Too Much Free Time

Discussion and reviews of games for NES, Intellivision, DOS, and others.

Miner 2049er

Posted by Tracy Poff on February 14, 2016

In the eighties, if any arcade game achieved success, there were sure to be dozens of home computer games ‘inspired by’ it released in short order1. Beginning in 1980, the California-based Big Five Software released nine games for the TRS-80, mostly very straightforward clones of arcade games (including three games based on Space Invaders). These achieved some success, but nothing compared to founder Bill Hogue’s first game for Atari home computers, Miner 2049er, released in 19822.

Miner 2049er (Atari 8-bit) - cover

The game’s manual lays out the gripping story: you play the role of Bounty Bob, “the most loyal, heroic, and charismatic mounty … ever known”, who has been sent to capture Yukon Yohan, a fur trapper wanted for murder. Bob chased Yohan all the way to a uranium mine once owned by Nuclear Ned, and an explosion has trapped both Bob and Yohan within. The radiation in the mine has turned the creatures within into deadly mutants. Helpfully, however: “Scattered throughout the mine are various articles that have been lost by previous miners. Capture them by touching them and you will be awarded points. Additionally, the mutants will turn into green happy creatures that are now edible.”

Miner 2049er (Atari 8-bit) - 01

The goal of the game is to walk over every section of the platforms. Doing so fills the section in, and once every section is filled in, you’re awarded points depending on how much time remains on the timer, and the next station begins.

Miner 2049er (Atari 8-bit) - 02

There are ten stations in total, and a number of difficulty settings which increase the speed at which the enemies move.

Miner 2049er (Atari 8-bit) - 03

Later stations feature a variety of extra obstacles: beginning with station 2, there are slides which you slide down when you walk over them; station 3 has transporters you may operate to move between sections; station 5 has moving platforms; station 8 has a lift you can use to reach high places; station 9 features that staple of platformers, the ‘pulverizers’ which descend to crush you; finally, stage 10 has a cannon which you must load up with TNT to blast yourself to higher platforms.

Miner was ported to more or less every game platform and released in many countries. Reviews praise its originality, though it clearly shows the influence of both Donkey Kong, with its platforms and ladders, and Pac-Man, with its powerups that allow the player to eat otherwise-fatal enemies. In its turn, Miner influenced later games, such as City Connection, which lifts the painting-every-square mechanic directly from Miner. Originality aside, it’s obvious why Miner was such a success; it’s fun, challenging, and has good, smooth controls.

A sequel to the game, Bounty Bob Strikes Back!, was released in 1984. A free Windows emulated version of both games is available from Big Five Software’s website3.

Bibliography and further reading

[Anderson1983]
John J. Anderson, Mastering Miner 2049er, Creative Computer Video & Arcade Games 1, 2 (1983-12), 103–108.
A detailed strategy guide with routes for each of the game’s ten stages.
[ConsumerGuide1983]
The Editors of Consumer Guide, Personal Computers & Games. (Beekman House, 1983).
A general overview of the computer systems available, short guides to a few games, and brief reviews of several more.
Miner 2049er is featured as ‘The Best Theme Game’ and a guide to the first five stages is given on pages 55-57.
[ElectronicGames1983]
The Editors of Electronic Games, The Miner 2049er Story, Electronic Games 2, 6 (1983-08), 26–42.
[MarentesFriedman2001]
Nickolas Marentes & Barry Friedman, Interview with Barry Friedman, The Miner 2049er Information Page (2001-06).
[MarentesKunkel2001]
Nickolas Marentes & Bill Kunkel, Interview with Bill Kunkel, The Miner 2049er Information Page (2001-04).
[MarentesLivesay2001]
Nickolas Marentes & Mike Livesay, Interview with Mike Livesay, The Miner 2049er Information Page (2001-04).
[MarentesRoss2001]
Nickolas Marentes & Scott Ross, Interview with Scott Ross, The Miner 2049er Information Page (2001-05).
[ReedHogue2008]
Matthew Reed & Bill Hogue, An Interview with Bill Hogue, TRS-80.org (2008-10-29).
Miner 2049er is mentioned briefly.
[SavetzHogue2015]
Kevin Savetz & Bill Hogue, ANTIC Interview 94 – Bill Hogue, Miner 2049er, (2015-10-29).
A somewhat directionless interview reminiscing about Hogue’s work on games, and particularly Miner 2049er and its sequel, Bounty Bob Strikes Back.

  1. As [ElectronicGames1983] says, Miner 2049er “is as important for what it isn’t as for what it is”. In particular, it’s not an arcade conversion nor based on an existing license. Noteworthy, as “until the recent crack-down on infringers by the coin-op manufacturers, all too many computer games were little more than knock-offs of pay-for-play machines. No one will ever know how many computer games came into being as a result of ‘fact-finding’ trips by designers to their local family amusement centers. More than one programmer has returned from the arcade after pumping a few tokens into a promising game, with the outlines of something awfully similar already percolating in his head.” 
  2. But you don’t have to take my word for it. According to [ElectronicGames1983]: “The phenomenon — and make no mistake about it, Miner’s publication is perhaps the most significant software event of this year…” 
  3. Hogue recalled in [SavetzHogue2015] that when producing the emulated version, he had to spend some hours or days cracking his own anti-piracy measures–he never imagined that the person he’d be protecting his game from would be himself! 

