Sirlin Games announces complete Yomi release schedule


New and revised characters for the popular fighting card game round out cast to 20 available characters throughout 2015


San Francisco, CA; January 16 — Today Sirlin Games announced that, having fulfilled all Kickstarter pledges, the entire roster of 20 characters for its fighting-game themed card game Yomi will be made available to the general public in a phased release beginning now through the end of July.

Yomi simulates the mind games and other strategic choices found in video fighting games without requiring that players have the dexterity required to master those games. In keeping with that theme, Yomi allows players to pick a character, represented by a unique deck of cards. Each deck includes different abilities and special moves, providing for a deep, asymmetric game with varied play as each match-up between characters has its own specific flavor and strategy.

Unlike customizable or collectible card games, however, each Yomi deck is complete and balanced, meaning once players have acquired a particular deck there is no need for them to buy more cards to improve their deck.

The total set of 20 characters includes 10 characters who were previously available in the first version of Yomi (published in 2011) and have been re-balanced and artistically polished for the new release. Currently, eight characters are available for purchase. These are sold in 4-character packs (called Yomi: Round 1 and Round 2) for a suggested retail price of $39.99.

Both Round 1 and 2 offer a great way to new players to begin playing, as they include everything needed for at least 4 people to play. Round 1 features returning characters Grave, Jaina, Midori, and Setsuki while Round 2 includes the new characters Quince, Onimaru, Bal-Bas-Beta, and Troq.

“I originally developed Yomi so that everyone could experience the kind of dynamics that happen in fighting games,” said Sirlin. “I’m thrilled that it’s appealed to the tournament crowd as well as casual players. This latest version has years of polish and testing behind it, so it’s a great time to dive in.” 

Players looking to expand their options can purchase more characters according to the following schedule:

  • March 15 (PAX East): Menelker & Persephone
  • April 3: Rook & DeGrey
  • May 1: Valerie & Geiger
  • June 5 (Origins): Gloria & Gwen
  • July 3: Lum & Argagarg
  • July 31 (GenCon): Vendetta & Zane

Menelker, Persephone, Gloria, Gwen, Vendetta and Zane are all new Yomi characters. Individual decks will be sold for $9.99 each. All cards are printed on high-quality card stock with reinforced core and protective coating, making them very resistant to wear-and-tear.

Sirlin Games raised more than $217,000 — greatly exceeding its $30,000 goal — for the new and updated characters in a Kickstarter project which finished July 17, 2014. As the company has with both of its previous Kickstarters, all rewards were shipped on time by the end of December 2014. Backers who pledged for them have already received the new and improved character decks and other items; the above schedule applies only for general retail availability.

Also available as a result of the Kickstarter is the Yomi EX Powerup, an expansion for Yomi that comes with extra-powerful special moves and character cards for each of the 20 characters. These cards are not intended for serious competitive play but provide a “just for fun” alternative style. The EX Powerup expansion also includes a box designed to hold and display all 20 character decks for $39.99. It is currently available for purchase.

Finally, limited quantities exist of a special two-character package called Yomi Panda vs. G. Panda. Originally designed as a Kickstarter reward, it features a version of the Yomi panda character Lum rendered the same minimal art style found in Sirlin’s previous game Pandante, along with an all new “Gambling Panda” or “G. Panda” character.

Yomi can be purchased online from (worldwide) and (US only) or from local game stores around the world. To find a game store that carries Yomi, see To try Yomi for free, visit to play online. It is also available through the App Store on both iPad and iPhone.

About Sirlin Games

Sirlin Games creates premium, designer games, specializing in asymmetric games—the kind where players control different characters, each with different abilities and personalities. Years of care go into these game designs to make sure they strike the right balance of depth and elegance and to make sure the characters are balanced against each other.

Sirlin on Game Design, Ep6: Character Bans

In this episode, we discuss the many problems with games where players ban each other's characters. We touch on League of Legends, DOTA2, and Heroes of the Storm, though this discussion is applicable to any form of in-game banning in any game.

An unusual guest talks about the worst clock in the world.

Hosts: David Sirlin & Matt "Aphotix" DeMasi

Guilty Gear Xrd

Guilty Gear Xrd is great. I'd like to talk about the design behind it, but we really have to start with the art.


Gameplay is what should matter. Starting by telling you about the art may seem superficial, and maybe it is, but the art in this game is so notable—such a breakthrough—that I think it deserves special attention.

Guilty Gear Xrd is far and away the #1 best use of 3D that looks like 2D I have ever seen in a video game. Certain in the realm of fighting games it blows away every other attempt. Arc System Works Team Red (the makers of the game) have created many specialized 3D models that help with this illusion. In a game with 2D art, the artists can draw anything they can imagine for every frame, and so 2D fighting game characters often have limbs that change size, swords that whoosh into a blur, and entirely new shapes that appear only on certain frames such as Millia's hair attacks in previous Guilty Gear games. But 3D games are usually stuck with a certain 3D model for a character and deforming the mesh is a pain to do. Modeling entirely new character shapes just for small parts of certain moves is usually infeasible, yet that's exactly what the new Guilty Gear has done.

Check out this comparison of 2D Sol and 3D Sol:

Here's a set of different 3D models for Millia's hair

Faust is a masterpiece of this model swapping technology. I love his new walk-backward animation that makes his feet rapidly shuffle around like a cartoon character. He actually has 5 or 6 feet modeled to make that effect possible.

The lighting and shading is also incredible. We've had toon shaders for years that can make 3D look like 2D, but Team Red has perfected it to such a degree that I sometimes can't even tell I'm looking at 3D. During various anime cut-scene closeups of the characters, I have exclaimed multiple times to onlookers, "Remember, this is 3D."

The Potemkin model on the left is a development shot of what he looks like without any shaders. The right image shows him with the shading used in the game:

The shadows on the characters are partially just a consequence of what it looks like when you shine a light on a 3D model and see what happens. But the shadows were all meticulously touched up an altered to give them a 2D drawn feel. For example, Ky's skirt (left version before shadows edited; right version with edited shadow):

Or Axl's face (top versions before the custom shadow map is applied; bottom versions after):

On to the gameplay now.


The biggest new feature of the fighting system is the YRC, or yellow roman cancel. I think this is a really bold move on the developer's part and also the most surprising feature that I've seen in a fighting game in years. Before getting into what it actually is, let's look at what came before it.

Roman Cancels (Red)

Long ago, the Guilty Gear series introduced an innovative mechanic called roman cancel that lets you spend 50% of your meter (the same cost as a super move) to cancel any attack instantly on hit or block. Why is it called that? I don't know. The announcer says "romantic" when you do it, though, and you get a red special effect on the screen. Anyway, roman cancels allowed for many kinds of freeform combos. They also created the potential for lots of tricks and rushdown sequences. For example, do a string of moves that ends with something you can't normally do much after, but roman cancel that last move to instantly cancel it, then continue your pressure.

Roman cancel is a really great feature because it's so simple to do and opens up so many possibilities. You just press 3 buttons any time you want to cancel your move. You might notice that the Street Fighter 4 series copied this in a clunky, bad way by adding more inputs. There, you press 2 buttons to cancel your move into another move you don't want to do, then you dash cancel that so that you can eventualy do the move you really want to do. Roman cancel is the original though, and the better implementation of that concept by far.

