Mandible Games

Phoenix Wright: Justice for All

Developer: Capcom

Completion level: Finished game

Spoilers: Plotline may be spoiled. Sorry. Finish the game first.

Many many years ago, there was a developer named Sierra, who made adventure games.

You played a character (and oh boy, some of them were Characters) who wandered throughout a world, usually a strange, bizarre, twisted world, generally with some goal in mind. (Not always with a goal in mind.) You collected random items as you went and jammed them in your inventory. There were puzzles. You solved puzzles, frequently using your inventory, the "plot" continued, and the games were well-received and quite enjoyed at the time.

In retrospect, most of the old Sierra games were terrible.

I don't mean the graphics weren't up to our current standards, because obviously they weren't, we're talking really old games. I mean the gameplay was atrocious. The games penalized you for exploring (by dying), they penalized you for logical deduction (by dying), they penalized you for taking a reasonable approach to the problems (by dying), and even if you somehow managed to pass all the hurdles and read the developer's mind you would still frequently end up in a spot where you couldn't possibly finish the game . . . with no way of knowing that you were stuck. And when I say "read the developer's mind", I really do mean "read the developer's mind" – puzzles were byzantine at best, and at worst they were an exercise in surrealism that has rarely been matched since.

Adventure games got more and more complicated, increasingly weird and unsolvable, and nobody realized it. Hell, I didn't realize it at the time – I loved those games, and it's only looking back on them that I realize how much sheer frustration and guesswork went into playing them. People stopped buying them, they died a grisly protracted death, and considering that Sierra was responsible for much of the genre at that point, I place most of that responsibility squarely on Sierra. Adventure games became entertainment non grata in the industry, and roleplaying games sort of awkwardly shuffled into the niche that adventure games had previously filled.

Many many many many many years ago, there was a developer called Infocom. They made "interactive fiction" games. You had an inventory, you solved puzzles, the puzzles got increasingly complicated and byzantine over time . . . you see where this is going?

Infocom doesn't exist anymore.

Yeah, I bet you did see where that was going.

Phoenix Wright is a game about a defense attorney.

Each game is broken up into four or five independent cases. At the beginning of the game, someone is murdered. Someone else will be accused of murder. You defend that person.

The gameplay consists of two segments, which often repeat several times within a case. Occasionally, you'll be in court, picking holes in the witnesses' testimonies, using court evidence and their own words to ferret out the truth. This is amazingly fun. The developers did a wonderful job of making it suspenseful, through music, dialogue, and fabulous art. Alternatively, you might have to inspect the crime scene, interview witnesses, interview people who you can call to the witness stand, etc etc. This part isn't quite as fun, for me at least, but it's still damn entertaining and it makes the first part all the better.

"I've never held any sort of weapon. I've never even touched one!" "OBJECTION! Why are your fingerprints on this sword, then?" "Where . . . where did you get that? That . . . it must be a mistake!" "From a broken locker . . . behind your car. With your fingerprints on the lock!" "Nooooooo!"

Phoenix Wright puts a lot of work into ensuring that you can't get yourself stuck. For example, there's no "I'm done, go to court!" button – if there's stuff left to discover, then you keep wandering around until there isn't. If there isn't, you go to court immediately. The same philosophy works its way into the entire game. If you've discovered all you can from a witness, the cross-examination ends. If you haven't, it doesn't. At all points, you know you have what you need to finish the next segment, because if you didn't have it, you wouldn't be here.

The end result of this is that, generally, it's obvious what you're supposed to do. Either you need to wander around the game world a bit more and look for more clues, or you need to find a contradiction in what the witness is saying right now. The upside to this is that it pushes you along in the game at a reasonably nice clip. The downside is that the game becomes rather linear, which exacerbated by the occasional "false choice" – you're given a choice, yet all the choices lead to the same path. Still, the writing is skillful enough that you usually don't notice these unless you're watching for them or replaying the game (and, let's be honest here, these games have zero replayability.)

The game almost pulls everything off flawlessly, and if I was writing about the first game in the series, I'd say it did – because the first game did. I ran into some trouble in this game, and it worries me.