Posted in Good, Platformer, Full Review, 1982, Atari 8-bit | Tagged: , | Leave a Comment »

Pitfall!

Posted by Tracy Poff on February 9, 2016

Since we’re done with Space Panic and [Donkey Kong]([souce 2874]) (for now, though it has many, many ports, clones, and variants), we’ve come to the earliest platformer that I really enjoy: Pitfall! for Atari 2600, released by Activision in 19821.

Pitfall! cover

Of course, one cannot talk of an Activision game without mentioning the game’s designer. Pitfall! was created by David Crane, co-founder of Activision and creator of numerous other worthy games, including Little Computer People and A Boy and His Blob.

The creation of Pitfall! is what you might term a deliberate accident. Crane did not set out to create a platform game about a man in a jungle. He had been planning a sports game (later released as The Activision Decathlon) which he shelved because he felt he couldn’t do it justice. He had, however created a subroutine to animate a running man, which he wanted to use somehow. So he started to build a game around it2:

OK, there he is, running across the screen. What now? So I might as well put him on a path. Jungles have paths — better throw in a few trees — always bearing in mind that I’d want to be able to do this for other machines. Basically, if you can do it on the VCS, you can do at least a shadow of it on other systems.

So anyway, what use is a jungle path unless it leads somewhere? So I pencilled in a few objects. How about some places to fall? A few holes. He’s got to land somewhere — I had to put in an underground level. Then I spent the next two months defining the game, saying where do I put the treasure, what kind of monsters lurk? Scorpions look pretty good. I thought I might have ghosts and skeletons in the tunnel — none of them looked good, so they didn’t get in. We drew a lot of these beforehand on squared paper, colouring them in and so on. But it never looks the same on the screen as it does on paper — never.

That game, called Jungle Runner during development, became Pitfall!, went on to sell over 4 million copies on the 26003 (spending 64 weeks as the #1 best selling game), and was the progenitor of the smooth running and jumping that would be seen in the Super Mario Bros. series and many other, later platformers.

Pitfall! 01

When the game begins, you have 2000 points and two extra lives (which the manual calls ‘replacement Harrys’). The first screen is a gentle introduction: a single pit with a ladder and a stationary log are the only obstacles present. Falling into one of these pits (rather than climbing down a ladder) will cost you 100 points, while hitting a log will cost you some points over time as you remain in contact with it.

From the beginning, and at any point during the game, you can go either to the right or the left (unless there’s a wall in the way).

Pitfall! 02

The screen immediately to the right is more challenging: it contains three pits, only one ladder, and two logs rolling toward you. We can see immediately why (as the manual suggests) it is easier to go to the left–the logs always roll from right to left, so by moving in the same direction as the logs, you never have to worry about jumping over them. But where’s the fun in that? Onward to the right!

Pitfall! 03

More obstacles. This time, the rolling logs are joined by a wide pit–falling in this kind of pit means losing a life. Fortunately, there’s a vine above the pit you can use to swing across, so it’s merely a matter of timing the jump correctly to grab onto the vine, and then dropping off on the other side. In later screens, these pits will sometimes open and close, so you’ve got to be careful–a screen that seems safe may turn out to have a pit that opens under your feet, if you just run across incautiously.

Pitfall! 04

In this image we can see the remaining (major) obstacles in the game: crocodiles4 and scorpions. The crocodiles periodically open and close their mouths. When the mouths are closed, you can jump on them to get across the pool of water. When the mouths are open, you can only stand on th far right side of the crocodiles’ heads, behind the jaw, or you’ll be eaten. The scorpions merely move from left to right in the underground section, but they’re very wide, so precise jumps are necessary to make it over them.

So, if those are the main obstacles… what’s the point of this game?

Pitfall! 05

Collecting treasure for points, of course! The gold bar you see above is worth 4000 points. Silver bars are worth 3000, money bags are worth 2000, and diamond rings are worth a whopping 5000 points each. There are eight of each type of treasure, for a total of 32 treasures worth 112,000 points. A perfect game would end with 114,000 points (all the points for the treasures, plus the 2000 you started with, and no points lost to mistakes).

The game would be difficult, but manageable if you could just take your time with each screen. You might lose a few points to logs and other hazards, but with enough care around the deadly obstacles, collecting all 32 treasures would just be a matter of time. But time isn’t something you have to spare: there are 20 minutes on the clock when you start, and that’s all you get. It might sound like a pretty long time, but there are 255 screens in Pitfall!, leaving you with less than five seconds per screen, if you must visit them all.

How ever could you succeed with such a tight time limit? That’s where a clever mechanic comes into play. You’ve seen that each screen has an aboveground and underground part, the latter reach by either falling down a pit or climbing down a ladder. Every screen that you cross in the underground section is equivalent to three screens crossed in the aboveground section. Of course, you could skip right over a screen with a treasure on it, if you take the underground shortcuts through the whole game. So what are you to do? The manual suggests making a map5.

