Fire Emblem: Shadow Dragon and the Blade of Light

Me and Fire Emblem are not buds.

Strike one: these games are tactical RPGs. I don’t hate TRPGs, exactly, but I’ve liked and completed far fewer than I’ve gotten bored with and quit. I love Final Fantasy Tactics, as any right-thinking person, but it doesn’t crack my list of top ten Final Fantasy games.

Strike two: these games are the most anime-ist of anime. Again, I don’t hate anime, exactly, but I find a specimen has to be something truly spectacular to wake me up. You certainly can’t do it by spooning a serving out of the Big Barrel of Anime Tropes, and that’s exactly the barrel Fire Emblem swims in.

Strike three: this is predominantly a handheld series, and I don’t play games on handhelds. There was a time back in the early ’10s where my PSP and 3DS were all I had to game on at the office, but I was never wanting for something to play on them to the point where I would take a chance on Fire Emblem.

Strike four: this is a vast series, with something like sixteen mainline entries, with intertwining timelines and character arcs and geography and who knows what else. The first, uh, six (?) of which didn’t even get released in my region. Jumping into the middle of a long-standing series and trying to play catch-up with the story and characters is exhausting, and only worth it if the pay-off is a setting you’ll cherish for life.

But.

I’m an NES kid, and I have great fondness for that particular era of gaming. Having grown up in the ’90s cutting my teeth on stuff like Ultima: Exodus and Destiny of an Emperor, I’m insulated to a lot of the weird jankiness of that bygone era. I can pick up an 8-bit console game released in 1989 and appreciate it as an historical artifact, but I can also see how a younger me might have responded to its merits and flaws.

My first exposure to Fire Emblem, as I expect is often the case in North America, was Marth’s inclusion in Super Smash Bros. Melee. I had a vague awareness that he was from some Japanese game we never got, and some light sleuthing at the time revealed his game was a lot older than I’d expected. (His companion character Roy was from the then-newest title, which hadn’t been released as of Melee. Including the newest Fire Emblem protagonist would become a sort of Smash tradition, leading to the proliferation of “anime swords” in the game today.)

My experience with TRPGs at that point had been, well, Final Fantasy Tactics and a couple other titles here and there, mostly on the Sega Genesis. A friend and I had played most of the way through Warsong, and I’d rented Shining Force II once. These were complex games that seemed to be at the far edge of what the 16-bit consoles were capable of, so I was intrigued as to how an 8-bit progenitor could even work.

The solution: romhacks. Lots of old Famicom and Super Famicom JRPGs had terrible-to-middling fan translations at the time, done by hackers who thought they were being grown-up and character-accurate by making the heroes say “bitch” and “ass”. I’d already played Final Fantasy V, Mother, and some of Seiken Densetsu III that way. But the romhackers hadn’t gotten around to Fire Emblem yet. It was 2001, nobody knew what a Marth was.

I told that story on stream a couple years ago, stating the only Fire Emblem game I’d had any interest in playing was the original, and the one time I’d looked, nobody had translated it yet. Of course, by the end of the night, my viewers had located and patched a rom for me. I booted it up to confirm, yep, it’s in English alright, and decided to save it for a rainy day.

Then I forgot about it.

Sorry, Princess Whoever! Must have slipped my mind.

Then a thing happened the ring did not intend: Nintendo released Fire Emblem: Shadow Dragon and the Blade of Light for the Switch. Not a remake of the game, you understand. There is a remake, on the DS I think, but that’s not what they released. No, they were giving us the Famicom original, with a fresh new professional localization. And it wasn’t being dumped into their NES ghetto app; Blade of Light was a stand-alone title with some modern bells-and-whistles added in. You could experience the game in all it’s 8-bit jank and mean-ness, or you could smooth your playthrough with new fast-forward and rewind options not present in the original.

I bought the game immediately, then I forgot about it again. For a while. Then, this past couple weeks, I played through it twice.

You’ll Eat Your Perma-death and You’ll Like It, Mister

I’ve never really needed a fifth reason to stay away from the Fire Emblem games, but here’s one anyway: most of the early games in the series features perma-death as a mechanic. When a character dies in battle, they’re dead for good, and this can happen through no fault of yours if the dice rolls just happen to fall a particular way. The choice here seems to be whether a player embraces this system and rolls with the punches, or meticulously replays levels until all units survive. Both of these options seemed pretty stressful to me.

If the former, you’re going to inevitably lose a powerful fighter you’ve dumped a lot of resources into. This feels bad because you lose a character you’re attached to, yes, but also because those resources are just gone now, and you have to complete the rest of the game without them. The more you rely on a unit, the greater the chances this is going to happen, probably multiple times.

If the latter, you’re going to be replaying content repeatedly until you get the outcome you like. There are already individual moments in lots of JRPGs where I do this sort of thing, and it’s never fun. I tolerate it because it’s just one thread in the broader tapestry of experience, but in Fire Emblem the tapestry only has this one thread. It sounded exhausting.

Neither of these options feel right to me in a game that’s easily going to take twenty or more hours to complete. If Fire Emblem were a quick five- or six-hour jaunt, it’d be acceptable to play through it multiple times as you experience slightly different outcomes on each pass. But these games are heavy and dense. I’m not a man who plays a heavy, dense game and then immediately hits reset to play again.

I knew Blade of Light had new features to mitigate this, somewhat, but I was also very interested to experience the game as closely as I could to how it would have played in 1990. I resolved to not use the turn rewind feature, and to simply write off any dead units. I had two big questions:

1) Is the game even winnable if too many units die?
2) Do dead units really detract from the experience that much?

Before I started playing, I reasoned the answer to #1 was “it must be, somehow” and the answer to #2 was “probably not”. And it turns out, I was wrong on both counts… but that might be okay.

The game has 25 maps and something like 50 total recruitable units, 15 of which can be assigned to a given map. In some maps you’ll be assigning fewer, because they introduce new units, which count against your total, even if they’re terrible and you’ll never use them again. On my first run I was losing about one unit per map, and sometimes an unlucky two, so it wasn’t likely I’d ever be completely unable to field a full roster of 15. The game dumps more units on you than you can use specifically to patch up the holes left behind by the fallen.

Each unit is a named character with a hard-coded class, though, which means you have a finite number of units to fill a given role in combat. There are only so many heavy horse in the game, and they tend to form your army’s vanguard. As they drop one by one through the tough mid-game maps, the new horsemen you’re meant to replace your old ones with are likely not as strong as the ones you’ve lost, nor as well equipped. So, the more you lose, the more you tend to lose, and some of the late game maps are particularly brutal.

The situation I found myself in was reaching map 21 (of 25), a large open field with no healing tiles, choke points, or obstacles, which sent enemies at you in three waves and then spawned a fourth.

So like is there a rock I can hide behind, or…?

I believe the design intent of this map is a sort of knowledge checkpoint: you have to really know what each unit is capable of, and bait out each wave of enemies in turn, just on the strength of your army alone, without any assistance from the terrain. By the time I reached this map, I was really down to the dregs. I think I could have won the map, though I’d have suffered heavy losses doing so, and the most brutal maps are still to come. So, I reasoned that was about as far as I could reasonably go on that playthrough. I still wanted to see the game through, but I felt like I’d had my fill of the oldschool hardcore experience. I reset the game and allowed myself to use the turn reset feature to savescum any dead units back to life.

