Monthly Archives: April 2016


As we approach true artificial sapience at unknown but increasing speed from an unknown but decreasing distance, our fears and assumptions about what will happen to us become repetitious. It is generally assumed, for various reasons, that the existence of an intelligence superior to our own will mean the end of us.

Why? Well, look what we’ve done to the species we share the planet with. If we imagine a species that is as gentle and attentive a caretaker of humanity as humanity has been of the world, it’s not hard to see why we find the idea dismaying. Regardless, our fiction repeats a kind of weird certainty that to be supplanted as the greatest, the smartest, is to be destroyed or made irrelevant, to cede our place in the universe to our creations. It seems tremendously short-sighted to believe that our, frankly, extremely stupid appraisal of the necessity and utility of violence would be passed on to our creation, at least if it were a superior intelligence. If it were, that would speak more to flaws in that intelligence than it would to the priorities of a superior being.

Sometimes, in our stories, the destruction of humanity is violent, a massacre, and there’s two forms these take: One is where the artificial intelligence isn’t really intelligent at all, just smart enough to find targets and eradicate them without the aid of a human operator. This is basically the wild beast, or Big Dog, hypothesis: A world of dragons, that hunt humans because they don’t know better, that breathe fire and lead and don’t get tired. It’s a frightening image, but hardly a threat of extinction. We’ve survived worse as a species. Moreover, since these robotic-beasts aren’t truly intelligent, they’d eventually fall apart or be destroyed, especially since the facilities to create them would inevitably be high priority targets.

The other form is where the artificial intelligence is truly, boundlessly intelligent, and willfully engages in the destruction of humanity – and this is the one I find most interesting, that this image keeps repeating in our art, because what it suggests is that we generally believe that humanity should be destroyed. We apparently believe that a being of comparable or greater intelligence to that of humanity would inevitably come to the conclusion that we shouldn’t exist.


Mostly, I think, we believe this because the assumption that eliminating a dangerous person is justified is something deeply ingrained in our culture. The same belief system that makes it okay to put our criminals in a prison that we know to be unsafe and abusive is what lets us assume that an implacable machine enemy would put us in The Matrix. The same belief system that lets us believe that torture and indiscriminate bombings are justified in a new era of war lets us believe that, if it were the machines who were advanced and we were the ‘savages’, their divine right would allow them to eradicate us.

We believe in the approaching supremacy of machine gods who care nothing for human life because it justifies the existing supremacy that lets us care nothing for human life. We believe it is, given the power and intelligence, the only correct way to be. Machine supremacy becomes the new colonialism, and we believe that they, those advanced and logical machines, will believe it is correct – justifying, in retrospect, centuries of human atrocity.

In case it was ever not obvious: Machines aren’t what we need to fear. Even if they get smarter than us, faster than us, stronger than us, those capabilities are in themselves utterly unimpressive compared to what we have already done to each other with everything ranging from exotic chemical agents to the controlled application of disease to buckets of water and an overactive imagination.

Whether they remain our tools or become our successors, what we fear in machines is mostly the reflection we see in the metal.


Another slow week, though I feel like I’m starting to get back up to speed. I’ve been going back to animation work, though with a number of false starts: First, I discovered that I’d messed up somewhere in the process of creating animations derived from my running animation and completely lost the source files for the left-facing run. Oops. So I decided whatever, I’d get the right-facing run looking good and recreate the left version from the right-facing source files; I started creating tweens for each frame on the right run, and only after a little while realized this would be creating a 60fps animation, or close to it, which, if I applied that to all of the animations in the game, would require an absolutely absurd and, frankly, hugely wasteful amount of extra work. The difference between 30 and 60 frames per second in an animation within a game that otherwise runs at 60fps is barely noticeable, and obviously completely invisible if the game ends up running at 30, which is not an option I’ve completely discarded. So, basically two days of wasted work planning an animation at way too high a framerate.

After that and some more polishing of the prototype animation, it was time to bring the animation into Aseprite, the pixel editing software I’ve selected, and get to work. And… it’s quickly becoming apparent that much of my careful prototype animation work was probably a waste of time. I’d always vaguely suspected this might be the case, but pragmatically speaking I’m almost certainly better off creating the entire animation in Aseprite. Don’t get me wrong, the work done planning the animation out was valuable and will make things go a lot faster now, but the total time I have to spend on each animation is probably higher now than if I’d just gone straight in with the intent of creating a finished animation. Oh well, live and learn.

I was hoping to have at least one animation to show for the end of the week, but it’s quite time consuming to build even one frame at the level of detail I’m working at. As I practice, get into a flow, and figure out ways to work more efficiently, this should hopefully become less onerous, but it’s clear that just the player character animations will probably take a month or two, since she has like 400 frames. In the meanwhile, all I’ve got for now is this one frame.


