Clyde “Tomato” Mandelin, translator and localizer of video games, including especially of the fan translation of Mother 3, has a website and blog called Legends of Localization. It had been sleeping for a couple of years, but recently has sprung to life again with two posts this year. A few days ago he linked to a 2017 live playthrough and translation in a series of Youtube videos he made of a Sega Saturn JRPG called Tengai Makyō: The Apocalypse IV, a satirical game that pokes fun at Western culture, that he refers to as kind of a cousin of Earthbound. At 31 videos, each about two hours long, it takes quite a while to get through the whole thing, but it sounds like fun. Here is the 62-hour epic:
Back in April he curated a collection of articles about bad game translations, including games like Twinkle Star Sprites and Breath of Fire II, but also has a collection of iffy translations of English games into Japanese. It turns out that Atari Games was notorious for them, inspiring a couple of long-lasting Japanlish memes.
At the top of the screen: “koin ikko ireru,” an awkward way of saying “Insert Coin.” (image from the Legends of Localization site)
I hope these posts are an indication of further writing from Tomato in 2025!
Recently I’ve been working on a getting-started guide on what I think is one of the most interesting games in UFO 50, Pilot Quest. (Other games I’ve really enjoyed, though I’ve by no means tried every game in the collection yet: Magic Garden, Waldorf’s Journey, Planet Zoldath, Attactics, Kick Club, Onion Delivery, Porgy, Valbrace, Grimstone and Mini & Max.)
Guides take time, so in the meantime here’s an hour-long talk by Ed Logg on the creation of Gauntlet, from GDC 2012!
Arcade Mermaid is our classic arcade weirdness and obscurity column! Frequently (no promises) we aim to bring you an interesting and odd arcade game to wonder at.
Yes! More Rampart! I’m still out of town at DragonCon so I’m repurposing an old strategy guide I wrote for Extended Play into a week of posts. Even I’m starting to get sick of Rampart by now, but please stick around. And if you’ve ever tried playing this infuriating game in the arcade, this might give you the advice you need to get through to levels you might not have thought possible! And this is even a fairly short post!
Level Strategies
Level One (“Recruit”)
Castles available: 6
It’s advised not to start on this level unless it’s your first game. You’ll not only miss out of 5,000 points by starting here, but it means the beginning of Level 2 will begin you with only two cannons instead of three, and you’ll face more ships.
This level is so easy that it barely rates mention. It has six castles, and only Gunships appear. Unless you’re purposely stalling, you’ll finish this map in two rounds. Even if you stall, the level auto-completes in three rounds, regardless of if you fire a single shot.
Another reason to avoid picking Beginner is that its selection is always the “J” map, which is one of the easier selections from the initial four boards. It’s better if you pick it later, when the opposition is harder, to balance out the difficulty.
Level Two (“Veteran”)
Castles available: 5
The Veteran level is randomly chosen from one of the other three basic maps, although it tends to be the “C” map most often. Ideally it won’t be the “Hat” map, it being the easiest, as you’ll want it to be your Level 4.
If you started on the board, the first round will always put you up against four ships, one of which may be a Lander. Later ships are frequently also Landers. As a result grunts may appear on this board, but even if some appear, so long as you focus on destroying enemy ships and don’t suffer major setbacks like losing your home castle, you’ll probably wipe out the attackers before they become a huge problem.
On a later round in this level you’ll probably encounter your first Red Ship. A Red Ship tends to get off two or three shots in a round, each producing flaming crater you must build around.
Level Three
Castles available: 5
You always get to pick this level yourself. I usually try to make it “C,” “J” or “Hat,” in that order. I usually avoid picking “N,” but it’s not really harder than “C,” so pick whatever is your preference.
Much of the fleet will be Landers here, so this is the first level where you must be diligent about thinning out their numbers. Especially try to take care of ships about to land; grunts are unwelcome visitors to deal with, and it’s best if they never show up at all. In the event of a landin, which is sometimes unavoidable, try to spare a few seconds to build a barrier to them, to wall them off from the rest of the board. If you can contain them in a small area around the landing, you can keep their numbers down that way and thus reduce the number of shots you must expend in cleaning them up. Don’t think you can just let them be! They will overwhelm you easily if you ignore them.
I try to pick a home castle that’s near another castle, so I can capture them both with one wall. If you can consistently get both each round, that’s three cannons each time. Use them to connect the space between the two castles, but try to keep that space as small as possible. Try to aim for the minimum-sized rectangle needed to save both of them, and then try to place all your cannons within that rectangle.
Level Four
Castles available: 4
Now the game starts to get serious. If you can get here on your first credit you’ve done very well, but your weak basic guns will become more and more of a liability for you. Dark ships start appearing on this level, and will continue to appear until the end of the game. They all take one extra hit to sink. That’s three hits for Gunships, four for Landers, and six for Red Ships.
Try to make Landers that have a chance of landing at a diagonal shore your top priority. If one makes it through, then at the start of the Building phase devote three seconds or so it blocking their progress with walls. If they get away from the shore and a significant number appear, then focus on surviving and shoot them with cannons during Battle. Red Ships start to be a significant threat here. Once all the potential landings are taken care of, try to pump six shots into a Red Ship.
Level Five (peninsula)
Castles available: 3
Regardless of how you got here, you have a choice between two new maps for this level, a diagonal peninsula going upper-right to lower-left, or one going from upper-left to lower-right. I call these the “slash” and “backslash” maps, respectively. I usually pick backslash, but both are about the same difficulty.
These maps have the property of having two seas, a small one at the top of the screen and a more typical larger one at the bottom. The small water region is both a blessing and a curse; the restricted movement makes it easier to destroy ships there, but landings are very likely unless you clear out the Landers arriving there. What is more, these maps are the only ones in the game that break the rule about landings only being possible on diagonal shores. Landers in the small sea are also counted as landed if they contact land at the very edge of the screen, even if there’s no diagonal shore nearby. This property doesn’t extend to screen-edge landings in the larger sea.
