The business entity currently calling itself “Atari” published a new tech blog today, to address their current status with delivering the long-overdue Atari VCS systems. As usual, things are not well.
At the time of this writing, the world is in the grip of a global pandemic of COVID-19, which originated in China, and has lead to major economic disruptions as authorities in China have shut down entire cities in an effort to contain the spread of the virus. Under such circumstances, it’s entirely understandable that this could cause delays.
Rather immediately, though, we see the usual disturbing signs that not all is well with the project, beyond these circumstances that are beyond anyone’s control:
…We have confirmed delivery of enough parts by the end of March to build our first 500 Atari VCS production units. A good portion of these first Atari VCS units are earmarked as dev kits for developers.
Four things leap out about this revelation:
One: Previously, Atari had been targeting March 31 for when the VCS would be available for purchase at retail, through GameStop and Wal Mart.
Two: Indiegogo preorders topped out around 10,000, and presumably Atari must have been intending to supply additional thousands or tens of thousands of units to stock retailers. But now they say they will have “parts” for 500 units by the end of March, well off the numbers needed to fulfill even the pre-order.
Three:“a good portion of these first Atari VCS units are earmarked as dev kits for developers.”
So… to be clear on this. Atari’s business plan is to:
Sell consoles through pre-order.
Design the consoles.
Build the consoles.
Provide the consoles to crowdfunding backers and retail and developers all at the same time.
Now developers start working on software for use on the console.
I don’t need to spell out why this is the wrong order to do things in, do I?
Getting “dev kits” into the hands of developers early so that there can be launch titles available when the hardware reaches consumers is vital to the success of the console. Games take months and years to develop. They’ve been working on the AtariBox project for a good 3 years now, they should bloody well have games on it when they release it to consumers.
But given that this is just a commodity Linux box in a fancy looking shell that is “certified” to run Unity engine games, what exactly does anyone need with a “dev kit” anyway? The only thing I can think of would be that the VCS’s classic joystick is different enough from a standard gamepad (which they also have for the system) that there’s some need for a dev kit. That’s plausible, but it still doesn’t excuse Atari from not getting dev kits into the hands of developers partnering with them well in advance of the console’s expected release date, which I will point out again is already about two years later than they promised during the crowdfunding campaign.
Four: Why are they receiving parts? Why aren’t they assembling everything in China? What the fuck?
The blog post goes on from there to go into minute details of manufacturing defect tolerances, for some reason. I guess to show that they have a lot of problems building defect free cases for the thing? That’s reassuring!
I guess they want us to believe that they are working hard with manufacturing to get the details right, but that this hasn’t been easy, and this has been part of the reason for all the delays.
And from there, some footage of people playing emulated Atari arcade titles such as Asteroids, Crystal Castles, and Centipede. And a video of someone playing Fortnight, allegedly on an Atari VCS system. Which, hey, great, but that’s something anyone can do right now, on a computer they already have.
“Atari” then go on to mention that their planned schedule of events has been disrupted as events such as GDC, SXSW, and E3 have been canceled due to the COVID-19 pandemic. Which, given the history of “Atari” attending these events in the last 2-3 years, maybe just means that the there wasn’t any point in them renting a hotel suite across the street from the convention and inviting attendees to swing by while they’re in the neighborhood, riding the event’s coattails.
Will “Atari” be at Pax, or Comic Con? Who knows? Who cares?
The bottom line is: You ain’t seeing your Atari VCS pre-order at the end of March. Surprise, surprise. And if you ever do receive your system, it’ll be around the same time that developers receive their dev kits. So while you’re waiting breathlessly for the next 2-3 years for them to crank out games that were actually designed for this system, you’ll be able to enjoy a library of existing games, many of which are already available elsewhere, and have been for literally the entire history of video games.
Today my inbox had an email from Atari in it, announcing that they are now ready for developers to start making games for the system. Which is supposed to be released to retail in March. Lol.
OK, so this might be slightly less ridiculous than it seems. Let’s recall that the real Atari gave Howard Scott Warshaw a whole 5 weeks to crank out E.T.: The Extra-Terrestrial. Which he did. Because HSW is a beast.
But mainly, it’s not unreasonable because millions of game developers already have projects that they’re developing in Unity3D, and it’s probably trivial in most cases to create a build that will run on Linux, which is what AtariBox’s OS will be based on.
The announcement reads as follows:
Hello Backers, Fans, and Followers!
In light of our recent confirmation of Unity compatibility with the Atari VCS, it’s time to start unlocking access and giving game and app developers the green light to start developing new games and making plans for porting existing content to the platform. We have recently contacted interested developers with our initial plans and are happy to now share more with our broader audiences.
Getting content onto the Atari VCS will be straightforward and easy for both large professional studios and independents alike. Most developers already have all the tools they need to develop for our Linux Debian-based OS and can start right away.
Atari will also have a path to get your games and apps into the Atari VCS store and make real money!
A couple hundred BackerKit survey remain incomplete. If you have not completed one, go to https://atari-vcs.backerkit.com now and request your survey. It only takes a few minutes and will make your Atari VCS shipment(s) run smoother and faster.