False Roman Cancel (Blue)

In the game Guilty Gear XX, the series added a new mechanic: the FRC, or false roman cancel. While the normal roman cancel uses a red special effect when you do it, the newer FRC used a blue effect. FRCs cost only 25% meter rather than 50%, but you can't do them whenever you want. Specific moves have specific timing windows on them where if you press the 3 buttons to RC, you instead get a reduced cost FRC.

This added a really great dynamic to the game. I mentioned before that one use of the regular red RC was to extend your pressure sequence against an opponent. You could also do tricks where you cancel a move they didn't expect and try to hit them after that. The problem is that while those two uses are fun, they were usually not worth 50% of your meter to do. That same meter could be used to turn an ok combo into a very damaging combo, so it was usually better to do that. But FRCs created a fair way to charge only 25% meter in many cases, and that let players use them for pressure strings and tricks. It took something that existed in the game and was fun but not used much, then put way more emphasis on it.

The Problem

It's easy to see why FRCs were added. While they've added a lot to the game, they've also been a black spot on the design. They've been such a problem that I think FRCs specifically have kept a lot of people from playing the game. "Guilty Gear is really great. You should play it," I say. "Yeah but I'd have to deal with FRC timings," they say. And I frown, "yeah, that's true unfortunately."

These timing windows are incredibly strict. Some were 1 frame (1 60ths of a second), though I think later versions of the game upped those to 2 frames. But 2 or 3 frame windows are common for FRCs. That means you have to do them with very precise timing, reliably, and often. Keep in mind that this barrier involves 0 strategy or thinking. In cases where you want to do the FRC, you already made the strategic decision, and it's purely an execution test—and an artificial one at that. These moves really do not need to be this hard. Even if they needed to activate within those tiny timing windows, they could have had a buffer or something where you can do them 10 frames early and have them come out at the appropriate point.

There are other possible solutions to making FRCs easier, but the real problem is in developer mindset, not in finding a solution. Guilty Gear is a complicated game and it got more and more complicated over time. Making the game more accessible and easier to get to the strategy part did not appear to be anywhere on the list of priorities. But amazingly, now it is in Xrd. In a big way.

I'm impressed that they had the courage to do something about this, knowing full-well that the players would complain about any change. I think the truth is that requiring a bunch of 2 frame windows all over the place to even play the game properly is off the table as a thing we can even POSSIBLY consider to be in a well-designed game though. No matter what dynamics such a thing brings, it should be rejected out-of-hand as a non-starter. Surely there is some new system that is good enough, or better even, that doesn't have this kind of artificial execution difficulty.

New Roman Cancel System

Guilty Gear Xrd changed roman cancel system entirely. Now, there are 3 kinds: red, yellow, and purple. They all have the same command of inputting 3 buttons, you just get a different one depending on what's going on. If the opponent is in hitstun or blockstun, doing the command will cancel your current move with a red roman cancel. That costs 50% meter and works similarly to how it did before.

If the opponent isn't in hitstun or blockstun, you'll get the new YRC, or yellow roman cancel, and it only costs 25% meter. This means you can cancel at ANY frame during the entire startup of any move, not just during a 2-frame window. This is a night and day difference.

To put this into perspective, consider the character Ky. Ky has a fireball, and it had an FRC point before. This FRC is absolutely critical for Ky to use. He desperately needs it to keep up the pressure and for him to even work as a character. I don't know what is up with that particular FRC, but I just could not do it. Even after lots of practice in training mode, I can barely ever hit it. I have a friend who likes Ky and always wanted to play him, but he couldn't do that move either. Sometimes we'd play Ky vs Ky and during the match we'd attempt to FRC the fireballs always, 100% of the time, but get it less than 5%.

I am a very experienced Guilty Gear player. I placed 9th at the singles tournament at Evolution one year. In a team tournament at Evolution, I beat an entire Japanese team of 3 myself once. I know what I'm doing, but even I can't pick Ky because of this stupid FRC. Meanwhile, in the new Guilty Gear Xrd, it's dead simple to yellow roman cancel his fireball. Anyone can do it. I can do it over and over without missing. It just works. It's how it always should have been. Whatever issues YRC might introduce, this is exactly the solid foundation we need and the developer can iron out any issues it creates.

If you're wondering what the purple roman cancel is, I think it's basically the first bug fix to YRC. Using a YRC too late in a move to cancel its recovery would often be kind of too good, so if you input the cancel command late in a move, you get a purple cancel instead of a yellow one. The purple cancel has more startup before you're able to do your next move and it costs 50% meter. It's basically a deterrent so you don't use the cancel in that way, which is fine.

Slow Time

It's also important to note the new slow time feature that happens whenever you roman cancel. This also goes to show how much the developer cared about accessibility. Let's say you hit the opponent with a combo, then you (red) roman cancel the last big hit so you can keep going with more hits. At the moment you RC, you character is back to neutral state and you can do any move you want. This works great and I never complained about it. do have to be really fast to use this feature properly. Hit, hit, hit, roman cancel, hit, hit. You're doing the "hit, roman cancel, hit" part of that incredibly fast for it to work correctly. I personally can do that just fine, but I have to admit that it would be difficult for new players. In Guilty Gear Xrd, the roman cancel actually slows down time briefly after you do it. This lets you see and visually process what is even going on and it also gives you more time to input your next move. Experts were going to input their next move correctly anyway, but this makes the feature easier to use. I was surprised and impressed that the developers cared enough to put this feature in.

Does it change which combos are possible? Yes. But really, who cares. Some new slightly different set of combos is possible, and that can be balanced around, but when it comes to having basic competence at using the feature and understanding what's going on, it's a big improvement.

A Quick History of Guilty Gear

Let me tell you briefly about several versions that came before this one.

Guilty Gear X

Some interesting ideas, including (red) roman cancel. Pre-historic times in the series though. Catastrophically, it has a 2-button sweep, something no game should ever have. (Looking at you, Persona 4 Arena).

Guilty Gear XX

This is the breakthrough. A 5th button added so there's a rightful 1-button sweep. FRCs added. The new Burst feature to escape combos is probably the most important fighting game feature in a decade or something. It's hard to play other fighting games without Burst. Marvel vs. Capcom 3 would later go on to win the award for "game that most needed to copy Guilty Gear's Burst, but didn't." Anyway, GGXX was the first time the series really came together in the form we know now. The guard meter system, progressive gravity and hitstun decay to prevent infinites, I think "alpha counters" were added in this version. GGXX was a milestone for Guilty Gear, and for fighting games in general.

Guilty Gear XX #Reload

A refinement of the previous version. Characters slightly more fleshed out and tuned.

Guilty Gear XX Slash

A refinement of the previous version. Characters slightly more fleshed out and tuned. For example, Chipp gets his new "behind and above" teleport that he desperately needed and Potemkin gets fun new punch with unlimited super armor. I think this version is a highpoint of the series, with just the unfortunate note that Ky and Anji are slightly too powerful.

Guilty Gear XX Accent Core

A disaster. What even happened? Dozens and dozens of changes that make no sense at all to me. At one point, I tried to find even one change I thought was good, then gave up.

New mechanic: Slashback, aka parry. It's difficult yet seems to result in worse dynamics overall. It should just be removed.

Potemkin has a new special throw in the air and semi-important combos with it, but there's something about the input detection of the move that seems actually broken or buggy, given my hours and hours of training mode spent specifically on that move.

