The idea; an initial list (Friday, 8 May 2020, 14:44PM)

The idea

Initially I thought I’d make a game called COMBATS that would be a direct continuation of PONGS and BREAKSOUT, with a very broad brief on what the variations might be. But my brain doesn’t seem to work with such an unconstrained palette (and it’s not impossible I’ve already “done” all the obvious takes on a basic arcade game I wanted to take?). Given that I was recently approached to write something for a special issue of a journal on “adaptation in film and videogames”, I suppose my mind jumped to adapting films to a specific game, and since Combat was already on my mind, here we are: adapting various movies into the basic framework of Atari’s game Combat.

Rather that try to totally exhaust the possibilities (by, say, trying to make 36 of them), I think I’ll try to put together a package of 10 (like a “top ten”) and then can add new sets as with Chesses if it seems relevant.

I think the general principle here is to be reasonable free with the adaptation, including single player versions (maybe even mostly), rather than require simply mechanical changes to the original game’s nature. In essence, then, I’m almost talking more about having the tank from Combat “star” in Atari-esque adaptations of famous films?

If I’m framing this from the outset as a research piece, what would I say is interesting about this activity? It continues my more general interest and work on adaptation of many kinds (Sibilant Snakelikes, The Shining, Breakout Indies, Ancient Greek Punishment, etc.). In all those cases a major point of interest is in that central tension between source material and destination material, or tenor and vehicle, or what have you. It’s a design challenge, but always ends up being about the available mechanics, as well as the (often comedic) relationships between the imagery of the two different parts? A tank in a wig etc.

Plus it just sounds fun.

An initial list

Should they all be single player? Kind of weird given that COMBAT is two player, but perhaps that’s okay? I’m bolding the ones I think could work for now…

So I mean, that’s ten. I’m almost certain it would be possible to do ten more with more thought, this was largely a list derived from some obvious choices and the BFI 100 list.

So I suppose next is to implement the basic concept of a tank.

Graphics; speech (Friday, 8 May 2020, 16:07PM)

Graphics

Have now put together the world’s simplest version of a tank moving around with keyboard controls. By no means a difficult thing, but interesting because of the ways in which Phaser’s default toolset/nature is kind of inappropriate to the replication of the Atari game. Most obviously the framerate allows for very smooth movement and rotation of a tank sprite, which completely mismatches the original game, which is characteristically jerky. Further, antialiasing and high resolutions mess with the original graphical style, so that had to be turned off.

Current problem is that with antialiasing off, the tank still doesn’t rotate in the same way as the Atari tank. This is presumably because either a) the rotation algorithm doesn’t match, or b) the Atari version actually uses custom sprites for each rotational position. If I had to take a guess, I might say it’s the latter. In which case I’ll need to mimic that myself if I want to pursue making something that genuinely looks like the Atari version.

This is all kind of important because I’m (at least for now) trying to make the game in the “proper” (very low) resolution. There’s not much room for error here because it’s so obvious if a pixel is out of place. This also has huge implications for what I may or may not be able to communicate in the games themselves. Movement and shooting are both very expressive, but with such a small resolution, the complexity of things I can show might be very low. I’m most concerned about text for the moment, as at least a couple of the games might require that.

Speech

ALTHOUGH, perhaps I could look into some super degraded-sounding audio of a voice instead? I mean, that’s not really something the Atari could do though is it? I’ll have to consider that. Well I just did a little research and it sounds like the Atari 2600 could do some version of voice stuff! Here’s a discussion of that on Atari Age. There’s this ROM hack called Berzerk Voice Enhanced, which I’m going to assume stays true to the Atari’s sound capacity, given the nature of these people. Apparently the game QuadRun was also one that had voice, just announcing the title.

Having listened to the voice: hyper distorted is the main character. Maybe I could work out how to filter a voice/my voice in Audacity to a point where it sounds enough like this? For Rosebud, You Talkin’ To Me? Like Tears in Rain.

Collisions, confusions (Tuesday, 12 May 2020, 16:01PM)

Well I finally got a version of pixel-based collision done, even though it is truly, truly horrible. Truly horrible. Like, I don’t know quite how to express how shit it is? Depressing really. The problems all arise because I’m trying to mimick how the Atari works and how Combat works, and those aren’t tools readily available to me with something like Phaser. So it’s all very stopgap bad solutions I’m not all that happy with.

But then on the other hand I’ve been really keen to get the game to at least appear relatively similar to the original game so that it’s not distractingly not Combat, you know? I’m losing track of the point of that though. To the extent I’m not actually making an Atari game, why am I trying to fake Atari-style elements? And if I’m not making an Atari game, then what am I actually doing and why am I beholden to Atari graphics, Atari collisions, Atari sound design, and so on?