Ongoing thanks again to every backer, fan, and developer. We can’t wait to see what you create!
— The Atari VCS Team
So, great. Potentially, tens of thousands of already-existing games built with Unity can be easily ported to the AtariBox, which if for some reason you wanted to play games on that platform instead of one of the half dozen or so platforms that you already own that has been capable of playing Unity games since forever, then hooray — you can!!
Let’s be clear, announcing “Unity compatible” is neither shocking, nor impressive. Your smart phone, your web browser, and your generic PC can all run Unity games. So can your Macintosh PC, and your iPad.
It’s rather difficult to imagine what hurdles Atari might have needed to leap over in order for their AMD x64 linux PC in a fancy case might have had to overcome in order to verify that it could run Unity games. Because, frankly, it is stupid easy to do. Which is one of the really nice things about Unity.
It’s yet another underwhelming announcement in a series of underwhelming announcements from Atari about the “progress” that they’ve managed to “achieve” with their New VCS project.
But let’s be clear: you or I could assemble a PC capable of running Unity games in about an hour with components that I ordered from NewEgg, probably for around the same cost as what Atari is selling the VCS for, and it wouldn’t take me 3 years to develop it.
Atari’s latest Medium blog update mentions that they are now working on establishing compatibility with other game development engines. But this should also be a trivial exercise as well, given that the Atari VCS is a generic linux PC under the hood. From what they’ve said, they’re only really developing a graphical shell environment (and even then, very likely all they’re doing is creating a “skin” for some existing graphical shell, with minimum customization, plus an app store and a launcher. But I expect that we’ll be hearing that Atari VCS will support Godot engine, GameMaker Studio 2, Unreal Engine, the Gnu C Compiler, and pretty much anything else that already has the means to build executable binaries for Linux.
But if you know what that means, it’s not even slightly impressive, because literally everything in the world can run linux, and can run software compiled for linux.
The one nice thing that I see in Atari’s announcement today is that they are taking only a 12% cut of your sales if you produce exclusive content for the Atari VCS, and just 20% for non-exclusive content sold through their app store. Which is a lot better than the 30% cut that is taken by Valve, Apple, and Google for selling apps through their stores. Of course, considering that Atari has only had about 10,000 units in pre-sale that we know of (based on their initial Indiegogo crowdfunding) it seems that limiting your game’s audience to AtariBox customers is going to cost you a lot more in sales revenue than that 8% could ever hope to make up for.
So… meh. It doesn’t look, then, like there will be a whole lot of exclusive titles for this system, which means that there won’t be any reason to buy this system, which means that there won’t be any reason to target this system for exclusive game titles, which means that this whole thing is unlikely to catch on. Pretty much as I’ve said all along.
If you’re a developer and interested in more information, they say to write to firstname.lastname@example.org. Which, maybe, hopefully, finally won’t bounce messages sent to it, as they have the last two times I tried to write to them over the past year.
Intellivision launched their Amico Club app for Android and iOS recently. I gave it a try, and got to experience… well, I’m not quite sure what I experienced. Let’s talk about it.
The app, at least so far, seems like a teaser advertisement for the Amico console. It gives a 15-second demo of a re-vamped Moon Patrol, which was a well-received game in arcades in 1982, and a game I liked to play back in the day. This version provides nicer graphics, some better jumping physics, and a power-up that gives the player more firepower.
The original was a game that is notable for being one of the first to employ a parallax scrolling background to create an illusion of depth, one of the first games to feature a full background music soundtrack, and was essentially an endless runner before there was such a genre named.
With such a brief glimpse at the game, I don’t want to jump to a premature conclusion, but so far I am not quite impressed yet.
Since this game is running on my smartphone, I have no idea what it’ll actually be like running on real Amico hardware. Presumably it should offer an even better experience. I would expect that the real Amico controller provide a far better experience than the provided touchscreen controls possibly could, but I can’t really guess how I will respond to an Amico controller until I’m holding one in my hands.
The actual gameplay was OK. I don’t think there’s really a whole lot to Moon Patrol. It’s jump, shoot, and not much else. You have to watch above your moon buggy for alien saucers that will try to bomb you, and ahead of you for craters to jump and boulders to shoot or jump over. You can slow down or accelerate, but there’s no stopping. It looks like there’s possibly some novel twists that they could add to the game to reinvigorate it, but from what I see so far, I’m not quite sold yet.
Hey everybody, remember Breakout?
Yes, Breakout, the mid-1970s game about bouncing a ball against a brick wall with a paddle because there’s no one around to play Pong with, remember? It’s back.
And, well, it brings some enhanced visuals, and some power-ups… nothing that wasn’t possible 20-25 years ago. If you remember Arkanoid, you’ll see a lot of the “enhancements” aren’t really new ideas, either. But it’s sideways. Because a 16×9 screen would be sadistic to bounce against in a vertical orientation.
And, I mean, this looks decent. I’m not sure that it’s going to light the world on fire… pretty sure it’s not, actually, but hey. If you like Breakout, now you have another version you could play, if you wanted to.
