the Horn Farm Paste Mob

match-3 research

Despite there being a million match-3 games out there, there are two things I’m having trouble finding. I wonder if you can help!

1. Mathematical data about Bejeweled-type games. If I randomly fill an n x n grid with gems of k colors, how many moves will there be, on average? How many existing matches (i.e. gems that break before my first turn) will there be? How many new possible moves per turn does ideal play create, and how many does playing at random create (or remove)? And so on.

I could do the math on some of those questions with a little work (and I could Monte Carlo them even more easily), but not all of them, and anyway I wonder if somebody’s already done it. Like that guy who proved there’s no strategy for Tetris that guarantees you won’t lose.

2. Match-3 games with strategic elements or substantial game mechanics outside the board. There’s the wonderful-but-slightly-grindy Puzzle Quest (and a bunch of failed followups by the same company), and Dungeon Raid for iOS. Oh, and you could count Gyromancer, but its RPG elements seemed like window dressing to me– the different creatures you summoned were interchangeable.

Maybe potential imitators of Puzzle Quest were put off by the fact that even the people who actually designed Puzzle Quest couldn’t do it a second time. But seriously, look around the App Store or Kongregate– the current state of casual match-3 development is the closest thing the gaming world has to an infinite number of monkeys. Something interesting must be happening.

By “match-3″ I mean the Bejeweled mechanic but also “Bejeweled Twist” (rotate 2×2 sections of the board) and path-tracing games like Azkend or Dungeon Raid. Anything that involves matching tiles on a static board which refills itself whenever a hole appears– as opposed to games where you fill the board yourself one piece at a time, like Tetris or Snood.


Super Crate Box (Vlambeer)

Download it free for Mac or PC.

Reviews of video games have a lot of somewhat disingenuous warnings about addictiveness. “You won’t be able to put it down!” But addictiveness is in fact just about the only thing wrong with Super Crate Box– it’s a really elegant tiny game that’s unfortunately tempting to play all the fun out of in one sitting.

It’s ingenious, though. You face a room constantly filling up with monsters; your weapon changes randomly every time you pick up a crate, and your score is the number of crates you pick up before a monster gets you. With a good weapon, it only takes a few seconds to clear the room, at which point… well, if the next crate sucks, hopefully the one after THAT will be good. Or the one after that.

The thing to realize is that although there are new weapons and levels to unlock, this isn’t a game with a constant supply of treats to give you. You should think of the weapon unlocks as an extension of the tutorial– the weapons that are hard to use will be introduced one at a time, but still all pretty fast. After about 20 minutes, once you’ve got them all, it’s just you and the high score board.

When Portal came out, Jerry Holkins called it “slapstick”, which struck me as weird. How can slapstick happen, if you’re in control of the protagonist and you aren’t trying to get hurt? But slapstick is about timing, and if Portal’s physics are such that you sometimes fling yourself into a wall with perfect timing, it’s funny! And so it is, sometimes, in Super Crate Box when a monster falls on your head, or you open fire with a gun whose recoil pushes you into a pit, or… well, I’ll just say that one of the best power-ups comes with a well-deserved apology from the designers.

Unquestionably worth a few minutes of your time, and impressively deep and well-balanced for how simple it is. It’s just not quite deep enough for how much fun it is.

(I’ve unlocked SMFT on the first arena, but not the other two, and I haven’t gotten Ambush Mode. So there may be gameplay I’m missing. The stuff I *have* seen doesn’t make it seem that way, though.)


Rock Band 3 experiment journal, part 1

I got one of the 102-button (6 strings x 17 frets) guitar controllers for Rock Band 3 and am trying to learn to play it. Which is the sort of thing I’m tempted to post about in painful detail, except–

See, I was going to say “everybody who’s playing Rock Band 3 must be doing that”, but that’s not many people. The scoreboard for “The Hardest Button To Button”, which the game tries very very hard to coax you into playing (possibly it’s the easier guitar song) has about 800 people on it. So, MAYBE a thousand people are playing Rock Band 3′s Pro Guitar Mode on XBox.

That’s not a large number.

That’s the question here, right? If Harmonix succeeded in making this game a way to learn actual guitar skills, then they perforce made a game which is really, really hard and requires a lot of repetitive drilling to get better at. Is that appealing to most people?

It is to me. Apparently. For the moment.

