Life after the Ph.D. – plotting an escape into industry

I don’t write much reflective personal stuff on here, and I haven’t blogged much about my Ph.D. itself. That 80,000 word monstrosity has been so dependent upon analysing the lives of a small pool of interviewees and even though there’s nothing particularly sensitive in there, there are still issues around sticking that up on a blog. So this has mainly just been a place to dump the odd game-related thing I write or, more recently, progress on things I’ve been designing or making. In other words, it’s a home for all the things I’ve been doing to avoid what I should be doing.

It’s coming up to the time when I submit the final draft of my Ph.D. and naturally I’ve been thinking about where to go from here. Until recently I had felt 50/50 about staying in academia, a career that demands I uproot my two kids and my already-employed wife to chase sessional teaching jobs around the country. My wife is off for the next 10 months or so on maternity leave, and I feel like this is a crucial time for me to be starting a career, rather than faffing about doing dribs-and-drabs of teaching for an institution that has just forked out tens of thousands of pounds on my scholarly training. I really enjoyed the teaching I’ve undertaken during my scholarship; it felt like the best parts of me were getting a workout and being of use to someone else – a rare feeling for me if I’m honest. I didn’t enjoy the cover teaching and supervising I did when I was working at a secondary school in my early twenties, mainly because of the constant need to enforce rules I hadn’t personally committed to. But teaching adults is great, and I look forward to the sessional classes I’ve got lined up this year, even if they aren’t really enough to build a career on. I feel a little less certain about research itself though.

Tarot-Card-Meaning-The-TowerThree years of writing a Massive Document No-one Else Will Read has really made me realize that I’m not a writer. Not in the sense that I enjoy writing and want it to be the main thing I do. It’s a necessary evil for communication, and I envy those who are a little bit braver at putting their thoughts and feelings ‘on paper’. Yes, I’m fully aware that writing a thesis isn’t about writing a book for someone else to read. It’s a process of transformation; the evidence that you’ve done the work and thought about not only the topic you’re studying but also the changes that you’ve gone through as a researcher. But the main thing it’s taught me about myself is that “this isn’t for me”. Research? Sure. I especially love going to conferences, meeting a bunch of cool diverse thinkers and sharing ideas. But researching the same thing, on my lonesome, for long periods of time? Nah.

Also, being in the sociological side of games and technology can get a little… heavy. I feel weighed down by a million journal articles on socio-cultural problems that are seemingly there only to be analysed and never actually resolved. I had hoped to bag a teaching position and then do some action research – something like making games with disadvantaged young people in youth clubs. But I don’t think that sort of secure post is going to materialise anytime soon – not geographically near enough to where my family has put down its roots. I can commute from Kent to London but no further than that for the time being.

People talk about the ‘writerly craft’, and I know what it feels like to be deeply engaged with a craft. I just don’t feel that same engagement when I’m writing as I do when I make games or music. It was different when I was an undergrad, working 3 days a week in a secondary school while doing my BA. Then, writing was something I felt I excelled at. Undergrad university was a mixed bag. I didn’t really make any lasting relationships during that period; being a slightly older student and a bit of a weirdo. But the essays came in short manageable bursts on a variety of topics. They were an escape from everyday life into theory – especially in the third year when my son was born – and a way to get intellectual pats-on-the-back that I hadn’t really had before.

I feel like that thirst for knowledge dried up a little when it was disconnected from the everyday (I had to quit my job as a teaching assistant to take on my scholarship). Since I started my Ph.D. in 2012, I confess I haven’t read an academic book from cover to cover. It all became a bit of a grind. So, feeling slightly disheartened after losing a recent interview for the only local job going for what I can teach, I had a little moan on Twitter and an acquaintance on there came to my aid. They pointed out that people from games studies – especially those with design and/or education experience like myself – can do well in Instructional Design (ID). So today I’m heading home from the library with a heap of books on ID, UX and other acronyms, ready to start learning about something I can actually apply outside the confines of the university. Something that will use my skills, satisfy my need for project-based work and won’t require me to do the ridiculous trial-by-fire that is the academic tenure-track.