The worst thing about these levels is that the game takes off the limiter on the number of Red Ships that can be generated each round. Red Ships here are nearly as common as Double-Sailed ones! You’ll frequently see seven or more on-screen at a time. While you must remain vigilant to prevent and/or mitigate landings, you must also work towards reducing the number of Red Ships. The large number of flaming craters on these levels made adopting a “summer home” strategy imperative, where you switch off between castles when one gets too difficult, or even impossible, to save. It’s usually a good idea to try to capture two castles in the first round because of this; then you can station some or all of your cannons for that round at the other castle, so you’re not defenseless when you’re forced to turn to it for survival. From there, you can focus on getting cannons around the third castle for maximum safety.
This level is a strong test of your Rampart skills, and often is a barrier to progression until you’ve played many games. But it’s not invincible. On Easy difficulty, I’ve passed this level still on my first credit multiple times. The biggest thing you have to worry about is the high probability of impossible situations.
Consider: when any castle is adjacent to other objects, you must capture the whole collection as a set. Chains of objects like this increase the difficulty of getting the whole thing, and increases the odds of getting an impossible save. I’ve seen this happen after even the first round on a peninsula board, and I’ve seen it happen on all three castles on one of these maps at the same time. The only substantive thing you can do, besides rotating between castles, is to shoot some of the dangerous walls yourself. Red Cannonballs don’t leave craters if the wall is already destroyed, but that takes time and often makes for a difficult repair.
So, what if you manage to clear this map? What happens next? The game scrolls the FINAL BATTLE banner, and you get a one-way ticket to:
Level Six (island)
Castles available: 2
It all comes down to this. This level is a special kind of hell, only slightly mitigated by the fact that it seems you can get away with sinking fewer ships here, and, because you’re surrounded on all sides, your shots often don’t have that far to travel to reach a ship. There are four prime landing areas, each in one corner of the map, so landings can only be delayed, not prevented. And you only get two castles. Enjoy!
When you start looking more closely at the map, some slight advantages become apparent. There’s a lake in the bottom-left of the board, which gets in the way, somewhat, of grunts approaching from the bottom. Both castles have a good amount of land around them, although there are so many Red Ships here that the “summer home” strategy is essential to survival.
The continue limit eases up a little bit here. If you were at the maximum of four continues upon starting Level Six, the game will grant you an extra continue when you get here. (You still have to pay your coins for it though.) And also, if you were at that final credit, the game will actually downgrade your cannons one step when you get here, so get used to expending two/three/five shots for Gunships/Landers/Red Ships again. If (ha ha, “if”) you do expend that last continue, you go back to only needing one/two/four shots to sink ships. (There is more information on the effects of continues on cannon power in the notes, later on.)
This level is an ordeal, but it can be finished surprisingly quickly. I’ve seen it cleared in three rounds, although that was after several continues (it had been a pretty great game up to that point). This level is the big obstacle to one-credit clears, I’ve managed to finish everything up to it on one credit only to bang my head against it repeatedly. Using save states, I’ve managed to finish it with a single continue, so it is possible to finish it relatively cheaply, but making do with such weak cannons is a real challenge.
Arcade Mermaid is our classic arcade weirdness and obscurity column! Frequently (no promises) we aim to bring you an interesting and odd arcade game to wonder at.
Because I’m out of town for a few days, we’re continuing our week-long coverage of Atari Games’ brilliant, yet really hard classic arcade strategy game Rampart.
Each of Rampart’s phases has its own considerations, so let’s take them one at a time.
Battle Phase
Rampart’s most traditional mode is its Battle sequence, which plays a bit like Atari’s old hit Missile Command. For a limited amount of time, players use the trackball to move a crosshairs around the screen, and the Fire button launches a cannonball from one of their cannons aimed at that spot. Cannonballs have a set speed that they travel as they move through the air; nearby shots go directly towards the destination, but distant targets travel in a high arc. Thus, more distant shots take much more time to reach their targets, due to the fact that they have to be angled upward.
The speed of your shots matters for two important reasons. First, in single-player mode, enemy ships don’t always sit still and let your projectiles sink them, but move around at different speeds, and sometimes even try to dodge your shots. One of the most annoying events in Rampart is when you launch a volley towards a ship, and it decides to pull anchor and float elsewhere just as your shots launch, wasting those precious missiles. Firing at close-up ships means you don’t have to lead your targets by nearly as much, and reduces the chance that a ship will just sail out from under your projectiles.
The other important thing is, each cannon can only have one shot in the air at a time. Firing at nearby targets means you can get more shots off during a Battle phase. The combination of the two, being more accurate when shooting moving targets and getting more shots off, means that, absent other issues, you should prioritize firing at ships near your guns.
You cannot entirely erase the chance that a ship will decide to start, or stop, moving after you launch shots towards it, but there is some finesse that can be applied. Ships always sail in a straight line in one of the eight cardinal/diagonal directions, and ships cannot move through other ships. If a moving ship runs into another one, it will stop. They cannot turn around while in motion, but must wait until their invisible captain picks a new direction, which could be immediately or never. They also stop if they reach the shoreline, which for a Lander could be very bad for you. Once in a while a moving ship will just stop for no reason. The only constant is, all ships stop moving at the very end of the Battle phase, once all cannonballs have landed.
The rules of cannonfire, the ones concerning shot speed and there only being one shot in play at a time per weapon, they also apply to the enemy ships. Nearby ships are able to launch more shots at closer walls than distant ones, another reason to prioritize shooting nearby ships. But this must be balanced against those times when you need to prevent a landing, or take out some Red Ships to make the Build phase more survivable.
The game continues no matter how many ships you sink in Battle, but to progress to the next level, and eventually win the game, it’s advantageous to sink ships. The exact qualification for finishing a level is unknown, but it seems to happen after you wipe out most of the ships on-screen.
There is a maximum of 16 ships that can be in play at any time. If that number is reached, but there are some weakened or Single-Sailed Ships on-screen, Rampart is known to remove some of them so stronger ships may enter the fray.