So far, what I see is that Intellivision is delivering, and that they are staying true to their word that they will be publishing simple games that are easy for casual gamers of all ages to pick up and play without having to invest a lot of time in getting to know characters, learn a background story, getting deep into the worldbuilding lore, or figure out controls on a 13-button gamepad. And I think that if they can deliver that, with published games in the $8 per title price point that they had announced a little over a year ago, this could be a fun system.
At least it has evidence of actually existing. It’s nice to see the INTV guys actually doing something and showing tangible progress on developing their product, in contrast to the AtariBox “effort” that has shown little beyond some conceptual artwork and pre-production prototypes of the hardware with nothing at all mentioned or shown when it comes to exclusive new games to be released for the platform.
This past Saturday, Dec 7, I had the joy of exhibiting my game, Ancient Technologies, at the Akron Art Museum’s Open World Arcade. It was an honor to have my game selected by the committee for inclusion in the event.
I made Ancient Technologies in 2016 for Ludum Dare 36, already quite a while ago. In preparation for the event, I wanted to make a few improvements and add features I’d wanted to include in the game, but hadn’t had time for due to the time constraints of the game jam weekend.
As a result, behind the scenes, I did battle with a few “ancient technologies” of my own.
Plenty of time to get ready
About a month prior to the event, I had to decide how to bring my game to the public. I didn’t want to use my daily driver laptop for this, as it is the machine I depend on for everything, and I just didn’t want to assume any additional risk with it being touched by anyone who I didn’t know. But I have an old AMD Phenom II PC that I built around 10 years ago, which had sufficient specs to play the game, so I decided I would bring that.
I don’t run this box a whole lot, so the very first thing I needed to do was get it updated. I had wiped it some time ago, and reinstalled Windows Professional 7.1 x64, Google Chrome, a few other minor apps, and then hadn’t used it for much since then. So when I powered it on, the first thing I wanted to do was run Windows Update. WU detected about 150 updates that needed to be downloaded and installed, which took quite some time to complete, and I expected that it would take an hour or so, but it ended up taking far longer than it should have. My PC ran updates overnight, and I woke up the next day to find that it had failed to apply updates.
I then had to spend a few days researching the failure, and trying various things to get the system to update. Eventually, I found a Microsoft tool that fixes Windows Update when it breaks in the way mine was broken, and was able to install updates. Well, almost all of them. There’s still one update that just will not install, no matter what I do. Oh well. Typical Microsoft garbage.
I didn’t need to have the PC connected to the internet for the event, so I wasn’t that worried about not having it 100% updated, but I always like to run systems that are updated and maintained.
Once the hardware was updated, I turned my attention to the game. There were a number of features that I had to drop from the original Ancient Technologies, so I took the opportunity to work on adding these.
The most important to me were the UFOs and the player’s special ability (hyperspace, shields, or flip) that in the Atari 2600 version of Asteroids that I was re-creating inside of Ancient Technologies, is activated by pressing Down on the joystick. Fortunately, getting these implemented wasn’t terribly difficult, and I found it enjoyable to add these things in and get to see the fully realized vision of what I had wanted this game to be. It took me only about four evenings to do it, and it mostly went smoothly.
And, actually, adding these features wasn’t what took most of my time, it was fixing numerous minor bugs that I kept finding in the game as I tested it, and I tested the game extensively. None of these bugs were game breaking, and probably wouldn’t have been noticed by players in a short session, but, being a perfectionist, if I know it’s there, if I can fix it, it needs to be fixed. Numerous minute details, like logical conditions that should prevented certain sound effects from being heard, such as the TV being turned off, or the game console being unplugged or the cartridge being removed from the slot, needed to be checked and tested and fixed.
I figured out a simpler way to do this: create an audio group for all the sound effects that play through the in-game TV, and then set the gain to 0 when the TV is off.
The last minute stuff
The day before Open World Arcade, I tested out my game one last time. This time, I did more extensive play testing with my gamepad. Mostly, I’d been testing the game with keyboard input, since it was more convenient. To my horror, I discovered that the D-pad on Xbox 360 gamepads is absolutely terrible. It wasn’t so noticeable before, but now that Down does something, I kept accidentally hyperspacing when I just wanted to turn. It was completely unacceptable. The engineers who fucked up the D-pad on the XBox 360 gamepad should be flayed and their families put to ruin. Seriously, you guys, how hard is it to include a bit of tech that’s been around since the mid-80s, and not make it total garbage?
Suddenly, on short notice I had to find a better controller, a wired gamepad that worked via Xinput, and that has a really good D-pad.
I researched and couldn’t find any reviews that I considered reliable enough for alternative wired XBox 360 controllers that have a D-pad that doesn’t suck. I did find a number of tutorials on how to do surgery on your controller to make it work better, but I wasn’t willing to take a risk on it so close to deadline. I’ll probably do it later, though.
I did end up finding a few controllers for Xbox One that had reviews for good D-pads. I wanted to pick up a Hori Fighting Commander, which looked like it would be ideal, as it lacks analog sticks, it therefore must have a good D-pad. But I couldn’t find a physical store to buy one at within 100 miles, and I couldn’t order one to be delivered and have any hope of getting it in time. This was super frustrating.