Is the point of this game that it’s literally Combat, and therefore an Atari take on these things? Or is the point that it’s move scene re-expressed using a tank and Combat happens to be an iconic tank game that gives it some kind of grounding? And also provides constraints of other things like graphic and sound and so forth.

I feel a little confused by the project right now. Can you tell?

Death to tilemaps (Wednesday, 13 May 2020, 11:55AM)

It’s not reflected in the codebase because I just discarded all the changes (thanks Git!), but I spent a good (bad) hour trying to use a tilemap as the basis for drawing levels in the game, to no real avail. It’s probably some dumb little thing I’m doing, but in the end who cares.

My new plan: draw tiny little levels in Pixen, do pixel collision as with the tanks, and hey presto.

This stuff is SO boring.

It’s probably a good data point showing that not every single thing that happens during development is actually of interest or really worth dwelling on. This was just a technical moment that led nowhere.

Dead ends and questions (Wednesday, 13 May 2020, 14:47PM)

Just discarded a bunch of changes where I tried to make simple image-based collision stuff work and… it didn’t. A frustrating time indeed.

So let me ask me some questions then: does it need to be Combat? If so, does it need to be a close match to the Atari style at the level of things like collisions? Rotation?

What would the game be like if the graphics were just literally the standard tank moving around, rotating per the standard approach in Phaser, etc.? And we just went from there. Not totally Atari, but in the spirit of it in terms of basic graphic restrictions and basic mechanics?

Fundamentally: why does pixel perfect collision detection and the like even matter? It fucking doesn’t, right? And if it doesn’t matter, why not make the game the “easy” way, probably still with a tile-map involved, since that seems like a helpful way to draw scenes, but relying on Phaser’s physics and velocities etc.? Try to lean into what the library gives me while making it “enough” like the original game to be a clear tribute.

So that would mean standard collision detection for sure, which would mean applying a more standard velocity to the tank, which might eliminate the jerky style of motion which would be a shame, but probably not the end of the world??? Maybe you can stagger the updating of rendering or something weird?

This all suggests that I should perhaps start a new branch, reimplement the game and see what happens?

OKAY.

Tiled (Thursday, 14 May 2020, 12:00PM)

OKAY WELL. I did what I said I would do in a new branch this morning and have already merged it feeling fairly successful. We have:

The result is something that ACTUALLY LOOKS LIKE COMBAT, with the one significant difference being a lack of pixel perfect collisions. It’s not impossible I could revisit this specific issue, I suppose, but I kiiiind of doubt I’m going to do that, as it’s pretty strenuous and it’s not THAT clear it’s vital to what the game is “about”.

So frankly I’m kind of happy. I need the tanks to be able to shoot (each other? with scoring?), now, and then after that I think I can actually look at the titles of films I was thinking of and actually work out how they might fit this frame?

Amazing… maybe I’m actually allowed to start working in a way that doesn’t suck?

The basics (Friday, 15 May 2020, 10:14AM)

Well the basics are in now I think. Shooting now mostly works, including when you shoot and they teleport into a wall (it tunnels out in the same direction) or they go off the screen (it wraps). So I think we’ve now got a close enough framework to continue with.

Next is some kind of text display, some experiments with producing sound a bit like the Atari voice playback thing, and then also obviously some honest-to-god design of the levels themselves. Phew. At which point we might actually find out whether this even works or not.

Audiophile (Friday, 15 May 2020, 11:21AM)

Spent some time in both Audacity and LMMS trying to work out some way to make my robot voice as it would potentially just obviate any need for text at all, as I’m not sure about text and matching resolutions. Combat itself doesn’t display any text other than numbers, and I’d want to be potentially able to display things like “Rosebud” and “You talkin’ to me?” etc. Especially for the longer phrases which help to ground the game in the movie, it would be a tall order potentially to display. I doubt it’s impossible (possibly breaking it up or scrolling it?), but I like the idea of at least pursuing an audio version first, especially since actual speech is so important to films (usually - talkies and all that), and because it’s funny that the Atari did actually have this capacity in a super limited way.

In the end I couldn’t get anything going with various chains of filters (though LMMS was way better for experimentation because of its non-modifying effects stuff making experimentation faster). I found interesting distorted sounds, but non of them really sounded much like the robot-y stripped back sound from Atari. I guess it’s not that it sounds like a robot but just insanely “low resolution”?

In searching for tutorials (which mostly focus on echo and multiple samples with different pitches and tempos), I re-ran into the idea of a vocoder, which I didn’t use the first time because I had a mono recording (stupid reason, but this is the way it goes when you’re hurrying around looking for quick fixes - it takes forever). After all, a vocoder is kind of exactly a version of what I want - a voice encoder. And Kraftwerk used them. So after playing around with the parameters I absolutely don’t understand in Audacity’s vocoder, I got something that (at least for me saying “Rosebud” in an uninspiring way) did actually seem to produce something kind of nice. I also tried it on the actual audio from Bladerunner’s “I’ve seen things” seen, but in the end that audio was too messy. It’s likely that it’s worth rerecording these texts with some kind of simple speech synthesis voice, or even me doing a monotone or something.