Potemkin is ruined in other more important ways, too. His "fall on face move" that knocks the opponent down now puts them in a very special state, one of the only things in the game like this, where they are knocked down yet any hit will fully launch them instead of doing the usual weak off-the-ground (OTG) hits. This combined with changes to his heat knuckle air grab mean that he has a new loop: (OTG kick, slash, heat knuckle to grab them out of the air, RC, let them fall, wall super (doesn't hit), fall-on-face, repeat) that he can threaten to do forever, even from mid-screen. It's possible to get out of, but very very difficult. Some good players I faced just couldn't get out at all, for example. This whole thing is annoying to do, requires different timing on Potemkin's part against every character, and has nothing to do with his core gameplay of being a grappler. I hated every moment of doing this powerful thing, and wished this whole technique would go away.

Here's another bad thing. Before this version, in the corner Potemkin could combo after his Potemkin Buster throw. Midscreen, he couldn't. Ok that's fine. But in this version, he has a new FRC on his Potemkin buster and he must do that to combo in the corner, and also to combo midscreen. So now there's a new, difficult timing window of a just a few frames that he has to do all the time. Potemkin didn't used to be that hard to control at a basic level, but now you need to do a difficult FRC all the time, an air combo with a very difficult (buggy?) new aerial special throw, and a finicky loop that's requires different timing against every character. He is thoroughly gunked up and ruined across the board here, and that's all just about one character.

The reason I'm telling you this whole history of Guilty Gear is actually mostly because of the Force Break feature that was added in this version. Every character now has new moves that cost 25% meter to do; they'd be called "EX moves" in other fighting games. At the time, I didn't really think they needed to add these moves. The game was already so complicated, does it need even more? Will it make the game better, and if so, is it worth the cost of complexity?

Chipp: gets some useless EX move that he might as well not have. Uh, great.

Potemkin: He got that great punch, Judgement Gauntlet, back in Slash that had infinite super armor. Now it...costs meter? This is a Procrustean Bed kind of thing where they really needed EX moves for everyone whether they made sense or not, so might as well make this one use meter now. Uh, ok. Not like it needed to before. Also that annoying-to-do aerial special throw I mentioned is an EX move. So 0 for 2 here.

Slayer: Now he's some totally different character, basically. Why? I don't know. He worked fine before. What is going on in this game?

I think some other team made this game.

Guilty Gear XX Accent Core +R

This is a refinement of Accent Core. Many changes in this game make sense, given the starting point of Accent Core. While it's an improvement, it's an improvement on a lot of stuff that was weird to begin with. It's built on top of a base that has EX moves we don't really need and FRCs that are just as fiddly as ever. While it's a good game, it's thick with weeds and desperately needs pruning.

Guilty Gear Xrd Is Simpler

I told you that whole history of Guilty Gear to give some context as to why it's a good idea that Xrd has returned to simpler times. It's like they scraped the gunk off of the design, started over, then polished up that simpler moveset. They also added a few new moves here and there where they made sense, rather than across the board as an EX system.

For example, Slayer has always had a move called Dandy Step where he quickly steps backward, then rushes forward into one of several possible moves, including an awesome one called Pilebunker. He's had that in every version, and still does. In Accent Core +R, my brain could not even keep straight all the variations of it though. He could Dandy Step into Pilebunker, or into EX Pilebunker. But he could also Reverse Dandy Step (an EX move) into Pilebunker. Or EX Dandy Step into EX Pilebunker. Yeah it's powerful to have all that, but I do not miss it at all. Instead, they reverted it all to how it used to be and added one new possible move you can do out of it. Cool.

If you want to complain about Xrd Slayer, you could make a good case. His Pilebunker doesn't reach as far, and nerfing one of the most fun moves in the game wasn't a good idea imo. They should have left it how it was and balanced around it. His Dead-on-Time super is much slower. It was absurdly fast before, which maybe caused problems, but it seems like he needed it. His Undertoe move used to be unblockable (and useless), then it got 1 hit of super armor in Slash (and was great), but I think is now reverted to the unblockable version, unfortunately. I can't make any balance claims, but if it turns out he's too weak, the fixes are all in the balance tuning. They don't have to do with the concept of giving us a simpler moveset. His current moveset can easily be tuned to be way overpowered, if desired. My point is I'm all for having a Slayer I can understand again.

Other characters follow that same trend of being simpler, yet having a little something new and useful. Venom can hold down the button on his ball summon to automatically teleport. Meawhile Chipp's 5 teleports were simplified down to 4 versions without losing much of anything, and he gets a new air super. Sol gets a new divekick. Ky actually works like he should finally, because he can easily YRC his fireballs. He's also the most boring character, so it makes sense he has a new mechanic of placing floating energy things that cause his fireballs to do exciting things when they pass through them. Potemkin doesn't have to worry about ANY of the stuff I mentioned in the Accent Core section, but he has a new air move where he fires himself like a missile (helpful to get close to a keep-away opponent), a new pillar-of-fire move, and a new dashing normal attack. Despite these 3 new moves, he's simpler to control and understand on a basic level than before.

Things I Don't Know About

Danger Time

Sometimes when you clash your attack with the opponent's, the screen flashes red, a countdown is shown, and Danger Time activates. During this time, any hits you land are extra powerful counterhits. I guess it's exciting? It doesn't really need to exist though. If it were deleted, no one would even notice. Maybe it pulls its weight as a feature if it ever generates any hype moments. I would lean toward getting rid of it, but I'm not actively against it.


It's called "Blitz Shield," but it's a parry so I will call it that. This is a surprising feature to exist. I mentioned the other parry called Slashback that was a new feature in Accent Core that I said should be removed. Instead, it's revamped here. I'm actually not sure what I think of it at this point.

The surprising thing is that someone finally copied the right kind of parry. The parry in Soul Calibur 1 is, in my opinion, the best parry in fighting games. It puts the victim in a vulnerable state during which they cannot block, but they can still do their OWN parry during that time. In SC1, it's a very great feature. If you parry someone, you then have the great advantage of threatening a mixup that's basically "attack right away OR wait a moment then attack OR wait 'two moments' and hope the victim tries to counter-parry." If you can make the opponent whiff their counter-parry attempt, you can then hit back with something guaranteed. This mixup is in your favor, and you deserve that for landing the parry in the first place, but there's a potential for a good mind-game here. There's also the potential for a really hype back-and-forth where you parry each other several times in rapid succession.

For some reason, no other fighting game has copied this, not even more recent Soul Caliburs! Soul Calibur 5, an overall great game, really flubbed this one. Instead of the great SC1 parry system, it implemented two parry systems: one that's super difficult and one that's super easy. The really difficult one requires very, very precise timing and the specifics of it are fiddly and wonky. I was not even able to make any use of this feature. The other parry is way easy but requires a huge chunk of your super meter. So much of it that there is no chance for the back-and-forth thing to happen here.

Anyway, the new Guilty Gear Xrd copied that original SC1 thing, the good thing. Is it good in Xrd though? It's hard to say because it's such a radically different game. You don't really stand there and strike each other the way you do in Soul Calibur, so the back-and-forth thing probably won't actually happen in practice, even though it theoretically could.

Another aspect of this type of parry is that Its existence in Soul Calibur helps slow characters like Astaroth be more viable because no matter how much they are overwhelmed by fast attacks, they can still attempt to parry. Meanwhile, Guilty Gear has a history of doing a great job with universal defensive maneuvers which then allow the developers to get away with crazily varied types of offense for each character. So in that tradition, this Soul Calibur 1 feature makes sense. Maybe it's one more universal defensive feature that helps the crazier types of offense possible to put in the game.