I ended up picking up a PowerA Enhanced Wired gamepad for Xbox One.
Then I found that because Microsoft is a horrible company, they couldn’t be bothered to ensure that Xbox One controller drivers were available and easy to install for Windows 7, and had to spend several hours trying to find a method to install drivers so that my PC would recognize the damn thing. I still don’t have it working on my laptop, but fortunately, for whatever reason, I found a driver that will install and work on the PC I intended to use for the Arcade. Why the exact same driver won’t install on my laptop, I have no friggin’ clue. Goddammit, Microsoft, why? Why do you have to be so abysmal when it comes to supporting your own shit?
One nice thing, though, the D-pad on the PowerA Enhanced Wired gamepad for Xbox One is better than the horrible D-pad on the otherwise pretty darn good Xbox 360 controller.
Another last minute idea that I had was that I thought it would be fun to put Ancient Technologies side-by-side with the original. I had the equipment, I only needed a small enough CRT TV set that I could easily transport it and set it up. I didn’t have one, but I put the word out and one of my friends had a small TV that I could borrow.
Game day started out smoothly, until I realized during setup that I had forgotten to pack my PC’s speakers. My desktop system has a nice set of 5.1 speakers which were really more than was called for, and I knew I had a set of regular stereo speakers in a closet somewhere, meant to grab them, got distracted, and forgot. This necessitated a last minute run to a store where I could pick up a set. Fortunately, I had plenty of time before the opening of the Arcade to do this, and it ended up not being a huge deal, but I had to drop another $25 on that, which, after the $25 I had to spend on a gamepad yesterday, left me annoyed at myself.
When I came back from the store, I happened to check my phone, and saw in one of the Facebook communities that someone had posted a photo of their old living room gaming setup:
Seeing this hit me emotionally, and it made me feel that the game I was about to exhibit belonged and had relevance and cultural resonance. So may people in my generation grew up in a house with a big wood cabinet color TV set, and hooked up their game systems to it.
I also discovered, to my dismay, that the TV set that I borrowed wouldn’t work. I didn’t have time to test it prior to the event, but when I looked at it more closely, I discovered that the power button was gone. There was a hole where it used to be. A remote control was taped to the TV set, but when I tried it, it wouldn’t work, and then I noticed that the batteries inside had gone bad and corroded the contacts, so there would be no way to use the TV. Regretfully, I put the TV and the Atari back in my car, and they would not be a part of the show.
Turning my attention to the PC, I booted it, launched, and tested that everything was working as it should. Of course, it wasn’t. Why should it be? I had only tested extensively and fixed every visible problem that I could see for a week. But somehow or another, the “put down” sound effect that I had coded wasn’t working. When you unplug the cartridge, or the power cord, it’s supposed to play the sound, but it doesn’t for some reason. I had never noticed this previously to the day of the event. My first thought was, “OK, somehow or other, I accidentally added the “put down” sound effect into the group of sounds that play through the in-game TV set, which is off, and that must be why the sound isn’t playing.” Nope! When I turn the TV on, the sound still doesn’t play. And when I went back to review my code, the “put down” sound effect isn’t in the TV sounds audio group. What’s more, if you unplug the joystick it also plays the same sound, and that’s the one place where it’s still working properly. Looking at the code in that object vs. the others, it’s the same frickin code: when you click the mouse on the object, play the sound. It works there, but not in two other places that are identical. No clue why. Maybe it’s because the joystick stays in the same place, while the cartridge and power cable move, putting them away from the mouse cursor, which somehow retroactively fails the audio_play_sound() function, even though it’s in the same god damn code block as the code that moved the object? WTF, GameMaker? What the Fing F?
Ancient Technologies is coded in GMS 1.4, which is no longer supported. I suppose the next thing to do would be to import the project into GMS 2, compile it, and see if it behaves the same way. I had been enjoying picking up this old project and working on it again, but with weird shit like this breaking for no explainable reason, it really puts me off wanting to do anything serious with GameMaker.
This problem wasn’t a showstopper. Most people who played the game probably wouldn’t have noticed it at all, not knowing that there was supposed to be a sound played when there was none, or in a lot of cases not even doing the action that would have triggered the sound effect. But after looking into it and finding no fault in my code whatsoever, I’m more frustrated with GameMaker Studio than ever.
The Open World Arcade
The event itself was great. I sat in my chair all day, and people came up and gave Ancient Technologies a try. I tried to give each person the experience from the start of the game, rather than leave the Atari console hooked up and playing Asteroids already. Players were about 50%-50% on having owned an Atari or played one before. I had people of all ages try it out, from about 4 on the low end to a gentleman who looked to be in his 60s.
Most of them were uncomfortable with the controls at first, having forgotten the function of the Down direction, or never having known it. I observed that nearly all of them immediately went to the dual analog sticks on the gamepad, a conditioned reflex that explaining the controls to them would not undo. About half of the players opted to use the keyboard controls rather than the gamepad, and it occurred to me that I could just as well have not bothered providing the gamepad at all, and no one would have really missed it.