So this did end up being a productive little session in the end, though it was extremely frustrating for much of the time. I think that, in keeping with the obvious limitations on Atari hardware, it would also make sense to keep audio quite short. So rather than have the entire “I’ve seen things” speech I’d rather just have the first sentence or something… or that and “time to die” before the death spin? Or not. But anyway, play around with that brevity.

Design; Other movies?; Materials? (Friday, 22 May 2020, 10:27AM)

Okay well I think I’ve spent enough time fiddling with the base case of Combat and getting some of the little flourishes that will make this thing seem more or less like close-ish versions of the original game. Now there has to come a moment where I actually think about each of the movies and work out what I’m supposed to do for each one.

Designing

Citizen Kane

The idea is to do the rosebud death scene because it’s the most iconic (is that my criterion overall? I suppose it needs to be to at least some extent to overcome the potential that the scenes won’t be all that recognizable otherwise!). In that scene he’s lying down in bed, he whispers “rosebud” and then he dies and drops the glass globe that contains the little scene referring back to his childhood home. (Here is the relevant clip)

So the two keys here are that, first, the tank should say Rosebud, either with text or, perhaps better, the hyper compressed voice that the Atari seems to have been able to do (and which I seem able to maybe approximate with Audacity’s vocoder). And second, the tank should die immediately afterwards, which is represented by the violent spinning from the game. Comedy gold?

How do you trigger that final reminiscence? I suppose you just use the “fire” button (which can double as “action” in some of these games). And I suppose you’re stuck in place, so you can’t navigate anywhere. How do I convey the idea that you’re in bed in your giant mansion? Some kind of primitive mansion drawing with the player’s tank at the centre of it?

Do I need to think about the bauble? Could that be represented by shooting out a final bullet as you say rosebud? Or is it best left out because it’d be pretty unclear unless someone knows the movie well enough.

(On a stupid note, could I make the entirety of Citizen Kane out of Combat? Recreate a version of each scene? Probably not.)

The Godfather

I’m thinking of the tollbooth scene because it’s quite a standout in terms of drama. (Here’s a clip of it.) So he gets caught behind another car (tank?) at the toll area, and then other people (tanks?) emerge and everyone shoots the shit out of his car, and then him as he emerges, including shooting him when he’s dead.

It seems obvious I’d have to identify the car and Sonny as one thing, so there wouldn’t be the idea of actually getting out. But beyond this the player could drive down a road, get stopped behind another tank (or maybe just stopped by the toll arm? How do I explain that you can’t shoot? I guess he doesn’t appear to have a gun with him…). Then when you reach the right point, the tank in front rotates and other tanks come out from the edges of the screen, and they all shoot you.

It’s part of the scene that Sonny doesn’t die immediately, so is the idea that your tank gets buffeted by the bullets at first, and then at some point it dies and spins? And a tank shoots it even then? How difficult is all that choreography going to be? Perhaps with tweens it’s not so bad actually?

2001: A Space Odyssey

The monolith scene for this one. In essence we can just have our tank and then a bunch of other tanks shooting at a monolith in the centre of the scene? I’m wondering about the colors of the tanks - if the player tank is different to the others, it gives the idea that they’re against each other? But actually isn’t that what happens anyway now that I think about it? It starts of “peaceful”, then the monolith kind of teaches them violence? I mean, it’s very appropriate… let me check it out. (Here’s a clip.) So they wake up in the morning, it’s there and they shout at it, they form a ring around it, then a couple touch it, … and actually no violence erupts! Completely misremembered this scene? They just crowd around it, touch it a bit, and then that’s it.

Maybe that’s fun though anyway. You’re just one tank among many. The monolith is there, and you all kind of bump into it over and over and then at some point the scene ends? But the monolith is also ultimately about giving them the idea of a weapon, so it would be nice to think about some way that it connects to the tanks shooting? Maybe there’s a “moment”, the monolith disappears, and then you can all shoot? And they can all just shoot each other? Hehe.

Au Hasard Balthazar

Given that the film is basically just about the fatalistic journey of a donkey, it would make sense to imagine this as Balthazar just showing up in a Combat level? So just a donkey instead of the other tank, and it just stands there not really doing anything. You can shoot it or not (I guess it spins?). How does the scene end if you don’t shoot it? Just a timer (like the real game).