The trickiest part is handling transparencies well: If there were no transparency, I could just eliminate the alpha channel to clip off all the pixel smoothing artifacts from Photoshop, but as it is I can’t do that without turning the clothing solid. I suppose that will be less of a concern when it comes to enemy animation than it is here, so that’s something to look forward to. Mostly, now, I’m just going in and drawing over prototype using it as a template, rather than being able to lock its alpha channel and use it directly as a mask.

Expect to see a lot more frames in progress over the coming weeks, though hopefully many of them will be in the context of finished animations.



Playing through Titan Souls, I was impressed by the constraints that the designers had chosen to work within. The concept of Titan Souls is that it’s a game where you can die in one hit or win in one shot, and the difficulty emerges from avoiding that hit and setting up that shot. Now, technically, in order to get that shot sometimes you have to shoot to disable a defense or otherwise change the scenario, but you never have to shoot quite the same target more than once. This is a game design that centers around fighting a large number of unique and differentiated opponents – ‘bosses’, by the standards of most game design – while removing the main tool designers have used to make boss fights long and, if you will, ‘epic’.

And yet, within these constraints, it flourishes. Because each ‘boss’ needs special defenses, special weaknesses, special movement, they’re far more memorable and distinctive than bosses in most other games. A battle can be over in a second or take place over a minute or two of fierce dodging and frantic counter-attacking, but either way you feel like you have to know your opponent to achieve victory, and you know that slipping up for even a moment means failure.

This is an unusual game: its gameplay and its content far more closely intertwined than is common, especially given its relatively traditionally gamey presentation. You couldn’t use these boss designs with any other gameplay; you couldn’t have these fighting mechanics with any other boss designs. In this, we can see how the designer started from certain principles to design the game’s content, and used those to enable a very specific set of mechanics. What are these principles? First, a boss can have no hit-points: Each successful attack either creates a vulnerability or kills the boss. Second, the boss must have some recognizable weakness: Ideally something that is easy to perceive but difficult to reach. This is one reason why the game has so much eye-poking, which I still find amusing.

There are other principles to the design, but most of them are shared with other games, and I’ll get to those in a minute. In thinking about designing from principle, I tried to think of other games that took this approach. Unfortunately, it can be difficult to tell, outside of extreme cases and instances where the game’s design is well-documented, how a game’s particular design choices were made, but certain examples do spring up: The first Deus Ex tried to always give the player three possible ways to clear each obstacle, which in turn enabled the gameplay design convention of having multiple upgrade paths giving the player different world-navigating abilities, from fighting to swimming to hacking to acrobatics; the original Thief ensured that killing was never necessary to complete a level, and this allowed them to create gameplay that made the player character physically weak and rewarded them for avoiding confrontation; Undertale gave each battle multiple routes to peaceful resolution, ensuring the player’s decisions, violent and non-violent, were actually decisions and could be treated as such.

Most principles, though, aren’t really unique to a given game, and are just generally considered ‘good game design’ now – of course, saying ‘good’ game design doesn’t have any more intrinsic meaning than saying a ‘good’ game, denoting personal preference and enjoyment and no more. These principles come and go, based on our current understanding of what players want, but are still often treated as somehow being unassailably good or bad depending on whether they’re currently in or out of favor – the possibility rarely being acknowledged that, as with all things, the conventions of genre and game design shift with time and context. Still, some are evergreens: “If the player dies they understand why”; “All damage can be avoided”, “The path ahead shall be clearly marked”. Titan Souls shares principles like these: taking pains to ensure that everything we understand about ‘fair’ game design is strictly observed – which makes sense, since having a game violate these unspoken rules when the stakes are always instant death could quickly become very tedious indeed.

Because most games follow these shared principles, to create a new design principle and really stick to it is a bold and striking choice. Indeed, it could be argued that, just as the principles by which a game is designed largely tend to fall in line with its genre, choosing a new principle and designing to that is intrinsically a genre-defining choice. Of course, many genres get defined and never really catch on, so that may not be as impressive as it sounds.

In looking for precedents to this approach to creation, I noticed something else interesting: ‘Traditional’ games design, such as in sports and board games, doesn’t really have this concept, since these games don’t have content which the players play through but, instead, pit the players directly against each other, whereupon older concepts such as fair play and sportsmanship keep the game interesting. These are related but distinct ideas: Teasing out the relationship between sportsmanship and game design principles could be an interesting essay to pursue in the future but let’s just leave it for now.

Rather, let’s look a bit further afield, where something like principles of game design do exist: Literature. If you’ve read Strunk and White’s The Elements of Style, certainly the idea of adhering to aesthetic principles must sound familiar: Avoid cliché, cut out unnecessary words, and so on. And, if you’ve studied creative writing at all, surely other injunctions have shuffled themselves in alongside those in your mind: Don’t resolve your plot with a deus ex machina, avoid these tedious character archetypes, show rather than tell, and so on. If you carry on and look at each individual genre, they usually have their own principles. Most notably and obviously the murder mystery genre has many formalized conventions for creating a story that is satisfying to the reader’s expectations of the genre, but other genres have similar conventions – if, perhaps, less clearly marked and advertised.