Almost everyone started out playing the Hyperspace variation, and everyone’s immediate reaction to starting the game was to rapidly touch every control on the gamepad to figure out what did what. In virtually every case, the first thing they did was blink out of existence, into hyperspace, and then re-appear, confused and often about to collide with an Asteroid, which would promptly kill them. I realized pretty quickly that Hyperspace was probably the worst ability to start players off with, and that Shields or Flip would have been a much better choice. I needed to explain to most players that there were several variations, and how to access them using the Game Select switch on the console by clicking it with the mouse.
Many players seemed to hesitate after losing a life, reaching for the mouse for some reason. I’m not sure what they were thinking in the moment, but perhaps they weren’t aware of how many lives they had remaining, and thought that they would need the mouse to restart the game. But most players didn’t play more than once, even though I told everyone they could play as many times as they liked. It occurred to me that using the gamepad’s start button would have been convenient, but this wasn’t really a design decision given that I was trying to replicate the experience of hooking up and using an Atari 2600, and the Game Reset switch is on the console, not on the controller. A few players left their game in-progress, but I didn’t let it bother me.
I had left a stack of flyers with information about the game, but if anyone took one, I’d be surprised. I don’t think anyone did, although a handful at least looked at it. I also left a notepad for players to leave feedback or a comment, and only one person did so. I’m not sure what to take away from that, to be honest. I didn’t push anyone to leave me feedback or take a flyer or one of the business cards that I had, unless they engaged me in conversation and seemed like they would like one, then I gently offered that they could take one if they wanted.
Most of the “feedback” I obtained was through silently observing players and noting common patterns in how they engaged and interacted with the game. This gave me a bunch of ideas of things that I wish I would have thought of to put in the game, but couldn’t have thought of without watching someone else experiencing the game. Any time I noticed anything that could have been added or changed to make the experience better, I took note of it. Much of this was help/tutorial text on screen. Although, I wanted to avoid that, since the first part of the game is figuring out how to hook up the Atari. But I think something, like a ? icon in the corner that appears after several seconds of nothing happening, could have possibly helped players get into the game. As it was, I found that I had to explain to most of the players how to set up the Atari, and I think that defeated much of the purpose of letting them figure out the “Mystery of the Ancients” for themselves. Another feature I would have liked to add was an idle timer that reset the game after a period of inactivity, so that it would always start the next player off with a fresh experience even if I forgot to reset the game after the last person.
In all, it was a great experience for me to show my little game to random people and see how they interacted with it. Thanks to the Akron Art Museum and their wonderful staff for organizing this event.
This Saturday, December 7, from 11:00 am to 5:00 pm, Akron Art Museum’s Open World exhibit will host the Open World Arcade. Tickets are $10 for musem members, $24/nonmembers, Free for children 12 and under*
*Children must be accompanied by an adult.
A special edition of my game, Ancient Technologies, will be presented, along with games by other developers.
Cathedral is a nostalgia-driven homage to the action-adventure platform games of the NES. If you enjoyed Capcom platformers like DuckTales, or Ghosts N Goblins, or Wizards and Warriors (by Acclaim), this is one to check out.
It looks and sounds like a NES game, but with a little extra — bigger sprites, fewer palette limitations. And the game play is pretty polished, and works with either keyboard or modern gamepad, although I definitely recommend the gamepad option. The map system, menus, and character dialogs and story scenes are all a few notches above a typical NES game, while still retaining a distinctly NES inspired flavor.
I would rate the difficulty as medium-hard, most of the challenges do not require extreme skill to overcome, but you start out with very low health, and stuff that hurts you tends to do a lot of damage, so there’s not much room for error. But it’s certainly not Ghosts N Goblins hard. The one boss fight I’ve encountered so far wasn’t too bad, and I enjoyed beating it.
I’ve been struggling through it, unable to figure out how to get past various roadblocks, to get anywhere to do anything. This is a bit frustrating, and I wish that there was an issue of Nintendo Power magazine laying on my desk that could tell me what to do.
It feels like I’m missing an item or ability that will enable me to clear these obstacles, and I have no idea where I need to go to get them. There’s a lot to explore, and a lot of missions. I’ve barely scratched the surface of it so far.
If I could fault the game for anything, it’s for not being clear enough in what to do. Visually, there are subtle indicators of what is a solid wall and what isn’t, what’s a destructible object and what isn’t, or what is a platform you can stand on and what isn’t. The visual language isn’t always as clear as I would like it to be, which leads me to feel uncertain at times. The game does have some nice in-game hinting/tutorial moments, which helps, but I feel aspects are still a bit too ambiguous.
So far, I’ve managed to exit the Cathedral, where you start out, and explore the town that you encounter next, and then everywhere else I go is an apparent dead end. There’s a graveyard, which I managed to explore a bit, a scary woods, which I can’t make any headway with, a sewer, same deal, and a library with a key that I can’t figure out how to reach. I can’t figure out how to go back to the cathedral, or if I’m supposed to.
There’s some other annoying things about the menu system, like, half the time I can not figure out how to get to the menu area that lets me exit the game. I’ve discovered it on accident 2 or 3 times, and I still can’t figure out how to get there. I finally figured it out. It’s accessed through the inventory screen, as one of the icons that you can select from your inventory. This is a pretty strange way to hide the exit menu.