Taxi Driver

I liked the idea that there would be a mirror you could drive your tank in front of, and if you shoot the mirror you’d have the sound effects from the movie? Just “You talkin’ to me?” over and over again. That would be all. Does the overall apartment need to be there? Just a mirror? Will it be clear what it is? (To what extent to do many of these end up feeling like just triggering movie quotes?) Again, there’s an appropriateness to a tank doing this routine, with its protruding gun and threat of violence? The idea that it’s trying to “act tough” in a mirror is quite nice? Would you actually shoot at the mirror, or just trigger the sound? I guess it’s just the sound because in the movie he doesn’t shoot he just pretends he will, or menaces the reflection person with the gun. While admiring himself. I like the idea of the tank admiring itself. (Here’s a clip. There’s reference there if I did want the apartment too, but I suspect it might end up just visually confusing…)

Blade Runner

Well it’s more sound stuff right? Ultimately the point is that you shoot the other (replicant) tank and it does the “I’ve scene things” speech. How can you tell it’s dying? Could it give the speech while spinning? Or is it a better payoff if it says “time to die” and then spins? It feels like the spinning indicates that you’re dead, not dying, given that at the end of the spin it immediately resets? (Here’s a clip.) In the movie he rescues Harrison in a human act, then sits down and does the speech, then dies. He dies on his own, so there’s no obvious reason for you to shoot him? In fact he’s in pretty nice shape when he dies.

This could be pretty confusing. I’m not sure how to represent the idea that one tank saves another (especially not from falling in a world with no height). And if it just kicks off with the speech and then he dies (spins), then what is that, really? Perhaps the scene just has too little activity for the player to really work?

THX1138

I mean, this would literally just be a single player tank in a white world driving around endlessly until the round ends? (This makes me wonder if it makes sense to have a visual timer of the round so that people know it’s going to end?) Given the whole “high brow” vibe of the other movies, I kind of wonder with this one fits the mould all that well, but it’s a pleasingly iconic scene.

Other movies?

In current excluding a couple of blockbusters (Indiana Jones, Star Wars), I’m aware my list is a lot shorter now. Especially if Blade Runner doesn’t work either. There’s also just no real unifying thematic across all the little scenes. Which I don’t think has to matter, it could just be the project to make whatever seems fun - but it could also be plausible to pick scenes about death and violence (which are the two things that Combat is basically about) in order to highlight those elements and their lack of subtlety?

Some other thoughts (without thinking hard about what would be the content of the game)

The key is that I think it’s interesting to represent unlikely movies (and perhaps for that reason looking at classic “great movies” is good because they give us a canon) specifically because of the awkward fit. It can’t be the case, though, that every time the punchline is just “oh x shot y” or similar. It needs to leverage different parts of Combat (movement, spinning, voice, walls, shooting) and ideally even reveal something in the game or make available some element that’s not usually there.

Materials?

Looking at the design and development work so far (Friday, 5 June 2020, 13:40PM)

Having now actually made a bit of headway into the game I can probably actually say a few things about what it’s been like and what I think might be interesting about the process.

I’ve now put together versions of Citizen Kane, L’Avventura, Rashomon, and Au Hasard Balthazar. I’d originally been assuming that the majority of the work would be about the specific relationship between a tank and its actions and the potential of the scene, and that has been important, but it’s been nice to run into other elements as well.

Citizen Kane

This has generally been the one I’ve had the clearest bead on from the beginning, though I’m not too sure it’s the “best” of the ideas. It’s just punchy and amusing. It does leverage the core idea of death so central to Combat, and in particular the visual and audio representations of death (most compellingly the spinning). I think there’s something to the idea that you press your shooting button to die instead of to kill, though I don’t know that it makes sense in the context of Citizen Kane to say that he chose to die in that moment?

An alternative would be to say that the objective of this game is to say Rosebud before you die? To “realize” or recall the importance of Rosebud before death takes you? In which case I would just have a timer running (visible or not?) and if you hit the action button you say Rosebud and then die, perhaps you get one point for saying it.

I actually quite enjoy the bullet that is fired by the Kane Tank flying off screen as it genuinely does end up mirroring the bauble/globe from the movie for me. It’s a little thing, but I think it’s fun that there’s that correspondence, and it does make me think a bit about the underlying Atari implementation of the missiles and how the missile sprite is used for different purposes in different Atari games (the ball, the bullet, something I can’t remember in Adventure, etc.).

So anyway maybe I rethink the timer idea - since a timer is part of the original game, and it perhaps gets more accurately at the idea of death taking you when it will. Would it be funny/fun if you could say “rosebud” multiple times and kind of build up points with that growing force of memory? Perhaps so.

I do think even in this ultra simple idea there’s something to it.

L’Avventura

This was a latecomer to the thoughts, spurred on by wanting something more like a “great movies” list rather than just “movies I can think of”. I also think it’s good to maintain some consistency in the list so I could make other games looking at specific genres or more popular movies and so forth.

The core point in L’Avventura is that the guy is searching for the missing woman fruitlessly. It’s actually been a long while since I saw it (and perhaps I should rewatch), but I think the central representation of the tank on an island looking around is solid. The timer would be present here again (and perhaps it needs to always be present really, given it’s a major part of how Combat works), to signify the end of the search.