I’ve got a sack of books and I haven’t felt this stoked about reading/learning something in over three years. Thank you, random Twitter buddy.

theme tune from “The Great Escape” intensifies

Crow Hill Kids: an (abandoned) RPGmaker game about superheroic British teens


the protagonists’ walking sprite from Crow Hill Kids

Some games start life as sets of rules; get the thing in the thing and you will receive a thing and get enough of those things and you win the thing. Other games germinate from narrative and setting ideas; there’s a dragon and some ninjas and it’s all cyberpunk AF and cuties make out. Crow Hill Kids was a project very much in the latter category. The ‘setting’ came from a time between around 2007-2012 when I was working in a local academy school (for non-UK readers, “academy” sounds fancy but it’s sort of UK code for “failing school which has been privatized so that the government are no longer as responsible for its failings”.) I had some ideas about what it would be like if some of the kids I was working with had superpowers, and what sort of wish-fulfillment those powers would involve. These ideas came out in various short comic scripts (that was my thing at the time) and were then forgotten for a while after the series Misfits came along with a very similar premise.

I haven’t really done enough game-making to give out much solid advice but I would say the number one tip for new folks is to work out whether your game is a ‘pure’ game (urgh @ this phrasing) or whether it is an interactive story. It is very unlikely that you’ll be able to produce a set of mechanics and write a whole narrative/setting and produce visual assets. This is why programs like RPG Maker are so great. Sure, lots of games of low quality get made, but I think it’s a given that commercial-quality, original-art RPG Maker projects are the minority. The same is true of any tool that makes a difficult process more accessible. I’d used one iteration of RPG maker during my teens and bought it again at the end of 2014 with the intention of messing with it some more. I returned to my previous setting of Crow Hill; a fictional council estate somewhere in urban Britain (again, for American pals, a council estate is social housing like the projects). What follows is a garbled, illegible account of the world I was building; somewhere between X-Men, the original UK version of Shameless and Earthbound/Mother


Crow Hill is a suburban area; a small council estate outside of any major urban center. It has one large towerblock called Curwen House. The local high-school; Crow Hill High, was knocked down in 2006 and rebuilt as the Huxley Academy with the backing of private funders. It turns out that Paragon – the academy’s financial benefactors – are particularly interested in rooting out kids with any unusual abilities. The player is given control of a group of 5-8 teenagers and pressed with the task of uncovering local legends and global conspiracies, with the help of the odd cybernetic augmentation or injection of power-granting sludge. The protagonists are generally aged 14-17 and from a diverse array of backgrounds, and part of the writing ethos was to give each one of them at least one major flaw that would bring them into conflict with their peers in ways which might jeopardize the player’s overall mission (stories like X-men are, after all, about working out differences in order to overcome a common enemy).


from left to right: Brendon, Elijah, Miriam, Jasmine, Owen, a “prefect” droid, and Flo.

Player Characters

Elijah was initially the protagonist; the new kid who has just moved to town with his father. I initially toyed with making him a shapeshifter for this reason. The idea was that his driving desire was to ‘fit in’ and that manifested as the power to take on any other person’s form, without really making much of an impression on people as himself. This turned out to be a little tricky to implement and also, given RPGmaker games’ emphasis on turn-based combat, not very good for fighting. So in latter versions, Elijah’s power involved becoming a mass of liquid shadow, similar to Venom or something from The Darkness. He’s torn between wanting to make friends and just wanting to disappear into nothing, and the form he takes during combat is an approximation of a creature that began to visit him in nightmares around about the same time as his mother died.

miriam all

all the miriams (we went for number 3 in the end)

Miriam is a British Nigerian girl and is, in terms of the role she plays and her relationship to other characters, very much like the “racebent” Hermione Granger described here. She comes from the most affluent household among the group, is very studious, and sees some of her peers as somewhat beneath her, both culturally and academically. Miriam’s moveset revolves around psychic blasts, confusion attacks and mental domination. For all her skill at controlling the minds of others, Miriam does not know how to read them.