Basically, the cases are getting more complicated.

There's more stuff going on. There's more surrealism. The puzzles aren't byzantine yet . . . but they're sort of nudging around the edges of it. They're considering it. If I was a history major you'd be getting a cute historical joke involving "not being byzantine yet", but I'm not, so just pretend there's one here.

This creates some issues with the linear Phoenix Wright gameplay – namely, that you can occasionally logic things out better than Phoenix did, and you get penalized for it. And sometimes, even though you know exactly what you want to say, you can't figure out how to say it within the confines of the game.

I'm going to spoil the hell out of the third case here, so, y'know, consider yourself warned.

The third case takes place in a circus. The ringmaster was found dead, the magician is a suspect, you're defending the magician, blah blah blah. The real criminal is the acrobat, and at a late point in the game you've figured out that he had both motive and opportunity, but you're still pinning down the details on how it happened.

Well, I wasn't pinning down the details. I'd figured it out. His pet monkey helped him. (This is not abnormal in a Phoenix Wright game.) So when the judge asked if the acrobat had an accomplice . . . well, yeah, he did. It was the monkey. Duh.

But you're not supposed to realize this at that point in the game. Despite being right, that was the wrong answer. I was not conforming to the exact pattern they wanted, and the game penalized me for it, and I had to work gradually through the guesswork they wanted me to guess at . . . eventually coming to the conclusion that, hey, the monkey helped him. The entire process was extraordinarily difficult, as it's very hard to figure out what they want you to say when, in fact, you know the right answer but aren't supposed to.

In a game like this it is vital to playtest thoroughly – ridiculously thoroughly – so you can see where people get stuck, and where people think too much and come up with an answer they're not yet supposed to have, and figure out how to design the game so that neither of those are a problem. And this is really really hard, especially when you're trying to make a game which is essentially linear.

There's two more games in the series that I haven't played yet (okay, the most recent one is Ace Apollo, not Phoenix Wright, but it's still the same series) and at least one spinoff being produced. It is entirely possible that they've recognized and fixed the problem by then.

But it's also possible they haven't. And this worries me, quite a bit. We're finally rejuvenating the old adventure game genre, after Infocom damaged it and Sierra did its best to finish the genre off. It's a good genre. There's a lot of fun to be had, there's a lot of entertainment, and I don't want to see it gone . . . but it's also a genre that's very easy to do badly, and very hard to do well, and painfully hard to tell the difference.

Still, I'm looking forward to the next game. We'll see.

I am, of course, a perfect programmer. You can tell because I never make mistakes. Since I'm obviously perfect, this release is clearly not a bugfix for an incredibly stupid and important mistake, but rather I have decided that implementing one or two very minor features clearly justifies a new release, for excellent reasons that nobody else could hope to comprehend.

Therefore, this release includes a few things:

• A little codebase cleanup
• The EMP weapon's glory device resistance has been increased significantly, and I've added a visual effect to show it
• Small UI improvements, including a note to emphasize the glory device resistance a little further and some improvements to gamepad setup
• I've heard that some people might have found that the game just didn't run. Obviously this is impossible since I never make mistakes but, you know, if you did have an issue running the game . . . you, uh, might want to download this version.

Time for another release!

Version 0.1.4.0 is now here, with two new exciting features.

First: Linux support! I've done my best at getting a functional Linux package together, and I think I have one. This ought to plug straight into a vanilla Ubuntu installation. If you want to install it on any other distribution . . . well, hopefully you can figure it out. Is there a better way to package it? Maybe. If so, please let me know, since I'm really kind of in the dark here. You will need some kind of 3d acceleration support, just to warn you – software rendering is, shall we say, slow.

Second: Editor support! I've included the level editor I use for all the levels and much of the 3d geometry. It's undocumented and largely unsupported, and there are many issues with it (such as the lack of copy/paste) but it does work, and it is usable. If you want to make your own levels, or edit the existing levels, now you can. I'm certain there are ways you can make invalid levels that will cause the game to crash, but I'm OK with that because invalid levels should really be spitting out useful error messages instead of just crashing.