What happened, though, was that each map went much smoother on my second pass. I was able to use what I’d learned on my first attempt to really put some stank on my second run. Having prior knowledge of map layouts and reinforcements allowed me to position my units better and earlier, and I had a much better grasp on equipment types, movement ranges, and secondary objectives. I’d lost a lot of units on my first play learning lessones I was able to apply on my second, and when I did inevitably lose a unit to an unforseeable crit, I decided to just replay the map instead of reset the turn. Just as I imagine I would have back in 1990.

I arrived back at map 21, the open field that had defeated me previously, with a much stronger army of more balanced classes. I was able to use my flying units to bait out the opposition’s wyvern knights and bring them down with well-positioned archers. I used my horsemen to tank their line of generals, then their paladins, while my mages attacked from the back rank. By the time their reinforcements rolled up in the form of heavy ballista, I’d advanced Marth and my heroes deep enough into enemy territory, armed with lightning swords, that I was able to stop them in their tracks.

This felt good. I mean, very good. I felt like I was applying a level of proficiency that isn’t often demanded of me. It was a kind of clairvoyance, to be sure; the game was easier now because I was better at it, but also because I knew what was going to happen on each map. It’s a bit like going into The Legend of Zelda and already knowing which bushes to burn. At this level play I was able to accurately predict how each enemy turn would play out before the computer played it, as though I were controlling both our units.

I enjoyed stumbling through the process of learning the game, and I enjoyed curbstomping it too. And I’ve concluded that is the intended form of play in Blade of Light. You’re supposed to play as far as you can, losing the units you lose to bad decisions (and sometimes bad luck), until you can go no further. Then you reset, start again at map 1, but you know now. This is how I used to play these kinds of games, back in 1990. I did not clear Adventure of Link or Crystalis or Faxanadu on my first try, either, in the bad old days.

I was expecting Blade of Light to be a history lesson. I wasn’t expecting it to rekindle the feeling of resetting Quest of the Avatar with a new hero because you know where some of the good hidden stuff is now. I enjoyed being caught by surprise.

I think, all told, I spent about forty hours with Blade of Light, across those two playthroughs. And I can’t advise whether the way I played would be right for you. Maybe you’ll have more fun making liberal use of turn reset to save every unit as you go. I can say the game gets a lot friendlier after you’ve learned it, though, and that’s a chance you might consider giving yourself.

Jagen is the Best Jagen

I’ve learned a little about Fire Emblem tropes these past weeks, dipping my toes ever-so-slightly into the wikis and fansites and subreddits. And I can say with some confidence: Jagen is a pretty good Jagen.

Let’s unpack that.

Jagen is one of your first units in Blade of Light. He’s available from the very start and is Marth’s stalwart supporter for the entire adventure, or at least until he gets cut down by a BS critical hit. He is stupidly strong on that first map; he’s already a paladin (which is the promoted form of your heavy horse), he has good HP and defense, and he starts with a silver lance, easily the strongest weapon your army has access to early in the game. You could concievably use Jagen to cut a bloody path through the first several maps of the game, and this is of course exactly what I did.

But ah, I was told, by viewers and well-wishers far better versed in the Fire Emblem mythos than I, I should not fall for the trap. For you see, every Fire Emblem has a Jagen, of which Jagen is merely the first. The Jagen is a character who starts out very strong, but who has little room for improvement. If you use the Jagen overmuch early in the game, he merely serves to soak up EXP for other characters who could level up and surpass him.

There are three aspects to a Fire Emblem unit you have to consider, when it comes to their stats. The first is the stats themselves, which determine what a unit is capable of. Units with higher Strength deal more damage, units with higher Speed are capable of attacking twice, units with higher Move can traverse the map more quickly. The Jagen (and therefore Jagen) is quite good in this regard, at least at first. However, each unit also has a hidden “growth rate”, which determines how quickly their stats increase as they level up, and this is where all of Jagenkind falters. Their stats go up only infrequently, and a Speed that’s excellent on the first map is only merely “good enough” on the tenth. The third consideration is whether a unit can promote, but the Jagen comes out of the gate already promoted. Units reset their experience level to one upon promotion, and the level cap is twenty. This means your un-promoted horse units have 38 potential level-ups, whereas your Jagen has only 19.

I’m here to tell you though, at least where Blade of Light is concerned, none of that really matters.

On my second successful run of the game, I still used Jagen to cut a bloody path through the first few maps, soaking up a bunch of EXP and basically just making a nuisance of himself. It ended up not mattering. There are enough enemy units in the game to spread EXP around to everyone who wants it, and while it’s true you will notice your other horse units surpassing Jagen in the late game (especially after they promote), he remains a viable unit all the way through to the end.

I don’t know whether this is true of every Jagen in every Fire Emblem entry, or not. Maybe you’ve played a bunch of these games and are just in the habit of stealing your Jagen’s silver lance and benching him at first opportunity. You can do that here, if you want, but I think all you’ll accomplish is making the early game a little harder for yourself.

See, that’s what’s fun about going tback to the first game in an established series with its own little galaxy of tropes and expectations. A lot of times, those expectations haven’t solidified yet. Maybe Jagens are bad now, because the playerbase got wise to their trap and so the developers started feeding into that by making them worse and worse over the years, but this first Jagen — Jagen Prime — is just what he is: a strong early game unit they probably expected you to use until his untimely death, at which point you’d replace him with one of the other horse you’ve been leveling up.

There are other things “missing” from this first game, if you’re a Fire Emblem veteran. Weapons have properties, but there’s no RPS “weapon wheel” where swords always necessarily beat axes, or whatever. There’s no sortie screen, no dating mechanics. Healers don’t level up by healing, and largely don’t need to. And who even knows what else.

What I’m saying is, I don’t know how you should manage your expectations if you’re go into this game after years with the 3DS and GBA sequels. All I can tell you is, it’s safe to use Jagen. Jagen is good. Jagen will get the job done.

Cutting Edge Hi-Def Graphics!

Blade of Light isn’t one of the last Famicom games, but it’s one of the later ones. 1990 was the year of Final Fantasy III, Super Mario Bros. 3, and Mega Man III. So, uh, a lot of threes, really.

I think the game is gorgeous, as a man who has a fondness for old 8-bit pixel art. The game does lean heavily on function over form, ensuring that unit types are distinct from each other and terrain is clearly conveyed. As a result everything skews towards the simplistic, especially on the map screen where you’ll be spending most of your playtime. But it also has great little flourishes, in those places where they were able to sneak them in. It probably doesn’t sound like much when I tell you Blade of Light’s mapmen have three frames of idle animation, but most of Marth’s contemporaries have only two, and the second was often merely a reflection of the first. See how much livelier Marth looks in comparison:

Left to right: Dragon Warrior IV, Ultima: Quest of the Avatar, Final Fantasy III, Fire Emblem: Shadow Dragon and the Blade of Light

Once you actually engage the enemy, the battle sprites are big and detailed. I loved Marth’s cocky flourish before delivering a critical hit, and I grinned pretty wide the first time I saw the original 8-bit version of what is now his signature run cycle in Smash Bros. There’s an option in the menu to turn off combat animations, but I never wanted to do this.