So far, I like this game pretty well, and am enjoying its puzzles and missions, although I’m not sure how to proceed, and I wish some of the options available to me were a bit more obvious.
I had one more idea for an interesting map. This time I wanted to emphasize the importance of the Bridge to the map. So I thought, I would split the map into two halves, and put the bridge between them, as the only way to get from one side of the map to the other.
This was my prototype:
I thought that this map had interesting potential, but I also had some concerns. I wanted to make sure that the traffic flow would still work, and that by splitting the map in this way, I didn’t make it likely that random movement would tend to collect everybody in one area of the map, and I wanted the random distribution of characters to not be unevenly distributed between east and west ends of town. I also wanted to ensure that the subway system would be evenly distributed, both in terms of entrances and exits, and that the subway provided useful shortcuts.
As I walked through this map, I quickly decided that a less obvioulsy symmetric map would be more interesting. I re-arranged screens and quickly came up with this:
The connections between the screens are a bit different from my previous versions. Moving horizontally, the map wraps, shifting up a row if you’re at the west edge, and down if you’re at the east edge. Vertically, the columns wrap around without shifting. The Bridge screen is different, when moving vertically it wraps around back to itself. This serves to keep the Bridge screen isolated, so bridge pieces will be somewhat protected from the helicopter when placed here.
Finally, the extreme corners of the map, the northeast and southwest corner screens, are connected to each other horizontally, creating a second junction between the two halves of the map. This helps provide a route for Clark Kent to walk to the Daily Planet at the end of the game, without being forced to use the Subway system, although this overworld walking route is very long.
The subway exits are again unchanged, and the subways provide several routes for traversing from one side of the map to the other. I arranged the subway system so that each colored subway screen has two exits on the opposite side of the map, and one exit on the same side of the map.
Thus, despite the broken bridge in the center, this map has very nice traffic flow between both halves of the map, many interesting shortcuts, and a challenging layout to learn, without the confusing one-way vertical borders on the Phonebooth and Bridge screens that vexed many beginner players of the original. After playing this map a few times, I think it’s every bit as good as the original, and might even be more fun to play. And aesthetically, I love that the Bridge is now the centerpiece of the map, and truly joins the two halves of Metropolis together.
Here’s the map again, with the wrap routes indicated:
I don’t think there’s much more I can do with the map after this. So I think this is where I will leave the evolution of the map variations.
I would still like to introduce randomized bridge piece starting screens, but to figure that out will require more understanding of the source code than I currently have.
I also think it would be neat to make a super-rom that includes all of the map variations in one file, switchable via the Game Select switch. Again, this is beyond my current capabilities with my very limited understanding of the source code and 6502 asm.
You can download the entire collection of romhacks here:
After publishing part 2 of this series, I played my grid map romhack a few times, and I think it’s rather good. I don’t know that I would say that I prefer it over the original map, but this is certainly a viable alternative map.
After a few playthroughs, I notice a few things:
The helicopter moves bridge pieces away from the Bridge screen much more frequently. In the original, once you’ve placed a piece at the Bridge screen, the helicopter would only rarely remove it.
The helicopter seems to move pieces from their starting screen much sooner, as well.
I like both of these differences; they make the game more challenging, and I think, more fun. You really feel like you need to race the helicopter to complete the bridge task. But I think it’s worth experimenting with the map to see if there’s a way to restore the original design intent, by reducing the number of paths into the Bridge screen. It occurs to me that an easy way to do this is to make the two screens above and below the Bridge screen connect to each other, rather than to the Bridge screen. So the Bridge will exit vertically to those screens, but reversing course will skip over the Bridge, but still place you near the Bridge; players would need to figure out that they are “around the corner” from the Bridge in this map, which is potentially disorienting, but should be less of a problem than the original map, which puts you clear across into the middle of the other end of the city.
I do think that this map has a few peculiarities that make it sub-optimal.
Some of the subway entrances are right next to their exits. There’s little point in a subway trip that results in you re-emerging on the overworld map one screen over from where you started. For example, the Red subway entrance is one screen away from two of its exits, due to being on the top left corner of the grid, and the wrapping effect. The Blue and Yellow subway entrances are about optimally located relative to their exit screens, and the Green subway entrance is directly above one of its exit screens.
It’s possible to remedy this by moving the subway entrances around, or by changing their exits to more useful screens for this map layout. I was reluctant to try this, because I wanted to keep the changes that I am making to the maps minimal, but it seems that these changes are necessary for the good of the game.
Another thing that strikes me is that I didn’t need to remove one screen from the overworld in order to have even rows. 21 = 3 x 7, so I could have made an overworld map of 3 rows of 7 screens. This might make the subways more useful, since it would make it easier to place the subway exits such that they are 3-4 screens from their entrances.
A 7×3 grid will allow all 21 overworld screens to be included in the new map. Again, I went to an image editor and re-arranged the screens to make a layout, then visually reviewed and analyzed the map for playabiltiy concerns, and made tweaks.