The message I’m getting at here is that these common, shared design principles should be actively acknowledged by the designer, and new, untested principles should be undertaken with caution. Either adding a new principle or removing an old one could be a bold new take on game design, but also could easily lose those players who, like frustrated whodunit readers, have very specific expectations of the genre. Just realize that these invisible pillars of game design exist and serve a purpose – and that, furthermore, that purpose isn’t to be ‘good’, to prop up some perfect ideal of game design, but to serve the needs of their audience and genre. Just remember that every one of the principles of good and clear writing has been broken by great authors in works of genius, that each principle serves a purpose and each purpose has a scope – and that, if we want to push the boundaries of what is possible and expressible, some day we may have to move past that scope and into the unfamiliar.


Well, I brought this up in yesterday’s post, but this week has basically been a long exercise in not getting anything done, and I’m still figuring out how I feel about that. On the one hand, i recognize that sometimes I have to push a bit outside of my comfort zone and do things that are hard for me, and that this will necessarily reduce my capacity to pursue my regular interests, if only, perhaps, temporarily. On the other I’ve maintained a cherished practice of freaking the fuck out whenever I spend more than a week without making substantive progress on my game, and who am I to flout tradition?

So, basically, nothing to see here. No progress on the site: I still need to get a mailing list set up so that people can subscribe to it the way they can subscribe here. No progress on the game: I’ve made tentative steps towards finishing up the player animations and made numerous false starts towards getting a stone tileset done – it’s probably going to be more than one, actually, since it turns out the kind of stone formations typical of most caves are so strange and otherworldly they’d look completely out of place out on the surface world – but definitely nothing substantial. Generally speaking, zero substantial progress for two weeks.

Eh. I’ve had worse dry spells, games-work-wise, and generally for worse reasons.

The fact is, at the point I committed to a streaming schedule for the week, this became what was going to happen. I just got tired, and once I got tired I found it impossible to work. I’m going to be rescheduling my streaming schedule soon – I’m thinking maybe Sunday, Monday, Wednesday, Thursday, 8am-9:30am – and also I’m going to be trying to take some steps to both increase my available pool of energy and to consume it more efficiently. The conclusion I’ve come to from my experience with exhaustion and procrastination and productivity and everything in between is that if you want to actually get shit done you have to be honest about how things affect you and under what circumstances you can work. It’s pretty obvious that, as I am now, I can’t maintain the work schedule I want and the streaming schedule I want at the same time. This might change with practice! We’ll see.

In the meanwhile: Changing the streaming schedule, maintaining a regular sleep schedule, and possibly looking into more ways to address the anxiety issue, since I think that presents a big background drain on my energy levels.

Sorry I don’t have any nifty tilesets or animations to show off this week. So it goes. Hopefully next week!


This week I did an experiment: I streamed playing a game for two hours every day, 8am-10am PST. Originally I’d intended to stream Dark Souls 3, but it turns out my computer can’t run it. I thought my computer would be up to the job because it ran Dark Souls 2 quite well, but…

It’s not certain, at this point, whether the poor performance is due to the limitations of my system, a lousy port by From Software, poorly tested drivers by nVidia, or some combination of the above. What is clear is that I can’t play this game yet, much less stream my playthrough.

So I went ahead and streamed other games, but the fact I can’t play Dark Souls 3 nags at me, something incomplete, something never-began, a hole in my expectations. I created this streaming schedule with that game in mind, believing that I would be playing it two hours a day regardless and hoping to get some extra mileage out of something I’d be doing anyway: However, rather than having to hold back on playing a game I’d been anticipating, I found myself, with no Dark Souls 3, having to push myself to play games I wouldn’t be playing otherwise. That was fine, at first: I’d been meaning to play Titan Souls, so I finally got around to that, which covered my first couple of streamsthen I spent a couple of hours playing Nuclear Throne, which is a pretty cool gamethen, yesterday, I started in on a run of the original Dark Souls, since that’s generally a fun game to revisit and a nice way to take a bit of the edge off of my discomfort with not being able to play Dark Souls 3.

But I feel off-kilter. It’s not just about not being able to play a video game: It’s about not being able to participate in a thriving conversation that’s coursing through games. It’s about, not only being unable to stream the game myself, but unable to watch anyone else’s streams. It’s everyone in the world watching the next episode of That Show You Like in the next room, and being three episodes behind and unable to join in with no opportunity to watch those three episodes for the foreseeable future. It’s not not playing the game that’s eating at me, it’s being unable to play the game, feeling unable to reach out and grab a fairly modest goal, something created to be available, and feeling like less of a person due to that lack.

It’s very strange the ways the tendrils of consumerism have bound the edges of our social and intellectual lives.