As for the game itself, it's mostly only been changed with a few small bugfixes and usability fixes. Finally, though, I'm moving onto actually improving the game itself, so with luck it'll get more exciting soon.

As a side note, I'd love to make an OSX version. However, I don't have a Mac. If anyone has an ancient OSX-capable Mac system that they'd be willing to donate or loan, I promise I'll get an OSX build working just as soon as I have a platform I can run it on :)

Download the Windows version and the Linux version, and if you're getting the Linux version, please tell me if it works!

Patapon

Developer: Pyramid

Completion level: Finished game, not 100%

Spoilers: I am not going to spoil the plotline. I will be spoiling the gameplay mechanics. If you're planning to play the game, however, you may want these spoiled for you.

I swear, it took me a week to figure out what I wanted to say here.

I keep notes on games as I play them, y'see. Anything that annoys me, anything that impresses me, any thoughts I have, it all goes into the notes. Eventually I finish the game, and I write up a dissection based on my notes.

Patapon has more notes than every single dissection you've seen so far put together – as well as two you haven't. To say that I am divided on this game would be an understatement.

So let's start at the beginning.

Patapon is a sidescrolling rhythm game. You control a bunch of little mobile eyeballs with weapons named Patapons, and you "control" them in a moderately indirect manner that takes the form of a rhythm game. You have a set of "commands", and if you punch in the commands with the right rhythm, your little eyeballs do things.

Once you finish a level – whose goals are virtually always either "get to the end of the level" or "kill a boss" – you are returned to the Patapon Village, where you can play various minigames, buy and upgrade Patapons, and go out to a new level.

That's the game.

First off, the game is pretty – I mean, look at that picture up there, that's almost exactly what the game looks like. You fight giant enemies, ten times the height of any of your warrior eyeballs, weapons visibly stick in them as you fight, the animation is brilliant, etc etc etc I don't really have a lot to say about the graphics besides "yum". It's worth buying just for the awesome visuals.

Besides that, though – Patapon has issues. Big, humongous issues. And it took me days to figure out why.

First off, your units do not have a vast repertoire of abilities. They have, for example, "move right", and "attack". They can also "defend", "run away", and "charge up the next attack". You'll never use "charge". You may notice this gives you four useful abilities . . . in the entire game . . . and you would be exactly right. You will be doing those four things over and over again. There's one more ability – "magic" – but to be honest you'll use that one perhaps twice in the entire game. You see, it ends Fever Mode, and that's something you never want to have happen.

So there's the first problem – there's no variety. Fundamentally you just don't do many things in the game, and you do them over and over again.

The next problem is Fever Mode.

Patapon is, as I mentioned, a rhythm game. Each of the abovementioned "attacks" is four drum beats, which you press in rhythm. If you get the rhythm right, consistently, then your Patapons eventually enter "Fever Mode" and become useful.

Yeah, read that again. If they're not in Fever Mode they are basically useless. Archer units fire three times as many arrows in Fever Mode and I think each of them does more damage. Mounted units gain the only ability that makes them worth bringing along. Units run away faster and go further – without Fever Mode they kind of run away, a bit, and then get stomped by the huge range of the enemy you're currently fighting. Their defense gets stronger, their speed goes up, everything your units do is vastly improved, with the end result that your performance is directly correlated to how long you can maintain Fever Mode.

And Fever Mode is a fractious, unruly beast-queen. The manual is unclear on how it starts and ends. Sometimes you'll enter it after a mere three commands, sometimes it will take ten. Sometimes you'll have no trouble staying in it for long periods of time. Sometimes it will end for no obvious reason, even when the commands seem to have been input correctly. A frustratingly large amount of the time it will end on the very next command after you enter it. The timing that you need to push buttons is extremely tight, and there's no visual or auditory clue as to whether you're too early or too late. There is an auditory clue as to how close to the beat you are, but it's subtle and if you start concentrating on it you're almost certain to miss your timing a little bit – which sort of defeats the point of concentrating on it. On top of that, the mechanics involved with Fever Mode are byzantine and complicated, and never explained anywhere. More than once, especially at the beginning of the game, you'll be fighting a boss, and you'll think "oh, maybe I will beat him this time!" and then you'll drop out of Fever Mode randomly and get slaughtered.