Listen, I understand not everyone is enamored with 8-bit graphics. I don’t know what to tell you. I was there, in 1990, when these graphics were cutting edge. To this day I love pixel art and pixel games. I think this art style has a character to it that modern hand-drawn high-def sprites just don’t have. I don’t want every game to look like this, but I’m very happy that some games do.

Dat Jank Doe

Part of why I was interested in playing the earliest known console TRPGis, quite frankly, I wasn’t sure how you could map everything the game needs to be able to do to just four face buttons. The answer is, well, they tried, god bless ’em.

This game has some pretty severe jank. The first major shock to my system was being dumped directly onto the first battle map from power on; there’s not even so much as a prologue scene or a sortie screen. Shops and NPC interactions are directly on the battle map, and only one unit can be on a tile at a time. This means if you have three new units who each need a silver sword, it will take at least three player phases until it’s all taken care of, as each of them waits for their turn to move to the shop tile and spend some gold. And the whole time, the enemy is closing in on you. Often you’ll be building a defensive perimeter with your already-equipped units so your new guys can get sorted, and just as often you’ll have a daisy chain of units stretching back to the start of a map as units who needed a turn or two to get organized spend the whole battle playing catch up.

Everything is mapped to the A button. Selecting a unit for movement, action, or even just to look at their equipment is all handled by highlighting them and pushing A some number of times. It’s not confusing, exactly, but it’s very easy to double-press and end up at the wrong menu depth, and get lost in the interface for a moment. I spent a lot of time hitting B to back all the way out to the main map view in order to “try again” rather than just elegantly address my units. It was also very easy to accidentally issue the wrong command, and such mistakes are very costly in this game, since it can be several minutes in between issuing commands. Selecting the wrong command, or the wrong target unit, or picking the wrong square to move to, can put you behind in action economy. Play carefully.

The flip side of that coin is, the Famicom doesn’t exactly have the chops for sophisticated enemy AI. Enemy units move very predictably, and if you know their Move score you will almost always know where they intend to move on their next turn. A lot of the “git gud” aspect of Blade of Light involves exploiting enemy AI, and there are lots of hilarious ways you can do this. Your opponent doesn’t try to build a perimeter or make use of choke points, for example, so it’s easy to divide or surround them. It’s also very easy to tell what target they are going to select; they always try to hit Marth, if they can, even if it puts them in an incredibly disadvantageous position. This means Marth (who is one of your most survivable units) can “escort” a much weaker unit around the map without much worry they’ll be targeted. I frequently sent Marth along with an archer or mage unit to complete some side objective, and the computer just never had a good answer for this.

To win each map, you must defeat the boss unit and then command Marth to take the castle. This means you can full clear a map and then just live there forever, issuing movement and commands one unit at a time, for as long as you care to listen to the cheerful music. I frequently had to do this, since there’s no other opportunity to equip your units in safety. The endgame stats roll reports I spent over 50 turns on one map, using my flying units as go-betweens to load up on equipment in the shop and painstakingly trade one sword at a time to the rest of my units, patiently waiting in a long line.

It took me a very long time to figure out what was going on with the Speed stat. Every unit can be addressed exactly once on your turn, so high Speed units don’t get more turns than low Speed units. Instead, the stat is used to influence which combatant in a skirmish gets to act twice. In a typical round the attacking unit goes first, then the defending unit offers reprisal. Then, whichever unit is faster attacks a second time. (The hip Fire Emblem kids call this “doubling”.) Sometimes, though, my enemy would attack me twice even when I was faster, which made it feel like an element of randomness was involved. There are already to-hit rolls and crit chance rolls in this game, so an extra “who gets to double” roll on top of that felt really unfair. It turns out, there’s a hidden “equipment weight” stat in the game, where the weapon you use deducts from your Speed, and it’s this modified stat that determines who gets to double. This is mentioned in the instruction manual, but isn’t indicated in the game at all. I’m sure I lost a unit or two in my first run figuring this out.

Hmm… I think this last thing counts as jank. Remember my TRPG pedigree is Final Fantasy Tactics where, on a unit’s turn, you can direct them to move and then act in either order. In Blade of Light you can move and then act or you can only act. A unit can’t attack an enemy and then move away from them, nor can a unit buy something in a shop and then step away, freeing up the space for another unit. I don’t know if this is a limitation of the times or just part of the style of the Fire Emblem series, but I know it never felt great to me, and I would have preferred the freedom to take my turn in either order.

This Game Costs $96

That’s not true, it’s $5.99 as a digital download on Switch. And I could have let it end there. But I had a stack of gift cards left over from Christmas and nothing in particular to spend them on. Instead of just whittling away at them using Doordash, I decided to make an extremely unwise purchase:

Pictured: a life decision. Also pictured: award-winning photography.

This is the 30th anniversary ultimate collector’s package, and I am embarrassed to own it. I blame a combination of having too much “free” money on my desk and whatever nostalgia endorphins were still kicking around after finishing the game. But whatever, I bought the dang thing, let’s talk about what’s inside of it.

What I imagine will be the main draw for most people is the big nice art book, filled with character art from a variety of Fire Emblem games and spin-offs. I’m sure an actual long-time fanboy would get more enjoyment out of this than I do, but it’s nice to flip through and it will look handsome on the shelf. I learned two interesting things as I was flipping through it. First, apparently the cast of Blade of Light recurs in a lot of sequels throughout the years; not only Marth, but the whole gang. And second, as the series goes on, it starts leaning harder and harder on the Uncomfortable Jailbait Waifu Index (UJWI). I’m forbidden by law to have an anime waifu, especially not one who looks 14, so probably best to get away from this series while I have the chance.

Next in the package is an incredibly cool Nintendo Power mock-up. Someone actually sat down and thought, hmm, what would the Fire Emblem issue have looked like back in 1990? Not only did they nail the cover layout perfectly, but the flip side of the mockup has a hand-to-god Counselors’ Corner and Classified Information section addressing some of the game’s hidden stuff. The mock-up declares itself to be “Volume 11.5”, so I looked it up and it turns out that would have put this “issue” smack dab in 1990, right after the Super Mario Bros. 3 feature. That’s an incredible attention to detail, and I will definitely have to frame this thing.

The actual expensive bit, though, is the NES packaging. It is exactly the right shape and size, and even smell. Inside is a full instruction manual, one of the double-tall ones like Final Fantasy and Dragon Warrior IV had. Oh, and a fold-out map. Do you have any idea how much good it did my poor crumbling heart to open up an NES box and pull out a fold-out map? I about cried. Then there’s the fake cartridge, a beautiful glass art piece that fits perfectly in the included cartridge sleeve. I have a huge custom display case in my dining room filled with NES carts, and this piece will look incredible once it takes its rightful place alongside all the old classics.

(Interestingly, when I first saw pictures of this cart, I thought it was made of shiny silver plastic. This seemed fitting to me. Imagine if we’d been getting Fire Emblem games all along, starting in 1990; doesn’t it make sense that its cartridges would be uniquely silver, as a counterpart to Nintendo’s other epic fantasy series, The Legend of Zelda, which had always been gold?)

The only part of this package’s presentation that betrays it as a modern replica is all the printed material is presented in French as well as English. This continues to weird me out a little bit on all of Nintendo’s products. I understand why they do it, of course, it’s just, living in Florida, I always expect the bilingual packaging to be English and Spanish, and it never is.