I’m starting to wonder whether I need little titles at the end of the game telling you the outcome beyond just the points total. “You didn’t find her” or similar. Which is what would always happen.

I contemplated having an invisible and unfindable tank also on the island, but I don’t think that really makes sense.

I think there’s a decent poignancy to the idea of one Combat tank looking for another, presumably to fight to the death (perhaps mirroring things about the relationship in L’Avventura? Does he even want to find her, really?). As an image I think it’s a pretty strong one on loneliness, futility, nobody to kill, and so on. Just realizing it probably does make sense for the tank to be able to shoot, a way to express frustration? Reaching out?

I guess you’d never score any points.

Should these things also have titles at the beginning telling you what you’re meant to actually do? (I mean, Atari games wouldn’t normally print instructions to the screen would they? I’ll have to check up on that a bit. I do like the idea of producing a proper little manual, but it might be overkill and too much work for little old me?)

Rashomon

Here’s an example of one that ended up being as much about the API of Phaser as Combat itself. I think this is a fairly “obvious” idea in terms of representing multiple versions of the same story by having multiple cameras simultaneously. It would have been possible, too, to present four separate minigames, each of which would present the idea of combat in a slightly different way, but I think that the four cameras probably does a better job or being a bit confusing, showing the idea of perspectives and so on. I like that this enabled me to recognize that Combat itself essentially has this idea of multiple “perspectives” in that it presents multiple game modes you can cycle through. I’ve leaned on the idea of differing kinds of visibility in particular (walls, tanks) to that end. It would be harder (maybe vaguely impossible) to create the same kinds of mechanical differences (e.g. bending bullets or not, bouncing bullets or not) based on the camera.

In the end this should probably be a relatively playable game, reminding me of Bernie’s game I can’t remember the name of right now (the platformer with the jumbled up views). I’ll need an AI tank I suppose, to make this fun, but I can probably get away with largely random behaviour as it’s not really the point (I suppose I could make it two player, but then it raises the question of the various other one player versions. Does one player get to be the sled? The nurse? in Citizen Kane? The missing woman in L’Avventura? Maybe.)

It’s appropriate that Rashomon is about an act of violence obviously, perhaps it could be too on the nose, but it’s fun to draw a multi-camera tank game into comparison with one of the greatest movies ever made etc. etc.

Au Hasard Balthazar

Bit of a one-liner, but actually one with some teeth I think. You simply have Balthazar the donkey there instead of another tank. If two player then I suppose the other player controls the donkey, if not then I think it’s the player as tank and the donkey maybe just wanders around or stays put. You could shoot the donkey (if you’re a prick) but you wouldn’t get points for that (or for anything? Random points for the “au hasard”?). I think there’s something good about pitting the tank against the donkey (or the potential for it to be “against” anyway). It has pathos, it’s that classic story of “choosing not to shoot” that videogames leverage from time to time. I like the idea that it could be peaceful.

So far

Well I think it’s going slowly, but perhaps a little better than anticipated. I’m finding relationships between Combat and movies. Yadda yadda (lost the thread on this because I wrote the majority of this entry while waiting for my emergency passport, and now it’s the next day and things have been full on). So whatever it says above, let’s say that’s what I have to say.

Hi; my evil twin (Monday, 29 June 2020, 17:14PM)

Hi

Been a while obviously. Quite a lot has been going on in our lives in the last few weeks, notably emergency travel to Aotearoa, 14 days in managed isolation at the Auckland Novotel, and then family time in Wellington.

I’m finding it pretty hard to get back in the saddle to be honest, but I’ve at least chipped away a bit and got a couple of small pieces put in place, including a very basic Taxi Driver with a mirror.

My evil twin

One somewhat interesting thing about the mirror is that I wasn’t able to just create a mirroring camera (the flipY property of the camera in Phaser didn’t appear to do anything), and that meant I had to create an actual second tank that mirrors the movements of the player’s tank, and then train a different camera on that tank to create the appearance of a mirror.

In one way it’s disappointing because it’s not a proper mirror, arguably, or at least the metaphor doesn’t extend down into the engine itself. On the other hand, having the other tank “really there” gels with the traditional structure of Combat having two tanks (and frankly of the Atari having its memory organized that way etc.), and in this case is especially apt as the scene literally is him pretending to face off against an imagined enemy. So there’s something there which I do find pleasing in the end, beyond the base level idea that there’s something fun about seeing a tank threatening itself in the mirror without actually doing anything about it…

Where are we now?; 2001 thoughts (Thursday, 2 July 2020, 12:00PM)

Where are we now?

So I have eight movies in different states of completion and satisfaction:

Phew, that’s a lot. So it’s clear I need to think through 2001 further, but for the rest I think we’re quite far advanced?