Yes, there are bosses that will one-shot your entire army.

Unless you're in Fever Mode, of course.

And to cap things off, there's the Patapon Village. You can buy new Patapons, but apparently randomly you'll get a different kind of patapon – maybe one with bunny ears that can't use armor? Maybe one that looks like a hedgehog! Or, hey, this one has angel wings. Unless you're extremely observant you're just not going to figure out what causes different Patapon types until you – like me – go and check a walkthrough.

It took me about a week to figure out what the underlying cause to all of this annoyance was.

There's a concept I've heard of which is occasionally called an "expert interface". The idea is that it's an interface designed explicitly for experts to use it – not for novices. A lot of professional 3d software has this sort of interface – it has a grueling, brutal learning curve, but once you learn it you're able to work incredibly fast – far faster than you would be able to work with a "novice interface". Often these interfaces include many byzantine and inexplicable key combinations, and every aspect of them is chosen for speed of work rather than intuitiveness.

Patapon is an expert game.

The game isn't designed for newbies. It isn't designed for casual gamers. It's designed for people who are willing to sit down and absolutely master the interface, and it's designed to still give them a good gameplay experience once they do so. Experts don't need to be told whether they were just a little too fast or a little too slow on a button-push – they just know. Experts know that dropping Fever Mode is probably death, and they just won't drop it. Experts will understand the nooks and crannies of the interface and, honestly, probably won't even notice them.

Patapon does a really good job of being an expert game.

Once you figure it out – which takes quite a while, admittedly – it has amazing flow. Yes, there's only four things you'll realistically be doing, but it like you're coordinating the movement of all your little suicidal patapons rather than simply giving them orders. Enemy ahead! Pata pata pata pon! Attack! Pon pon pata pon! Keep attacking! Pon pon pata pon! Dodge, pon pata pon pata! Pata pata pata pon! Pon pon pata pon! Chaka chaka pata pon, pon pon pata pon, pata pata pata pon, pon pon chaka chaka, pon pon pata pon!

(Don do-don do-don.)

And that's when the game shines – when you're no longer fighting with Fever Mode, when you're not trying to decipher what the hell a "Mofeel" is and where it came from, when you're just assaulting these ridiculously gigantic and fantastic monsters with your army of little eyeballs.

It's worth talking about Expert Games a little more, because I expect that this is going to come up again. I don't think the people who made Patapon intentionally made an expert game, because if you intentionally make an expert game, you generally think to include a good detailed tutorial.

Accidentally making an Expert Game is unfortunately easy. It's a common trap to fall into in game design. The game is yours, therefore you know everything about it. The mechanics are clear to you (since you know them all by heart) and therefore you see no problem with learning them. You can make a game which is fun, balanced, and polished, and then release it to the world and . . . nobody can figure out how to play it.

This is, incidentally, sometimes I've tangled with constantly in Devastation Net. Devastation Net is an expert game. You're meant to get to the point where you fundamentally know the weaponry, and where you fundamentally know the abilities of tanks, and that is when the strategy takes place. Partially I'm trying to solve this by making all of the game balance numbers available to you, and in your face – move the cursor over a tank, you instantly see how tough it is and how fast it is. Choose a weapon and you should quickly see how it works. Partially, though, I'm having trouble with the learning curve, because teaching people things is hard, especially when it's an uncommon game style.

Unfortunately it's really not something game designers have much experience with in multiplayer games. Generally, the way you teach the game to someone is you lead the player through a single-player campaign that unlocks things one step at a time. That just doesn't work with a multiplayer game.

I'm still trying to find a good solution, to be honest.

I've rambled on long enough at this point.

Patapon is a beautiful game. It is also a fun game, once you get past the initial learning curve. Don't be afraid to check a walkthrough on this one – read everything except the mission descriptions, and you'll be thankful.

New version of D-Net.