My first iteration of this was to lay the screens out in the original map’s horizontal sequence, and see how that would play. That map looks like this:
Without actually playing it, it’s easy to miss things, but I can see that this map has a few potential issues:
The bridge pieces starting screens are all on the west end of the map, and are nearby the Bridge screen.
The Phonebooth, Bridge, and Jail are all on the bottom row, which makes the bottom row feel too important, and the top row feel unimportant.
The “critical path” relationships in the original map are broken in this map. This may not be a problem per se, but it will make the new map feel less familiar to experienced players.
Otherwise, I think the map looks pretty decent. I’m not sure how I feel about the potential issues I mention above.
I do think that the Bridge should be the center piece of the map, so I shifted rows and columns in order to make it so, but the rooms are still in the original order, using the horizontal progression from the original map.
It’s interesting how shifting things around can change perspectives on things. With this view, I’m able to notice different things:
Now all the “important” screens (Jail, Bridge, Phonebooth, Daily Planet, and all but one of the bridge pieces) are on the west end of town.
But balancing that, all the subway entrances are on the East end of town.
Subway exits are somewhat evenly distributed, but there’s a few interesting things to point out: The Red subway exits all line up in a column at the far west edge of the map, while the Yellow subway exits are all distributed across the top row. The Green subway exits follow a diagonal from the northwest corner of the map, moving southeast. And the Blue subway exits are scattered about as far from each other as they could be.
Shifting the screens changes the loopback point at the top right, and as a result the screen that normally is to the left of the Phonebooth screen no longer is. Which, despite what I thought when I first decided to shift the screens to put the bridge in the center of the map, does fundamentally change the map. Then I realize… I did it wrong!
I can’t just grab the top row and move it to the bottom, and then slice the right couple of columns and move it to the left. That will not preserve the horizontal order of the screens. To do that, I need to grab the top-right screen, move it to the bottom left, and then let all the other screens shift right by one, snaking around. This approach gives this result:
Overall, it’s actually very similar to the first two iterations. It’s literally the same horizontal sequence as the first iteration, but it’s not all that different from the second iteration, either; the “important” screens are still mostly to the west, the subway entrances are still mostly to the east. I think the subway exits are a bit better distributed than in iteration 2.
The vertical path is quite different. I think this map will run slower, due to the placement of the Jail in the corner of the map, and its distance from most of the subway entrances, and the Phonebooth/Bridge screens. There might yet be some quick paths that will become apparent through repeated play, but for now I feel that the map feels disorienting because the Jail isn’t north of the Yellow subway entrance screen any longer, and I struggle to locate it as a result.
Certainly there are many other possible arrangements for the overworld map, and I haven’t even begun to re-design the subway. But I think this is enough to satisfy my curiosity, at least for the time being. I may come back and revisit this further if I feel a need to after playing the maps I already created more extensively.
There’s much more to do in a hack of Superman, as well. I would like to figure out how to randomize the starting places of the bridge pieces, for example. I would also be interested in figuring out how to put up a hard barrier around the outer edge of the map, rather than having the map wrap around at the edges. But these projects, if I ever attempt them, will have to wait until I have a deeper understanding of 6502 assembly.
For the second alternative map, I needed to carefully re-organize the screens to put them into a grid.
There are 21 overworld screens, and so I will use a 5×4 grid, omitting one of the overworld screens, since there can be only 20 screens in a 5×4 grid.
Selecting the right screen to omit is important. The removed screen must not be:
A subway entrance
A subway exit
The phonebooth, bridge, or jail screen
A bridge piece start screen
One of my goals with this redesign is to keep the layout of the screens reasonably familiar to players of the original game. This is a formidable task, and technically impossible since I am changing the layout, but my goal is to make the changes minimal, and to preserve as much of the critical paths through the map as possible.
But just as important, I want the redesigned map to be balanced and fun to play.
To achieve this, I want the subway entrances and exits to be distributed about the map, and not clustered together too much.
There are 11 unique subway exits, and 5 subway entrances (counting the Daily Planet), so 16 of these screens have something to do with the subway system, and another three are the Jail, the Daily Planet, the Phonebooth, and the Bridge screen. (Both the Daily Planet and the Jail are also subway exit screens, so are already counted among the 16 overworld screens that are related to the subway system, but adding the Phonebooth and Bridge brings the total of “must keep” screens to 18, leaving just 3 redundant screens for possible removal.) I didn’t realize how tight the map design is until I noticed this.
After considering the matter for a bit, I decided that this screen would be the one to go:
But I just as well I could have picked this one:
Or this one:
I note that the three “dispensable” screens all have a similar color scheme to them, green and white, and wonder if there’s something being communicated through a visual language here, or if it’s just a coincidence… there are two other green/light grey screens, the Yellow Subway entrance, and the screen above the Jail, which is one of the bridge piece starting rooms. So… probably just a coincidence, then.
At the edges of the overworld grid, the map will wrap around to the next row or column. As an experiment, I think I’ll also create a variation where the map will wrap around to the same row or column, but I think this variation will only be playable on Easy difficulty — because, when Superman is struck by kryptonite, he will be stuck in the current row of the overworld map, and if Lois Lane isn’t present on that row, he’ll be unable to get revived until she wanders in, or the helicopter brings her, which could take ages, and would be unfair.