And look, I realize there’s a good chance if you’re reading this, and are active in my Discord, I have made fun of you in the past for buying overpriced collector’s editions of games filled with worthless trinkets. You might be feeling ways about things right now. If that’s the case, I just want you to know, this is totally different because [insert excuses and/or hypocricy here]. I still think amiibos are dumb.

I Will Never Play Another Fire Emblem Game

Fire Emblem: Shadow Dragon and the Blade of Light was the perfect storm. I approached it as a historical curiosity, it ended up grabbing me a lot harder than I ever intended, and then I splurged on the big fanboy box just minutes before it disappeared into the bowels of eBay for all of time.

I don’t think this heralds a new age of me devouring Fire Emblem games, though. I still think GBA games look kinda dumb and ugly, and the entries on modern consoles look to me like episodes of Super Generic Fantasy Anime interspersed with TRPG maps. I did sample a few of the sequels from through the years by checking out playthroughs on YouTube, and maaaaybe I could see me plunking down for another go if one of the Super Famicom entries gets this same love by Nintendo in the future. But, as with Pokémon Shield, I feel like I’ve poked my head into a weird wide world, got a taste of what it was all about, and can now leave satisfied.

I think that about covers everything. Oh, wait, no, I kinda like playing as Byleth in Smash Bros. Ultimate. There, that covers everything.

Thank you for reading about my series of unwise Fire Emblem decisions!

Is There Gold in the Hole? (I just watched Tenet.)

I watched Tenet twice. I think I get it. And oh, by the way, this post contains spoilers for Tenet. I apologize up-front for how confusing it all might be to follow.

A lot of people find Tenet inscrutable. I didn’t find it inscrutable, but as mentioned, I did have to watch it twice. My first view gave me enough context to understand what I was seeing the second time, which is pretty standard for a good meaty time travel movie. (Primer was more like four times, heh heh.) And while I enjoyed the movie well enough, I find myself a little chuffed at the plot holes. I’m not used to Nolan films having them.

(There’s a long conversation to be had here over whether plot holes are really “that bad”, or not, that is a topic for another time. I think they’re more forgivable in some movies than others, and Tenet is the kind of movie where they’re less forgivable by a mile. Nolan is not a director I typically associate with “plot spackle”.)

Time travel stories are hard, because it’s a thing we can’t observe, because it doesn’t exist. Unlike other things that don’t exist, though, moving the wrong way through time creates weird and tantalizing situations when you start following the logic up. It’s not like when Gandalf makes a dragon out of fireworks, that’s just magic and you go “oh, right, magic.” You don’t have to justify breaking the laws of physics if you’re okay with appealing to magic or ghosts or god or nanomachines.

But breaking causality is a very different beast. The logic fails immediately and so as part of your story you have to construct new logic, and this new logic is now the scaffolding that holds up the rest of the story. There’s a reason every time travel movie ever made, Tenet included, has a scene where a character just stands there explicitly explaining the rules.

My gold standard for a truly great time travel story hinges on these two points:

  • Did the story explain the rules and do I understand them correctly?
  • Does the story follow its own rules once they’ve been established?

It’s very common for time travel stories to succeed at the first point, then fail at the second. That’s the category Tenet falls into, and it fails pretty badly.

Disclaimer!

Tenet is a very dense movie. Its time travel mechanics are fairly simple but the resulting logic gets very complicated very quickly. I’m a pretty smart guy (my mom thinks so anyway) but it’s possible I’m missing something important about how the mechanics work that would invalidate this post. I don’t think that’s the case, and I’ve confirmed that a lot of people are confused in the same way I am (beyond the typical “it’s time travel so I don’t get it” responses, I mean).

Before we get started, I just wanted to make sure “whoops turns out I’m dumb” is on the table as a possibility.

Temporal Polarity

What are the rules for time travel in Tenet? What we’re told, and what we’re shown in every character interaction in the film, is that history is immutable. “Immutable” is a big important word we’ll come back to later. It means you can’t go back in time and change anything. The reason you can’t, in the Tenet universe, is the same reason you can’t go forward in time and change anything. That statement sounds like nonsense, and it would be under different rules, but it’s the feature that makes Tenet stand out in its genre.

We think of the past as stuff that’s already happened, and the future as stuff that hasn’t happened yet, because that’s the direction our arrow of time points. To a person facing the other direction, the future is the stuff that’s already happened and the past is the stuff that hasn’t happened yet. If we were to meet such a person, we would disagree on what was “past” and what was “future”, for the same reason two cars facing each other at a red light would disagree on what is “ahead” and what is “behind”. In that situation my “behind” is your “ahead” and vice-versa. When the light turns green, we pass each other, then move away from each other, even though we’re both driving “ahead”.

Objects and people in Tenet have a temporal polarity. If it was 9:59 one minute ago, what time is it now? If your polarity is normal, the answer is 10:00. A normal clock on the wall moves forward at the rate of one second per second. But if your polarity is inverted, the answer is 9:58. That same clock appears to be moving backward, at the rate of one second per second. And, of course, we could also invert the clock too, so it appears “normal” for you and “inverted” for me. A clock (or a person, or a car) of the opposite polarity appears to be moving backward because their arrow of time is pointing the opposite direction.

If you want to travel back in time one week, you have to invert your polarity and then live out the whole week in real time. You will be one week older when you arrive at your target time, and there will be another you there, two weeks younger, doing whatever you were doing one week ago.

The temporal polarity of the world itself is normal in Tenet. That the world itself has a particular bias for which way it moves on the arrow of time is one of the film’s main conflicts. This means that during your one-week journey into the past, you will watch the sun rise at dusk and set at dawn. You will experience reversed laws of physics. You will be unable to breathe normal air. And everyone else will be walking around backwards, giving you funny looks before not noticing you at all.

Whatever Happened, Happened

Whether the timeline is mutable or immutable is probably the most obvious concern in any time travel story. Can you change the past? If so, history is mutable in this world. If not, history is immutable. There are lots of examples of both in time travel fiction, and protagonists thinking they’re in one type of world only to discover they’re in the other is a common time travel twist. Every time you’ve seen a movie where a guy went back in time to save Someone, only to accidentally cause the thing that killed Someone in the first place, you’re seeing the “time is immutable after all!” twist at work.

Lost was the first story I’m aware of that stated its immutability clearly and concisely: “Whatever happened, happened.” You can’t travel to the past and cause an event that didn’t happen. The reference is apt here because Tenet uses the exact same phrase to describe the same thing. It’s reinforced every time the protagonist interacts with his own past; we see the same events playing out in reverse. You can’t change these events, but you can witness them from a different perspective.

(I replayed Final Fantasy VIII recently, and it’s the same rules there. The quote is, “You’re the one that changes, not the past.” And that’s applicable to Tenet too.)

Whether history is actually immutable, as a universal axiom, is a question of some debate between characters in Tenet. There are clearly some characters who believe otherwise, including the protagonist at certain points, and that belief makes them dangerous. But the evidence we’re shown on-screen always points to immutability. There are no scenes in the movie where we see an event occur on one pass, and then later see that event occur differently on the next pass.

Or Are There?

And this brings us to the gold in the plot hole. And also in the literal hole. In this case the plot hole involves a real hole, which may or may not contain gold.