I haven't had time to really wrap my mind around this feature in real matches yet. I will reserve judgment on if it helps or hurts the game overall.

Option Selects Are Bad

An option select is when a certain input has more than one possible outcome, and the computer picks the most favorable one for you. I wouldn't say every option select is bad to exist, but most of them are. There's a fallacious belief amongst some players that it "makes a game deeper" to have option selects, but that just isn't true. Playing rock, paper, scissors by being able to input something that counts as rock and paper—or worse yet that counts as rock AND paper AND scissors—is not deeper. It's degenerate and removes strategy.

I'll tell you two option selects that are bad to exist in Xrd.

First, the throw option select. Throwing with one button is good, rather than bad, because it's simple and it works. The "whiff" you get if you tried to throw but the opponent wasn't throwable is your f+heavy attack, which is generally so big of an attack that you're going to get dragon punched or supered or counter-thrown or whatever else. That's exactly what should happen to you.

But in Xrd, you can input f+K+HS to throw and you get f+HS (a throw) if the opponent is throwable and K (a kick) if the opponent isn't. For Sol, that kick is one of the fastest startup moves in the entire game and recovers very quickly too. So there's really no reason not to use this option select. It's totally stupid to exist and changes the "real" command to throw from f+HS to f+K+HS. That makes the command to throw more complicated than it has any reason to be and creates the awkward situation that the explicit stated command by the developer for throw (f+HS) is WRONG to do. As in, you are playing badly if you do it. How embarrassing from a design standpoint.

If this were simply an oversight, we could forgive it and hope it's fixed. After all, this option select isn't a consequence of some necessary system. It does nothing useful except create this throw exploit. But here's the real problem: IT'S IN THE TUTORIAL. The same makers who boldly reworked the entire roman cancel system to have YRCs and slow time to make the game easier to play have also gunked up the throw command, they know they did, and they put it in a tutorial to tell you all about it. F- grade on that. It's really upsetting and disappointing.

(And of course I know that previous Guilty Gears had this option select, but to not fix it and put it in the tutorial is extra damning.)

Next, there's an option with YRCs and Bursts. When you combo someone, they can use a Burst to break out of the combo. Usually, the Burst will hit you (the attacker) so that the person you were comboing now gets some momentum going and a chance to go on the offensive.

Now imagine if you have more than 25% meter but less than 50% meter. In this case, you should always input the YRC command during each hit of your combo. If you do your combo correctly, the opponent will be in hitstun the entire time and no YRC will ever come out. It will just be a bunch of useless inputs on your part. But if the opponent Bursts, then they will not be in hitstun, your YRC will happen, then you'll be able to block the Burst rather than get hit by it. This technique won't work if you have 50% meter or more because you'll get red roman cancel in that case, which will waste your meter and change your combo.

This option select is very narrow. It only even does anything when you have specific amounts of super meter and in those cases, helps but not a crazy amount. It's just that there's no drawback to doing it in those cases, so you "should." And that's annoying. It would be nice if this were fixed up somehow, I think making Burst startup cause a red RC would do it. I should emphasis that even though I do not approve of it, I think it will come up about 0 times in the next 100 games I play, while the ability to YRC will come up constantly in every one. So overall YRCs are very great even when factoring in this unfortunate problem.

EDIT: A lot of option select-apologists are upset and have aggressively argued that junking up the game's inputs is good, usually because it "gives you an advantage to use." That is, of course, irrelevant to whether it's a good design idea. Here's an in-depth explanation of that.


Guilty Gear Xrd is a fighting game released in 2014 that uses input delay rather than roll-back code of GGPO. In other words, it's incompetent in its implementation of netcode. There is just no other way to say it. This is a solved problem so it's baffling that this is even happening. The developers must simply be unaware that input delay-style netcode is vastly inferior while GGPO is readily available to them, is a fantastic method for handling online play, and is in several shipped games such as Skullgirls and Street Fighter HD Remix.

If anyone out there has any way of contacting the developers, let them know that you demand GGPO's roll-back style netcode rather than input delay.


Ok so the netcode is bad and there's some minor option selects. The core gameplay is the most solid foundation for what is possibly the best fighting game series around though. And with graphics that are a new milestone, untouchable by anything else in the genre so far. Overall a great game, and I would kill for Arc System Works to make a Fantasy Strike fighting game.


Sirlin on Game Design, Ep5: Now Playing

We discuss the Street Fighter 5 trailer as well as the games we're playing now: Super Mario 3D World, Smash Bros. 4, Transistor, and Counter-Strike: Global Offensive. Sirlin gives an award for worst menus in a game while Aphotix gushes about a shooter.

Hosts: David Sirlin, Matt "Aphotix" DeMasi, and Sean "MrGPhantome" Washington.

Sirlin on Game Design, Ep4: Tournament Rules

In this episode, we discuss the kind of things that go wrong when creating rules for a tournament. The rules for high profile tournaments are sometimes surprisingly bad, yet even when we avoid the obvious problems, it's still very difficult to make completely fair rules. Sirlin explains the rules he chose for his own games as well.

Special guest Garcia1000 and Sirlin discuss the curious case of Nintendo.

Hosts: David Sirlin & Matt "Aphotix" DeMasi

Yomi Christmas Shipping

The Yomi kickstarter is shipping to backers right now. All shipping will be completed in the next few days, which means it's my third on-time kickstarter out of three. Yay!

If you missed out on the kickstarter (or if you participated, but want to order even more Yomi stuff), you can order now from Right now you can get Yomi Round 1 (4 decks from the base set), Round 2 (4 expansion decks), the EX Powerup expansion, the minimalist Panda vs. G.Panda 2-deck set, and a full set of foil Aces. 12 more decks will come out two at a time over the course of several months. The first release of those will be Menelker and Persephone for PAX East 2015.

Yomi products you order now will ship by December 10th. If you want Yomi before Christmas, it's best to put in that order now. If you order by December 11th, we can confidentially say that it will get to you by December 24th in the US. If you order after that, it might not make it, though you'll have a better chance if you're on the East Coast because that's where it will ship from. Go ahead and order now. ;) If you have any questions about shipping, please contact

If you're in the US and prefer to order from Amazon, you can also do that. Keep in mind that it will come from the same place and get to you just as fast either way. Here are the Amazon links:

Yomi Round 1
Yomi Round 2
Yomi EX Powerup expansion
Yomi 2-deck set of Panda vs. G.Panda

The Panda vs. G.Panda set is a limited edition, and has a really minimal graphic design style inspired by Pandante. Will it be a collector's item?

Also don't forget the online versions of Yomi. You can play online at and also on iPhone or iPad. We recently did a huge update to the iOS version, which you can read about here.

Pandante and Puzzle Strike are also great gifts for the holiday season!

Sirlin on Game Design, Ep3: Uneven Playfields

In this episode, we discuss uneven playfields in competitive games. That's when a competitive game gives some players a material advantage, rather than being fair. Although fairness should be a basic premise of a competitive game, MOST competitive games have unfortunately become uneven playfields, and players seem to accept this. That's tragic to Sirlin and Aphotix. 

Garcia1000 explains that his friends like kingmaker and collusion in games, which are usually considered bad properties.