Building Phase
This is the meat of the game. While shooting ships eventually clears levels, you can’t do that without cannons, and cannons are awarded from skill at building. Also, if you fail to capture at least one castle in a round, you lose.
The way it works is, the game hands you an assortment of building pieces of various shapes, one at a time. The pieces you’re given first exist as just a flashing outline that you can maneuver freely over the land with the trackball or joystick. You can move it anywhere while you’re getting it into position, or rotate it clockwise with a press of the “Rotate” button. (If you start out your first building phase without using the button, the game’s voice will suddenly intone, “Use ROTATE for a better fit!” Rampart doesn’t have nearly as much digitized speech as Gauntlet, but it sure is memorable when it does speak.)
The difficult bit is, you cannot place pieces anywhere there is something per-existing. A piece cannot be placed if any part of it overlaps with another wall, the shore, a castle, a cannon, a crater, a grunt, or the edge of the screen. Everything visible beneath the piece’s outline must be the clean, green checkerboard of countryside.
You must place the current piece before you get the next one. Intermediate Rampart players ruefully observe that the game tends to hand out just the wrong piece they need at a given time. A lot of this is observation bias, you don’t tend to notice the times when the game hands you just the right piece, but it’s true that the pieces the game has to hand get larger and more complex as the game continues.
When a piece is plopped down it turns into wall. If the wall adjoins wall pieces already extant, they’ll join together pleasingly into a solid-looking structure. Your task is to connect wall pieces together like this until you have a complete loop all the way around and enclosing at least one castle. Usually you want this castle will be the one you start with, called your “Home Castle.” It’s worth extra points and cannons if you capture it, but any castle will allow you to survive.
There are two particular hassles in the Building phase that deserve special mention. The first are those grunts the Landers sometimes drop off if they reach the shore, and that have the power to move around during Building. They look like tiny tanks, or little battering rams.
Grunts move during Building, meaning they can actually block pieces just as you’re trying to place them. They not only move, but slowly multiply as they go, tending to form lines between the shore and your castles. Worst of all, grunts that make it to your castles have the power to destroy them during the following Battle phase, the only thing in the game that can harm a castle itself.
Grunts are so dangerous that landings must quickly be dealt with or else your game will soon be over. You can shoot them during the Battle phase, or you can surround the territory they’re on, destroying them instantly with a satisfying crunching noise.
Keep Landers away from diagonal shorelines, or you’ll have a bad time!
One important thing to note about grunts is that they can’t be dropped off just anywhere. Landers (here circled) can only leave grunts on diagonal shorelines. Highlighted above are the landing sites on the “Hat” map. Guard them at all costs.
The other problem you must worry about during Building is craters, flaming pits in the very ground left when the red cannonballs launched by Red Ships strike your wall. A crater is an implacable obstacle; there is no way to remove a crater from the board other than time. Near the end of the game, depending on the randomized ship generation, you may end up with rounds where you must face six or more Red Ships at once, which is horrifying.
Caused by too many craters left by Red Ships, this situation cannot be survived. Can you see why?
When a crater is formed, it’s a full-strength pit of fire. Every Battle phase, each crater diminishes a bit. They all do this in lockstep: all the craters formed in one round will advance to the next state at once.
How does the game decide how quickly to make craters disappear? It’s a function of how recently you’ve put in money and continued the game! If you’ve recently continued, then for a while the game makes craters disappear in two rounds instead of three.
Cannon Phase
The Cannon phase is much less tense than the others. You get ten seconds to place additional cannons for subsequent Battles. But placement of cannons is one of the most important decisions of all. In a sense, the reason you’re capturing castles, beyond mere survival, is to support your artillery.
For the first Cannon phase of the whole game, the game gives you three cannons to get you started. On the first Cannon phase after continuing, you get four, as part of the “and more firepower” the game promises. On the first round of other levels, you start with just two.
Beyond that, on each successive Cannon phase, you get two more cannons to place for having captured your home castle, and one more for each other castle. But this only happens if you have room to place them. When surrounding castles in the Building phase, you must not only survive, but try to leave extra room for placing cannons, that is, if you want more artillery.
The more cannons you have, the more shots you can get off in the Battle phase, and thus the more of a dent you can put into the enemy navy and/or other players’ walls. This, in turn, decreases the shots coming in towards your walls, lets you do something about Landers before they drop off Grunts, lets you shoot at Grunts directly, allows you to sink more Red Ships before they set your whole countryside ablaze, and ultimately depletes the opposing navy faster.
Lots of cannons mean lots of shots! Shots can stop landings, clear out Red Ships, and hasten the end of the level.
But cannons are also a big problem. Each cannon you place is a 2×2 square of ground that cannot be otherwise built upon, which restricts the shapes of pieces that can fit on the ground. Having lots of cannons progresses you through the level, but increases the area you have to surround to survive. Practically, you must build at least some cannons directly adjacent to your castle, and other cannons adjacent to those, in a big sulfuric blob. Thus, to capture that castle, you’ll also have to capture all the cannons that adjoin it, a mass of guns that grows in size with each round. And in case you come up with the idea to build cannons away from castles to avoid this, you should know that cannons that aren’t “owned” during a battle phase sit idle, doing nothing.
A “summer home” is a secondary castle you turn to in order to survive, usually with cannons attached, when your home castle is too difficult to recover.
You must have cannons to survive through each level, to punch back against the enemy, to take care of threats and eventually to advance towards finishing it, but having them makes later Building phases harder. Once a cannon is placed, it cannot be removed except by losing and continuing the game. And not having any cannons is a counter to using the “summer home” strategy for every round: while you can often abandon your home castle and its armaments in favor of a small unadorned dwelling out in the sticks, unless you’ve built cannons there on previous rounds, you’ll have, at most, one gun to thwart landings. A lot of games have ended due to this.