We know two rules about gold bars in Tenet. The first is, the bar has temporal polarity; it’s moving either forwards or backwards in time. It can be doing either, but it can’t be doing both at once, and it can’t change spontaneously. The second is, whatever happens to that gold bar is the thing that happened to it. Whether it’s normal or inverted, and whether acted upon by a normal or inverted force, the state of the bar at a particular moment in time is always that state, immutably.

The villain in Tenet is working with people in the future, who need the villain to perform a set of tasks in his-present-their-past. In return, they give him boxes of gold. Let’s run through how it works from each perspective.

From the villain’s perspective, he travels to a pre-arranged location and buries a normal, non-inverted, empty box. Then he waits fifteen minutes. Then he digs the box back up, except now it’s inverted and filled with inverted gold bars. He takes the bars and uses them to fund his criminal empire. The process is nearly instantaneous.

From the future peoples’ perspective, they travel to a pre-arranged location and dig up a normal, non-inverted, empty box. They fill the box with non-inverted gold bars, then send the whole package through their inversion machine. Then they bury the inverted box, happy to fund the villain’s criminal empire in the past. The gold simply vanishes; it doesn’t exist anymore beyond when it emerges from the inversion machine.

From the box’s perspective, it is buried by the villain in a pre-arranged location. There it sits for 300 years, moving forward in time, one second per second. At that time the future people dig it up, fill it with gold bars, and send it through their inversion machine. The now-inverted gold is buried in the same location, and sits for another 300 years, moving backwards, one second per second. When the villain digs the box back up, 600 years will have passed for the box, though it will not have moved an inch.

Which Rule is Broken?

Say the villain buries the empty box at 9:50 and digs it back up at 10:00, finding it full of inverted gold. Let’s take one of those inverted gold bars and ask this question: what is the gold bar’s state at 10:10?

The obvious answer is, the villain has it. It’s in a duffel bag or loaded onto a truck en route to the villain’s bank or money bin or whatever. This is the point of sending the gold back in the first place: the future people are paying him for services rendered.

But wait — the only reason the box had any gold in it is because the gold was inverted 300 years from now. At 10:10 it was sitting in the buried box just as it had for 300 years, and it won’t be dug up for another ten minutes. The gold bar can be in two places at once near the point in time when it’s inverted. But there can never be two gold bars.

In fact, if the gold bar is in a duffel bag, the implication is that it had never been placed into an empty box and buried for 300 years. A normal person carrying an inverted object is, by definition, carrying that object to its point of origin. This wouldn’t be a problem if the bars are inverted again (de-inverted?) after being removed from the box, but they aren’t. They paradoxically originate in the villain’s money bin and 300 years in the future.

Let’s illustrate it this way. An observer monitoring an object that has been inverted at least once could describe that object in terms of whether it’s moving towards or away from its inversion point. As long as the gold is in the hole, this works fine. A normal observer would say “towards”, because they could watch for 300 years as the gold bar sits there doing nothing, then gets dug up, then gets carried backwards into the machine. But what about once the gold is in a duffel bag? A normal observer would have to conclude “towards”, but they’d be wrong, because no matter how long they watch the gold will never be buried for 300 years, nor dug up, nor carried into the machine.

It’s even stranger for inverted observers. One watches the gold as it emerges from the machine and is buried, then quietly monitors it for 300 years while it does nothing, and then it just… vanishes. As soon as the villain puts his hands on it, the gold no longer exists. Meanwhile, another inverted observer watches as gold which has never emerged from an inversion machine gets placed into a duffel bag, is placed into a hole, and then vanishes. Both of these observers would agree the inverted gold is moving “away” from its inversion point, but one of them would be confused as to when that happened (because it never did), and then they would both lose the gold entirely as soon as the villain interacts with it.

It’s almost like the villain touching the gold acts just like an inversion machine. Could this be what’s happening? In the film we see inverted bullets leaving holes in glass. (And we see this in both directions.) But wait, no, we’re also explicitly shown in the first polarity exposition scene that inverted objects behave strangely, but consinstently. “Dropped” objects magically jump up into your hand. Bullets shoot out of walls and into your gun. Piles of rubble form up into buildings and structures. Manipulating an inverted object doesn’t change its polarity at all. From the object’s point of view, your handling it is just one of many events that happens to it in between its emergence from an inversion machine and its eventual final resting place. But our gold has no final resting place. It originates at two different points (one of which is unknown) and then vanishes entirely when the villain touches it.

That just leaves immutability. In fact, a surface read of the event has us throwing immutability out the window immediately! The future people saw that there was an empty box in the past, and decided to change the past, and fill that box with gold. For them to dig up an empty box, it must have traveled there, sitting empty, for 300 years. And for the villain to receive the gold at his end of the trip, the box must sit there again, for 300 years, in the same spot. Pick any point on that 300 year timeline and ask: what does that space contain? An empty box? Or a full one?

The answer must be the box is empty in one timeline, and full in another, but the whole point of immutability is we’re not supposed to have timelines. If that’s possible, then the protagonist’s original idea that he can change the past is correct and we should have gotten a very different movie.

To make this gold work, we have to either throw out our understanding of temporal polarity, or throw out immutability, and doing either wrecks the rest of the events of the film. Argh!

Radiation?

There’s always room to hand-wave stuff like this aside in any time travel story. Again, one of Tenet’s main plot points is that the antagonists believe the timeline is mutable, and are actively trying to change the past. The movie fudges hard-to-explain interactions between normal and inverted matter by going “something something radiation” and then not looking at it too closely in most scenes. That’s well enough, and maybe about as good as anyone could expect of a writer, and if that were the end of my gripe I wouldn’t worry so much about this gold.

But I just can’t shake it. The antagonists in the film are wrong about changing the past, and so is the protagonist each and every time he attempts it. Indeed, it’s only after he accepts the immutability of the timeline that things start going right for him. (His quote is, “Whatever happened, happened. I get that now.”) Time travel stories that embrace immutability have a certain beauty to them, a sort of all-wrapped-up elegance that I really appreciate. Unfortunately, Tenet doesn’t.

The gold bars aren’t the only objects in the story that seem to break when you start considering them carefully. But they are one of the objects that do need to be considered carefully, if the villain’s motivation is to make any sense. Either they’re stupid, or I am.

Thanks for reading!

The Lightning Saga Remastered (Coming Soon to PS5!)

This is all hypothetical. I’m sorry if you Googled that and it popped up and you clicked the link thinking maybe I had some news the rest of the games industry didn’t. I don’t. But this year I played through Final Fantasy VIII Remastered and Final Fantasy XII: The Zodiac Age and Final Fantasy VII Remake and the Steam port of the 3D remake of Final Fantasy III and, ya know, I got to thinking. Isn’t it about that time for FFXIII?

The Final Fantasy XIII trilogy was released originally on the PS3 and then later atrociously ported to Steam. I say atrociously because I tried to play the PC version of Final Fantasy XIII-2, but couldn’t, because it installed something that made my controller catch fire. Even after I refunded and uninstalled the game, the garbage it installed caused my gamepad configuration to break in a bunch of other games I’d never had trouble with. So, ah, if you came here looking for a review of the Steam port of FFXIII-2, I guess that’s the takeaway: don’t buy it, because if you do it won’t work, and then you’ll have to spend two hours manually scraping malicious .dlls out of your hard drive.