Pitfall! 06

You don’t have to get every treasure, of course. I was pretty happy getting just under 32,000 points, after a few tries. Back when the game was released, Activision offered to send an Explorers’ Club patch to anyone who got at least 20,000 points and sent in a picture of the TV screen to prove it.

Pitfall! is a great early platform game, and its sequel Pitfall II: Lost Caverns was if anything even more impressive and ahead of its time. Anyone curious about where platformers came from should absolutely give it a try. And even if you’re not a game historian, it’s a fun game well worth playing.

Further reading


  1. The date of April 20 is given by allgame, though I know not on strength of what evidence. In an interview I see the release dated to September. The year, at least, is correct. 
  2. This excerpt is from an interview in Big K #1 (April 1984). 
  3. This sales figure is given by IGN
  4. The crocodiles were inspired by the introduction to The Heckle and Jeckle Cartoon Show
  5. Of course, if you don’t want to make your own, you can use someone else’s. This map by Ben Valdes not only shows the contents of the rooms, but also suggests the best route to take. 

Posted in 1982, Action, Atari 2600, Full Review, Good, Platformer | Tagged: , | Leave a Comment »

First Impressions: Naughty Mouse

Posted by Tracy Poff on February 8, 2016

Naughty Mouse is a 1981 collect-em-up by Amenip.

Naughty Mouse (arcade) [nmouse] - title

First, a point of contention: Arcade-History describes this game as a platformer, and mentions that “The player has a single button with which to make Naughty Mouse jump over [enemies].” In truth, the game is no platformer, and, as far as I can tell, has no buttons. Okay, enough about what it isn’t. What is Naughty Mouse?

Naughty Mouse (arcade) [nmouse] - 01

It’s a collect-em-up, like Pac-Man (and, apparently, runs on the same hardware). The player controls the titular mouse and must touch the eggs on each of the houses while avoiding the birds in order to complete the level, racing against the timer. When touching an egg, the player scores the number of points remaining on the countdown timer. When the timer reaches zero, or when the player touches an enemy, a life is lost and the level is reset.

In the first level, there are five eggs to touch and two enemies.

Naughty Mouse (arcade) [nmouse] - 02

In the second level, there are three enemies, instead.

Naughty Mouse (arcade) [nmouse] - 03

The second level is as far as I got, though. I was never very good at Pac-Man, and this game seems a bit more difficult to me. Also, I don’t really enjoy this kind of game, so I’m going to call 11,810 points good enough.

Amenip also released a very similar variation of this game called Woodpecker.

Posted in 1981, Arcade, Arcade, Bad, Collect 'em Up | Tagged: | Leave a Comment »

Brickout!

Posted by Tracy Poff on February 3, 2016

Nearly ten years ago, now, I wrote a ‘first impressions’ review of Brickout! for the Intellivision; as it happens, it’s the only Intellivision game I’ve ever reviewed here. I was pretty hard on the game, and I’ve learned some things in the years since that review, so in the interest of fairness, I’d like to take another quick look at it.

The story of Brickout! starts with another cartridge, Triple Action, a multi-game compilation programmed by Rich O’Keefe, containing games inspired by Atari products, developed under the working title Some of Theirs. Originally conceived as containing six, and later five, games, in the final cut two more games were removed as being too similar to Atari games (thus inviting legal trouble), one a Pong clone, and the other–Brickout!.1

Unlike the Pong clone, which to my knowledge does not survive, the excised Breakout clone found its way onto the 1998 Windows and Mac compilation Intellivision Lives!, along with about fifty other games and numerous extras.2 It has since been included in other products such as the Intellivision Flashback reproduction console.

So, does this history lesson change my opinion of the game? Not really. The ball is still tiny, the collision detection is still bad, and the lack of a paddle controller is still disappointing. It’s more forgivable in an unreleased prototype, though. The video game market was already well on its way to the surfeit of low-quality clones that preceded the great video game crash, so it’s in some ways comforting that this game was kept back, even if it was for pragmatic rather than artistic reasons.

So, my prior recommendation stands: if you want to play a Breakout clone, Arkanoid is a much better choice. But, maybe, if you’re in an academic mood, it wouldn’t hurt to take a glance at Brickout!, too.


  1. This history is thanks to Intellivision Productions
  2. Which is still available for purchase in an updated edition, and which was also ported to PS2, Xbox, and Gamecube. There was a compilation under the same title for the Nintendo DS, though I believe it included only games, and no historical extras. 

Posted in 1981, 1998, Bad, Breakout, Intellivision | Tagged: | Leave a Comment »

Gender and video games: How is female gender generally represented in various genres of video games? by Xeniya Kondrat

Posted by Tracy Poff on November 8, 2015

I’ve written about game-related academic papers a couple of times in the past. I think that the general gaming public would benefit by being more aware of the academic research, and the research would benefit if it received some scrutiny and discussion from the public. If I can ever arrange my schedule to suit, I’d like to do an ongoing series looking at some of these in detail (as with my previous posts) and giving briefer commentary on others (as this post will be).