I take a picture of the overworld, cut out each room, and put it into an image editor and move them around until I have them arranged in a 4×5 grid that I think will be satisfactory. I end up needing to move things around a good bit to achieve this, but I think I managed to do it while still preserving some of the critical relationships between certain rooms.
Here’s what I ended up with:
Here’s the original map for comparison:
So you can see, there’s a bit of re-shuffling that I had to do in order to balance the map out, and many of the vertical and horizontal connections are different in the new map.
The major changes are:
The one-way vertical boundaries for the Phone Booth and Bridge screens are two-way.
One screen is removed from the overworld, to allow for an even grid.
The Phonebooth and Bridge screens are in the center of the map. (Although, since the edges of the map still wrap, center/edge is all relative.)
The important non-changes are:
The subway system exits are all to the same screens as before. (But, due to the shuffling of these screens relative to one another on the overworld, this does still mean there’s some significant differences. The key exits to get to the Daily Planet and Jail screens are the same as in the original, though. And several of the subway exits now get you close to the Bridge screen.)
The bridge pieces are still found on the same screens as they were in the original. I didn’t try to re-locate them. (Sometime, I’d like to introduce randomization into their starting positions, and see how that plays.)
Now that I had the design determined, I had to hack the rom to make the changes for real. To do this, I went through the original code and noted the names of each screen, and figured out from the boundary change code the correct addresses for each room, and mapped them together. This was straightforward, if a bit tedious, but fortunately the map is small.
One thing I noticed as I coded the new map connections was that there’s a peculiarity at the top-right and bottom-left corners of the map, due to the way the rows and columns wrap; if you’re in the top-right screen, going up or right will both take you to the bottom-left screen; and if you’re in the bottom-left screen, going down or left will take you to the top-right screen. This doesn’t make sense, unless you understand the way the overworld wraps. If you get to the end of a row or column, continuing in that direction wraps you around to the next row or column, and at the very end of all the rows and columns, it wraps you around to the opposite corner, and so the top-right and bottom-left screens are connected to each other in two directions.
I have only played through this map once, and mostly just to test that everything was connected properly. I think it’s a viable, and interesting variation, but I wouldn’t call it better than the original. Certainly, due to the re-positioning of the screens, the quick path to the bridge pieces isn’t as short as in the original map, which means speed running potential for this map is reduced. Beyond that, the Phonebooth and Bridge screens are not low-traffic screens any more, meaning that you’ll often find gangsters on these screens, and the helicopter may find bridge pieces that you’ve left at the Bridge site more readily, which will again make it more challenging as the helicopter removes them again.
You can download the entire collection of romhacks here:
So for the first alternative map, I will just make the Up screen the same as the Right screen, and the Down screen the same as the Left screen. This means that there will only be the Horizontal progression through the map, and that Up/Right will be “forward” and Down/Left will be “backward” through this progression.
I playtested this, and found that it works exactly as intended. The map is far simpler to navigate, and getting lost is now virtually impossible. Fly up/right to go forward, down/left to go backward, and all overworld screens appear in order. The confusing one-way vertical borders on the Phonebooth and Bridge screens are eliminated.
The downside of this is that traversing the overworld map is slower, since there are no shortcuts to be gained by using cornering techniques. You still can rapidly advance two screens in the horizontal map sequence by going diagonal from near the corner, transitioning on both the vertical and horizontal edge transitions. But compared to the original, where a vertical transition would typically advance you between three and five screens in the horizontal sequence, it’s much slower, and offers no real opportunities for useful shortcuts.
The only real shortcuts possible are via the subway system, which I left unmodified. Despite playing this game over nearly four decades, I’ve never bothered to memorize the subway in its entirety — there’s little reason to. The only exits of real importance are the ones that go to the Jail and Daily Planet. But with this map, knowing the rest of the exits is potentially more valuable, because it can get you to the other side of the overworld in less time than any other method. It may be interesting to study the exits and try to figure out alternate layouts that might make the map more interesting by providing additional useful shortcuts.
I can reliably win Superman typically in 1:30, +/- 15 seconds, but this variant took me 3:04 to win on my first play. I’m sure faster times are possible. The gangsters seemed to be bunched up more, and rounding them up preoccupied me at first, leading to the helicopter moving bridge pieces away, which I think happened earlier than normal. But because the map is essentially a two-dimensional line, finding the helicopter was trivial, making any challenge increase by the bridge pieces moving around quite minimal.
My impression is that this alternative map might have made the game more playable for very young players, ages 3-5, and as an “easy” version for beginning players, much like Superman‘s sibling Adventure offered players Game 1 with a simplified map. But overall, the simpler map makes the game less interesting and less challenging to play. But if you’re interested in the game’s design, playing this variant to see why it’s less interesting is… well, an interesting exercise. I invite you to download it and play it for yourself and make up your own mind. If you do, please drop a comment and let me know what you think.
My next Superman romhack will present the overworld map as a grid.