For one thing, it’s fun. But I’m also driven by curiosity… can you actually learn guitar this way? So I’ve been tempted to borrow my housemate’s guitar and try transferring skills immediately, I’m resisting in the name of science. I want to get to the point on plastic guitar where, if the skills were 100% compatible, I could pick up a real guitar and play a song. That, obviously, won’t happen. I’m pretty certain, though, that at that point I will also be a lot closer to playing a song on real guitar than to playing a song on real saxophone. We’ll see.

== November 14 ==

I played the first few lessons before even trying a song– if you start the game with a MIDI guitar plugged in, it says “Hey, the trainer will help you play with that. Want to be taken straight there?”

After a few lessons I tried “The Hardest Button To Button” on Easy. “Easy” indeed– one note every few measures, always on the downbeat, the kind of thing which isn’t a challenge even if you can’t do it. If you see what I mean. Okay, I thought, maybe I was too cautious. Let’s try Medium… WHOA HEY WHAT ARE CHORDS

So I went back to the lessons. Whenever I could memorize something, so that I could play it while looking down at my hands, that made things a million times easier. Even so, I hit a difficulty wall as soon as the lessons introduced real three-note chords. (Power chords were fine, at least in the basic form presented.) It took two solid sessions for me to play G-C-D-C consistently enough that the game would let me move on.

Not that a few hours is that long, but the trainer doesn’t make it easy to alternate between two lessons in order to stave off despair/boredom.

== November 16 ==


I’m used to ignoring the bass option when playing GRYBO, but Easy Pro Bass is fun. So is Medium. I still miss so many notes that going up to Hard sounds like a bad idea pedagogically, even though more complex rhythms would be nice.

(“Pro Bass” is another big step away from verisimilitude, since it is in fact just playing the guitar controller and ignoring two of the strings. But anyway, bass parts mean more hopping around the fretboard and fewer chords.)

On guitar, I’m alternating between the trainer (still deeply stuck on three-finger chords) and playing actual songs on Medium, where it turns out almost the only chords, on easier songs, are power chords and… barre chords? Chords where you stop all the strings at the same fret, most of which seem to secretly be power chords anyway, just with the fifth played lower than the… root? Tonic?

Someone taught me most of these words a long time ago.

Also, there’s no good name for this object yet. Branding different kinds of guitar controller with the names of real guitar models (this is the “Mustang”) is an elegant solution to the problem of needing names for them. It’s just also inane. Calling it a “pro guitar” is also fatuous; nobody has yet offered me money to play Rock Band 3 for them. Maybe “MIDI guitar”, even if technically that’s also irrelevant to using it as a game controller.

== November 17 ==

My fingers seem much more willing to do G-C-D-C, with a little warmup. Power chords are suddenly an issue, though. My ring finger is no longer ever in the right place if I try to switch position quickly.

The charts for Easy and Medium might be at the right spot on the difficulty curve, pedagogically speaking, but it’s much easier to learn when there’s a close fit between what you do and the sounds you hear (or don’t hear). On Easy, you pluck one string and hear five or six notes, or a chord. In songs with the guitar mixed low or heavily processed, it’s hard to even tell which sound is ‘you’.

Proposal for RB4: a switch for “play my notes louder” (or maybe two switches, so you can decide whether, on lower difficulties, the game should also amplify notes from your instrument that don’t correspond to your actions).

Might get into trouble in setups with a lot of lag, though.

== November 18 ==

The new drums showed up today. I spent longer putting them together than playing them (45 minutes and 30 minutes, respectively) and am not sure what I think yet.

It hadn’t occurred to me that most songs just don’t use the toms much except during fills. I wasn’t planning to pay the $10 to import all of Rock Band 2′s songs, but right now I’d really enjoy playing “Go Your Own Way”.

Subjectively, the speed at which the notes move seems wrong. At first I found myself speeding up or slowing down to adjust, but that just got me out of sync– the problem’s in my head. It’s as if I can still sight-read the charts at high speed, but there’s a tiny bit of extra thinking I have to do *after* I read it.

Come to think of it, that’s pretty much exactly what’s going on.

I’m sort of glad this isn’t much of an additional challenge; I only have so much time, and have already been learning two new skills concurrently, with the MIDI guitar and keyboard.

Not to mention occasionally playing music with people for real.