Jasmine is the popular, fashion-conscious prom-queen character. She works as a subversion of the vamp/femme fatal trope: rather than having powers geared toward seducing boys, she repels them with an arsenal of disgusting bio-chemical weaponry. (Her power revolves around a desire to deterr unwanted sexual advances). Jasmine’s moveset thus revolves around poison or acid elements, administered through clawing and projectile vomiting.

Owen is the muscular sports ‘lad’ of the group, with super strength and speed. He is also a wheelchair user, despite his superpowers. Owen is desperate to be seen by his peers as the alpha male, and sometimes this can manifest itself in crude misogynistic or homophobic talk.

Florence is the hacker of the group, with the ability to influence digital technology with the power of her mind. She controls an up-gradable drone, can deal bonus damage to mechanical enemies, and also modify the powers of comrades’ cybernetic augments. Her father runs the local PC Repair shop. Florence is a self-confessed emo/scene girl and wouldn’t normally mix with the other members of the group.

early examples of sprites in-game (sprites are 32x96; 3 times the height of regular RPGm sprites)

early examples of sprites in-game (sprites are 32×96; 3 times the height of regular RPGm sprites)

Quest Structure

Crow Hill Kids‘ story follows two intertwining quest-lines which can be completed one after the other or simultaneously. Each carries its own set of rewards as well as a particular flavor of enemies.

Quest Thread A: Paragon

As the kids explore the estate, they will be set upon various times by hooded youths in Academy uniforms. These turn out to be androids, not unlike the Foot Clan in the TMNT cartoons. On further investigation, the gang find that these droids; the Prefects, are being produced at a secret facility underneath their school. Throughout this questline, players will come to learn about the history of the Paragon initiative and the ulterior motives behind the schools’ reconstruction, as well as acquiring numerous cybernetic upgrades which they may equip on any character of their choosing (I worked on the premise that traditional RPG weapon systems were an ill-fit with a superhero world, so instead cybernetics are permanent equipment which augment existing powers in similar ways to Spiderman’s web-shooters or Wolverine’s claws).

Quest Thread B: Rift

In various dark recesses of the estate lurk large, sluglike creatures which drool acid. As players explore, they will encounter more and more of these creatures, and hear stories about other strange happenings. During this questline, players will meet an elderly resident of the estate; a retired comic-book writer who was among the first generation of locals to obtain powers. He explains to the gang that their powers are the source of an inter-dimensional rift, and posits several theories as to the cause of such a disturbance. His power was to open gateways from this world into that one (something he exploited for inspiration and financial gain to great success during the 1980s). He offers the use of his power to allow the gang to traverse dimensions and stop the alien menace.  Enemies in this thread have an emphasis on poison and mind-control, and there is a bias toward harvesting XP mutagen items here.

Quest Thread C: Oracle (optional)

At the top floor of Curwen House lives a single mother with the power to show individuals their future. Visiting her will trigger an interactive account of one possible future where, according to the player’s actions thus-far, different members of the gang will be placed on different sides of a conflict in a middle-eastern country.

Elijah's battler sprites. The left is the finished thing; the right an early sketch.

Elijah’s battler sprites. The left is the finished thing; the right an early sketch.

Additional Notes on Design

1. I really don’t like grinding. It’s an abuse of players’ time. Have a little bit of grinding but make the main source of XP these mutagen jar injections (later turns out the mysterious slime in these comes from the Riftslugs described earlier). Making the player root out a limited number of upgrade items (as in the Deus Ex series) just feels more interesting to me, and makes for a smaller, less bloated game.

2. Mutagens are supposed to be gross. It’s 15 year old kids injecting themselves with slugslime. Maybe they have a tradeoff like lowering HP (interesting way of ramping difficulty without scaling enemies) or they give a woozy effect of some sort until the end of the day?