Enough navel-gazing. Today, we’re taking a brief look at a recent paper in the Journal of Comparative Research in Anthropology and Sociology (Compaso) by Xeniya Kondrat, titled “Gender and video games: How is female gender generally represented in various genres of video games?”. The abstract:

Gender representation in video games is a current sensitive topic in entertainment media.
Gender studies in video games look at the difference between the portrayal of female and
male characters. Most video games tend to over-represent stereotypes and in general use
extensive violence and cruelty (Maietti, 2008). Some video games use wrong, disrespectful
and sometimes even violent representations of both genders. This research paper focuses
on the current representation of female gender in video games and how they are
represented, stereotyped and used as characters in games. Results show that there is a
difference between portraying women in the past and present. This research paper is
based on previous academic research and results which were achieved with online
questionnaire among game players and two interviews with professionals in the field of
game design. The results show that there is still negative stereotyping of female gender.
However, at the same time, the answers of the respondents show that the target audience
of video games desires improvements in presentation of female gender as well as male.

On accuracy

A study on the changing portrayal of female gender in video games is right up my alley, but this one sets off red flags from the word go. Compaso claims to be a peer-reviewed journal, but the “Introduction and background” section contains some errors that even the most passing familiarity with games (or the most cursory examination of the paper’s sources) would reveal.

I’ll look at just the first paragraph, sentence by sentence. Since Compaso is an open access journal (for which I am grateful, even if I’m being harsh on the paper), feel free to follow along with me, here.

Video game history started in 1940 when Edward Condon designed a computer that could play a game called “Nim” with one player (Video Game History Timeline).

That’s an odd place to begin. If you’re going to include simple games like Nim, you might as well include other electromechanical games of the era. I found a patent for an electrical shooting gallery with a lightgun that was filed in 1936, for example.

I won’t split hairs, though. You can choose whatever starting point suits you–there’s not one right answer. The reason the author chose this particular starting point, I believe, was that her sole source for information on the history of video games was the Video Game History Timeline cited, and it begins there.

The first home video game, “Space Odyssey”, was created in 1972.

I don’t know what to say about this one. I can’t find any game from that year with that title. I think Kondrat is confusing Spacewar! (released in 1962), one of the earliest computer games, with the Magnavox Odyssey (patented in 1968 and first sold in 1972), the first home game console.

In 1993, a release of “Mortal Kombat” forced the US government to start rating the games based on their violence level (Video Game History Timeline).

Again, this is just not right. The ESRB was established shortly after the release of Mortal Kombat (though the way I heard it, Night Trap was more responsible for its formation), but the ESRB is not part of the US government (which isn’t in the business of rating video games, or movies or anything else, for that matter), and games had been rated prior to its establishment, though not in so organized a fashion.

The first game with a female protagonist appeared in 1996: “Tomb Raider”.

Now this is just a whopper. Not only is it incredibly false, missing even hugely popular games like Metroid (and MobyGames lists at least 400 other, earlier games with female protagonists), it even contradicts the Video Game History Timeline that Kondrat has relied upon.

For 1996, the VGHT says “Lara Croft debuts as the star of Eidos’s adventure game Tomb Raider. Players love her, but critics charge that she’s an example of sexism in video games.” No mention of ‘first game with a female protagonist.’

On the other hand, the VGHT’s entry for 1980 mentions that “Two years later [than the release of Pac-Man], Ms. Pac-Man strikes a blow for gender equality by becoming the best-selling arcade game of all time.”

It became one of the most popular games in video game history.

No complaint, Tomb Raider really was popular.

Afterwards, Will Wrights created a game called “The Sims” in 2000 which became the most popular game amongst female players.

First, the man’s name is ‘Wright’. Second, this is just repeating part of an entry on the VGHT: “Will Wright’s The Sims models real life. It is not the first simulation game . . . but it becomes the best-selling computer game ever and the most popular game with female players.”

Further comments

When I first read this paper’s abstract, I’d hoped to be able to write a post reporting on some interesting results, but the sheer density of factual errors in the introduction has quite put me off this paper.

Frankly, my skimming of the rest of the paper has left me shocked that a publication that purports to be a serious, peer-reviewed journal would print it. This paper “was a part of
[Kondrat’s] graduation project for her undergraduate study in International Communication and Media at the HU University of Applied Sciences Utrecht.” Formatting or language issues I forgive easily, but the factual errors and lack of clarity and organization is more what I would expect from a high school student’s rushed term paper than a postgraduate student’s paper submitted for publication.

I’d rather build up than tear down, but it’s my policy that if I decide to review something, I follow through on negative as well as positive reviews. In the interest of fairness to the journal: it appears, from a brief look, that other articles published in that issue, if not excellent, are of far higher quality. Some, indeed, look quite interesting.

Bibliography

[Kondrat2015]
Xeniya Kondrat, Gender and video games: How is female gender generally represented in various genres of video games?, Journal of Comparative Research in Anthropology and Sociology 6, 1 (2015), 171–193.