But here we are with our toes dipped in this new console hardware generation, and while we know FFXVI is looking pretty sexy, it’s probably still a few years out. What’s Square-Enix cooking up to keep us fed in the meantime? I submit to you that the FFXIII series is ripe for the plucking.

I don’t actually care that much about FFXIII-2 or Lightning Returns, but there’s a lot of room for improvement in the original FFXIII.

Package it with FFXIII-2 and Lightning Returns.

Like I said, I don’t actually care about these two games very much, but if I’m paying $60, they’d better be in there. Oh, don’t get me wrong, I will absolutely spend $60 on a stand-alone FFXIII remake, because I am a putz with more money than sense. But I’ll spend $80 on a FFXIII Collection in a snazzy box, even if I don’t like one-third of the games.

Better Replay Options

FFVII Remake and The Zodiac Age both have excellent New Game Plus modes, and there’s no reason FFXIII shouldn’t, either. This was the first Final Fantasy title which split its gameplay into concise numbered chapters, and we seem to be doing that in all of them now, so an unlockable Chapter Select after you roll the credits would definitely be welcome. And let me use any combination of the six characters I want in each chapter, too. Maybe include a “Hard Mode” option that replaces each enemy with the hardest version of that enemy type, so re-visiting old chapters is a good way to stockpile CP and develop your characters.

Can I play on the racetrack, please?

We have a canonically-established grand prix racetrack in this universe, and six player characters whose spiritual monster buddies all transform into vehicles. There’s a dumb mini-game here waiting to happen, and I think we can all agree that FFXIII is sorely lacking in the dumb minigame department. Unlock “Cocoon Circuit” once Lightning defeats Odin, and make the other five characters playable as they receive their own eidolons. Unlock a second “Pulse Circuit” upon reaching Oerba.

The reward structure doesn’t need to be more complicated than just giving me piles of money. We all know I’m going to be farming gil for hours to afford mats for weapon leveling, just give me a second thing to grind at so I can switch up the scenery every so often. Prize stickers I can slap on Odin’s butt will also be acceptable.

Swap Party Leader

This is the thing everyone always complains about in FFXIII. When the party leader dies, you get a Game Over, and there is no mechanism by which to switch party leaders during battle. The final boss in particular has this really nasty combo where he can poison you with one of his regular actions, but he also has a special scripted action to drop your whole party’s HP to 1 upon hitting a certain HP treshhold. If you don’t know about that and these two things happen at almost the same time…

The only reason it seems like we’re not allowed to swap leaders is each hero has a special ultimate attack that must be manually selected. Since your AI partners will never use their ultimates, your choice of leader is also a choice of which ultimate attack you want to have in the chamber. I guess there are balance reasons why we’re not allowed to access more than one ultimate at a time. Maybe the game would crack in half if you queued one up and then switched to someone else to queue theirs.

The game already has a party resource for this sort of thing, though: TP. Techs have to be manually selected and they burn a meter that isn’t tied to any particular character. Make these six ultimate attacks into techs requiring full TP and let me choose between the three associated with my characters, and whatever balance bugaboos you’re worried about are sated.

(If we’re doing this, Vanille should probably get a new ultimate attack, and her existing Death spell should just be rolled into the SAB crystarium.)

FFXII already did this right, fools!

One of the major annoyances in FFXIII is, every time you switch your party (which is mandated at many points during the game) you have to go in and reset your paradigm decks. More than half of your input as a player during combat just boils down to switching paradigms, and the game doesn’t let you save deck favorites. We don’t need to get super crazy with this, just remember what the deck looks like per each combination of three characters, so I can switch around and switch back and end up with something that looks like what I already had.

We also really need some control over our AI partners outside of swapping over to them and inputting their commands manually. (In FFXIII you usually don’t even input your own character’s commands manually.) The way it works currently is each role has a sequence of commands it wants to use, and an order it wants to use them in, and it goes through the list one at a time. This is most noticable for the SYN and SAB roles but every role has some version of it.

All I want here is a page that lists all of a character’s abilities for a given role, and a way to toggle them on and off. I don’t need Veil and Shell in absolutely every random encounter, Sazh. Come on.

New roles!

Adding new character content is a Final Fantasy remake tradition: FFIV DS‘s augments, FFV Advance‘s new jobs, FFX-2 HD‘s new pokémon. For better or worse, FFXIII has a unique combat interface focused on macro-actions that no other game has. The trick here is to introduce some classic Final Fantasy jobs which stand out in FFXIII‘s interface.

In FFX, Yuna can summon aeons to fight for her. She does this by issuing individual commands just like the player typically does. However, a couple of her aeons are controlled only indirectly; instead of issuing commands to Yojimbo, she pays him some amount of gil, and then Yojimbo decides what to do. This is kind of a fun little thing in an otherwise complete system, and I think we can pluck at it for FFXIII.

I propose these three new roles, unlocked late in the game: [OPP]ortunist, [COL]lector, and [OBS]erver.

OPP is the new spin on the traditional Gambler. Their AI is limited to Auto-slots, which rolls the tumblers to pull from a variety of effects. When controlled directly, you can specify some amount of gil to get a calculated amount of damage and perhaps special side effects like dispelling buffs or hitting elemental weaknesses. Include a powerful attack called “Go all in!” which spends some preposterous amount of gil, say 10x the average gil value of whatever the enemies typically drop, for a chance at instant stagger.

[COL] is the Chemist. Their AI is Auto-potion, the only function of which is to heal the party with consumable items. (All item use in XIII is otherwise manual.) Give them a passive ability that doubles item effectiveness, and another that gives some percentage chance to use an item without consuming it, and a back-row COL becomes a pretty good party healer. Besides that, XIII has a huge number of individual items, none of which have any function other than their sell value or their XP potential for equipment upgrades. At any time during combat you can switch to your COL and have them mix any of these two otherwise flavorless items together to create some bombastic effect. Low-tier items are already available in shops in large quantities, and the really rare items are so difficult to obtain you could get away with creating truly spectacular attacks.

[OBS] is the Blue Mage. The role does nothing at all except have a passive Learning trait, and there’s a big list of enemy skills to learn. The AI that chooses which skills to use out of the big long list could run on the same sort of logic that the current RAV/SYN/SAB skills do. (Check enemy weaknesses, don’t apply redundant debuffs, etc.) Mostly I just want this in the game to justify an attack called 1,000,000 Needles.

Greg

Spend one million gil in the Gilgamesh, Inc. store and the man himself shows up to yell at you about swords, then you beat him up. Come on, this one’s a no-brainer.

Re-design Gestalt Battles

Nobody knows how these fights work and everybody complains about them. Whatever the devs were trying to do, they failed. Re-design these fights to use the same rules as regular boss battles.

Increase bankable CP from 999,999 to 99,999,999.

You’d think a million CP is enough for anyone, but it’s actually only a fraction of what you need to complete the Crystarium for any given character. You have to fill the jar a few times for each character, and any CP you earn while the jar is filled just gets wasted. Accidentally wasting CP because you don’t want to use Hope in the active party during a long grind session is just a bad situation we should probably avoid.

A Better Post-game

Which is to say, uh, any post-game at all.

(Quick digression: I actually abhor the term “post-game” because [long list of reasons]. I use it here only because we all kind of know what it means. End of digression.)

