Original image

Gaming History: Discovering Deep Ms. Pac-Man Secrets

Original image

I played a fair amount of Ms. Pac-Man in my youth, sometimes plugging my entire $3 allowance into the machine at the movie theater, or the one at the pizza parlor near my grandmother's house. I was never any good at it -- never much good at any arcade games, really -- but my father and brother were skilled Ms. Pac-Man gamers. They preferred Ms. Pac-Man to the original Pac-Man, but I don't recall ever getting an answer about why. (I must have assumed that her little red bow was the killer feature.)

Now, the original Pac-Man was famous among game nerds for its nonrandom ghost behavior. There were stories of people playing the game for days at a time by using the "hold" position -- a specific location on each board where you could park Pac-Man and he'd never be hit by a ghost, since the ghosts moved in a repeating pattern. (You'd use the hold in order to go to the bathroom, get a bite to eat, and then continue playing -- assuming you had a compatriot who would make sure no one touched the machine.) I heard legends of Pac-Man players (think Billy from King of Kong) who had racked up unbelievably high scores by memorizing the ghosts' nonrandom movements and liberally using the hold positions over multi-day marathon sessions.

But the nonrandom ghost behavior is specific to the original Pac-Man. Ms. Pac-Man was supposed to be different. I just came across an article from 1984 revealing how a secret "hold" position was actually discovered (through extensive trial-and-error) in Ms. Pac-Man when it was thought impossible due to randomized ghost movement. Here's a tidbit:

Ms. Pac-Man ScreenshotPac-Man was a game you could beat. You could beat it by memorizing patterns. The ghosts, you see, weren't programmed for randomness. If you zigged and they zagged, they'd do the exact same thing in a similar situation. It wasn't long before everybody knew the patterns to beat Pac-Man.

Ms. Pac-Man is a different story. The ghosts are programmed for randomness, so there isn't a pattern that exists to beat it-the ghosts behave differently in each game. But there is one technique that will earn a player an incredible amount of points [called] "Grouping." If you can induce the ghosts to move close to one another, you can stay alive and get 1,600 points when you gobble them near a power pill. This is the story of three guys from Montana who got together and figured out how to give Ms. Pac-Man a beating she'll never forget.

Read the rest for a nice story of kids overcoming obstacles to achieve the "impossible." (Note: the original article is by Paul Stokstad from Computer Games magazine, June 1984. It's reprinted in the linked blog with a source credit at the bottom.)

(Via the most-excellent Anarchaia.)

Original image
iStock // Ekaterina Minaeva
Man Buys Two Metric Tons of LEGO Bricks; Sorts Them Via Machine Learning
May 21, 2017
Original image
iStock // Ekaterina Minaeva

Jacques Mattheij made a small, but awesome, mistake. He went on eBay one evening and bid on a bunch of bulk LEGO brick auctions, then went to sleep. Upon waking, he discovered that he was the high bidder on many, and was now the proud owner of two tons of LEGO bricks. (This is about 4400 pounds.) He wrote, "[L]esson 1: if you win almost all bids you are bidding too high."

Mattheij had noticed that bulk, unsorted bricks sell for something like €10/kilogram, whereas sets are roughly €40/kg and rare parts go for up to €100/kg. Much of the value of the bricks is in their sorting. If he could reduce the entropy of these bins of unsorted bricks, he could make a tidy profit. While many people do this work by hand, the problem is enormous—just the kind of challenge for a computer. Mattheij writes:

There are 38000+ shapes and there are 100+ possible shades of color (you can roughly tell how old someone is by asking them what lego colors they remember from their youth).

In the following months, Mattheij built a proof-of-concept sorting system using, of course, LEGO. He broke the problem down into a series of sub-problems (including "feeding LEGO reliably from a hopper is surprisingly hard," one of those facts of nature that will stymie even the best system design). After tinkering with the prototype at length, he expanded the system to a surprisingly complex system of conveyer belts (powered by a home treadmill), various pieces of cabinetry, and "copious quantities of crazy glue."

Here's a video showing the current system running at low speed:

The key part of the system was running the bricks past a camera paired with a computer running a neural net-based image classifier. That allows the computer (when sufficiently trained on brick images) to recognize bricks and thus categorize them by color, shape, or other parameters. Remember that as bricks pass by, they can be in any orientation, can be dirty, can even be stuck to other pieces. So having a flexible software system is key to recognizing—in a fraction of a second—what a given brick is, in order to sort it out. When a match is found, a jet of compressed air pops the piece off the conveyer belt and into a waiting bin.

After much experimentation, Mattheij rewrote the software (several times in fact) to accomplish a variety of basic tasks. At its core, the system takes images from a webcam and feeds them to a neural network to do the classification. Of course, the neural net needs to be "trained" by showing it lots of images, and telling it what those images represent. Mattheij's breakthrough was allowing the machine to effectively train itself, with guidance: Running pieces through allows the system to take its own photos, make a guess, and build on that guess. As long as Mattheij corrects the incorrect guesses, he ends up with a decent (and self-reinforcing) corpus of training data. As the machine continues running, it can rack up more training, allowing it to recognize a broad variety of pieces on the fly.

Here's another video, focusing on how the pieces move on conveyer belts (running at slow speed so puny humans can follow). You can also see the air jets in action:

In an email interview, Mattheij told Mental Floss that the system currently sorts LEGO bricks into more than 50 categories. It can also be run in a color-sorting mode to bin the parts across 12 color groups. (Thus at present you'd likely do a two-pass sort on the bricks: once for shape, then a separate pass for color.) He continues to refine the system, with a focus on making its recognition abilities faster. At some point down the line, he plans to make the software portion open source. You're on your own as far as building conveyer belts, bins, and so forth.

Check out Mattheij's writeup in two parts for more information. It starts with an overview of the story, followed up with a deep dive on the software. He's also tweeting about the project (among other things). And if you look around a bit, you'll find bulk LEGO brick auctions online—it's definitely a thing!

Original image
Nick Briggs/Comic Relief
What Happened to Jamie and Aurelia From Love Actually?
May 26, 2017
Original image
Nick Briggs/Comic Relief

Fans of the romantic-comedy Love Actually recently got a bonus reunion in the form of Red Nose Day Actually, a short charity special that gave audiences a peek at where their favorite characters ended up almost 15 years later.

One of the most improbable pairings from the original film was between Jamie (Colin Firth) and Aurelia (Lúcia Moniz), who fell in love despite almost no shared vocabulary. Jamie is English, and Aurelia is Portuguese, and they know just enough of each other’s native tongues for Jamie to propose and Aurelia to accept.

A decade and a half on, they have both improved their knowledge of each other’s languages—if not perfectly, in Jamie’s case. But apparently, their love is much stronger than his grasp on Portuguese grammar, because they’ve got three bilingual kids and another on the way. (And still enjoy having important romantic moments in the car.)

In 2015, Love Actually script editor Emma Freud revealed via Twitter what happened between Karen and Harry (Emma Thompson and Alan Rickman, who passed away last year). Most of the other couples get happy endings in the short—even if Hugh Grant's character hasn't gotten any better at dancing.

[h/t TV Guide]