Hosts: David Sirlin & Matt "Aphotix" DeMasi

Interstellar and Coherence

Interstellar and Coherence are both science fiction movies. They each use science as a jumping off point to tell an unusual story. They each require our suspension of disbelief in order to think about some aspect of humanity the story is trying to illuminate.

In some ways, they’re very different though. Interstellar uses lots of expensive special effects and employs lots of expensive movie stars. Its budget was $165 million. Coherence uses no special effects and no movie stars. It was filmed in the filmmaker’s own house, the cast outnumbered the crew, and it had literally no budget. The budget is officially listed as “not applicable.” Another key difference between these films is that Interstellar is an excruciating disaster while Coherence is a gem.

Scientific Accuracy

I’m not claiming that a work of fiction even needs to be scientifically accurate. It’s possible to make a compelling, interesting story with fantasy elements, and often “sci fi” is really just fantasy with sciency sounding stuff in it. Or inspired by something from science, but not completely accurate to real science.

Interstellar tried very hard to do science right. Physicist Kip Thorne was an advisor on the film, and he had the special effects team create the visuals for the black hole using a complex set of real physics equations. Wikipedia says: “The resulting visual effect provided Thorne with new insight into the effects of gravitational lensing and accretion disks surrounding black holes, and will lead to the creation of two scientific papers, one for the astrophysics community and one for the computer graphics community.”

Coherence, on the other hand, took the (real) idea of parallel universes from quantum physics and used it as a fantasy element. It explores the idea of multiple universes in a way that’s very interesting as a story, but that doesn’t make sense as hard science.

The irony here is that Coherence is a much smarter film. Interstellar is in the uncanny valley of science. In visual arts, the term uncanny valley refers to a character that’s just human enough to look kind of realistic, yet just “wrong” enough to be creepy and scary. You can make a character look realistic or cartoony, but if it’s almost realistic, it’s really uncomfortable to look at. Interstellar is trying so incredibly hard to be legit science that its missteps are cringeworthy and difficult to watch. Coherence, by contrast, takes one leap of fantasy-inspired-by-science then follows its own rules well and has its characters think about their situation in a smarter fashion than the awful dialogue in Interstellar.

Sensible Dialogue

There’s a reason that the characters in Coherence talk to each other in a believable way: because they are actually talking to each other. They aren’t reciting lines of a script to each other because there is no exact script. Each actor was told what their character would know at that point and given a page of notes from the director such as “you really want to go outside” or “you want to tell a story about X thing.” The actors did not know what events would happen around them during each scene. Instead, they are genuinely experiencing those events and they are genuinely talking to each other about what it means and what they should do about it.

That’s an interesting experiment when it comes to acting and filmmaking, and it didn’t have the result I expected. Having the whole thing improvised like that might make it feel “more real” (and I think it does), but this would be at the cost of clever scripting and plotting. If the author can precisely control every line of dialogue, they can work in more nuances, more double meaning, more foreshadowing, and have a tighter plot in the end. Or at least I thought that would be the case, but somehow Coherence has all of that too. It’s a movie that reveals more the second time you watch it, and I think it would reveal more even the fifth time. It’s a tightly packed puzzle of a movie, yet also a freeform improv.

Interstellar has an actual script, and in that script are lines that no one involved in science would ever actually say. The dialogue from the fictional world’s leading scientist (played by Michael Caine) are kind of insulting to our intelligence, almost like the writers of The Big Bang Theory snuck in. When asked how he was making advancements in his research, he said that once a certain gravitational phenomenon happened on Earth, it made him realize that such a thing was possible. So then he was able to develop new equations and better understanding and so on. But that he hadn’t yet “solved gravity.” Once he does, he says, then he’ll be able to do the fantastical things with it that the characters need. When asked how close he is to “solving gravity,” he says very close, and promises that he’ll solve it by the time one character returns from a trip.

This whole thing is so face-palm-inducing. Imagine if a new chemical compound was discovered, and you want me to synthesize more of it. I tell you that I’m working on “solving chemistry” to do that. Then before I “solve chemistry,” I’m able to give you some kind of progress bar like “chemistry 85% solved!” and a projection that it will be solved in one month. The problem is that I’m not able to know how close I am to discovering something I haven’t yet discovered. I’m not able to tell you when I will discover it. These are the wrong kinds of questions to even ask. Furthermore, the binary notion of gravity (or chemistry) solved / not solved is simplistic and childish.

Kip Thorne may have lent credibility to visual depiction of black holes and so on, but on a much more basic level, he really should have been consulted about how smart people attempt to solve a problem, how they would talk about solving a problem, and what solving a problem even really means.


The rest of this post contains spoilers for Interstellar, followed by another section of spoilers for Coherence. If you haven’t seen either movie, I’d recommend reading the spoilers for Interstellar, because seriously whatever, but not for Coherence. You can watch Coherence on Amazon Instant Video, here.

Yes I know that Michael Caine’s entire plan to “solve gravity” was intentionally a hoax. That doesn’t excuse anything I said above though. He went to absurdly great lengths to make everything about his actions and research as believable as possible. In fact, too great of lengths. Apparently, the entire building NASA is housed in he designed to be a giant centrifuge capable of being launched into space to function as a space station. Even though he knew that this was all a hoax and that the building would never be used that way, he seems to have actually made it really capable of that, and it ends up working correctly after his death. Yet he talks to other scientists with a childish, elementary school concept of what it means to do research.

That said, I think his whole hoax is interesting, story-wise. All his life’s work is a lie, and intentionally so because he knew that humans needed to be lied to in order to get them to do the work needed to save the human race from extinction. In the specific scenario his world is in, I think he’s probably right. It was probably morally correct of him to do so, and at the very least, it’s arguably defensible as such, and that’s interesting to think about.

And while I’m saying some good things, it’s cute that John Lithgow told Michael Caine that Murph makes the teachers at school look like fools, and if she joins NASA then they’ll make Michael Caine look like a fool. At the moment he said that, Murph was just a child, but she grows up to do exactly that. And for more foreshadowing, John Lithgow also told Matthew McConaughey (his stepson) that Matthew had an incredible skill and then the world changed and he never got to use that, and he’s truly sorry it worked out that way. Yet Matthew McConaughey did get to use his piloting skills and engineering knowledge, and he saved the entire world because of it.

Badly Written, Badly Edited

Let’s talk more about how badly written and edited Interstellar is.

That whole thing about how Michael Caine’s work is all a lie was revealed to us when he was on his deathbed in the hospital. The music told us that this was important moment and so we should listen carefully. I did listen carefully. I strained. Michael Caine mumbled some incomprehensible words for this big reveal. I wasn’t sure if he said what I thought he did. Murph (incidentally, that’s a stupid name) reacted as if she heard it though, and that it was a big deal. I pieced together what was going on, but couldn’t they have shot a second take of this? Or given me some subtitles? I just don’t understand why such an important moment is delivered with mumbly slurred speech.

Instant Friends and Instant Life Decisions

Matthew McConaughey goes from meeting NASA and being afraid they are going to kill his daughter to suiting up for a mission into space that will have him leave Earth forever in the space day. At least I think it’s one day, I was unable to tell due the poor pacing and editing during this part of the movie. Did he really become chummy and start palling around with the crew THAT quickly? Like the same day or something or the very next day?

So Like, When Someone Buys Too Many Scratchy Lotteries?