Posted in General Commentary | Tagged: , , | 2 Comments »

Hanami no Yopparai wo Osaeru

Posted by Tracy Poff on October 6, 2015

It’s October, and that means IFComp! Naturally, therefore, I’m reviewing… not interactive fiction. I’ll get to that Real Soon Now. Instead, I’m continuing my (announced and immediately ignored) series on independent (doujin) games.

The game we’re looking at today is Hanami no Yopparai wo Osaeru (花見の酔っ払いを抑える), meaning (very roughly) Stop the flower-viewing1 drunkard.

Hanami no Yopparai wo Osaeru title

The developer was attempting to make a game in the style of old handheld games, and in my judgment had absolute success.

Hanami no Yopparai wo Osaeru 01

The player character, on the left, must reach the drunk, on the right, while avoiding the thrown bottles.

Hanami no Yopparai wo Osaeru 02

Reaching the drunk successfully adds points to the player’s score.

Hanami no Yopparai wo Osaeru 03

Failing to avoid the bottles costs the player time.

It’s hard to tell from the screenshots, but the bottles are moving from position to position like they would on a Game & Watch or other LCD game. A video will show it better:

I only played the trial version of the game (available on DLsite). I’m not sure what might be different in the full version. This isn’t a great game, even for 108 yen. For a similar, and rather better, game, you can play Dave Baskin’s Bouncing Babies for DOS, itself a clone of the Game & Watch title Fire.


  1. For the cultural significance of hanami, see Wikipedia

Posted in 2015, Action, Arcade, Bad, Full Review, Windows | Tagged: , , | Leave a Comment »

Nintendo Action Games

Posted by Tracy Poff on July 2, 2015

Nintendo Action Games by Christopher Lampton reviews twelve action games for the NES.

nintendo-action-games-cover

It’s incredible that not so long ago a book such as this could be published. There’s nothing wrong with the idea of it–brief reviews and miniature strategy guides was also the format of Jeff Rovin’s excellent How to Win at Nintendo Games series, after all–but there are only twelve games included in the book, and the entries are far from thorough.

The average entry contains one full-page screenshot and about two pages of text, half of which is a description and review, and half tips, passwords, and trivia. The reviews are brief sketches at best. The tips are often useless, too; for Tetris, Lampton offers:

The rows that you fill in don’t have to be at the bottom of the screen. They can be anywhere in the pile. If there are holes you can’t reach because of the pieces on top of them, see if you can’t remove the rows that are in your way.

Another tip informs us that the line piece, though rare, is useful. Gee, thanks.

The ‘Fascinating Factoids’ are no better. From the entry for Ninja Gaiden:

The Japanese word gaiden means “telegram” or “message.” At the beginning of the game, the Ninja Ryu receives a message from his late father telling him to go to America. Hence, the title means, roughly, “Ninja message.”

Nice story. It’d be nicer if it were true. The gaiden (外伝) in the title means, roughly, ‘side story’. There is a word gaiden (外電) meaning telegram, but it’s not the one that is used in the game’s title. Better luck next time.

In the future, I’ll take a look at Rovin’s work for an example of this kind of book done right. As for this one–well, don’t trust it more than any random blog on the internet, I suppose.

List of games

  • The Adventures of Lolo 2
  • Batman
  • Bubble Bobble
  • Contra
  • Double Dragon II: The Revenge
  • Duck Tales
  • Ninja Gaiden
  • Super Mario Bros. 2
  • Super Mario Bros. 3
  • Teenage Mutant Ninja Turtles
  • Tetris
  • 1943

Bibliography

[Lampton1991]
Christopher Lampton, Nintendo Action Games. (Brookfield, CT: Millbrook Press, 1991).
[Rovin1988]
Jeff Rovin, How to Win at Nintendo Games. (St. Martin’s Press, 1988).

Posted in 1991, Book | Tagged: | 2 Comments »

Classic Nintendo Games are (NP-)Hard by Aloupis et al.

Posted by Tracy Poff on June 21, 2015

I came across an interesting paper on arXiv, “Classic Nintendo Games are (NP-)Hard” by Greg Aloupis, Erik D. Demaine, Alan Guo, and Giovanni Viglietta. The paper is originally from 2012, but it’s been updated in February of this year with additional results. It’s been mentioned a few times on the web, but I didn’t see anyone give a satisfactory explanation of what the paper is really about. I’m not an expert, but I’ll give it a go.

First, though, the abstract:

We prove NP-hardness results for five of Nintendo’s largest video game franchises: Mario, Donkey Kong, Legend of Zelda, Metroid, and Pokemon. Our results apply to Super Mario Bros. 1, 3, Lost Levels, and Super Mario World; Donkey Kong Country 1-3; all Legend of Zelda games except Zelda II: The Adventure of Link; all Metroid games; and all Pokemon role-playing games. For Mario and Donkey Kong, we show NP-completeness. In addition, we observe that several games in the Zelda series are PSPACE-complete.

What does it all mean?