Super Meat Boy (Team Meat, for XBox)

Super Meat Boy has made me realize I don’t know what ‘difficulty’ is. How long it takes to finish a game? No, there are long easy games. How frustrating it is? That’s definitely part of the way a hard game feels, but frustration is draining in some games– the ones that don’t feel like they SHOULD be frustrating– and not in others. How many tries each goal takes? No, otherwise trial-and-error puzzles would be the hardest ones, as opposed to just the most tedious.

Super Meat Boy is very hard. I think. It seems hard, anyway; I’m close to 4000 deaths (it counts for you) and that’s at not quite halfway to the real ending. A lot of levels I’ve finished seemed impossible, which makes me feel good for beating them but just proves my sense of these things is highly fallible.

Is the game forgiving? Maybe that’s the key. It’s brutally unforgiving about basic failure/success; if you’re jumping over a sawblade, a split-second mistake usually means doom. On the other hand, if you have the right plan for a speed run, you can bobble the execution in a few places and still get that satisfying “GRADE: A+” at level’s end.

Plus, the more you focus, the more forgiving you can make it. Meat Boy and his friends are all very steerable in mid-jump, though this will do you more harm than good until you internalize just what rates of acceleration you can expect from various characters and situations. Most of the levels are designed to reward precision recklessness– take every jump at full speed and everything lines up but the floor is always falling out from under your feet; jump cautiously and you can catch your breath, but the angles are awkward.

It’s a very technical game. You have to enjoy seeing how the little details fit together, because there is no big picture other than “beat enough levels and you win”. If you’re capable of feeling like a winner for getting through level #28 out of 300, though, even knowing that level 29 will be even harder, you have to play Super Meat Boy at least once.


Ellen Willis died on Thursday. (Times obituary)

It’s easy for people who believe in the radicalizing power of freedom and pleasure to become self-centered; Willis was anything but. She was also the first writer to make me care about how the 60s were still relevant, and this without herself sounding stuck there. She’s had a bigger effect on my politics than any other writer.

Time to reread Beginning To See The Light, I guess, and think about the fan letter I never wrote.



Not much in this documentary would be new for anyone who’s paid attention, over the years, to the MPAA ratings board’s particular form of sickness: a little sex is worse than a lot of violence; women enjoying sex is worse than women being forced to have sex; gay sex is worse than straight sex. “Worse” is a slightly unfair term, since the first defense any MPAA bigwig raises to any criticism is that they aren’t censors– just regular folks advising parents. In theory the ratings board should be of no interest to an adult picking a movie to watch him- or herself, and so no value judgment is implied.

Filmmaker Kirby Dick fortunately doesn’t spend long belaboring how stupid this argument is. If most chains won’t exhibit your movie, Wal-Mart won’t sell it, and Blockbuster won’t rent it, then large segments of American society will never know it existed. Even so, you get the feeling the many filmmakers interviewed in This Film Is Not Yet Rated would accept the system if it were sane, or at least fair. It is not.

So we already knew how bad things were, but the MPAA’s amazing lack of integrity will still piss you off if you watch this. What little information the organization gives the public about the members of their ratings board (that they are the parents of children under 17 and serve fixed terms on the board) turns out to be flatly untrue. Longtime MPAA head Jack Valenti is depicted as incapable of speaking sincerely about anything*. And Kirby Dick can do some things that print articles can’t, like actually showing the scenes which the MPAA star chamber objects to alongside near-identical scenes from other movies that got lower ratings.

In the end Dick doesn’t have much of an argument to make; he puts information in front of the viewer and treats the conclusions as obvious. You might call it preaching to the choir, but I think he’s uniquely positioned to justify that as a documentary strategy. The fact that the ratings board is completely dependent on lies and obscurity to maintain their position of power speaks as poorly of them as anything else you could say. If you leave This Film wanting to punch Jack Valenti (which, his advanced age and the potential hypocrisy of doing so notwithstanding, you will), spare a thought for the people who ought to see it but will never hear its name except in ads for other movies.

* And Dick leaves out many unrelated obnoxious things Valenti has famously said, like his statement 24 years ago that the VCR was to his industry “as the Boston Strangler is to the woman home alone” and his like-minded recent comparison of file-sharers to terrorists.



I’ve switched database servers to make everything a lot faster. This, in turn, makes the archives usable, so those are back.

The move involved a disagreement over which character set Pastemob uses, so if you notice while reading that something untoward has happened to a letter with an umlaut or accent, let me know.



new design

Emails or comments about problems with the new site will be quite welcome.