Matthew McConaughey then leaves Earth in the old-style Saturn V-looking rocket, headed for an outpost near Saturn. Yet later they have spacecrafts capable of easily launching off planets into orbit, so why didn’t they use these here? Anyway, the reason they’re going to Saturn is to access the wormhole that appeared there 40 years earlier and that has been there ever since. Through the wormhole is another galaxy containing new worlds we could potentially colonize. Some are near a super massive black hole (SMBH) called Gargantua, and this has very important time-dilation consequences that are actually a real thing in science.

Ok fine, but here’s the thing. We’re really supposed to believe that the first time Matthew McConaughey even heard about this black hole or heard the word “Gargantua” was after they had already reached Saturn and were about to go into the wormhole? I actually said “what?” out loud at this. It’s just really poor writing to present it that way, because surely he would have been briefed on wtf they are even doing way, way before that. I also understand we need an audience-surrogate so the film can explain things to us, but using Matthew McConaughey this way compounds the bad-writing problem I just mentioned. When the other scientists explain what’s going on to him and he says lines like “oh so it’s like looking around with a periscope on a submarine,” it’s almost like he’s intended to play the same role as Ice-T on Law & Order: Special Victim’s unit.

Did You Hear That Music Too?

At one point, Matthew McConaughey watches a video of his son who says that this is the last video he’ll be sending. The music is very loud during this, really awkwardly and noticeably loud. The exact instant the son’s video ends, the music suddenly ends too. Was the music intended to be IN the son’s video?? It was edited as if it was, but that doesn’t make sense? Would the son really add incredibly loud dramatic music to his own webcam video? Did someone from NASA do it for him? Are we intended to think the music was in our movie, but not in his video? Why is this edited so poorly that I am asking myself these questions instead of paying attention to the movie?

The loud music thing happens a lot in this movie. It seems like they’re trying to contrast sounds with silence, which was an effective technique in the movie Gravity (sidenote: go see that movie instead of this one), but is so over-the-top here as to pull me out of the experience. Several times I thought to myself “why is the music or background soundscape so loud and distracting right now? Is something wrong with the theater’s speakers? No, because all the rest of the dialog and sound effects are the proper volume, so really it’s intended to be this distracting right now? Shouldn’t I be thinking about the situation going on in the movie instead? Is this a first pass edit rather than the theatrical release? Hmm, but I’m in a theater.”

Tidal Forces and Lack of Foresight

The situation with the watery planet makes no sense. First, a minor point where the science might make no sense, then a huge, glaring plot-destroying problem.

The water planet is in a gravity well of a super massive black hole such that one hour of time passing there is equivalent to 7 years passing on Earth. Time dilation from differences in gravity is a real phenomenon, and even measurable in fractions of a second per year when real astronauts go from the low gravity of an orbit to the standard gravity on the Earth’s surface. Ok so far.

If gravity is that intense though, it seems like it would be pretty hard to blast off from the surface of that planet to escape the high gravity region. It already makes no sense that this journey began with an old-style rocket when later on they had spacecraft that could easily take off a planet surface, but now the craft can escape the mega-massive gravity here easily?

It’s possible I just don’t understand enough of the real physics to explain that. I also thought that they’d just be crushed by the gravity, and that later on when Matthew McConaughey crosses the event horizon, he’d be spaghettified, as Stephen Hawking called it. “Tidal force” is a term that refers to the difference in the pull of gravity on one part of something and another part. It’s what makes our tides go in and out. Here’s a video about that:

The difference in gravitational force on your head is different from the force on your feet because your feet are closer to the center of the Earth. This differential, called a tidal force, would be so large when nearing the event horizon of a black hole that you’d be stretched like spaghetti and destroyed. That said, this isn’t true for all black holes. Super Massive Black Holes (SMBH) actually have less density, not more density, than their less-massive counterparts. They can be so large and so diffuse that a person on the surface of the Earth and one at the event horizon of a 10 million solar mass black hole experience about the same tidal force between their head and feet. So Matthew McConaughey might have been able to cross the event horizon without being spaghettified and maybe, somehow, this explains the gravity on the water planet being normal enough that you’re not crushed just by being there. I still question whether it explains the spacecraft being able to escape this super intense gravity well though.

We’ll let that one go. But not this next one. The characters see that the explorer on the water planet was already dead, then realize she must have been dead for only a few minutes from her perspective. She was only on the planet long enough to send the data from it once, not over and over across a period of many years.

Think about that. Before the characters reached this planet, they knew all about the time dilation effects of going there. When analyzing the data they got from the planet, they knew about this time dilation situation. That means they knew how many packets of annual data they expected to receive: one packet. What they actually received was many packets which would indicate years of data sending. They specifically mention this because it was an indicator that the explorer was still alive because she had been sending data all this time.

So a) they would have known that they only had the first year of data, not other year’s worth, b) they would not have expected to receive lots and lots of years of data, c) they DID receive years and years of data, which they should have realized makes no sense at all. The script hand-waves past this by saying the discrepancy between “data was sent once” and “data was sent many times” is explained because the data was caught in some sort of “gravity echo” or something and it was just the same signal repeating itself over and over. Mega face palm, and insulting to the audience. It’s a stupid non-explanation and even it were real, it still wouldn’t explain why anyone expected to gets years of data in the first place. This whole thing just doesn’t work, plot-wise.

The Data-Revealing Robot

Matt Damon’s robot. Matt Damon tells the other characters that he took his robot apart to get to the fuel cell so he could continue with his mission. When the crew’s robot offered to fix his robot, he discouraged this and said it needed “a human touch.” We later learned he meant that literally, that it needed the thumbprint authorization from a human to activate the decommissioned robot.

I don’t really get any of that, but I’ll give it a pass anyway. Here’s the thing: what was Matt Damon’s actual plan with sabotaging this robot? I get the general idea that the robot contained data showing that Matt Damon was lying about the status of the planet, so apparently he rigged the robot to self-destruct in a big explosion if anyone tried to access that data. When did he do this? It seems impossible for him to have had the time after he was revived, so I guess he did this a long time ago before he went into hibernation.

Ok but, then why did the robot immediately display the most damning data when a human did put their thumb print on it? I assume he programmed it not to explode if he used his own thumb, but if it’s someone else’s thumb, then he set it to automatically show them the exact data he needs to hide, THEN explode? This is really sloppy writing. The writers want to let us know that the data is in there, and have the drama of a character discovering this, but it makes no sense for Matt Damon to have set it up that way.

Incidentally, I really like that Matt Damon’s character lied. Michael Caine (who also lied about his entire mission) told us that the 12 explorers were the bravest humans who ever lived and that Matt Damon’s character, named Dr. Mann, was the greatest of them all. But then Mann turned out to actually be a coward who would do anything to save his own skin, at any expense. (Mann is mankind, get it?)

Way, Way, Way Too Many 0s and 1s

Behind the bookcases. So inside the black hole, is a crazy 5th dimensional space that far future humans constructed so that Matthew McConaughey could talk to his daughter. Uh, well, ok. There are just several awkward things here.