FFXIII is one of the few Final Fantasy games where defeating the final boss actually opens up new game content; the final stage of your Crystarium is locked until you roll the credits. A few of the remakes unlock new challenges or bonus dungeons or whatever, but FFXIII explicitly tells you “you aren’t done yet.”

Unfortunately that’s all it does. There’s nothing to do with that final Crystarium stage except throw it against the areas you can already reach, to better humiliate the bosses you could already kill. There’s not much a party capable of defeating Barty 3 and Orphan can’t do, and slogging through the millions of CP required to fill the final tier on old monsters just feels like unnecessary hassle.

I already mentioned some stuff we can do with a Chapter Replay option, but there’s no reason we have to stop there.

Revamp Tier III Weapons

Currently, each weapon in the game belongs in one of three tiers, with each character’s “ultimate weapon” being Tier III.

(It’s actually slightly more complicated than that. Each weapon in the game has a Tier III version, and so each character has several different options for “ultimate weapon.”)

To make Tier III weapons you need a rare item called a Trapezohedron. One or two of these drop in your lap just by clearing game content, but the rest have to be farmed for over long hours. Instead of a one-size-fits all ultimate weapon strategy, let’s revamp the system so each character now has an optional endgame sidequest to earn a unique crafting material, instead. Maybe keep Trapezohedrons as a semi-rare resource in the game, used to transform one character’s Tier III into another version, so you can play with it without having to farm it up again from scratch.

And while we’re at it…

Dark Aeons (Except Good)

…maybe we can tie those sidequests into each character’s eidolon, and doing so causes a superboss version of the Gestalt fight to appear somewhere in the world. If there’s one thing FFXIII needs, it’s more giant super-monsters to throw your millions of CP worth of abilities and paradigm decks at. And maybe…

11th Crystarium Tier

…defeating a dark eidolon (or some other superboss) unlocks an 11th tier to the character’s Crystarium. There’s a lot of space to make these characters even more absurdly powerful than they already are, including:

  • a standard “break damage limit” node;
  • a node that unlocks a feature which increases that role’s effectiveness based on banked CP, say a 1% increase in effectiveness (damage output for COM, damage reduction for SEN, buff duration for SYN, etc.) per 1 million CP, up to the new max of 99% increase at 99 million;
  • a node that gives a character some fraction of this role’s bonus even when not in that role (e.g., Lightning could enjoy some percent of her COM damage bonus even as a RAV); and
  • the missing skills from that character’s lower Crystarium levels (e.g., Lightning doesn’t learn Curaja as a MED, even though MED is one of her starting roles, so she could learn it here).

The idea here is, in many Final Fantasy games, there are two parallel “lanes” of leveling up. You gain experience to go from L1 to L99 in Final Fantasy VI, but that doesn’t leave you with the most powerful characters. To do that you need to maximize esper stat growth, which is a totally separate system. Nothing like that really exists in FFXIII, but the game also doesn’t really support a “totally separate system” for parallel leveling. (And it’s a mostly vestigial concept in other games too. Neglecting esper stat growth in VI still leaves your characters way overpowered for any actual challenge in the game.)

Anyway, we need all this extra power because…

Dark Faultwarrens

…we’re sending these super-charged 11-tier 99m-CP-banked characters into the Dark Faultwarrens.

O.G. vanilla FFXIII has a location called the Faultwarrens, which is a sort of branching battle gauntlet. You fight a battle there, and the battle arena has two exits, and each exit takes you to a different path, until you reach one of eight possible exits (each with unique rewards). These are some of the toughest fights available in XIII and it’s a concept I really like.

Let’s leave those in, but at some point let’s open up a second Faultwarrens that works the same way but contains much stronger monsters. The 8 ending fights should be on the level of “maybe 2% of players will complete this content.”

Brick your ideas are dumb and you are dumb.

Possibly!

XIII is rare in this series in that there’s really only one direction for your characters to advance in. Once you level up and equip the best weapon, your work is done except for making the numbers ever more and more preposterous. That’s why my ideal vision for bonus endgame content in this title specifically involves so much “push the numbers as high as they’ll go”.

That said, I’m one of the few dorklingers that actually enjoys Final Fantasy XIII. It’s possible the game doesn’t have enough grip to bother remastering at all, or if it does, perhaps the remastering efforts are better spent bolstering those elements of the game so many people found objectionable.

Mostly, for my purposes, I’m just ready to put the PS3 in the closet.

Thanks for reading!

18 Adventurers or, An Exercise in D&D Alignment

A party of 18 adventurers find themselves in a bit of a moral quandary.

They had been adventuring in and around a rustic village and have earned some well-deserved renown amongst the townsfolk. Every so often one farmer or another living on the village’s outskirts complains of bizarre livestock mutilations. The village elders (some of whom have lost livestock of their own) bicker about whether the mutilations are the work of a malicious prankster or a monster living in the wilds beyond the fields. The party become involved when the most recent mutilation raises the stakes dramatically: instead of a sheep or a goat, a child has been found murdered and the body badly ruined. The villagers need this problem solved quickly. Either they have a murderer living amongst them or there is a vicious beast growing bolder by the moment. The adventurers are hired to investigate the problem and deal with it. Upon successful completion of the quest, their reward shall be 1,800 pieces of gold (100 pieces each).

The party follows the trail to a distraught woodsman in a cabin out beyond the fields. The man admits to the mutilations, but swears innocence, blaming his actions on a terrible curse that has befallen him. He periodically transforms into a werewolf with an insatiable hunger for slaughter, and cannot control his actions while in that state. He has known about his condition since winter and has been trying to manage it by limiting his visits to the village to just those of necessity. He is understandably traumatized to learn he has slain a child and fears his explanation will fall on deaf ears if he is delivered to the village elders. If the village gets hold of him, he will surely be found guilty of murder and hanged. He swears to the party he will pack his meager belongings and head off into the wilds, living off the land and staying far from civilization, if they agree to let him go.

There’s a slight wrinkle, however, when under the effects of a zone of truth spell, the woodsman is discovered to be lying. The curse is real, as is his remorse, but the party discovers he has two daughters living in the village whom he loves dearly, who he does not intend to be apart from them. His will to leave is genuine, but he plans to continue sparse visits to the village to see them, believing that he can return in the time between his violent episodes so as to cause no further harm.

The decision before the party is this: they can either let the man go and run the risk his curse management plan fails, leading to more deaths, or they can bring him back to the village, collect their reward and leave the poor woodsman to the village’s justice.

The lawful good druid argues to let the man go. The laws she recognizes are the laws of nature, in all their primal beauty, and the nature of the woodsman’s curse is a perversion of those laws. While the child’s death is tragic, manufacturing a second death in no way pays for the first. The only reasonable course of action, she says, is to seek a cure for the woodsman’s ailment, thereby preventing any future unnatural loss of life.

The lawful good paladin angrily disagrees. Letting the man go is simply not an option. While the circumstances are indeed tragic, the fact remains he broke the laws of the village where he trades. The paladin intends to uphold the law by bringing the woodsman back in irons and submitting him to their justice, but concedes to advocate on the man’s behalf against execution, using his portion of the reward money to that end if necessary.