One subtle aspect to Rampart strategy is the cannon firing order. Cannons fire in the order in which they were placed. Any uncaptured cannons, and also cannons with shots in the air, are ignored in the queue. The difference in lead time between a cannon near the shore, and one way back at the screen edge, is significant, and can result in a lot of missed shots. To thwart this, it helps to try to build cannons in one mass, instead of spread out, and to try to build in more centralized locations, but truthfully it’s difficult to consciously make use of this strategy during a game.
This is a danger spot, a location where, if a single Red Cannonball hits, it renders the whole castle and its cannons impossible to capture.
One thing that players cannot afford to ignore is the danger of building cannons too close to the water or the edge of the screen. This is a huge concern. Building cannons close to hard barriers like those greatly reduces the number of pieces that will fit between the gun and the obstacle, and increases the chance that a red cannonball will hit in a such a place as to create an impossible rescue. To the game’s credit, Red Ships do not appear to target these locations intentionally, but neither do they avoid them, so It’s important to try to leave at least two spaces between cannons and any hard barriers, and more if you can. Especially look out for one-space diagonal separation, because it isn’t as obviously dangerous as horizontal or vertical. The more free space around your installations, the easier it’ll be to build walls there later, and the better off you’ll be.
So what if you decide you have enough cannons and don’t want to place some or all this round, do you have to sit and wait out the clock? Why, no: if you make five consecutive illegal cannon placements in a row, the game will take that as a signal that you’re done placing them and end the Cannon phase immediately.
That’s three parts so far! Yes, there’s more to say about this diabolical game, but we’ll take a break for a couple of days, for other posts. See you soon!
Arcade Mermaid is our classic arcade weirdness and obscurity column! Frequently (no promises) we aim to bring you an interesting and odd arcade game to wonder at.
We continue our lengthy, obsessive coverage of Atari Games’ brilliant, but really difficult, arcade strategy game Rampart.
Overview
Building Phase
Rampart is really two separate but related games, the single-player game and multiplayer. Both are excellent. Even so, this article mostly concerns itself with single-player, but towards the end I will have a few things to say about its multiplayer mode.
Before we begin, you should know that Rampart is a highly abstract game, and as such it might not be easy to see how it all fits together from a text description. I have uploadedseveralcompleteplaythroughs of the game to YouTube, the first such on that site of the arcade version. You might watch a bit of one of them, it should make it pretty obvious how the game works.
The Board
In both single-player and multiplayer, after selecting a home castle and placing initial cannons, the game repeats a sequence of rounds, each consisting of a Battle phase, Building phase and Cannon phase. In the Battle phase, players shoot at ships or the other players’ walls, while those ships or enemy cannons fire back at their own walls. In the Building phase, players place a variety of shaped pieces in an attempt to fix their wall and/or construct walls around at least one of the castles on their land within a time limit. Then in the Cannon phase, players get a number of new cannons to place within the territory they captured in the Building phase, adding to the number of shots they can get off in Battle.
Select Home Castle, then Cannons -> Battle -> Building -> repeat
This cycle continues until the player wins or loses. Losing happens when a player fails to capture a castle in time in Building. Winning happens, in single-player, when the game decides the ships don’t have a reasonable chance of defeating the player, and advances to the next level, out of six in total. In multiplayer, it happens when all the other players have lost and don’t continue, or for a set number of battles. If the fight goes the full distance, the winner is the surviving player with the highest score.
A defeated player can put more money in the machine to continue the game “with more firepower.” This is allowed four times in single-player, so up to five credits in all, and losing after that results in a solid Game Over. The dynamics of continuing in Rampart are really more complex than this, and I’ll explain those later, but for the time being, you should know “with more firepower” signifies several important changes that are not all clear during play. In multiplayer, losing players can choose to continue individually twice. That can make for a decently long game if all the players choose to buy all their continues. But eventually, the game will declare a round to be the “Final Battle,” after which no continues will be offered to losing players, and the game ends regardless. The number of rounds a game is allowed to go is operator-adjustable, and continued games usually add a few battles to the overall length. There also exists one semi-secret game variation. If just two players play Rampart, and at least one of them is joining in again after a completed game, there is a map selection screen that offers the chance to play a composite game mode, featuring both two players and ships! In this, the ships belong to one or the other side, and are color-coded based on who they belong to. They function like extra cannons, generally firing where the player aims their cursor, but are more autonomous. This mode is an interesting variation, and is fairly obscure. Could there be other such secrets hidden in Rampart’s code?
Basics
For now, we’re most concerned with figuring out and defeating the very difficult single-player game. The basic play of both games is similar, but there is a lot of difference between attacking ships or walls, and the structure of single-player is quite different, so it’s worth treating each mode separately. When you put a coin in and press Start, this appears:
This ornate lobby screen is accompanied by booming drums and realistic, although synthesized, trumpets and fifes. With the volume turned up (as all Rampart machines should rightfully have) the effect is startling and bound to attract some attention. The purpose of this screen is to give other players the opportunity to join the game, for unlike the Gauntlet-style, “join any time” play model Atari pioneered and most arcade games used, once underway a game of Rampart cannot be entered by others. Players can only leave a game, by failing and not continuing.
An interesting thing about Rampart is that, after a game concludes, it always returns to the lobby screen and adds the player’s score to a total across all their games in the session. This encourages players to play repeated games, to see their overall score climb higher and higher. I’ve seen the total go up to over 900,000 points, which takes around 18 full-length games. I know of no reason to go for a high total score other than vanity, but if someone else has information on this I’d certainly like to know!
Your Domain
Assuming a single-player game, the screen clears and shows an overview screen of an island. There are four possible maps that can be selected for play here. At the start of the game, two areas along the shore are boxed, one marked “RECRUIT,” the other “VETERAN +5000 points.” Most of Atari Games’ arcade releases had such a selection, for it was a major part of their house style to offer a basic and an advanced difficulty, with a score bonus for starting harder. With Rampart, this choice is mostly an illusion. All choosing Recruit does is start you off with a very easy first board (the Recruit map is always the “J” map), which usually takes two rounds to finish.