3. Having a massive inventory of swords and guns feels somewhat off-brand. Equipments are replaced with semi-permanent cybernetic augments which charge via a “rage meter” (e.g. they can be used more later in a battle). Chloe’s hacker skills can be used to fast-charge someone’s Energy. Cybernetics can include passive advantages (like making a character immune to a dangerous status effect like Confuse) or active skills like arm-mounted machine-guns or flamethrowers. The idea would be to have these be generally quite powerful, but only allow each character to have, for example, on “attack” aug and one “defence” aug, or some similar limitation.

4. toward the end of my time with the project I was still struggling with standardized RPG terminology. Do I really want the energy that supes use to deploy their powers referred to as “MP”? Or would it be better to do away with it entirely and have cooldowns? Or a universal fatigue system that’s used for physical attacks too?

Closing Point: my love/hate relationship with RPG Maker

GubiD's 2.5d-looking Tactical Battle is pretty impressive, allowing RPGm folks to make games similar to FF Tactics or Vandal Hearts.

GubiD’s Tactical Battle System allows RPGm users to make neat games in the style of FF Tactics or Vandal Hearts

really don’t like JRPG combat much. And even if you use a script for sideview animated battlers, it takes a lot of hacking away at Ruby to get the player characters on the left and moving correctly to enemies on the right. To me, the direction of most games is left-to-right, ergo player characters should be attacking from the left. But the RPGm community at large seems happy enough to stick with the Japanese convention of player-characters on the right. This infuriates me a disproportionate amount.GubiD’s 2.5d-looking Tactical Battle is pretty impressive, allowing RPGm folks to make games similar to FF Tactics or Vandal Hearts.

If I had the time and the effort, the ideal sort of engine for something like this would be more similar to the Infinite Engine. It’s not that I dislike jRPGs entirely, it’s just that they tend to abstract everything happening on the battlefield out into numbers; big obnoxious numbers. They can often feel a bit more like playing a board/card game and although some tactical battle systems exist – allowing you to make an RPG-maker game where battles occur on a tactical grid – these tend to be more difficult to implement, and not necessarily compatible with my 32×96 sprites.

Battleheart is a neat mini RTS/cRPG thingy

Battleheart is a neat mini RTS/cRPG thingy. maybe you can help me rip it off?

working on making Beatopia more transparent to players

Beatopia has been through a number of iterations in the 3 or 4 months since I started toying with the idea. So in lieu of a proper devblog post, I’m basically just going to describe the current features/play-patterns of the game and then run through some of the issues I’m currently wrestling with.

When I first started this, I said I really wanted to avoid using too much RNG (random number generation) in the game. A few devs on Twitter said things like “oh you can’t do that, it’s really hard to keep the game interesting that way” but I’m really thinking back to classic arcade puzzle games like Tetris, Puzzle Fighter and Bust-a-Move, where the player is usually given the same pieces to work with in the same order for every level, thus reducing the frustration (but also the interesting variation) that can come out of RNG. I still think this is a workable goal and I find myself gravitating back toward it even more now that the game is working but too difficult to explain. My current task is to root out all of the ‘invisible’ variables working behind the scenes in the game and make them more explicit to the players (taking a hint from this great GDC talk on transparency in game design). Transparency is really important in board games where the players have to remember the rules, but also in sims like Civilization which just contain an overwhelming amount of in-game information. But I’m making a game you play in 3 minutes on the train, not Civilization, so I also need to take into account the fact that players aren’t going to RTFM in order to play.

One of the biggest challenges has been making the relationships between the player’s buildings and their neighbouring districts both strategically important and relatively obvious. I’m still not quite there yet but here are the ‘rules’ I currently have in play. A lot of this will sound complicated but my goal at the moment is really to pare a lot of this down so that it is complex enough to be interesting to play, without being needlessly complicated and oblique. As the game has become gradually more like a game and less like an experimental toy, I find myself returning to the “no RNG” rule, and in doing so I need to make a more explicit 1:1 relationship between the numbers driving everything and the info pop-ups.