Matthew McConaughey first uses the bookcases to see himself in the past before he went on the mission, and he tried to tell himself to STAY, as in not to go on the mission. He fails though, and his earlier self does go. Then he realizes that he can use this whole setup as a way to communicate information to his daughter Murph. It’s awkward that he must communicate quantum data from a black hole though. Probably like terabytes of data in who-knows-what-form, converted to just 0s and 1s? Does this take like 9000 years? How does she even make sense of that? I feel like if he had sent her the 0s and 1s for the computer code behind Microsoft Word, that she wouldn’t have been able to understand that and create Microsoft Word any more than she could make heads or tails of “quantum data.” Why wasn’t the message something more human, like an inspiration for her to research the right kinds of things? Or an equation, at most. I guess because it had to be something they discovered *on* the mission, which means data from the black hole. That’s a tough writing problem, and the writers took the way out that required the least thought and delivered the most disappointment to the viewer.

Love Conquers All

Let’s jump back to an earlier part of the movie involving Love Conquering All, then return to the bookcases. Matthew McConaughey, Astrophysics Black Guy, and Anne Hathaway must choose whether they visit Mann’s planet, or Edmunds’s planet, but they can’t choose both with the fuel they have left. Literally the fate of all humanity rests on this decision. Matthew McConaughey calls out Anne Hathaway as being biased because she’s in love with Edmunds. She admits that this is true, but argues it’s still the correct decision. Both planets sent good data, and the one she wants send slightly better data. The drawback is that they stopped getting communication from Edmunds at some point, while Mann has still been sending data. Maybe Edmunds’s communication equipment broke, or maybe he died, but either way that planet looks a bit better. Furthermore, love is powerful and worth something and might give us access to mystical information that we can’t otherwise divine. Something deep is telling her she’s right to choose this planet, not Mann’s, and that should be considered too.

I actually think that’s an interesting argument. It reminds me of my work in balancing games when people argue and claim opposite things. Sometimes people on one side bet everything on some sort of data or calculation, saying that proves their stance is more solid. The problem comes when the other side is relying on intuition or factors they can’t verbalize, when the first group is *neglecting* subtle factors that weren’t part of the calculation. Matthew McConaughey decides to “go with the data” (by valuing Mann’s data more because he’s still sending data each year) and yet it turns out to be the wrong choice in the end. Maybe Anne Hathaway just got lucky with the right answer, or maybe love really did tell her something. Thematically, I like that.

Hamfisted Love Conquers All

Love Conquers All again back at the bookcases, but this time it’s much more hamfisted. McConaughey’s dialogue about this is just corny. When confronted with a bizarre, surreal bookcase-world that crosses boundaries of time and space, he immediately claims he can navigate correctly because of his love for his daughter. Come on now. I’m fine with “love conquers all” but not when expressed in such a hokey way.

The story from Murph’s perspective is just as silly. From her perspective, she has gone her entire life without even considering that the “ghost” she remembers from her childhood was actually her father from the future. So she returns (as an adult) to the place where the ghost told her things, but can’t seem to get anything from it. Then she suddenly realizes it’s her father. Again, I actually said “what?” out loud. She had no clues and no information at all what was going on, or even that anything was going on. She suddenly and magically just knows.

I understand the movie’s intention here. The idea is that gravity is one of the only things that can cross from a 5 dimensional space to our world and potentially be used to send messages. (That actually has some real basis in physics, btw.) And the implication here is that perhaps *love* is able to do this as well. Perhaps it also can cross the boundaries that we thought were uncrossable. Ok, that’s thematically appropriate if you’re doing a Love Conquers All story and it need not be scientifically accurate. But it should be shown in an understandable and at least emotionally believable way. Here she is staring at the same bookcase she had known for 20 years, but now, for no apparent reason, has exactly the correct revelation about her father. This part of the movie really should have shown her noticing some clue that helped her connect the dots.

Carelessness With the Most Important Object in the World

Matthew McConaughey uses the wrist watch he gave Murph to transmit the quantum information. Then soon after, Murph talks to her angry brother. This guy is pissed off, told her to leave forever. Ten seconds ago het got back from putting out a fire in his fields that she started on purpose. Meanwhile, she holds up the watch—THE WATCH. It’s literally the most valuable object in the entire world and the only thing that can save the entire human race from dying. It is LITERALLY that. She waves it around in front of the emotionally unstable and angry brother as she tells him some nonsense about the ghost of her father sending info to her in the watch. I actually cringed at this, expecting him to grab it from her and smash the watch. Then she tries to hug him and I expected him to punch her lights out and somehow ruin the watch. This isn’t even intentional dramatic tension though, just case #127 of bad writing in this movie. Case #128 is when the brother’s reaction to all this is to passively hug the sister instead of saying “wtf are you even talking about?”

Time Paradoxes

Matthew McConaughey claims (suddenly and randomly, by the way) that the 5 dimensional bookcase zone was built by far future humans, not by aliens. How did he know this? I have no idea. I didn’t see any indication of this, but I guess he just knew.

Ok, so when did they build it? In the far future. So let’s think that through. After the end of the movie, humanity was in a position to continue on. Great! So they do, and then way in the future, maybe like millions of years or more, humans “evolve into 5th dimensional beings.” Let’s not even question that. Ok so the 5th dimensional being thing happens, and they are all hanging out. Maybe at a 5th dimensional Starbucks, drinking Pumpkin Spice Pure Energy. One of them says, “hey, remember in the history books—well I know we don’t have a concept of books or even of history since we can now experience all time as a physical dimension, but you know what I mean. Anyway remember how the humanity was going to be destroyed way long ago? The only reason that didn’t happen was that at some point we got around to building a 5th dimensional bookcase thing for this one guy. Should we get that going? Or just wait a while on it?”

This is never even addressed in the movie, but it’s just really head-scratchy. Like what the hell is the timeline even supposed to be here?

The Great and Needless Heist

At the end, Matthew McConaughey goes to visit Anne Hathaway’s planet to join her. This is thematically very appropriate because his character represents the human spirit of exploration and it’s exactly what he’d do. Come to think of it, I guess he embodies the American spirit of pioneering and exploration, and he ends up in a new world with an American flag in it. Very American.

Anyway, in order to make this happen in the plot, he STEALS a spacecraft to go to her. He is living in a world foreign to him that’s 100+ years more advanced than what he’s familiar with. Does he really know how their docking system and aircraft hangar system work? Will he be shot down? What weapons does the space station have? Can they track him? Wouldn’t there be some kind of mission control room with radar and alarms that go off if he tried to do this? But the actual way it’s conveyed to us that anyone knows the spacecraft was stolen is that a guy with a clipboard wanders by the part of the hangar where that spacecraft was supposed to be and notices it’s missing. Sloppy writing #935.

So Wait, What's Happening Now?

And let me end all of this madness by going back to an earlier moment of confusing editing and unfortunate writing. Just before Matthew McConaughey enters the black hole to get to the bookcase area, he and Anne Hathaway start their journey back to Earth. We are told they need to get rid of some weight to do this, so they get rid of the robot. Matthew McConaughey then suddenly reveals his plan to leave himself behind too, so that Anne Hathwaway can get back home.

The moment of this reveal is so poorly edited that I thought, “wait, I thought he was on the same ship as her? They are on different ships actually? And wait what is going on here, he’s suddenly sacrificing himself?” I later asked a friend about this exact moment and she also though they were on the same ship until he revealed that he was leaving himself behind.

This is not some minor plot point though. Matthew McConaughey’s character is driven, even obsessed, with saving his daughter and returning to his family. He wants to save the world, but he uses the drive to save his family and see them again to push himself to keep going. Every action he’s taken up to this point has reinforced that and this is the FIRST moment that he abandoned his goal of seeing his family again. In his mind, he is sacrificing himself and he’s about to die because of it. And yet this is kind of sprung on us with not enough time to really process it or to even understand that he could have done that because it’s unclear where he is relative to Anne Hathaway.