So I’m off balance. I feel helpless and incompetent, like all of my work is at just high enough a level of quality to keep people from sneering at it without actually being good enough to make anyone want to support it. I’m not sure if what needs to change is me or the world, if what I’m making is wrong or not good enough or I just haven’t figured out how to make it appealing… or if I’m just terrible at figuring out how to make money.

Mostly I’m okay with it. Mostly. Generally mostly. I am worried about money and my future, whether anything I’m doing or have ever done has been worthwhile, but I deal; because I feel like I’m moving forward. I’m trying new things. And I know that that’s actually part of why I’m freaking out right now, why I feel like everything’s closing in, collapsing over me and binding me in place, is because I’m expanding beyond the structures that have sheltered me. Because it’s easy to never feel the rain when you just stay under the same rock. If I’m getting rained on, I’m at least going somewhere.

You can’t beat the game by staying at the first bonfire.

It’s stupid to act like the trials of finding a place in the world and building around it are in any way like playing a damn video game, but. But. We all take strength from stories. From parables and from history, we construct an idea of what’s possible. So maybe it’s okay if, just for a bit, I take the story of stasis broken, of dying suns, of giving up and becoming empty or not giving up and becoming consumed, and use that to justify and reify my own problems.

Just keep trying. As long as you can do that, eventually something will change. Isn’t it odd that such dark and foreboding games teach such an optimistic lesson?


This was not a good week for me.

Progress-wise, I have two tilesets that are maybe about 25% done on average, the tree wood tileset and the stone tileset. A new challenge has emerged with both of these: Due to their placement in the world, the surface lighting is going to differ from the model I established with the grass and dirt tilesets, and I’m still figuring out how. Basically, the viewing plane of the player would directly intersect with the terrain of the hills, so it made sense to have the terrain tilesets fade to darkness to represent where that intersection lay. That is to say, most terrain expands horizontally indefinitely, so for the abstraction purposes of a 2d platformer, so that the ‘camera’ doesn’t just get a face full of dirt, we just kind of fade out and imply that it keeps on going behind us. However, something like a tree is different. It doesn’t intersect the viewing plane, it’s just a free-standing object. Now, logically the player character would be able to just walk around objects like this, so the abstraction gets a bit weird at this point, but it’s still part of the gameplay language that objects like this are as solid as the ground. Either that or they’re just foreground/background objects that you can just run past, but either one or the other: You can’t switch between the two because that would be confusing, the player would no longer know what was interactive terrain and what wasn’t.

So, freestanding objects like this can’t fade to black as they approach the viewing plane the way the dirt and grass tilesets do. Thus, I now need to figure out a new lighting scheme for these freestanding world objects that looks nice, communicates the lighting model, and doesn’t feel incongruous with the existing tilesets.

That’s the bad news. The good news is that one of the tilesets I need to work on now, the stone tileset, is an ideal test case because it encompasses both scenarios: In a cave-type area, stone behaves both as terrain, intersecting the viewing plane, and as a freestanding object, comprising pillars and other formations for the player to interact with. Thus, I can make a single tileset to look consistent with itself while achieving both of these tasks, rather than trying to coordinate between two separate tilesets while I figure it out. It’s probably going to be another one like the grass tileset though, with a lot of false starts before I get something I really like. The early progress on the terrain-type stone tiles seems promising, so it’s primarily a matter of making the freestanding ones look good within that context at this point.

I’ve been cancelling a lot of work on AnxEdit both because it’s almost done, so the remaining tasks are the ones I’m most inclined to put off, and because I’m having a hard time maintaining general enthusiasm and morale so that’s the first task that gets bumped. I did a few interface improvements this week, but I’m likely going to be tabling it pending the testing paces that it will go through when it comes time to use it to start building new animations. I’ll probably be replacing that time slot with another job soon, maybe streaming or working on a small game jam game.

Aside from that it’s just been a very stressful week. I found out that due to circumstances largely beyond my control money’s going to get a lot tighter for me, so I’m probably going to start trying to make some of the work I’m doing, here on the blog and elsewhere, a bit more profitable soon. Speaking of soon, I bit the bullet and shelled out for a newer, hopefully more professional site: It’s still not quite up and running, but you can check it out now at . I”ll probably spend a month or so double-posting blog entries between here and there before I shutter this site and start posting there exclusively. I want to ensure as many followers as possible make the transition. I should be able to get subscriptions set up for the new site soon. If you have any thoughts or suggestions regarding the new site, feel free to post them here or email me about them.


In this post I take a more in-depth look at the design of Undertale. If you haven’t played the game yet I’d strongly recommend doing so before reading, since I pretty much spoil the entire story, and also just generally recommend playing the game because it’s fantastic and also important. If you’re still on the fence, perhaps you might want to check out Austin Walker’s excellent review of the game. I also get into kind of spoilery territory with The Beginner’s Guide, which I also think is very good and worth playing. I also spoil Chrono Trigger: If you want to play Chrono Trigger you really should have gotten to it by now. It’s been 20 years. It is also a very good game.