I have to say, first off, that this is going to be one of the least interesting releases ever. There's basically nothing changed that you'll notice. What there is, though, is quite important.

First off, I rewrote the build system entirely, in scons instead of make. This is entirely meaningless to the end-user, but it means that modifying the build system in the future will be much easier. This is a good thing. It makes Zorba less stressed.

Second, I added a small intro screen explaining what exactly you can expect from the game. If you're reading this, you probably already know what to expect, but the average dude coming in off the metaphorical street probably won't. So while that's not exciting either, it's important.

Third – and most importantly – I finally have a good crash reporting system. If the game crashes, it will beg permission to report the crash dump to my servers, and if given permission, give me information on what went wrong. I actually have no idea if it's been crashing for people – with luck it hasnt been – but the reason I don't know is because, fundamentally, nobody ever reports crashes, they just roll their eyes, say "oh, indie developers" and delete the game.

Which is not ideal from my point of view.

There's a lot of subtlety in a crash reporting system. For example, I've got mine rigged up so I can return messages, based on the game version, before the potentially large crash dump is sent. So if I start getting flooded with crashes that I've fixed, I can tell people to go download the new improved version, please, and stop bothering me with things I already know. (Perhaps not in those words.) Not only that but I can also return messages after the crash dump is sent, so I can analyze it server-side and return things like "your graphics card sucks" or "your RAM is bad". And it transmits the data in a compressed form to save on bandwidth, and it records as much information as I can without violating the privacy of my users.

Oh, yeah, that too. I don't violate user privacy. I don't send a single byte before the user has said "yes, I am fine with sending my debug log to Mandible." I've carefully checked the debug file to make sure I don't include anything that's identifiable (and removed a few debug printouts that went a bit too far, in fact) – the worst I can do is tell you what kind of graphics hardware you have, and what kind of joysticks you have plugged in. Which is obviously kind of important for debugging. I do not feel bad for including this.

Any kind of reporting of this type, of course, has to be agonizing careful about privacy issues. And, inevitably, someone is going to get annoyed at what I'm doing.

At least this way, I can tell them to just click "no".

I am a firm believer that every one of the current consoles has something to recommend it.

The Wii, of course, has a novel control method and a good number of interesting family games built around that method.

The XBox 360 has XBox Live Arcade, an increasingly solid game lineup, and the industry's best multiplayer.

The PS3 has a Blu-Ray player and, as far as I am concerned, four games. Those games are: Ratchet and Clank, God of War 3, Flow, and Everyday Shooter.

However. God of War 3 isn't out yet, Flow is a small game at best, and Everyday Shooter is not only a small game but is now available on Steam as well.

I am being completely honest when I say that Everyday Shooter is one of the reasons I was looking forward to buying a Playstation 3. And now it's on my computer, and I have one less justification to buy that Blu-Ray player with a few games attached to it.

I swear, every time I have a reason to get excited about the Playstation 3, Sony does their absolute best to nullify it as quickly as possible. I have no idea what's going on with Sony right now, but this is one of those cases where they should have offered an extra chunk of money just to get exclusive rights. Of course, if I were Jonathan Mak, I wouldn't have taken it . . . so there we have it.

Yes, there will be a dissection, and yes, I am hard at work on another version of D-Net. Right now I'm off to play Everyday Shooter though.

Quickly on the heels of that alpha test is 0.1.0 official release. Grab it (nowhere, sorry, removed!).

What's different? Not a lot. There's an installer, and that's it. I haven't gotten any bug reports, so I'm calling this a release.

But there's a new site feature – I've added a site forum for discussions. Take a look, start a thread, talk about D-Net. It's still under very heavy construction, so beware that the layout and look might change drastically as you browse. But it's officially open. As always, I'd love to hear any commentary you have about D-Net or the site, and if you can't find an appropriate thread to post it in, head for the forum.

This is probably not the most exciting post you've ever read, but there'll be more coming. Not to worry.

This has taken far longer than I hoped.

I've told you about the interface issues. Those were problematical. Once I finally figured out how gamepads should work, I had to figure out how keyboards should work, and that was another issue. Pretty much every interface element has had to be redesigned at least twice as I gradually understood what I needed better.