2001 thoughts

Looking at the movie summary of the appropriate sequence:

In the prehistoric African veldt, a tribe of hominids are driven away from their water hole by a rival tribe. Later, they awaken to find an alien monolith has appeared before them. Seemingly influenced by the monolith, they discover how to use a bone as a weapon and return to drive their rivals away.

So a plausible setup is that we have the player’s tanks on one side of the screen (matching colours), or even just the player alone, and then the enemy tanks on the other side guarding a watering hole? If I want to be that literal? I mean that at least gets at the idea that one side dominates the other prior to the monolith?

Then the monolith appears and afterwards the player side (either them alone or all their cronies) gain the ability to shoot and thus kill the enemy? This works with or without the watering hole obviously. Could develop a level where a single enemy tank blocks some specific way out or a clearly valuable zone (watering hole doesn’t make sense, but maybe there’s some kind of symbol of a flag or a coin or something?). Then the monolith comes, the player approach is (they would have to touch it I think), gains the ability to shoot, kills the enemy tank, and takes the flag?

In which case it could say CAPTURE THE FLAG at the start, which is semi-classic combat style play (if not Atari Combat play), and you can’t accomplish it until the monolith grants you the ability to use your weapon.

Although I think it would be visually fun to have multiple tanks all wandering around like apes, it’s not super practical for me to implement, and may not actually be more effective conceptually that sticking with the Atari two tank thing (not to mention that it’s kind of bullshit to have more than one tank in the Atari context - which throws The Godfather a bit into question too… should it just be one tank that blindsides you in fact?)

Okay I think that’s the “correct direction” for right now.

Two more needed! (Monday, 6 July 2020, 13:41PM)

I still need two more films to have a kind of “top ten” list vibe, which makes the most sense for films I think? I’ve been looking through the BFI top 100 list…

So…

So if I’m picking the ones to pursue it’s La Jetée, Beau Travail and The French Connection, all of which have kind of potentially interesting ways of engaging with the original mechanics (duplication, dancing, and listening respectively). I guess I just try making all of them and see where it leads me.

Shooters (Wednesday, 8 July 2020, 15:11PM)

Just popping for a very small note that popped into my head while working on Beau Travail, which is the frequency with which I’m finding myself overriding the shoot() function so that it doesn’t do anything, essentially disarming the tanks. This is true in at least three of the games/films, and in all the other cases the act of shooting tends to be significant if taken (perhaps with the weird exception of Some Like It Hot, which comes across more as a gag about the movie than a representation of the movie or a scene from it).

Return of the guy (Tuesday, 28 July 2020, 14:46PM)

I’m scrambling back into this saddle. Importantly, I’ve decided to make this game and its ideas the subject of a journal paper for /arts/ and a grant application, meaning I need to get my head back into it, feel like it’s a good idea again, and generally improve and think about What It All Means.

Probably the best way to do this is to restate my intentions, which I can probably do in the why document, and to just get some more work done on the damn thing.

Citizen Kaned (Tuesday, 28 July 2020, 15:58PM)

Mostly anyway.

It was good to actually work specifically on making a scene come together rather than my recent focus on the AI Tank which was boring and dissatisfying because writing a bad AI is just not a fun activity I suspect. Working on Citizen Kane made me think a bit more about a few things, which helps, and here are something I did/thought

Redrew the symbolic castle I had with a sort of loose attempt at the “actual” Xanadu from the movie. I was initially concerned about the resolution of the tilemap in combination with the complexity of Xanadu, but in the end I think it perhaps works out, looking like a sprawling structure of some kind, and it’s not inaccurate to the movie. A key here is that question of whether you’re going for the spirit (in which case a symbolic castle is fine) or the actual thing (in which case it’s “important” to have the proper castle represented, albeit within the confines of the tilemap’s resolution and color palette). Both options have their merits and I think it’s interesting that in the end I was able to (more or less subconsciously) split the movies about 50/50 between authentic and spirit. Both are worthwhile experimental modes. Funny that it comes up everywhere, from the overall design approach to the visual aesthetics and so forth.

Added point scoring when you remember Rosebud (by shooting). Now you shoot, it “fires off the memory” (ha ha!) and you say “Rooooosssseeebuuuuud” and then get a point for remembering. I guess that’s justifiable if only in that you accurately recreated the scene, but perhaps there’s also a kind of victory for Kane-tank himself in remembering? The mechanical fact of scoring a point is one of the few things we can leverage out of the Combat set of rules and representations, so it’s nice to use it wherever possible.

Added instruction screens with text. Added that to the whole game of course, but implemented it specifically for Citizen Kane for right now. The instruction is just “REMEMBER”, which is nice because it contains no “spoilers” about Rosebud? I like that it’s ambiguous and pairs so poorly with the shooting mechanic? (Other than synapses firing? Ha!) I should really think/check whether we should interpret Kane saying “Rosebud” as him remembering Rosebud or something else - there might be a better word. Whatever the word is, it gets straight to the heart of this project of representing more complex ideas in Combat. While of course, of course falling prey to just the same kinds of stupidities games suffer from when they have “Press X to pay respects” or similar, right? The fact you shoot and die, however, is what I think makes this worthwhile (as opposed to a cutscene or lame animation without consequence).