Matt Damon’s character dies suddenly too, but that’s fine and on purpose, while the hasty treatment of the main character’s ultimate sacrifice just feels like an oversight.

This post is long, but that’s just because the movie is long. Long and filled with badly executed story. The best thing I can say about Interstellar is that if it were presented in the form of a one page pitch, it would probably sound really compelling. It’s all in the execution.


You should watch Coherence first before reading this.

Whose story is it?

There are eight characters in this movie, and it might be unclear at first whose story it really is. It’s Emily’s story. If you rewatch the movie, this is even more clear. We start with Emily, and we end with Emily. What you might not have noticed though, is that we follow Emily and only Emily throughout the various alternate realities. Other characters are replaced with alternate reality versions of themselves and we visit alternate reality versions of the house, but it’s the same Emily all the way through—the story tracks her.

Everyone is in the Wrong House

Near the beginning, all the characters go outside, then come back in and there’s a broken vase, which someone says they don’t remember breaking. That’s a clever trick because at this point, we don’t know that going outside into the dark zone sends you through a roulette of universes. EVERYONE is now in a different house than they started in, before we even knew that was a thing. It doesn’t matter plot-wise, but it’s just kind of funny.

Random Events Are Markers for Each Universe

Hugh and Amir visit “the other house,” but a different Hugh and a different Amir return to our house. They don’t realize they’re out of place for a while, but the different color glow sticks tip them off. They have red glowsticks, but the box of red glow sticks in this house has never been opened. When they realize this, they leave and take the physics book and box of pictures with them because they figure they should take all the important information for themselves in an effort to return to their “right” reality.

The glow sticks, the band-aid being cloth vs regular, and the phones that have cracked screens vs don’t are all indicators of which universe is which. The idea is that infinitely many things could go this way or that way in quantum physics, each spawning a new reality, but these specific things are easy for the audience and characters to see and understand.

What’s smart is that the characters take things into their own hands by creating a marker for their house. They realize that it should be something based on randomness so that it will come out different in all the other houses. I think they cheat a little bit on the plot here, because the characters’ stated goal was an address for the HOUSE, but they actually created mini-addresses for each character too. I would think they’d roll several dice to create a house-address, but not an individual die for each person. We’ll let that go though; it turns out to be important to the story that we can tell which individual people are from which reality.

Who is Who?

Mike’s wife complains that Mike is drinking too much and says Mike is “not the man I married.” He literally isn’t because she’s now with a Mike from a different universe than hers.

Why Did Someone Break the Car Window?

Mike tells Kevin that he went to the other house to blackmail himself. He says he considered breaking the window of Hugh’s car in the other reality so that he could steal the physics book and prevent the others from having the conversation about other realities. But he says he didn’t actually do it. He didn’t, but later we see that another Mike from another reality DID do that.

Or did he?

When Emily returns to the house, the Mike from that house is trying to figure out why anyone would smash Hugh’s window when the book clearly wasn’t in there. Anyone could see through the car window that the book had already been removed, so there’s no reason to smash the window anymore. Someone else says maybe getting the book wasn’t the motive. Could it be some other reason, maybe intimidation?

By the end, we know why. Emily does it as a diversion to lure everyone outside in order to kill or incapacitate the other Emily.

Like A Ping Pong Paddle? Exactly.

Amir explains the idea of the quick codes, that putting a random object in the box helps create a better “address” code for the house. It could be a candle or a book, says. Or an oven mitt. Then Emily says “or a ping pong paddle.” Amir off-handedly says “yeah or a ping pong paddle, exactly.” Emily says, emphatically “exactly….”

What the characters and audience don’t know there is that Amir didn’t say oven mitt because that random object occurred to him at that moment. He said it because in his universe, the box he opened earlier had an oven mitt in it. Emily doesn’t know that, so she says “ping pong paddle,” as in “oh I get it, we opened a box with a ping pong paddle so that’s the quick code for that box.” Amir doesn’t realize that though and he thinks she’s naming a random object. She’s confused why saying “exactly” doesn’t get a better response from the other characters, but she doesn’t yet realize she’s in the oven mitt house, not the ping pong paddle house.

The Dark Versions of Us

Near the end Mike says, “This whole night we’ve been worrying there’s some dark version of us out there somewhere. What if we’re the dark version?”

This turns out to be true. Let’s look at the whole story now from Emily’s perspective.

Early on, Emily tells us how she created a dance production, but then she didn’t get to star in it. Katherine Meriss ended up in that part, became famous, and “stole her life.” Note that the reason Emily didn’t get the part is that when she was offered the understudy role (that would later become the real role), she took too long to say yes, so the yes became a no.

Emily’s husband Kevin will go on a business trip to Vietnam for four months. He wants Emily to come, but she’s hesitant. He says ok how about coming for just one month rather than the full four? She says she’s not saying no and that she’ll think about it. This is the same mistake: he says that doing nothing turns into a no.

Later in the evening, Emily takes more control of her life and her relationship with Kevin. When Kevin is going to go outside without her, she becomes flustered and demands to go with him. Even later, Kevin wants to go out without her again, and she physically blocks his way and very forcefully says that they need to stay together.

Eventually, Emily figures out that various people are from various different realities. Specifically, she and Mike are visitors in this reality, meaning Mike’s wife isn’t the wife he knows (and her husband isn’t the husband she knows either). Emily believes they must leave the house and take their chances with the reality roulette in order to get back to their own realities. Mike disagrees though. He says he’s made certain choices in life and is stuck with them. He’s stuck here.

This brings us to, “what if we’re the dark versions of ourselves?” Everything goes wrong, people are bickering and arguing and Emily sees it as a nightmare to be stuck in this reality. She takes charge and chooses to go shopping for a new, better reality. It’s implied that the clock is ticking too: when the comet passes, she’ll be stuck wherever she is.

The problem is that the happy reality she stumbles on has HERSELF already in it. There’s a line of dialogue in this “good house” about wondering if Katherine Meriss will understudy Emily’s life when she’s gone for a month. This tells us that in this reality, Emily didn’t wait too long to get the dancing part: she really got it and became famous. And she didn’t wait too long to decide to go to Vietnam with her husband either. These are the choices she “should have made.”

Will she kill in order to get that better life? The movie asks us an even deeper philosophical question here: will she kill HERSELF to get a better life? Is it ok to do so? Is it ethical or not? Does it even matter what happens to anyone, even “yourself” if they are in a different universe? It doesn’t seem ethically bad to stumble into these various realities, but by doing so you’re actually preventing other copies of yourself from getting to the good ones, so is it really even that much different if you force one of yourselves out of a good one? And if you kill yourself, is that even detrimental to you given that you the killer perceive that you live on anyway, and with no consequence?

I won’t try to answer those, but I will point out something Hugh said early on. He thought that his other selves would want to be left alone, so he thought it was best to leave all the other universes alone. Meanwhile, Mike thought it was best to murder his other self before his other self murdered him. Because there are so many Hughs and so many Mikes out there, this means that most Hughs are safe and most Mikes are not. It’s like karma made tangible.

In the end Emily is willing to kill to get the life she wanted. Even though her journey was one of growing confidence and taking charge of her life, we did see the dark version of her.