This behemoth is finally in a good state to be released.

Download Devastation Net, version 0.1.0 Release Candidate 2

(Note: download removed since it was technically not complying with some licenses.)

Now, just to warn you: this is not a final release. It's not even an alpha release. It's an alpha release candidate, and you'll notice it's Release Candidate 2. There's no installer either – you'll have to just decompress it to a directory yourself. (Installer is one of the next steps.)

I'm releasing this, probably a day or two at most before the actual alpha release, only to people who read this dev journal. I'm hoping people will download it and give it a try. If it crashes – and it's entirely possible it will – I'm hoping to fix problems before the official alpha release.

But this is, of sorts, a release. It's playable. There's AI you can play against, and there's enough support for multiplayer on a single computer (which, incidentally, I highly recommend.) The AI is terrible, and I know it's terrible, but it should at least give a sense for the game. Some of the interface is unfinished, still, but it's generally good enough to figure out what you should be doing.

Let me know in comments, or in emails, if it works or not (and, for that matter, how well it works.) Your feedback is very appreciated on this one.

As for why this has taken a while:

My unofficial goal is to post here weekly, at least. But, as always, problems crop up and things get delayed, and unexpected opportunities arise. In this case, the unexpected opportunity was this thing:

That's a small part of the Babbage Difference Engine. You might think it looks pretty cool, but you'd actually be wrong – it's far cooler than that. I had the opportunity to watch a good deal of the setup and tuning process, as well as stay out of the way of the people working on it, and honestly even staying out of their way was quite an honor. If you're in the Bay Area, I recommend coming and taking a look at it once it's officially open – although needless to say, the exhibit launch is likely to be packed beyond all comprehension.

Hopefully the delay is understandable.

Super Smash Brothers Brawl

Developer: Nintendo

Completion level: Beat Subspace Emissary

This is going to be an extraordinarily short one.

I wasn't even sure I would write on this topic for a while. What do you say about SSBB? It's got standard Mario-style graphics (I touched on this in the Mario Galaxy dissection), it's been painstakingly balanced to a knife-edge, and it's hugely popular. I was originally just planning to note that, yes, I played it, and really didn't have anything to say about it. They don't do anything particularly notable extraordinarily right, and they certainly don't do anything wrong. So there we have it.

But there's one thing I decided I wanted to say.

SSBB has a single-player mode called the Subspace Emissary. In this mode, Mario and Co team up to defeat the Bad Guys. Most of the missions include a rendered cutscene at the beginning and the end, showing the interactions between the characters and the inevitable action-movie-esque "Hey! You really are on our side!" moments.

The cutscenes are fantastic.

They're funny. They're entertaining. They're beautiful. They do a phenomenal job of setting the stage without ever actually interfering with the player's enjoyment of the game. There's no Final Fantasy "okay, go get a snack, you're going to be here for twenty minutes" moments. There isn't a single cutscene that becomes boring. They're just all excellent.

And they're done almost entirely without dialog.

I think Snake says something when he shows up for the first time. That's all, though.

The interactions are shown with body language – and considering that we're talking body language between an anthropomorphic fox, a mobile pink marshmallow, a monkey, and a lot of robots, this is a nontrivial task. These aren't even simple interactions. There's betrayal, there are turncoats, there are characters whose motives are unclear and contradictory in the beginning . . . and all of it is explained by the end. Nonverbally. It's really incredibly impressive, and if you don't plan to play the game, I actually recommend watching them. 1 2 3 4 5 6. Yeah, that's about an hour of video. If you don't want to watch that much, at least check out the Donkey Kong/Fox arc – 6:00 to 7:30, 1:50 to 3:40, and 7:55 to the end.

You'll notice that this clip is in Japanese. You'll also note that it doesn't matter. The only thing you miss is what the names of the various characters are. Even with those short Japanese clips, you get a reasonable idea of their personalities and interactions.

That's damn impressive.