Each level pits the player against a larger and more dangerous enemy navy.
At first, only weak Single-Sailed Ships attack, which go down with two hits from basic cannonry. Here, I call these Gunships.
With level two comes Double-Sailed Ships, a.k.a. Landers, which take three hits, and if they manage to get close to a diagonal shoreline, they drop off a swarm of evil little Grunts.
Grunts are a huge danger! They are the only enemy that has the power to move around during the Building phase. I’ll have more to say about them later, but for now just know that a grunt beachhead, left unopposed, has more credit-ending power than anything else in the game.
Starting with level three, Red Ships join the battle. Although few in number at first, these take five shots to sink, and their red cannonballs leave fiery craters where they land. Craters block piece placement and persist for multiple rounds. Usually a Red Ship can get off two or three shots each Battle, and they add up.
Levels four through six feature the same kinds of ships, but they’re darker in color, which in game terms means they take one additional hit before sinking. Dark Gunships take three hits, Dark Landers require four hits, and Dark Red Ships only go down after six cannonballs have struck them.
Every time you complete a level, you’re returned to the island screen to pick a new map to play. After the first level, three remaining unplayed options are offered. If you chose a Recruit game you’ll eventually have to face all these boards, but if you picked Veteran and started on level two, you end up skipping one of them, an opportunity to avoid a disliked map.
The “C” Map
The “Hat” Map
The “J” Map
The “N” Map
The maps are not even in difficulty. Generally, easier maps have castles with land around them on all sides, and more straight horizontal or vertical coastline, giving Landers fewer places to land. My opinion is that the “Hat” map is the easiest, followed by the “J” map, then the “C” and “Lowercase N” maps. A good strategy is to try to get one of the harder two maps out of the way as first selection, so you can later play the J and Hat maps to offset the harshness of the third and fourth levels, but sometimes the Veteran selection level ends up being the Hat, and you’ll have to play them in a different order.
The “Slash” Map
The “Backslash” Map
Level five is selected from one of two unique maps, both peninsulas with water on both sides. Both are difficult, although I think the “Backslash” board is a little easier. As you progress, each level starts you out with fewer castles to capture. By this point you’re only getting three castles, and you’ll probably have to use all of them to survive.
The final, “Island” Map
The last level is an island to itself, with a lake on it and only two castles! Here, ships attack from both sides of the screen. This level is a trial; although you have to sink fewer ships here to win, it’s difficult to prevent landings and grunts from overwhelming you, and the craters from Red Ships can easily give you an inescapable situation unless you mitigate them.
If you finish the last level, you get special ending music, and an illustration that few arcade goers have seen, before being returned to the High Score and Lobby screens. It may seem anti-climatic, but final victory and safe shores are their own reward.
Arcade Mermaid is our classic arcade weirdness and obscurity column! Frequently (no promises) we aim to bring you an interesting and odd arcade game to wonder at.
The next week is going to be pretty busy for me, so I’m reusing a long long piece I originally wrote for the short-lived fanzine I made, Extended Play. It’s been seen before, but only by a very small audience.
Rampart has been an obsession to me since it came out around 1990. I still think it’s one of the finest arcade games ever made, actually one of the best video games period. People tend to describe it as either an early version of Tower Defense, or Missile Command meets Tetris, but neither description is very good. Simply, there had never been a game like Rampart before, and despite some attempts to clone it, there’s never been a game like it since, either.
We’re all as obsessed with weird old arcade games as I am, right? Right?? Most (but not all) of the next week of posts will all be about this wonderful, but extremely difficult, game, one of the best from Atari Games at their heights. Please bear with me, and please, try to understand why I’ve been so into this unique game for so long.
In the late 80s Atari Games was recovering from 1983’s Great Game Crash, which decimated the US arcade industry and resulted in a great dying out of companies. Atari Games, newly split away from the consumer electronics company that would make the Atari ST, weathered it better than most, primarily through developing ingenious kinds of games the likes of which were seen nowhere else.
This period, 1984-1990, is the age that began with Marble Madness and the System 1 hardware, followed by hits like Paperboy, 720 Degrees, Gauntlet, Toobin’, Cyberball and others. It lasted roughly to the release of Street Fighter II, which sparked the fighting game craze and made arcades inhospitable to most other genres for a while. At the tail end of this period came Rampart, designed by John Salwitz and Dave Ralston, who also created three of the games in the foregoing list.
It’s a bit obscure now, but it was quite a hit for Atari at the time. For home systems Rampart received no less than thirteen distinct ports, plus a few notable emulations. It got very good reviews from Nintendo Power (SNES version) and Videogames & Computer Entertainment (arcade). More recently, it was released emulated with network play for free for the PlayStation 3. And yet, still, no one talks about Rampart anymore.
Battle Phase
That is a weird thing about what I call Atari Games’ “Silver Age,” that time after the classic arcade boom of the late 70s/early 80s but before the rise of one-on-one fighting games, its games, while still fun and surprisingly ingenious, and widely ported, don’t have the star power, other than a couple of things like Gauntlet, that Japanese releases from the same time have garnered.
Part of the reason may be trademark-related. The name Atari still carries considerable nostalgic weight, but another company, the one formerly known as Infogrammes, bought the trademark and the classic Fuji logo, and still releases games under it. They own the rights to everything before Marble Madness. Games after that were owned by Midway for a while, but Midway then, in a crushing indignity, renamed them Midway Games West, before closing them outright in 2005, getting out of arcades altogether. Many of its assets were sold to Warner Bros. Entertainment, who presumably now owns the rights to Atari’s post-Marble Madness catalog.
I’ve written before that, at their peak, I think Atari Games was more ingenious and original than Nintendo themselves. They made some stinkers, sure (few remember Thunderjaws, or their arcade take on Tim Burton’s Batman movie) but their best games were, and still are, amazing. I think that Rampart, while little talked about now, is at the top of the gilded heap.