I’ll go into some detail shortly, but I’d like to describe in broad strokes what this is (and isn’t).

The authors look at how hard it is to answer the question: given a particular level layout, a starting point, and a finish point, is it possible to get from the start to the finish?

This isn’t about the levels actually in the games–in fact, it isn’t really about the level design at all. What it is about is the rules of the games. Super Mario Bros., for example, has several well-known rules:

  1. Mushrooms make small Mario big.
  2. Touching enemies (or some other hazards) makes big Mario small, or kills small Mario.
  3. Mario can break bricks only if he’s big.
  4. Stars make Mario temporarily invulnerable.
  5. Mario can jump to a height of 4 tiles, or 5 when running.

There are more rules, but you see the point. Importantly, the authors specify that their results are about an idealized version of the rules. In other words, glitches don’t count. Sorry.

What the authors prove is that if you turn someone loose with a level editor, they can design a level for which it is very hard to determine whether it’s even possible to complete.

They do this by showing that you can implement a problem called the 3-satisfiability problem (or 3-SAT) in each of the games. Think of someone using redstone to make Tetris in Minecraft, and you’ve got the general idea. 3-SAT is known to be NP-complete (read as: difficult), and the game it’s implemented in must be at least as hard as 3-SAT. Those problems are called NP-hard.

Details

The 3-satisfiability problem is to determine if, given a boolean formula of a certain form, it is possible to set the variables (which can be either true or false) in such a way that the whole formula comes out true. A two-variable example of 3-SAT might be:

(y is TRUE | x is TRUE | x is TRUE)
    AND
(x is FALSE | y is TRUE | y is TRUE)
    AND
(x is FALSE | y is FALSE | y is FALSE)

(Don’t be alarmed by the duplication. The reason for writing that way will be apparent later.)

The 3 in 3-SAT means that each clause (the parts in parentheses) can have at most 3 components. In this case, each has only two distinct components–one is repeated.

All three clauses must be true for the formula to be true, and a clause is true if any of its components (which are separated by pipes) is true. In logic-speak, it’s a conjunction of several disjunctions.

It’s pretty easy to see that this expression is true if y is true (satisfying the first part) and x is false (satisfying the other two). But what if there were fifty or a hundred or a billion sub-expressions? That would be harder to tell.

Requirements for 3-SAT

To be able to implement 3-SAT, you need a few components. You need:

  1. A starting point
  2. A finishing point
  3. A way to pick whether a variable is true or false
  4. A way for those choices to either impede or allow progress
  5. A way to let paths cross without being able to change from one path to another

That last one is so that you can build big, complex paths. Without it, you’d be able to backtrack and mess things up.

The authors implement these components in ‘gadgets’, which are basically self-contained rooms or screens that can be connected to each other. They’re the building blocks of the satisfiability puzzle.

Gadgets in Super Mario Bros.

I’ll take a brief look at how the authors build these gadgets for Super Mario Bros., as an example. This is going to be pretty much the same as what the authors say in the paper, so if you’re feeling industrious, you can look at the paper and see how they put it.

Start

The start gadget is just where Mario starts the level–there’s nothing to it. In general, the authors want Mario to be big throughout the level, so they put a mushroom at the beginning that will be required at the finish.

Finish

Again, this is trivial. It’s just the flagpole. To make sure Mario stayed big throughout the level, you have him enter the screen with the flagpole in a corridor with a brick he must break to reach the flagpole.

Finish gadget

Variable

To pick whether a variable is true or false, Mario is given a screen with two possible paths–a vertical drop to either the left or right. Say left means “x is true” and right means “x is false”. It doesn’t matter which. The drops are long enough that Mario can’t get back up, so once a choice is made, Mario can’t go back and also take the other path.

Clause and Check

This is the most interesting (and important) gadget in Super Mario Bros. Take a look at it:

Clause and Check gadget

After making a choice and dropping from the Variable gadget, Mario will be led to a series of Clause gadgets. He enters one of the small open areas from the bottom–each one of them will have a path from a Variable gadget that leads to it.

Each block contains a star, which will be trapped in the ‘fence’ when Mario hits the block. Later, Mario will come through the top part (that’s the Check part of the gadget) and collect the star so that he can run through the Firebars. If Mario gets to the Check gadget without having hit the star block (thus satisfying the disjunction), he can’t proceed–the Firebars will kill him.

Remember our example of 3-SAT from earlier?

(y is TRUE | x is TRUE | x is TRUE)
    AND
(x is FALSE | y is TRUE | y is TRUE)
    AND
(x is FALSE | y is FALSE | y is FALSE)

Imagine that the gadget pictured corresponds to the top disjunction, (y is TRUE | x is TRUE | x is TRUE). Then Mario will reach the left star block if he picked “y is TRUE” by dropping down the left side of the corresponding Variable gadget, and the middle and right star blocks by picking “x is TRUE” by dropping down the left side of that particular Variable gadget.

But what if Mario picked “x is false” and “y is false”? Then he’d never get to the blocks and release the star, so when he got to the Check gadget, the Firebars would kill him. That means that both variables being false isn’t a solution to the expression, and (equivalently) taking the right path at both variable gadgets won’t let Mario finish the level.