Two weeks ago, discussing Undertale, I said the game wasn’t really about violence: At the same time, I dedicated the better part of a small essay on how its treatment of violence affected me. This might seem a bit hypocritical, so I’d like to expand a bit about why the discourse around a game that I say isn’t about violence tends to center so much around violence. Ironically this means I’ll have dedicated even more writing to this aspect that I say the game isn’t really about. So it goes.

In the immediate aftermath of playing the game, and in light of my brief and traumatic experiments with a no-mercy-kill-everything playthrough, I was powerfully reminded both of my experience playing Hotline Miami for the first time and, more recently, playing The Beginner’s Guide. These may, perhaps, seem like strange and tenuous connections: these three games are, to put it mildly, somewhat different from each other. However, what makes them similar in my mind, and gives them a similar emotional tenor in retrospect, is that all three have a kind of reflective quality, a style that holds a mirror up to the player and makes them question themselves.

Now: This is not the same thing as interrogating the player directly, which has recently become a somewhat trendy way of adding a sort of facile depth to game narrative. While it was genuinely a breath of fresh air when Braid and Spec Ops: The Line started directly questioning the narratives we build game genres on, and our relationship as players to those narratives, it’s quickly become trite and co-opted, a way to have a cake and eat it too, a way to pretend to be above the crowd while still being firmly part of it, superficially questioning the morality of the player’s actions while still locking their actions into place and encouraging the player to take them, rewarding them for doing so. Hotline Miami doesn’t ever say the things you do in the game are wrong or that you’re bad for enjoying them, though this is for some reason a popular interpretation: It merely takes pains to make it clear to you, the player, that you have just killed a shitload of virtual people, and leaves it to you to decide how you feel about it, while occasionally pointing out that you must be getting something out of it since you keep doing it. This mere act of acknowledgement, though, is surprisingly ground-breaking in video games. Just taking a moment to look back on what you did, and think about it and about its consequences, without telling you what that means… that’s unusual! Still!

Do you like hurting people (yes)

That’s what I mean when I say a reflective quality. There’s something to the game that makes you look inwards, to examine your own motivations, rather than merely questioning the narrative conceits of the game itself. Another word for what Undertale does might be ‘responsive’. It anticipates a large range of player input and responds to that robustly, creating the feeling of a directly interacting with a world, if perhaps within rather tightly constrained terms. The consistency and levelness of this responsiveness, combined with a lack of overt value judgment on the results, creates the reflective effect: Leaving the player to say, “Oh, this is the effect that my choices have had. How do I feel about that?” Now that I’ve said all this, it should become clear to those of you who are familiar with The Beginner’s Guide why I would be reminded of it, a game largely, at least in its latter half, dedicated to introspection and the questioning of motivations. Though there’s no guarantee that the player of The Beginner’s Guide would be the sort of person who tries to write interesting critical thoughts on games, who tries to curate and present interesting new content to their friends and followers, who is jealous of other artists for their creative ability and self-sabotages their own creativity in consequence, it’s a pretty safe bet that the player has had some impulses like that, and then by their resemblance to Davey is led to look within.

Undertale is reflective in many ways – though, unlike my other examples, it allows you a range of choices with which to express your intent before it shows the results of those actions back to you in a way that makes you think about why you performed those actions. In this context, and the context of violence still being a largely uncontested necessity of game design, one which we’ve all been trained to participate in in so many ways, it’s no surprise that the strongest and most overt response that people have to the game is how it reflects their acts of violence within it. Undertale doesn’t make violence seem bad by saying that violence is bad, but rather than echoing the commonplace narratives of just war that we use to create and maintain our structures of violence it shows realistic, if not explicit, consequences of violence. In a context where the hypocritical justification of violence is made so mainstream as to be invisible, the mere act of depicting it as an action with consequences becomes radical. Given that, I kind of wish the game had included more tacit interrogations of the sort of pacifism it encourages engagement in: being kind to people so they’ll do what you want, being the good guy to get the ‘best’ ending. There are moments, Flowey mocking you for trying over and over to create a happy ending, Papyrus lecturing you on privacy when you try to snoop around his shed, or Undyne questioning your motivations for showing mercy after earlier being willing to kill opponents, but for the most part this path goes unwalked – perhaps a good theme to expand upon in future games.

When nearly every character in the game starts out as an at least nominal enemy, it’s only natural that a violent path leads to a painfully lonely experience. In this context, every boss battle becomes an important turning point, central to the game not just for being challenging but because the decisions you make here will affect how the entire remainder of the game plays out. The background music for these fights is the same as the character themes that play when the character is doing something fun or goofy, though changed a bit, given heavier percussion, made to feel more consequential to reflect the weight of the situation. Listening to the soundtrack, you hear the history of the game through its characters: The characters, and the music which expresses them, form the core of the plot, rather than just decorating it like leaves and branches. It’s incredible just how deep this thread of musical narrative runs in the game: One can listen to the soundtrack a hundred times and, if not listening carefully and expertly, not notice the subtle references character themes have to other character and location themes, all together laying out the story of the game like a musical tapestry.