Rampart is great, but it’s difficult. For a game I’ve been obsessed with since its release in arcades in 1989, I can only complete it on default settings about a third of the time, and usually only after expending all the continues the game allows. This was common for Atari Games games, who tended to make arcade machines that mocked the idea of one credit runs, but even among those, Rampart is hard. You are not going to master this one overnight.
I view my purpose here usually to be a secondary source. That is, I prefer to summarize, encapsulate, explain and preserve information available elsewhere. But the thing about Rampart is, there is very little of this information to find. For a game that was popular at the time of its release and with so many ports, googling it will not turn up much to aid you. As far as I know, and I’ve searched for it many times throughout the years, Rampart has never had a good, comprehensive strategy guide written for it.
It is because of this, favored reader, that I have decided that this shall be that guide. Since most of it is self-discovered, I have no one to blame for errors but myself. I have made some efforts to make sure it’s correct, but, well, sometimes even well-considered assumptions turn out to be false. I’ve done what I can.
Following on from the Defender tip video I linked, here’s a high-level tip video for a “radically” different game, Atari Games’ wonderful yet challenging arcade skateboard adventure 720°.
It’s really different, not just from space shooter games, but from just about everything else, even from other skateboarding games. In some ways it’s much like an early, 2D version of a Tony Hawk game, with an open world to explore between events called Skate City. But it also has a bit more going on than that: you have to earn points in Skate City doing tricks in order to earn Tickets, which allow you access to the four parks at the edges of the big isometric area, and you can earn money by doing well in the events to earn gear upgrades for your skater. Yes, there’s an equipment system in this 1986 arcade game!
Exploring Skate City isn’t a laid-back experience, however. It’s timed, and when that expires a now-iconic voice proclaims SKATE OR DIE, announcing the arrival of the killer skateboarder-hating bees, which get angrier, faster, and more Warner Bros. cartoon peril-like as further time elapses.
The only way to “die” in 720° is to be caught by the bees, all other defeats and injuries are harmless, but the bees still end lots of games: on default settings, you need 10,000 points to earn each Ticket, which is quite a lot! You are spotted a Ticket when you continue, but it’s not a gift or purchase, but a loan: the game will increase the points needed for the following Ticket by 10,000 when it happens, so you’ll have to score even more points to make it to the following Ticket. But lest you think this is a naked ploy by Atari to force players to credit-feed to see the later parks, you only get two of those continues! It’s best to think of continues as failsafes, in case you have a Ticket but get stung before you make it to a park.
720° is particularly interesting because of its unique joystick. It’s a standard 8-way stick, but its mechanism forces it to point in one direction at all times, so instead of pushing where you want to go, you spin it. The video contains a lot of that spinning. 720° is a very physical game because of it. In most games, arcade or otherwise, the controls could be considered just a way to communicate your intentions to the game, and missed inputs where you had intended to act feel like a betrayal by the hardware, but games like 720°, and Namco’s Alpine Racer, and Atari Game’s Marble Madness, the controls feel like an intrinsic part of the fun.
In the first class of these games, a brain interface might possibly be welcome for getting the controllers out of the way and removing any question of what your intent was, but 720° would be an entirely different experience that way. Physical execution is essential to the experience. It’s also a different game when played in emulation, because of it.
720° was designed by one of Atari Games’ most successful teams, John Salwitz and Dave Ralston, who also designed Paperboy, Cyberball, and my favorite of all of Atari’s output, probably my favorite arcade game of all: Rampart. Sadly, it looks like Rampart, while successful (at least judging by how many ports it got), was the title that marked their departure from Atari Games.
So relax for half an hour, or however much of it you can stand, and watch a 720° master demonstrate how to win over this uniquely challenging arcade game, on a physical cabinet no less. It’s a world where you earn 500 points from knocking over a bodybuilder, and isn’t that a place we’d all like to live in?
Escape From the Plant of the Robot Monsters (I’m just going to call it Escape Etc. from here) is a game I’ve always been curious about.
It’s weird to think now about the time frame of Atari arcade games. 1972 saw Pong; 1979 was Asteroids, signalling a new direction for Atari in arcades; 1984 was Marble Madness, their first post-crash hit; then, 1991 was Street Fighter II and the start of the fighting game craze, forcing Atari to change direction yet again. They would have some hits from there (like Primal Rage and Area 51), but nothing with real cultural staying power until the era of Gauntlet Legends and San Francisco Rush.
Escape Etc. I don’t think did badly, but it wasn’t a huge hit. You can kind of get an idea of the popularity of one of Atari’s arcade games by how many ports it got. APB, for example, Dave Theurer’s last game at Atari, only got Lynx and European home computer ports, while Rampart (John Salwitz and Dave Ralston’s last game, if we’re noting such things) got a ton of ports to lots of platforms. Escape Etc. didn’t even get a Lynx port, although one had been planned.
This isn’t an Arcade Mermaid post, just another link to a Youtube video review. It’s done in an old style, without a lot of flash, but there’s good things about that too, and the information is both interesting and thorough.
More information on Escape Etc. can be found in this post from Vintage Arcade Gal. It’s text!
Posted to Metafilter by Going to Maine, it’s definitely the kind of thing of interest to me and probably our readers. A link is a link, no matter where it came from. This site is an archive of messages made by Jed Margolin from the Atari Games corporate email server, from July 11, 1983 to when he left the company on September 15, 1992.
Note: we saw the Nintendo Direct with the trailer for the (newer) Super Mario Bros. movie, but there’s not really to comment on? Maybe we’ll talk about it later, when we’ve learned more.
It’s been known for a while that three classic-era Atari arcade games were produced by HAL for the Famicom and NES: Defender II a.k.a. Stargate, Millipede, and Joust. Nicole Express recently did her typically excellent job discussing them and what makes them interesting. Such as, among other things: they use music that was later reused in NES Punch-Out!!, they are subtly different between their Famicom and NES versions, Satoru Iwata may have personally worked on them, and they were developed as part of Nintendo’s pitch to Atari to have them distribute the NES in the United States.