Added game over screen with text. Uses the same code unsurprisingly. The game over when you remember is “YOU REMEMBERED ROSEBUD”. Originally I’d thought it would just be “YOU REMEMBERED”, but it actually typing it in I added ROSEBUD both for the blank comedy of it, but also because it actually solves the potential problem of players not having their audio on. It makes the game more playable, and I should probably think about whether there are ways to do similar signaling elsewhere - though perhaps Kane is the only one that relies so directly on a specific audio file? Well Taxi Driver too.

Added failure timer if you spend too long without shooting. Then YOU DIDN’T REMEMBER ROSEBUD (funny to remember Rosebud in the failure message). Definitely like that there’s a fail state.

A couple of extra thoughts to go on with…

Should you be able to move a bit? Currently all you can do is shoot, should you be able to thrash a little bit in your bed maybe? Just back and forth? In the movie he looks pretty damn incapacitated, but maybe just a wiggle between the three adjacent frames?

A Combat tank looks like a sled? I mean, it does. Is there some weird possibility of leveraging that? Having the enemy tank appear as if it’s rosebud? Could it be hit by the bullet and also die (to emulate it being burned?). Or does that get too weird and abstract? Where would the Rosebud tank be? Why does Kane-tank then also look like a sled by implication? Suspect this doesn’t work.

First draft? Violence stocktake (Wednesday, 5 August 2020, 17:13PM)

First draft?

With today’s push I feel like I probably have a “first draft” of the entire game in the sense that it has a menu, a flow, and versions of every subgame I’m envisaging including.

By and large I find the total package pretty satisfying, barring a few obvious fixes that are going to be needed here and there. Some Like It Hot continues to be a bit of a lame duck, but I’m holding onto it just as a teasing kind of thing about men-as-women trivialization.

Some of the ideas that I assumed would be satisfying, like saying “you talkin to me?” into the mirror as a tank do indeed feel satisfying. In fact it was kind of only in playing that that I realized the game serves almost as a kind of prelude to a regular match of Combat, one tank psyching itself out to go and be violent. There’s something good in that.

Violence stocktake

In terms of “commentary” on violence then we get

So you know what? Not really too bad at all. In all these cases the violence is spotlighted by Combat, and Combat is softened/complicated by the films.

J+M response (Thursday, 6 August 2020, 9:50AM)

Overnight I got responses to the game from both Jim and Mary (my parents if you’re reading this and you aren’t me, but you probably are me, hi!).

So, the bad news is that the game did not come across well for either of them. I’ve sent a follow-up email, but the key issue appears to have been to do with a lack of agency and thus anything that happened seeming kind of random. At least some of this appears to be a lack of understanding of how to drive the tank. Although the menu system does lay out the controls (arrow keys and space), that clearly wasn’t enough. Further, not every game even lets you move in the first place (Citizen Kane is actually the only one now I think about it). Further, your movement is not always very relevant in other games (The Conversation, Taxi Driver at present). And even when it’s relevant it might not really be acknowledged by the system in any particular way (e.g. “dancing” is just dancing, it’s not something where you get points for it).

Now, a lack of feeling major agency isn’t a huge issue in terms of getting the ideas behind the various games, but it is a bit disappointing just in terms of wanting the game to feel fun/amusing/interesting rather than frustrating/boring/opaque.

There are a few bits and pieces I will do and could do that would mitigate some of this, but not the major issue of the games themselves just being low-agency/low-feedback in the first place…

So there are a few options here that might offer solace. Mostly I probably just need to get the damn thing finished and move on though!!

Going manual (Friday, 7 August 2020, 14:56PM)

Alright so after the sadnesses of lacks of understanding, the idea is to pursue far more detailed instructions more in keeping with the Atari manual for Combat.

Just found this amazing nugget of information: All games end after 2 minutes, 16 seconds. Which is 136 seconds. Wonder how they came up with that?

Here are some notes on the Combat manual that might pertain to my game:

So, looking at that, it might be more possible to take some of the format (font, headings, screenshots, etc.) but not really the content/writing itself. My premise would be for now:

What about game over screens? Do I keep the same kind of quip-based ending? Or do I describe the basic outcomes in the instructions and have no reaction from the game when the round ends? If so I’d generally need point to indicate outcomes a little better, and just some kind of game over with “press a key” to return to the menu.

Also should add escape to menu any time.