Rules at Present

– the lose condition is 0 population (the population is essentially an “HP” bar)

– players drag boxes from the conveyor at the bottom of the screen and onto the isomentric grid to build. this removes some of the financial management aspect of a true sim, and allows me to construct challenges by giving the player a specific string of buildings. it means the game works a little more like a classic arcade/puzzle with $ only increasing. (so money is a score rather than a resource, because capitalism. also because making a 3-minute session casual game)

– districts’ effects are triggered as the ‘pulse’ (highlighted tile) passes over them. When triggered, a district will play a particular drum sound, and have some sort of effect on the main resource stats (Energy, Population, Money) as well as its neighbouring districts. This is the core of the game.


– players score by generating coins. orange buildings are ‘financial districts’ and yellow buildings are advanced financial districts. Orange buildings generate 1 coin per turn, + 1 coin for neighbouring oranges and +2 for neighbouring yellows. yellow buildings operate the same way but generate twice as many coins once neighbours are counted. (Previous iterations had coins representing more than $1, with the BPM determining how much each coin was worth. With the newer system, the BPM stays the same, and each coin is only worth $1 (maybe $1k in future?). This way the player doesn’t have to read thousands of numerical popups and can just roughly judge their progress by how many juicy little coins are springing up. I am generally aiming for this kind of 1:1 relationship between game variables and icon pop-ups in every area of the game (see ‘population loss’ bulletpoint)

– teal buildings produce energy (and play a ‘clap’ sound) other buildings reduce energy. zero energy means buildings will not trigger their effects.

– population loss is indicated in a similar way to money accumulation, using rising red skulls. green buildings counteract this, playing a ‘tomtom’ drum noise. They are represented by a house icon, and each one restores the population by 1. reasons for population loss are: pollution, riot, basslazer.

– each ‘coin’ or ‘home’ pop-up means a certain amount of pollution is added to the origin district, so although the goal is to connect up numerous orange and yellow buildings to generate the most $, this also means a far greater pollution output, so (orange/yellow) financial districts have to be interspersed with the (dark pink) treatment plants (represented by an O2 icon)

– an invisible ‘crime’ variable, beginning as a negative, gradually increases until a district riots. when a district is rioting, it stops performing its usual game function and plays a glitch sound when triggered. rioting districts also lose population each time the pulse passes over them. holding the mouse (/finger) over a rioting district plays a wubwub bass noise, using energy and lowering the population quickly. It feels kinda cool to do, too (see this LINK). A game about making electronic beats just really needs a bass cannon, and the nastiness of the audio kinda goes with the insinuation that you are going Judge Dredd on your own peoples.

– the way crime increases needs to be more explicit. crime is currently the most awkward mechanic in this sense. there’s also a limit to how many popups I can use with any degree of meaningfulness (once you have red skulls for population loss, what do you do for “+1 crime here”?.

– I’m currently thinking of modelling crime/riots using a dramatically simplified model of relative deprivation. Each time the pulse hits a district, that district will count how many direct neighbors it has (This is 4 for a central district, 2 for a corner and 3 for an edge). Crime will increase by one point for every neighbour which is not a financial district, and an additional point for every rioting neighbor. The underlying concept is that wealth in one area means a less prosperous neighbour will feel negative effects. I’m trying to strike a balance between making something which is transparent and instantly understandable to players, and something which rings as sociological ‘true’ (at least in terms you might explain sociological ‘fact’ to a 14 year-old in).

– everything in the above bullet is irrelevant until I can think of a cool icon popup that means “crime has gone up here”.

– pollution is currently visible via a floating cloud over each district which becomes more opaque as the district becomes polluted. However, the point at which this will result in deaths is still not clear enough.

Procedural Text Generation in IF

Originally posted on Emily Short's Interactive Storytelling:

In the Missing Tools discussion some time ago, one of the things people mentioned wanting more of in IF was procedural text generation, which here is meant specifically as the ability to have the computer describe complex world model states or story events without having to hand-author every possible variation.