Something that one might overlook reading the article, however, is the news that this isn’t the only version of Millipede made for the Famicom hardware. As an in-house project at Atari Games/Tengen, Ed Logg himself implemented Millipede for it, before they discovered that the publishing rights in the US resided with the other Atari, the one that got bought by Jack Tramiel.
The source was recently found on a backup tape from Atari, and hand-assembled. The resulting rom file can be downloaded from the AtariAge thread on it. It has no sound, because that hasn’t been implemented yet, the colorful alternate palettes of the arcade version are missing, there are some bugs (I mean program bugs, not enemies), and the DDT Bomb objects hadn’t been put in yet, but a lot of the game is still there. There are even multiplayer modes, supporting up to four players playing alternating or simultaneously! It’s especially interesting since Ed Logg assisted Centipede’s creator Dona Bailey in creating the original game.
Arcade Mermaid is our classic arcade weirdness and obscurity column! Once a month we aim to bring you an interesting and odd arcade game to wonder at. Although this time, we’re expanding the purview to talk about an extremely rare game that has just become playable by the public through emulation for the very first time. Please note, this was written quickly and late at night. It may see minor corrections once I see it by the harsh light of day.
One of the best podcasts out there for classic arcade enthusiasts is The Ted Dabney Experience! Episode 15 of that was a talk with Bob Flanagan, who created Atari’s underrated Skull & Crossbones, and also headed the wonderful, prototype, obscure and unreleased, yet recently revealed on the internet Marble Madness II.
Bob Flanagan was mainly a programmer during his time at Atari Games, where he helped implement the original Marble Madness, Paperboy, and Gauntlet. Skull & Crossbones is the only Flanagan-helmed game that got produced, but MM2 could have been another.
Another reason to cheer for this game finally, finally seeing the light of day is the music, which kyuubethe3rdmentions was among the finest work of Atari composer Brad Fuller, who sadly left us in 2016. Marble Madness had very memorable music, and the 14 main levels of MM2 sound like they could easily have come from that game, including a remix of the original’s Beginner Maze track.
Astral
I haven’t seen anyone talking about the source of the roms. It’s been known that the MAME developers have had copies for safe keeping, but were forbidden by the person who let them dump them from releasing them to the public. A MAME driver, I hear, has been around for a while, but maybe only privately. I don’t know what that has to do with the efforts of David ‘mamehaze” Haywood, who has worked to get the game working over the past few days.
I’ve watched a lot of playthroughs of MM2 over the past couple of days, it has been quite a focus to speedrunners. Here’s a couple, that make the game look really easy: FlannelKat, and LeKukie. DumpleChan has a slower run that looks a lot more like a good player would have done in an arcade if the game had been made:
As mentioned before, no one seems to know how this game, long a holy grail for preservations and arcade enthusiasts alike, got released. Was it leaked? Did someone who happened to have the roms just decide one day to throw them online? There’s a thread at AtariAge that notes that the owner would release the roms in exchange for $42,000. Did someone raise that much money? They are now on the Internet Archive, and work on an official MAME driver is well underway, so in any event, stuffing this genie back into its lamp now is probably impossible.
How It’s The Same
The sound design is nearly identical to Marble Madness, using many of the same noises. The attention to detail on recreating the experience of the first game is admirable! And as mentioned, the music is terrific.
All of the original game’s enemies and most of its obstacles are seen somewhere in this game’s 14 levels. The only ones that come to mind that are absent are the Intermediate Race’s rolling wave and the infuriating cycling platforms at the end of the Ultimate Race.
North Pole
It’s still a race against time, with leftover seconds carrying over to later levels. It’s still a co-op/competition game, where players can interfere with each other, seek to scroll them off screen for a time penalty, or coordinate their movements to double-team the Evil Black Marbles* and help keep players in the game.
A few of the levels have names from the original game. Their layouts are different, though.
How It’s Different: Structure
Marble Madness’s greatest weakness was always its ultra-short length. It only had six levels! It could be excused around the time of its creation, since in 1984 those kinds of scrolling map games were still a new thing. Well here there are 14 mazes, and three bonus rounds too!
Instead of a straight sequence of levels from start to finish, they’re arranged into tiers, of roughly equal difficulty:
Each tier can be completed in a variable order, with the player(s) deciding which maze to tackle first. Like in the original game, each maze in a tier carries time over from the previous one, with a small bonus. When the last maze is finished, players get 1,000 points for every second they had left, but also their time is zeroed out.
Oasis
After each tier, the players enter into a pinball-themed bonus stage! The acceleration is tuned way up for these, making them feel properly chaotic. The players use their marbles to hit targets to spell either MARBLE or MADNESS, and hit drop targets to earn as many points as they can. Players get five extra seconds for every 5,000 points they score, on top of a large time award granted for starting the next tier.
Throughout all of this, if a player runs out of time in solo play, they can opt to continue to try the maze again with a large amount of start time. In group play, they can continue with the same time as one of the other players.
This pattern lasts until the final level, King Of The Mountain. It’s like a tier to itself with only that one level. Only two attempts are granted here: The First Try, and then, after a continue, the Last Try.
A lot of the game is fairly easy, but King of the Mountain is long and tough! Unlike most of the other mazes, the player starts at the bottom and must ascend to the top. At the very end the players have to climb a series of icy slopes that’s very difficult to make it through! It’s a suitable successor to the end of the first game’s Ultimate Maze, with its disappearing pathways.
Control
This one’s big. Marble Madness, in the arcades, a trackball game. Marble Madness II uses standard eight-way joysticks. It makes for a huge difference, it makes the game easier, and also negates the point of the game a little.
In an earlier stage of development (as “Marble Man,” see below) the game went out on test in a trackball version, but it happened around the time the company began to move away from trackballs as a control method. Atari had been associated with trackballs for a long time, going back to the classic Atari Football, and they had recently released the trackball-based strategy/puzzle game Rampart. It’s possible, had Marble Madness II made it to production, that it would have been offered as an upgrade kit for Rampart. There is talk in the arcade modding community of some people wanting to turn their Rampart machines into Marble Madness II, despite the hardware being quite different. (I think this is a shame, as a long-time outspoken fan of Rampart, but it’s understandable for this game!)