I think, in my next game, I'm going to try hard to make the game's plotline understandable with dialog removed. I do plan to have dialog – but I'll add the dialog after the game makes sense without it.

And that's really all I have to say on this game.

Yeah. I just wrote a page about Super Smash Brother Brawl, focused entirely on the cutscenes in Subspace Emissary mode. Deal with it.

Super Mario Galaxy

Developer: Nintendo

Completion level: 100% completed

This is NOT a spoiler-free review. In fact, in the next line I plan to spoil the entire plot of the game.

Okay. So there's this princess, right? And she gets kidnapped by a giant dinosaur named Bowser. I know you're shocked by this. I was too. But luckily, help is on the way! Some guy named Mario – who is apparently a plumber – rescues her.

That's the plot.

It may sound familiar to you.

At the moment, there are, as I would count them, three major lines of Mario games. First, and best-known, are the Mario sidescrollers, starting with Super Mario Brothers and continuing up through Super Paper Mario. Second are the roleplaying games, which I believe started with Super Mario RPG and somewhat branched with the Paper Mario series and Superstar Saga series. And last, there's the 3D exploration games, including Mario 64, Mario Sunshine, and Mario Galaxy.

You'll notice a bit of confusion – I'm calling Super Paper Mario a sidescroller, but I'm also mentioning how the Paper Mario series is a roleplaying game. Super Paper Mario is an experimental intersection of sidescroller and roleplaying game. Nintendo has never been one to keep its games locked in tightly-defined precise boxes – Nintendo's built around fun. They make games which are fun, and if a certain convention gets in the way of making the game they want, the convention gets thrown away.

For example: lush, spectacular graphics. Mario games don't have those.

Their graphics certainly aren't bad. The art is always good, and it's always reasonably high-end by the standards of the console. But it's designed to be effective. it's not designed to be spectacular. it's not designed to be flashy. it's designed to convey how the world is constructed, and hold to a theme, and be consistent. All of which it succeeds at, quite nicely, but nobody will ever say "Oh man, did you see the latest Mario game? I didn't know video games could look like that!"

We all knew video games could look like that. We saw it in the last Mario game. This one just has more triangles.

Plot is another thing that, with the exception of the roleplaying line, Mario games just don't do. There's a princess. She gets kidnapped by Bowser. Mario defeats Bowser. Everyone lives happily ever after, inevitably including Bowser, who, don't worry, will try again next game. I don't even want to think about how many times Peach has been kidnapped – I suspect she's playing along with it at this point. Nothing else could possibly explain it. (At some point I should write about how Mario isn't based in story, it's based in myth. This is not that entry.)

So. "3D Exploration Game"? What's that?

The Mario 64 series has a gameplay style which I honestly can't say I've seen in any other game ever. Your goal ("save the princess") is governed by a very simple game mechanic: a series of things you must collect. In Mario 64 it was stars. In Mario Galaxy, well, it's stars. In Mario Sunshine you had to defeat Shadow Marios, and the way you got to them was by collecting "shines" . . . which look exactly like stars.

The game inevitably consists of a number of major areas – from seven up to around fifteen – and each one contains a number of stars, generally from six to eight. On top of that there's some number of minor areas that include one or two stars each. In order to unlock a new major area, you collect a bunch of stars. In order to unlock a new minor area, you collect a bunch of stars.

You can probably see a theme here.

There are some variations. Mario Galaxy divides its "galaxies" up into six groups, and to unlock the next groups you have to collect the single Grand Star. There are green stars, and red stars, and comets, and star bits, and hungry Lumas who eat star bits and explode forming into new galaxies which you can travel to and, surprise surprise, get a star. But fundamentally, the game comes down to:

1. Collect a star.
2. Can you fight the end boss? If so, go do it.
3. Can you fight a midboss? If so, go do it.
4. Return to step 1.

And this is one of the series's greatest strengths. There's never any question on what you should do next. You should go find another star. The game is carefully balanced so that, once you get past the first few stars, you always have several options on where to go next. If you get stuck on one particular zone, or decide that you don't really want to dodge fireballs today and you'd rather go play with flowers or space stations, you can always take a break and try another area.