This is an area where there’s a lot to learn from work going on in academic research, but as far as I’m aware there’s relatively little communication. As I mentioned in my ICIDS writeup, James Ryan at UCSC and Dr. Boyang Li at Georgia Tech’s Entertainment Intelligence Lab are doing work on a) how to better represent a richly complicated world model and b) how to procedurally alter narrative features such as the tone of narration. One of the things we particularly don’t seem to do in hobbyist IF, perhaps for lack of resources, is experiment with large…

View original 1,906 more words

My First Twine is a randomly-generated scifi-horror detective thing (of course!)

I’ve been meaning to get stuck into Twine for a while but had been putting it off until I had a real idea to work with. I’m a big fan of randomly-generated flavour text, and I’d been messing about with the theme of memory-hacking in my writing for years. In We’re All in this Together you play a “neuroscope” operator in a near-future London. There is a plague spreading which causes some sort of psychotic breakdown, and its your job to read the memories of your “patients” and quarantine those whose memories seem… iffy. The memory read-outs are presented to you as little stream-of-consciousness type poems, and if you are unsure you can “re-scan” a patient, at the expense of using up more of the day’s energy. alibi The way it works is as follows. The game has one rather long list of phrases which represent “everyday happenings” in the city. If the “patient” is infected (based on a random 1-100% chance, depending on how well you’re doing) then an addition list of phrases gets added, which are the “tells”. They mostly relate to desperation, feelings of aggression, and icky black goop dripping off of everything (because I’m edgy, hell yes). When a patient is scanned with the neuroscope, four of these one-liners is delivered to the player. The tension lies in the fact that, because it’s shitty near-future London anyway, it can be hard to tell what is a hallucination. In the above image, the two passages are as follows: Not infected: “The Big Smoke crawls / Lo-teks around a garbage fire / Warm tea with friends / Spat gum into a passing drain” Infected: “Lingering smell of rotting fruit / Skin beneath fingernails / Incisor meets a shoulderblade / Dub bass from a window up high” In the infected example, the first three phrases are drawn from the pool of “infected” memories, whereas the last one is from the regular memory pool. Next task will be to make it more “gamey” by adding and tuning the resource-management elements, making those “memory pool” lists longer to avoid repetition, and possibly add conversations between days, and events that might change a few things in the memory pool (for example, a

riot breaking out, or the nature of the virus changing after a certain number of days). I’m thinking about dropping that to three phrases at a time. Might make it more poetic, and more challenging potentially? You can try it out here. Probably, unless I change the title and then this link will probably die.

Coding in the Classroom


A good overview of some of the issues facing computer science in US schools – still very UK applicable even as CS becomes a core subject (supposedly from KS1 upwards)

Originally posted on SocioTech'nowledge:

One need not look to superstars such as Mark Zuckerberg or Bill Gates to justify reasons for using code and programming logic in the classroom. There’s plenty of literature that illustrates its positive learning outcomes. Coding in the classroom is linked to improved problem solving and analytical reasoning, and students who develop a mastery of coding have a “natural ability and drive to construct, hypothesize, explore, experiment, evaluate, and draw conclusions.”

But there are other compelling reasons for integrating code in the classroom.

Reasons to Teach Coding

1. Coding is a new type of literacy.

Wired Magazine reported that reading and writing code is the new literacy. Those students who master it are better prepared for a technical revolution that spans cultures and language boundaries. That’s because coding isn’t just a language. It’s a way of thinking about problem solving.

2. Coding is a tool to improve…

View original 973 more words

Designing Without a Genre Template: Beatopia (a game about building beats and defending cities)

The game I’m currently working on is something I’ve had kicking around in my brain for a while. I like city sims and pretty much any game that involves some level of building, and I’ve been mulling over how to incorporate this with a drum machine in some way. So here’s something I started working on in March which I hope shows some promise.

