its death. As a sidenote, the elves and the humans didn't get along very well, and in the year 60 the elves actually drove the humans out of the goblin tower they had conquered. The elves had no desire to live in a goblin tower so they left it abandoned. At this point, the goblin refugees that had been wandering through the mountains for fifteen years reclaimed the tower. They held it for a total of two years before being defeated by the humans again, who were then able to maintain control of the site until play began. I didn't check if any of the goblins managed to escape again.[Image from In Our Hearts We Were Giants: The Remarkable Story of the Lilliput Troupe, which is currently out of print but reviewed here.]