Crossover

This is just a utility to let paths cross without Mario backtracking and making two different choices for one Variable.

Crossover gadget

If Mario comes in from the lower path, he can break the bricks and keep going up.

If he comes in from the upper path, he can run into the Goomba to become small, head to the right, pick up the mushroom from the block, break the brick, and continue to the right.

Small Mario can’t break the bricks in the middle to go up, and big Mario can’t get through the narrow paths to go to the right. Big Mario also can’t break the bricks and then go back, because the drop on the left is too far for him to get back up.

Putting it all together

You can string together as many Variable, Clause, and Check gadgets as you need to make as big of a problem as you want. Since we know that 3-SAT is NP-complete, and we’ve just implemented 3-SAT in Super Mario Bros., we know that SMB is NP-hard.

Congratulations! If you’ve made it this far, then you’ve got the major idea of (part of) the paper. Pick up your CS degree on the way out the door.

Other games, other problems

I won’t go into detail about the other sections of the paper, but if you’ve read this far, you should be able to read the paper and see what is being done. The proofs for other Mario games just make some modifications to the gadgets we looked at above.

The proofs for the Zelda games are different. They still work by implementing an NP-hard problem, but it’s a different problem, and they implement it using the sliding blocks in the Zelda series.

If you want to know about the other games, or get more details, take look at the paper.

Caveat lector

I’ve done my best, here, but I’m not an expert. I’ve got a BS in math, not a doctorate in computer science. If I’ve made some terrible mistake, it’s all my fault (and I’d appreciate it if you’d let me know!), and I apologize for anything misleading, confusing, or dangerous that I’ve written. Always wear proper safety equipment when proving theorems.

Bibliography

[Aloupis2012v3]
Greg Aloupis & Erik D. Demaine & Alan Guo et al., Classic Nintendo Games are (NP-)Hard, (2012-03-08).

Posted in General Commentary | Tagged: , , , , , , , , , , | Leave a Comment »

TIS-100

Posted by Tracy Poff on June 19, 2015

370360_2015-06-17_00004

Recently, I saw a new game from Zachtronics Industries, TIS-100, which was released on Steam as an early access title on the first of June. In some ways, calling it a game is overstating it: it’s little more than a collection of programming problems, with a little story to give it some structure. The catch is that you’re programming in an assembly language on a virtual machine with unusual architecture; problems beyond the simplest will generally require you to take advantage of parallelism (which is the primary distinguishing feature of the VM), resulting in novel solutions for ordinary problems.

370360_2015-06-17_00005

Obviously, a game like that has a rather limited target audience. Case in point: I have myself previously created a little VM with a fake assembly language to play with. The game is clearly made just for me, but how many others are likely to be similarly interested? About 11,000,1 so far. It’s a minor hit.2

The concept of programming as gameplay isn’t new. Indeed, Zachtronics’s earlier game, SpaceChem, is also an exercise in parallel programming, though dressed up in fancier clothes. Way back in the mists of time,3 Robot Odyssey challenged players to program the titular robots to solve puzzles. And on the more-programming-than-game end of the spectrum, we have Core War4 and a multitude of web sites in the vein of Project Euler or CodinGame.

I’ve been enjoying TIS-100, but more than that, I think it’s singularly impressive to release a game of this kind. Certainly, there are games that trade on their difficulty (Super HexagonI Wanna Be the Guy, etc.) and some that take pride in their difficulty of interaction (Surgeon SimulatorAmpu-Tea, QWOP, etc.), and simple ‘retro-style’ graphics are de rigueur for indie games, but the very minimalistic functionalism of TIS-100 is astounding.

TIS-100 is difficult because the thinking required to solve the puzzles is difficult. It is perhaps inaccessible, because it consists of nothing else but the tools to solve the puzzle. Its graphics are simple because everything you need to solve the puzzles is a text-mode interactive debugger, and that’s what you get. Like a sudoku puzzle or a crossword, TIS-100 is a completely pure puzzle game: the game takes place in your head, and the software keeps score.

It is not by chance that TIS-100 so distinguishes itself from other games. During the production of Infinifactory, Zach Barth, the founder of Zachtronics, wanted to make a game with a smaller team–something more-indie-than-indie–to get back to his roots as an indie developer. The project turned out to be too great in scope, but from its wreckage was salvaged a programming minigame which became TIS-100.5 Viewed as an indie developer’s attempt to make something even more indie, with the understanding that it was a small part of something larger, the design makes sense.6

The game’s manual, too, reflects the niche targeted by the game. Who reads a manual, you ask? When it is positioned as a technical document describing the instruction set of a virtual machine, the answer to that question is: programmers. The manual is presented as the in-universe manual for the TIS-100 computer, previously the property of the player character’s Uncle Randy, including handwritten notes and highlighting. This was part of Zachtronics’s attempt to make a game with “an irresistible value proposition. For us, that’s a game with a 14-page technical manual that we designed, printed out, marked up and scanned back in again.”7 The manual is reminiscent of the feelies accompanying Infocom games, among others, in years past.8

370360_2015-06-18_00001

Like its predecessor, SpaceChemTIS-100 encourages players to perfect their solutions, optimizing for either execution speed, least number of nodes used, or least instructions–goals which are often contradictory, requiring multiple solutions.