Beatopia is designed for mobiles, and is a mini-RTS/city building sim crossed with a drum machine. It’s been a challenge to make because of the ways that the drum-machine format constrain what you can do.

beatopia 1

The whole thing started with a 4×4 isometric grid. It’s 4×4 so we get a total of 16s. Drum patterns can be split up into smaller increments but this felt the most manageable. The “pulse” as I’ve called it, starts in the left most square, and moves right, then comes back around. In the image below, purple squares represent bass drums, cyan/teal squares represent snare drums, and orange squares represent hihats. This means that the beat in the above image goes: BM-ts-CLAP-ts-BM-ts-CLAP-ts.

The next stage was to work out what each of these color-coded tiles would *do* in the context of a city simulation. My initial idea (and I’ve managed to retain it) is that Beatopia is a 80s-90s styled cyberpunk city which runs on electronic beats. (Maybe there is some creepy subtext about the population being controlled by music? See final version for pretentious Theodor Adorno quotes)

So here are the (tiny) sprites for the tiles, which are now “districts”. Note that the first four were the initial base for the game and the two on the right were added in later (I’m keeping this log chronological so I’ll explain those further down). I tried, to some extent, to keep the buildings resembling the instruments they play, so for example the orange buildings play hi-hats and resemble a revolving restaurant building – which looks kinda like hi-hat cymbats.

Orange: financial district – produces money and pollution

Pink: environmental – reduces pollution in neighboring districts

Teal: security – reduces crime in neighboring districts


Money – crime – pollution. The way each of these “works” in the game has changed drastically since I designed the sprites, but they are still the three key elements of the game. As hi-hats were played by the city/machine, the player accumulated money, but clouds of green smog would also gradually appear on that district and to a lesser extent its neighbors. I wasn’t sure how to handle “crime” yet, but I felt it should be instance-based, rather than a numerical value (as I had done with pollution).


The above image shows a brief experiment where I was toying with the idea of having the player drop buildings in in-time to the beat. It was janky and I hated it, and I had really set out to make a more relaxing “5 minute train journey” type game. So I threw it in the bin and shouted “begone horrible build!”. (also those colours – yuck!)

beatopia topdown

Then I had a bit of a breakdown feeling like the isometric wasn’t working in the small-screen mobile phone context and that it would be too fiddly to accomplish what I did. So I spent about 2 hours working on a top-down version and people I showed it to went “nope don’t do that”.

This seemed like as far as I could take the game without deciding on a genre and a set of conventions that went along with it. I needed a way to ensure that there was some sort of strategic meaning to the building placement. So over the course of a week or so the game evolved into the state you see below.

beatopia skirmish

Orange districts produce a coin each turn (work the beats-per-minute of the current level: meaning faster = difficult but with potentially higher scores) plus some pollution. Pollution has a very direct visible effect; slowly killing nearby player units. Pink buildings clear pollution. Worker Drones are built at your HQ (the white building) and collect coins (they could fight in the first version but I’ve since made them passive and added a separate class of Enforcers). Glitchers (the red skulls) represent anarchic gangs who wanna take your coins and generally mess with your city. They initially appeared randomly but now they come in waves, as they do in Tower Defence games. This also allows for a nice back-and-forth between combat and building time, which is accompanied by a shift in music; when no enemies are on screen, each bar of beat is accompanied by an ambient chord randomly picked from a pool of 7, whereas when enemies are present, this chord drops out and is replaced by a funkier set of bass notes.

The whole thing started to lean toward a mini-RTS/defense game thing, and I think I’m happy with it this way.

The green buildings became the equivalent of “houses” in an RTS; each one lets you build an additional 2 units. Units are currently built instantly but I will put in a timed build-queue in future too. These little constraints are so bread-and-butter in strategy and it felt like something was missing without them there. Adding the Residential districts as a 4th buildable district-type also gives the player a little more of a puzzle in terms of how to use the available space. Residential districts also produce pollution but at a slower rate than Financial.


So what of blue/teal buildings? They’re still snare-drums, which means we want to them to be important and expensive and to require good strategic placement. I always thought the blue/teal color had the strongest associations with law enforcement etc. So now, these buildings are Armories which recharge the shields of nearby Enforcers (see the little dude with the additional blue bar over his health in the .gif). Enforcers with a shield will attempt to engage Glitchers, and they will flee to the nearest Armory either when a wave of Glitchers has been defeated, or when their shield is depleted.