In most other RPGs, these characters would be party members, recruited to travel with you on your journey: However, in Undertale your journey is ultimately a lonely one, though companionship may lie at the end. The player’s relationship with other characters is evocative of the character Magus from Chrono Trigger. Like most of Undertale’s characters, he is initially presented as an antagonist, and though you do later find that his goals were noble the fact remains that he committed many crimes to achieve those goals. In the end, you choose whether to forgive him for these crimes and have him accompany you or to defeat him once and for all after his plans collapse. There’s power in this idea, that someone could as easily be enemy as ally: It forces the character to be complex, to be both ultimately relatable and morally suspect, to be both believably befriended or justifiably struck down. It’s not a coincidence that Undertale’s design premise, that you can befriend or destroy your enemies, lead to some of the most charming and endearingly written characters in the medium.

Undertale is designed to be played through multiple times, but not to be played to ‘completion’. It makes a mockery of anyone who wants to see everything there is to see, but it also conceals parts of itself behind layers that can only be seen by revisiting the game. What at first seems like a plot hole turns out to have its explanation in part of a completely different story path, and though this creates the impulse to try to see everything, inevitably if you try to see it all you’ll end up more dissatisfied, if perhaps better informed, because each thread comes to a logical conclusion, and pushing past that end-point leads to pointed and unpleasant questions about why you, the player, are doing this.

Most first-time playthroughs result in a muddy and troubling ending, where the player character escapes but nothing really changes in the underworld, and the actions the player took inevitably lead to some sort of political turmoil with far-reaching consequences. There are, however, two ‘true’ endings that fully resolve the plot: The true pacifist ending, found through not hurting anyone and going out of your way to ensure the major conflicts are all resolved, and the genocide ending, achieved by killing everyone in your path, and then picking fights with monsters until there are none left, thoroughly cleansing the underworld of life. If all you want is to make peace and friends, the true pacifist ending gives you that, letting you escape the underworld and resolve the tangled mess of Flowey’s warring consciousness. If you want to conquer, to become powerful, to prove you’re the best, then the genocide ending brings that to its logical conclusion with you destroying the game itself and moving on to conquer new worlds (eat your heart out Alexander the Great). At the end of both paths, you find a character who lives by the philosophy you have followed: Asriel, gentle and understanding, unwilling to fight back even as an angry mob fatally wounded him, or Chara, angry and bloodthirsty, eager to grab power by any means necessary.

These endings could both be regarded as ‘good’ endings, provided that in each case you actually believe in the ethos you were playing with. If you want both? You can’t have that. The victories you earn along one path or the other become hollow: If you try to make peace after you murder everything in a quest for power, the peace you find will be a lie. If you go back to destroy the world after saving it, you’re forced to confront how shallow your first quest was, how self-serving the peace you created. The very first puzzle in the game says it all:


One of the most overwhelming sensations I get playing Undertale is a sense of it pushing at its own boundaries. This is a game that uses all the tools at its disposal to tell its story: Where traditional JRPGs would tell a story through the text in a box, Undertale tells its story through the text in the box, the color of the text, the sounds the text makes as it appears, the speed the text appears at, the typeface used… Where other games play character themes for that character’s scenes, Undertale speeds up and slows down the playback of those themes to express the character’s current mental state, cuts them off and interrupts them with other characters’ themes. Where other games might occasionally save secondary data to enable certain secrets or to subtly improve the user experience, Undertale has characters aware of the player, aware they’re in some kind of contained reality, and working against or with the player using explicit knowledge of the game’s systems, or some in-world equivalent. At every corner, this game bursts at the seams, so eager to express itself that it breaks down every boundary we expect a game to have from years of playing games. Consistency in art style? Broken. Consistency in input style? Broken. Yet it pushes past these boundaries judiciously enough that every time you see it happen it seems special – and, while these tricks might seem wearing or pretentious if they were presented as clever, when these things happen they happen utterly earnestly, without pretense, so completely audaciously comfortable with how ridiculous they are, that it seems at once completely natural and completely absurd – natural because it’s so unabashedly absurd.

This style of blurring the boundaries between the art and reality is something that has become more common recently, but is still striking and effective when used well. House of Leaves, the novel by Mark Z. Danielewski, uses a tangled layered narrative combined with a truly labyrinthine layout to make its story of a house that’s slowly getting larger and becoming a gateway to somewhere else feel simultaneously closer to the reader and more distant, taking on some style of unsubstantiated rumor rather than just fiction; even as we know it’s all a story, it bypasses our defenses and accesses parts of our brain that are still scared of the dark.