Perhaps unsurprisingly, given that it is a game about programming, the players of TIS-100 have created some auxiliary tools, including TIS-100 PAD,9 which allows users to more easily share solutions, and a variety of TIS-100 (the virtual machine, not the game) emulators.10

In addition to this unsolicited community participation, with the release of the specification editor, which allows players to make their own puzzles, a puzzle design contest was announced. Twenty-five puzzles will be selected from the submissions for an official bonus campaign.

The feel of TIS-100 is both nostalgic and quite modern. It’s an intriguing combination, and I recommend it to anyone still interested after hearing me call it “a collection of programming problems.” Coders, no prior experience with assembly is needed. Others, if you like this game–try coding. You’ll probably like that, too.

Bibliography

[Barth2012]
Zach Barth, Postmortem: Zachtronics Industries’ SpaceChem, Gamasutra (2012-06-13).
[Dewdney1984]
Alexander K. Dewdney, In the game called Core War hostile programs engage in a battle of bits, Scientific American 250, 5 (1984), 15–19.
[McIlroy1971]
M. Douglas McIlroy & Robert Morris & Victor Vyssotsky, Letter to Aleph-Null (1971-06-29).
[Wawro2015]
Alex Wawro, ‘Things we create tell people who we are’: Designing Zachtronics’ TIS-100, Gamasutra (2015-06-09).

  1. According to SteamSpy
  2. As of this writing, it has 270 positive reviews and 2 negative reviews on Steam
  3. 1984, actually. It was released a year later for DOS, according to MobyGames
  4. Also from 1984, described in a Scientific American article, [Dewdney1984]. It’s based on a still earlier programming game, Darwin, which was played in 1961 and described publicly in 1972. See [McIlroy1971] for more. 
  5. The details of TIS-100‘s inception, and more, are discussed in an interview published by Gamasutra, [Wawro2015]. 
  6. However, Barth wrote in a post mortem of SpaceChem, [Barth2012], that SpaceChem was too difficult and inaccessible. New titles were forthcoming: “New titles, I might add, that are hopefully more accessible than SpaceChem!” 
  7. From [Wawro2015]. 
  8. Back when you got something for your money! Even application software used to have much more bulk to it
  9. Source available on GitHub
  10. Just have a look at the results of this search. But watch out, if you’d like to avoid spoilers! The puzzle solutions are code, after all, so a number of people have posted those, as well. 

Posted in 2015, Full Review, Good, Puzzle, Windows | Tagged: , | Leave a Comment »

Bleed Out Sakuretsu

Posted by Tracy Poff on April 5, 2015

I don’t believe I’ve written about any remotely popular games for at least two years… and I’m not about to change that. But I am writing about a good game, at least: Bleed Out Sakuretsu, a vertically scrolling shooter for the Sharp X68000, by Gold Cats Project.

Bleed Out Sakuretsu - title

First: yes, the title screen says ‘sakuretu’. But the common romanization of this game’s title seems to be ‘sakuretsu’, and as goes TOSEC, so goes my nation.

Bleed Out Sakuretsu - ingame

Bleed Out Sakuretsu has a powerup system reminiscent of Gradius. As you continue scoring points, you can purchase a shot upgrade, a barrier, or other, even more powerful powerups. The opposition is a variety of ships, small shots, and guided missiles.

Bleed Out Sakuretsu - boss

The boss sprays out bullets danmaku-style, and periodically fires a pair of front-facing lasers. You’ve got to deal with this while avoiding guided missiles, and since you are destroyed in one shot (unless you buy a barrier), this, like the rest of the level, provides a good challenge.

Here, an unfortunate revelation: the boss is, as far as I can tell, invulnerable, because the available version of Bleed Out Sakuretsu is a trial edition, and I can find no evidence that any full version was ever created. I would love to hear otherwise, if anyone has better information on old doujin games than I do, because it’s a very fun game.

It’s hard to do a shooter justice with just words and screenshots. The video above shows (I believe) the entire game, which should give a much better picture of it.

If you like shooters, I’d say Bleed Out Sakuretsu is worth a try. It’s only about two minutes long, but it’s fun, and I believe there’s a place for brief games, too.

The Touhou series looms largest on the doujin shooter landscape, of course, but there’s a huge and fascinating variety of games (of all genres) stretching back decades, which I think could use some more attention from the English-speaking web. I’ll probably be looking at a few more doujin games in coming weeks, so I’ll try to shine a light on whichever titles catch my eye. Maybe there will be some hidden gems, or maybe just hidden failures. We’ll see what comes.

Posted in 1996, Full Review, Good, Vertical Scrolling Shooter, X68000 | Tagged: , , , | 1 Comment »

 
Follow

Get every new post delivered to your Inbox.