That’s it for now. I had a ton of other stuff to do on it before it’s even remotely playable. There’s currently no lose-state and no way for the Glitchers to actually disrupt your city, beyond taking coins and destroying Workers and Enforcers. That will be next on the list, and then I’ll move onto making a HUD that isn’t crud.

The below shows a ring of 12 districts. This potentially allows for a level/beat to be in 12/8 time signature. This means a sort of “galloping” feel; like try counting 123/123/123/123 quite fast and you’ll get the idea. One of the coming challenges will be to see whether changing the layout of the available districts really adds any strategic variation to the game. Or whether I should just keep it to 4×4.

beatopia 128

Procedurally Generated Kids’ Apps – is this a thing?

I’ve been thinking a lot recently about apps for kids, looking into the market and considering what I can possibly bring to the table there. I’ve got a three year old and Baby Unit Two on the way, and I’ve been mulling over the idea of making some interactive story-book things as a way to hone up my minimalist design chops and also potentially make something which is going to provide some passive income.

Most games I’ve tinkered with previously rely heavily on randomization to produce narrative content – for example a shop where customers have random(ish) traits, names, needs and biographical backgrounds. When most people hear “procedural generation” the obvious example is usually Minecraft, where the landscape is what’s being produced differently each time. But PCG can also be used to create the actual puzzles for the player to solve; such as DragonXVI’s entry for ProcJam2014: Inquisitor – a medieval “whodunnit” where the victim, murderer and other details are different each time.

In the context of interactive kids apps, PCG seems like a way to add additional value through replayability. A plausible business model seems to make a story/set of puzzles which change every time – use that as the premium app – and then release the always-different version as the paid app. Nothing in an interactive kids’ book would be as complex as generating terrain (I’m not sure if they even constitute PCG!) but some examples I’ve thought of are:

  • The player has to put 4 tools into 5 appropriately-shaped slots in a tool kit. A different tool is always missing, and has to be found in the next scene
  • The player has to build a robot using different objects for the three sections of body. They are provided with 6-8 items drawn from a pool of 12-15. The robot they have made is persistent through the rest of the story.
  • A sight-seeing “walk” through an area with a narrator character, with animals/things which appear along the way (picked from a long list) each accompanied by a “what is this called?” type multiple-choice question. The list would be skewed to have more unusual animals/objects at the bottom end, so that the game can keep track of progress of how you’re doing and adjust accordingly.

This is mainly a brain-dump kind of post but feel free to let me know if there are any apps you think I should check out.

Gloomwood: sketching out adaptive music ideas

Originally posted on BossLevelVGM:

Woo, a graph! Well, not really a graph, just an audio timeline from FL Studio. I made this as a mockup of how realtime-adaptive music could potentially work in Gloomwood. The overlaid lines represent the volume of each track, controlled by an in-game variable.


The bottom track, for example, is the walking tuba you hear at around 0:20, representing a police constable passing close by. It’s a bit of flavour and tension, and potentially a way of alerting the player if they’re about to do something nefarious and haven’t looked around first.

At about 1:25 there’s a version of the entire track rendered through a reversed-echo amp model, which gives a sort of twinkly “lots of spirits crying” effect; this is represented by the third (light blue) channel in the graph and could be used to inform the player that something nearby is resonating magical energy.

The green track contains the…

View original 78 more words

TOP 10 Posts of 2014


Here is a review of the top 10 most popular posts of 2014 on TGA. This list includes only pieces which were published in 2014: pages and previously published posts are not considered here.

Originally posted on The Geek Anthropologist:

You know the saying, better late than never! Here is a review of the top 10 most popular posts of 2014 on TGA. This list includes only pieces which were published in 2014: pages and previously published posts are not considered here.

View original 203 more words


Get every new post delivered to your Inbox.

Join 1,004 other followers