At around the same time as this novel became popular, in the early 2000’s, the practice of creepypasta started to become popular. Creepypasta – a portmanteau of ‘copypasta’, a goofy slang for copy/paste, and ‘creepy’ – is basically the same thing as a scary story told around the campfire as applied to anonymous image boards. What it loses in atmosphere it gains in a sort of vague credibility. You can be pretty sure your troop leader isn’t actually a hook-handed murderer, since you can see both his hands, but when a story is posted anonymously to an image board it could be anyone from anywhere, could be a confession masquerading as a story, could be the last testament of someone being hunted, could be written by a sapient artificial intelligence, could be anything. Several games have come out of the creepypasta tradition, most notably a few explosively popular games based on the well-known ‘Slenderman’ mythos, but Undertale is one of the first to incorporate the stylistic approach of creepypasta into a more traditional game design. In so doing, it harnesses the strong points of both, using the traditional gameplay and narrative hooks to grab the player’s attention and invest them in a world while using creepypasta techniques to blur the boundary of that world. Blurring the lines like this is a powerful technique for a static narrative work like House of Leaves, but when the player is responsible for making choices life and death choices within that world the complete effect can be downright uncanny, almost too effective for comfort.


Another interesting trait of Undertale’s design is the ways it harnesses abstraction in its storytelling. Though all games are abstracted to varying degrees, and game designers have learned to leverage this abstraction to achieve some remarkable narrative and experiential results, Undertale goes above and beyond here as well – both in obvious ways, such as its use of low detail pixel graphics and a combat system rooted in metaphors, and in less obvious ways such as its wordless but symbolically loaded protagonist and its clever use of double entendre. Let’s look at each of those in turn:

Though pixel graphics are often derided as a cheap nostalgic appeal, there’s a lot of storytelling power to them that isn’t apparent at first. By representing characters and objects in a low-resolution way, you can omit defining detail that would otherwise change the interpretation of a scene, such as a character in disguise who would be recognizable to the audience, or you can make dissimilar objects look alike, such as a pile of autumn leaves and a campfire flame. By having something written on a sign and having the character read it back to you through a text box, you can create an unreliable narrator, where what appears in the text box isn’t necessarily what’s actually written on the sign. By abandoning a claim to be a pure representation of the reality of the game, you have created a situation where the reality of the game is being interpreted by the player based on available evidence, and what was once a passive experience is now an active one. These techniques aren’t unique to pixel graphics – you could achieve similar results with any style that isn’t purely representational – but by couching them in pixel graphics, a style that pretty much anyone who plays games is used to by now, means that you can do them so stealthily that they become effectively invisible while still subtly shaping the player’s experience.

The combat system is abstracted out into something resembling a bullet-hell style shooter game, where after each turn you must defend yourself from your opponent’s attacks by steering a little heart, representing your soul, away from their attacks. This allows the game to express the emotional state of your character, through the behavior of the heart, and the personality and emotional state of your opponent through their attacks. When they’re full of righteous fury, their attacks will be rapid and direct. while if they are questioning themselves or distracted, their attacks hesitate or drift off course. One could express something like this in a more conventional and purely representational battle system, but it would be very difficult to do so in a way that the player would actually pick up on, rather than crediting strange hesitations to AI problems or balance issues. Admittedly, part of the reason why it would be difficult to convey characterization solely through a representative battle is that it isn’t done, which creates a kind of chicken-and-egg scenario, so perhaps abstraction isn’t the only way to achieve these effects. Nevertheless, it is an elegant one – and creating this degree of expression within a traditional JRPG combat system would be impossible, regardless.

Though you name a character at the beginning of the game, which is implied to be the character you play as, the designer does something rather tricky here. The character you play as is actually named Frisk, and the character you named is the human who fell into the underworld long before the start of the game, catalyzing its events. However, Frisk and the named human, usually referred to as Chara, are related in some obscure way. Is Frisk a reincarnation of Chara, or just someone similar who becomes the vessel for an idea? To some extent it depends on how you play the game, but no matter what it’s always left sort of vague.There are other games with silent protagonists, which mostly use them as a blank slate through which the player can act, but Undertale goes several steps further, alternately writing the character’s blank-slate cipher status into the story, alluding to certain characters’ knowledge that there’s another person behind the character controlling them, and using the player’s actions to ‘bake’ a personality into the character from one of several options. This contributes to both the boundary-blurring effect mentioned earlier as well as world-building and characterization: Though it might seem tempting to replace Frisk with a more expressive and fully formed character, much of the game relies on the actor at the center being a conduit through which the rest of the story can flow.

Perhaps most fascinating is the game’s use of the abstractness of language. At certain points, the game uses double-entendre, not merely for jokes, but for key plot points. At each save point the game tells you you are filled with ‘determination’: Later, it explains that determination is the vital component of the human soul that allows it to fight off death even as the body housing it weakens. Yet ‘determination’ also comes to mean the ability to determine fate, to fight against destiny and choose the one hopeful thread in a sea of tragic possibilities – or, if you choose to play a genocide run, it becomes ‘determination’ as in ‘termination’, the will to eradicate everything you meet. Similarly, ‘save’ as in saving data, is conflated with ‘save’ as in rescue, leading up to a climax where the verb ‘save’, used so prosaically through the rest of the game, becomes what you must ultimately do to complete it. Word-play becomes word-business, and what would be a throwaway gag in another story forms the core of an emotional arc in Undertale.