With digital joystick control, long narrow passages lose much of their danger, turning a frantic perilous roll into the holding of a direction and the tap of a Turbo button. But also, joysticks can be less precise for this kind of game. I’ve seen forum posts speculating about what would be necessary to reimplement trackball control in the game, even though it’s not a simple change to the code. There’s the advantage that the programming of the original, trackball-based Marble Madness is out there. Time will inform us of the feasibility of this.
Powerups
Marble Madness II was developed seven years after Marble Madness, and in that time powerups went from intriguing new idea to de rigueur. Throughout most of the mazes (not the first or last) there are crown-like structures with a cycling powerup atop them. It’s possible to bash through these and collect the powerup that’s currently on display. Two of these, Cloak and Crusher, allow the player to either evade or destroy monsters. Knobby gives a marble super-sharp control, and Heli grants flight for a while, allowing for huge shortcuts.
Sandbox
As speedrun playthroughs have demonstrated, these powerups allow for gigantic time saves! Heli can bypass whole sections of the maze, and Knobby allows a marble to tear around corners with pinpoint precision. In a multiplayer game only one player can get a powerup from each location.
Points Earn Extra Time
This really is a major change. In Marble Madness, other than the time awards for starting mazes and the occasional random award of 10 seconds, there was no respite from the steady advancement of the clock. You do get five extra seconds for beating another player to the finish line, but that’s only awarded in two-player mode, and the difficulties of playing with another player overwhelm that meager allowance.
In Marble Madness II, each player is awarded five bonus seconds for every 5,000 points they collect. The plethora of bonus flags scattered around mean that players will be earning bonus time frequently, the time awards often greater than the loss from going out of your way to collect them.
Flags & Hidden Flags
Aerial
About those flags. They’re in every level but the first and last ones. They grant from 1,000 to 5,000 points, with the number printed on the flag. These are all over the place, and make it easy to amass a load of surplus time. Fortunately for the game’s design, this extra time is converted into more points at the end of a tier, and the players must begin stockpiling all over again Also throughout most of the mazes are bonus flags, worth 2,000 points each. These rarely flash into visibility for a couple of seconds, but often will appear as a reward for reaching out-of-the-way regions of the map.
The game promises players a “Super Bonus” for collecting all the flags in an area, but this requires a lot of patience to track them all down, and sometimes there are mutually-exclusive branches in the path of a maze that make it impossible for a single player to reliably get every flag if playing on their own. The Heli powerup, particularly, is great for collecting flags.
That last level King Of The Mountain doesn’t have any flags! Killing the enemies on that level is worth good points, but it means that very little bonus time is awarded there.
Tons of new obstacles
Every level has its own unique peril! Sand, crushers, falling icicles, killer satellites, meteor crashes, fists smashing out of walls, floating frying pans (in Wacky, of course) and many more. At the very end, huge rock-throwing trolls guard the passages up to the top of the Mountain.
Faces and voices
Oasis
It’s known that in an earlier stage of its development, Marble Madness II was subtitled “Marble Man,” and the winner of a race would transform into a super-heroic humanoid body for a moment at the end. A digitized voice would exclaim “Marble Man!” at the start of a game and the end of a race. There is footage of this version of the game, taken from off of one of the surviving machines, on YouTube, demonstrating the voice.
The released version of Marble Madness II doesn’t have the Marble Man theme, but it does have voices, and they’re pretty nice I think! When you start a game, your marble develops a face and shouts “Marble madness!” in a way that never fails to make me smile. When you hit a letter or spell a word in a bonus level it’s announced aloud, and to me that’s almost reward enough in itself. It’s probably for the best that the superhero theme was dropped, but the faces, both on the marbles in their death animations and on the enemies, add some needed character to what might otherwise have been a pretty dry game.
So, that’s what we have to report on Marble Madness II, a game developed 21 years ago and is just now seeing the light of day outside of an extremely small number of collectors. 21 years is a long time. A not-insignificant number of fans of the original Marble Madness have died in the intervening time, never having had a chance to enjoy it. And even now, because it’s only playable in emulators, a lot of people who could otherwise enjoy it, but cannot overcome the technical hurdle of getting MAME up and running and getting its roms into the right shape (a formidable obstacle to many), will still be denied playing it.
Marble Madness II is not yet up in official MAME, but its driver is well along, and I think it’ll probably appear in the next release. I urge you to at least give it a try, and think of all the people who would have enjoyed if it had appeared back at the time of its creation. It was a long time coming, but at least now it is here.
* The “Evil Black Marble” was adapted as a character on turn-of-the-century website The Conversatron. So says me, apparently the sole surviving person who remembers The Conversatron. Memory hurts.
This is not some fan game made to play like Marble Madness, but the real deal, a legendary lost prototype from the Silver Age of Atari Games! Cancelled because of the great arcade fervor at the time around fighting games, meaning little Atari released in that era performed well on test.
Word is that the rom has been released somewhere on the internet, although I do not know where. It had been known that all the surviving Marble Madness II cabinets were owned by old Atari staff or collectors who were averse to allowing the rom images to be released. Whether one of them had a change of heart or, as has been speculated with Akka Arrh, another legendary prototype, they may have been obtained through nefarious means.
Technically the rights are still owned by Warner Media, I believe. I’ve long been amazed that the current rights holders haven’t seeked out the owners of these prototypes and offered them a big payday to dump the roms and release something like a Midway Arcade Treasures 4. Sure, it’d only be a matter of time before someone broke the roms out of such a package, but they’d still sell a ton of units and the prototype owners would be properly rewarded for both maintaining their machines and for lost collector value, and importantly, the games would be out there among people who would enjoy them and be protected against further loss and obscurity.