It's a brilliantly simple game mechanic, but unfortunately I think Mario Galaxy missed one of the things that made Mario 64 great.

As I mentioned, most areas contain multiple stars. But in Mario 64, you can frequently pick up the "wrong star". Maybe you're meant to go ice skating, but instead you explore in the wrong direction and end up on top of the mountain. There's a star on top of the mountain, but despite the fact that the game said "Time for ice skating!" you haven't done any ice skating. That's okay. We can work with this. You can grab the star, and head back into the zone, and it'll say "Congratulations! You found the MOUNTAIN CLIMBER star! Your next star is: ICE SKATING." And then, when it would have normally said "Let's go climb a mountain!", it just skips that one – after all, you already found the mountain star – and sends you off to fight a Yeti, or collect a ton of coins, or race a penguin or something. Many of which you could have done instead of ice skating or mountain climbing.

In Mario 64, exploration is heavily rewarded. Each zone has several stars you can get at any point, and while you're encouraged to get the "next one", there's absolutely nothing forcing you. In Mario Galaxy, this is no longer the case. The vast majority of the time, only one star even exists in an area at a time. There are a small number of hidden stars – precisely one per area – but that hidden star only exists if you choose the right "non-hidden" star to go after. If you choose ICE SKATING you can get to MOUNTAIN CLIMBER also. If you choose YETI SLAYING you're going to go slay that yeti, or fail, and there simply aren't any other options. Have fun, good luck.

And that's sort of sad. In Mario 64 I felt like I could just wander wherever I felt like. Some feature of the landscape look interesting? Chances are good there's a star there. Find a wall that looks challenging to climb, but still possible? Probably a star at the top. Whereas Mario Galaxy, once you choose what star to retrieve, is an annoyingly linear game. The exploration is gone, and for a game that balances right on the edge of having a glorious sense of wonder about it, Mario Galaxy stops just short of what I was hoping for.

(As I'm imagining the game I wanted Galaxy to be, I get much of the same feeling as I did with much of Aquaria – the feeling that there's a small universe out there just waiting for me to find it. I didn't get that feeling at all in Galaxy.)

There's two other things I want to mention, but they're both pretty short.

Like most Mario games, the developers have decided to spend the time to write quite a large number of minigames and game mechanics that only show up once or twice. There's a section with you balancing on top of a giant sphere, for example. There's a racing section with riding stingrays on top of a (completely awesome-looking) floating water course. There's a quite neat segment with "spotlights" that cause matter to exist – if you jump in an area without a spotlight, you fall endlessly to your doom. In fact, there are exactly two sections of each of these. Despite all the trouble that these are to implement, the game designers saw fit to only use the code twice, in the entire game.

Partially this is annoying. Racing is fun. I want to do more of it! But on the other hand, it also neatly prevents burnout. I'm sure everyone reading this has seen a game which was fun at the beginning, but frustrating at the end. (Puzzle Quest is my most recent example of this.) Mario Galaxy doesn't do that. It doesn't even come close.

"Leave the audience wanting more." There's no other way to say it, and there's no better way to get the player excited to try out your next game. After all, maybe there'll be more stingray racing!

And finally, I noticed some neat subtlety with the music. The music actually changes depending on what you're doing – some tracks fade in, some fade out, and the music rapidly morphs to emphasize whatever you're currently doing. This is an amazingly powerful and beautiful technique, and I do not know why more companies don't do this. Nintendo's been doing this ever since the Super Nintendo (go jump on Yoshi in Super Mario World, the music is slightly different if you're riding Yoshi) and it's something I never see in other games. For a simple example, enter any Bowser boss fight – if you didn't notice this when playing, I recommend checking it out. It's worth it.

Summary.

Good game. Polished to a phenomenal level. Effort and money was spent on gameplay gameplay gameplay – not spectacular special effects, not a riveting plot, but on making sure the game was fun in every possible way. Game lacks the exploration sense that Mario 64 had, which I feel is a loss, but gets just about everything else right.

There's a reason Nintendo is doing well right now, and Mario Galaxy is a great example of it.