The sheer number of ways that Undertale pushes against our understanding of what a game is or can be, how games work and why they work, is staggering. This is a game that both incorporates everything we have learned about how to make a good game and tell a story within it, and then builds upon that with innovations of its own, both realizing and expanding our group conception of how game design works. It’s a monumental achievement. This isn’t to say it’s a perfect game, but perfection is boring and disposable, just a version of something we already understood with fewer mistakes: Much better than perfect, this is a creative game, an expansive and loving game, a game that changes our understanding of what a game is and can be.


This week was largely spent developing a dirt tileset, the second of the three main tilesets I’ll be using in the first section of the game.


It’s not perfect. If you’re used to looking at tiles, the repetition in the solid areas is rather noticeable, and some of the edges are a bit off still, though mostly they’re alright. The low-color low-res style I’m using here makes it possible to conceal a lot of transitions which would otherwise be really obvious, something I’m coming to appreciate more and more about pixel art the more I work with it. There’s a power in abstraction to make bolder transitions and changes while maintaining a sense of consistency. I’m trying to experiment with colors: There’s a lot of blue in this dirt for how brown it is, leading to the bright pink highlights which suggest an unusual light source.

To help myself along, I also went ahead and made it so I could reload tilesets while I’m working, which should be a real timesaver in the long run, especially when I’m doing the finicky work of making sure the edges all line up.

I’m still playing with these colors to see what works best, and likely won’t make a final decision until I have all three starting tilesets complete and can see how they play against each other. Right now this whole tileset is just 5 colors, and each of those colors has a very specific meaning about where the tile is and how the player can expect to interact with it. Even if I were to go much higher resolution, I think I’d still keep this low color count approach.

Meanwhile AnxEdit is getting close to completion, though I’ve also been slowing down my work on it as other tasks come up to take my attention. I fixed the problem that was causing occasional frame loss when converting animations, streamlined a few operations, made things a bit more consistent… at this point it’s largely a matter of just using the program for a while, figuring out what bothers me, and fixing that. Probably once I finish the third tileset, the stone tileset, I’ll go back to the prototype animations I have and start preparations to turn them into the real thing: Add new frames to smooth out the motion, make sure everything’s consistent, remove extraneous feedback stuff like that white bar over the feet, and then export that out from Photoshop to Aseprite so I can start doing the actual pixel art part. Once I’m back in the thick of working with animations daily, any remaining flaws in AnxEdit will quickly become apparent.

Also, it’s still really ugly, so I might try to fix that too.

I mentioned this in yesterday’s post, but I still haven’t pulled the trigger on making the new site live, partially because it’s a lot of money for me and partially because I’m worried about losing the meager following I have here in the transition to a new site. It may just be a necessary evil and one I should deal with sooner rather than later, but it’s a concern nevertheless.

Anyway, next week another tileset and revisiting animation, probably.

I got the piece I wanted to post today up to around 1300 words, with 2/3rds of the outline left to fill, and realized I wouldn’t be able to finish it in time.That’s okay, though, because I just finished a piece of music; so now I can take the good ol’ musical coward’s way out once more.

While I’d wanted to do vocal tracks this year, I’m finding that more challenging than anticipated. I wrote the lyrics, I wrote the melody, and then when it came time to actually record some singing I hit a wall. I have a very hard time speaking into microphones, to the point where it’s kind of a phobia, and while I was hoping that this would be something I could just kind of power through in the moment it’s clear that I’ll have to go for a more circumspect approach. Towards that end, and for other reasons, I’m going to try to get into game streaming, particularly my playthrough of Dark Souls 3. I’ll be doing test streams this week: I’ve already tested my video and it’s working decently as long my internet connection holds (it gets a bit finicky), but I’ll need to get my audio set up, something I’ve been kind of dragging my heels on (for some reason). Once I get things set up in a way I’m comfortable with, I’ll start announcing streams on twitter. In the meanwhile I suppose I’ll be doing more instrumental tracks.

I’ve also been working on getting a new site set up. This is kind of a tricky decision, though, since as much as I’d like to get a more proper and permanent set up through which to promote the game and my work, I already have a fair number of followers on here, and am concerned that they won’t make the jump to a new site. Alternately, I could double down on this version of the site, put together some proper front pages, tidy everything up and make it look nicer and more professional, and then pay to have wordpress remove their part of the domain name and remove all the ads. That might be a better solution, but probably wouldn’t result in as nice a site and might cost me some fine control. I haven’t decided, but probably by end of month I’ll get a proper domain set up, on one service or another.

Anyway, sorry there’s no piece this week, but next week’s will probably be a doozy. By my standards anyway, since I tend to usually keep things pretty short.