Manual a-go-go (Friday, 7 August 2020, 16:17PM)

This is definitely going to work. Although it’s not perfect at all, I really like the new menu in the Atari Combat manual style. It allows me to get more information in there, which may be advantageous in a relatively confusing game like this? And it’s just a fun contrast (as with the original Combat) between the level of explanation and detail and then the actual game that emerges on the other end.

Having made a basic menu, the next step is to produce at least one or two instruction screens for the games and see how that might change things.

Some Like It Not So Hot (Wednesday, 12 August 2020, 7:41AM)

Well after some back and forth with Mary I think it’s time to admit that Some Like It Hot just isn’t working, as much as I wanted it to stay there just to take up the space, and as much as I thought it might somehow be a bit funny and say something about crappy re-skins and so on. No. Okay.

So, that being the case, I need to come up with something new. Here are some thoughts.

Tokyo Story could be a good one. It’s one of the great films and its sense of stillness is quite a good misfit/contrast with a combat game. I’m a bit unclear on the “violence” angle (though we could perhaps talk about the two “foes” sitting side by side? But that does work at all with the movie). Could have the wife tank die at a certain point and leave the other sitting there on its own. How would I a) make you feel like you are one of the tanks if you’re just sitting together? b) make you feel connected to the other tank like you’ve been together forever? It reminds me a bit of passage actually, so perhaps something along those lines with it being more like you drive along together for a while but then she dies? Hard to say.

The Passion of Joan of Arc definitely fits the violence bill. Could be too related to The Godfather in the sense of a powerless? death (I mean, Sonny’s no martyr, but there’s a mechanical similarity). Making it black and white would of course be pretty funny (though Citizen Kane is in black and white and so are other movies in the list, so that wouldn’t really make sense either). Hmmm. I suppose I just can’t quite see it. A close-up could be kind of good? Extreme close-up, you can’t move because you’re tied to the stake, and at some point the other tank finds and kills you? There’s maybe something there… it would be the only game where you have no control at all except maybe a bit of a Citizen Kane style movement? Does she move on the pyre in the movie or is just very still?

Apocalypse Now is violence personified. Perhaps some kind of thing where you have to make your way to a Kurtz tank arduously and then kill him? He says the horror the horror?

Stalker could be funny because it’s already a “film adaptation”. The idea of the room they can’t approach could be of interest? An invisible barrier? But I don’t know, that doesn’t sound all that fascinating.

La Jetée is still pretty amazing? Maybe there’s a way to resurrect it in some less complex fashion? Maybe you’re just fleeing down the “jetty” and there’s another tank the same colour as you that maybe is confined but it clearly moving in the same way? Maybe if the jetty is confined enough you can’t turn around (and spoil the idea…). And you’d have the more frame-based idea in there as well so that it would only render sporadically…

HMM. More thinking please.

Quite a while later (Wednesday, 25 November 2020, 11:23AM)

Guess it has been… 15 weeks since I wrote a process journal entry, which isn’t amazing of me. Not very rigorous, not very procedural. I also haven’t necessarily done a ton of work on the game since then, but have finally polished off a few of the more frustrating issues.

Some Like It Hot > Tokyo Story

In particular, responding to the key concern of the previous entry, I replaced Some Like It Hot with Tokyo Story which was fairly straightforward, possibly too straightforward? It’s just a lonely old tank on its own until the end of the round. An attempt to capture some of the pathos at the end of the film in which Shūkichi Hirayama sits alone in his house without his wife. It seems like in the Combat context it makes sense that a similar loss would be if the player tank lost its counterpart in the enemy tank and thus lacks grounded and motive. Of course the Combat tank doesn’t have a neglectful family etc., but we can’t have (or not have?) everything.

Endings and palette cycling

One of the final hurdles to complete that game revolved around the way that Combat does a quite alluring palette cycle at the end of any given game/round. I wanted to recreate that with some degree of accuracy, which turned out to be kind of fiddly but doable in the end.

Also added the flashing of the scores to warn about the impending end of a round due to its time limit. And generally spent time with the code making sure the various ending were all working correctly, which was a bit of a kerfuffle but worked out.

Splash page, preloader

Finally, I put together an opening title screen based on the Combat box art, modified to reflect the new game. It’s pretty cheesy, but I think relatively satisfying nonetheless.

That and a working preloader mean the game is essentially finished as far as I can tell?

Next up: press kit?

That seems to mean that the final stage is to write up the press kit and get the game out into the world, freeing me up to work on something new for the first time since I started this thing back in May. Bloody hell. So all told probably a development time of something like seven months? Ugh. Oh well. It’s not like it really took seven months, there were some big ol’ gaps.

Oh yeah: /arts/ article

Oh yeah I wrote an article about the game called Film Adaptation as Experimental Game Design for the journal arts and its Special Issue on The Art of Adaptation in Film and Video Games.

That can really serve as a kind of closing statement for this particular project I think. It’s a fairly deep dive into the design process and intentions behind the game.