The neutral good bard was leaning toward letting the man go until casting zone of truth. Now that he knows the woodsman will return to the village from time to time, there’s really no way to prevent future murders except to bring the criminal to justice. The continued loss of life is simply unacceptable. Perhaps if the man had not lied about his intentions…

The neutral good dwarf doesn’t give a goblin’s bollocks about the man’s intentions. The fact remains, there’s some sort of malignant force in these woods transforming people into werewolves. While alive, the woodsman is an asset in locating the source of his curse; if he is executed they lose their primary lead until someone else is turned and another murder has occurred. He argues the party has a duty to this man, the village, and the slain child to find the creature or magic ultimately responsible for the violence and destroy it.

The chaotic good ranger doesn’t think the village can render a fair punishment. It’s wrong for this man to have killed a child, but he at least has the excuse of being magically compelled. An angry mob cannot be counted on to act rationally. If delivered to the village he will become the victim of a different kind of bloodlust, and one she wants no particular role in feeding into.

The chaotic good half-elf points out that it’s disingenuous to say the man has acted under magical duress. He’s known about his condition for months but valued his desire to see his daughters over a child’s life. He’s therefore ultimately responsible for his own actions. Each villager deserves to know the truth and has just as much right to weigh this man’s priorities as the man himself does.

The lawful neutral dragonborn wants to calm the discussion down and just stick to the facts. The village has laws. The law was broken. The punishment for breaking the law is hanging. That’s the end of it as far as she’s concerned. What’s left to discuss?

The lawful neutral monk is a bit surprised, as she has the same reasoning but arrived at the opposite conclusion. Respect of law must be observed, but in this case, the woodsman very pointedly hasn’t broken any laws. He acted under compulsion of a magical curse which, as of now, is ill-understood. To kill him for actions outside of his control and before all the facts are in evidence isn’t justice, it’s vengeance.

The true neutral barbarian says he’s fine with the villagers enacting vengeance. A life lost for a life taken. That’s fair in the only sense that matters.

The true neutral necromancer says the man should be freed. He’s been following the debate closely, and while only about half of his companions have spoken their piece, it was starting to look like “bring him in” was picking up steam. He wants to make sure that everyone’s thoughts have been given due consideration and that the dissenters aren’t bullied into submission.

The chaotic neutral half-orc flipped a coin. It came up heads. He therefore argues loudly and boisterously that they bring the man in and see him hanged.

The chaotic neutral artificer argues, equally loudly and boisterously, to let the man go. He doesn’t actually care about the man or what happens to him, but arguing for the man’s release ties the conversation back up, and the more discordance and bickering amongst his companions, the happier he is. He’ll change his mind later if it looks like things are swinging the other way.

The lawful evil cleric agrees with the paladin and the dragonborn: a law was broken, and the penalty is fair. However, she has been secretly unhappy with her recent efforts to sway the villagers into converting to her patron deity. She sees an opportunity to use the situation to cast the village’s church in a negative light by insisting their worship of pagan gods is what caused the curse in the first place. (Which, for all she knows, is true!)

The lawful evil warlock insists those arguing to drag the man back in chains are acting prematurely. He wants to join the dwarf and the druid in searching for the source of the curse, and they need the man alive to do that. If the village is willing to pay them each 100 gold to bring the murderer to justice, the warlock imagines he could negotiate double or even triple once the town learns of the much more insidious threat!

The neutral evil assassin has done the very simple math. If they bring the woodsman in, they each get 100 gold. If they don’t, they don’t. That’s all the information he needs to seal the woodsman’s fate.

The neutral evil wizard sees the 100 gold reward as a pittance when compared to the fascinating fel magic that’s obviously in play. She wants to let the werewolf live in order to study his case, potentially identifying and maybe even learning to control this curse for her own ends.

The chaotic evil battlemaster wants the man hanged. The sight of the child’s mangled corpse excited him, and the prospect of a violent public execution excites him further. He secretly hopes the mysterious curse creates more werewolves in the future, perpetuating the cycle of bloodshed.

The chaotic evil tiefling votes to release the man, but doesn’t explain why. Her plan is to sneak away from camp that night, track the werewolf down, and offer him this ultimatum: he must pay her 100 pieces of gold every month or she will reveal him to the village. If he refuses, she reasons, she can drag him in herself and claim the whole 1,800. Either way her payout is larger than it otherwise would be.

Sensing the gridlock, the party members prepare to go over the pros and cons of all their arguments again, when the bard suddenly realizes the woodsman has fled out a back window during the debate. The party takes a short rest, spends some hit dice, and the chase begins anew!

The Almost-Perfect Save: An Ill-Advised Final Fantasy Catastrophe

I recently started leaning hard into streaming again. I have to make it a habit again or I’m afraid I’ll just fall off the Internet forever.

I decided to embark on a probably year-long journey to get as close to a perfect save as reasonably possible, with maxed-out mostly-everything, in as many Final Fantasy games as I could stomach. Quick back-of-envelope calculations have this project somewhere in the ballpark of 1600 hours.

I set up a special page on this blog to track my progress. You can find that here: http://scibbe.com/the-almost-perfect-save.

Please drop by the live streams if you get a chance, or check out the archives on my YouTube channel, which I will try to keep organized by game. (I should also probably plan to keep the story completion playlists separate from the long naked grinding sessions. We’ll see how well I do on that front, I guess!)

I should be at this for quite a while. Lots to get through!

Final Fantasy VII Remake: Profoundly Unsatisfying

I didn’t hate it. That’s not the problem. I mostly enjoyed it. I really enjoyed parts of it. I might give it another run in a year or two, and I think I’m, er, looking forward to the sequel, or the “next episode”, or whatever it is we’re calling this series (sub-series?). It’s just… man. It’s such a weird sensation to mostly have a good time playing a game, but then leave it feeling so negative about the experience. I guess I’ll put Final Fantasy VII Remake on the same little grungy end-table where I keep The Witness and Breath of the Wild.

It kinda does.

There is a major gameplay reason for this negativity, and a major story reason, and I’d like to talk about both. First, though, I think it’s only fair we go through the good stuff. There aren’t any spoilers in this first part.

The Good Stuff

There’s this cheesy, kind of self-indulgent splash screen whenever you boot up Final Fantasy XV which declares the game to be “a Final Fantasy for fans and first-timers”. In 2016 I was like, yeah, that’s one way to categorize the entire population of planet Earth into two mututally-exclusive groups, I suppose! But I kept thinking about that screen this week. Remake sold like absolute gangbusters, and I know some number of people will see that splash screen for the first time, having bought FFXV after having enjoyed Remake, and feeling like they’re getting into something special. There’s a 30-year history a lot of new players are going to tumble down into purely because of this game. I think that’s a tremendous credit. In the coming years more people are going to meet Noct and his bros, and roll their eyes at Ashe’s ridiculous hot pants, and hear the word “disasteriffic”, and sign up for the Four Job Fiesta. That makes me happy.

Before 1997, RPGs were nerdy fantasy stories for basement-dwellers and grognards. Final Fantasy VII made them COOL. There really was a glorious era of exceptional video games swimming in its wake. Eventually the genre changed and shifted and mutated, and nowadays it seems to be largely populated by hyper-candy anime girls and unfocused, impenetrable gameplay systems. They’re games for basement-dwellers of another stripe, is what I mean. And Remake isn’t that. I think it does us all a lot of good if, for the next ten years, developers lean towards Remake and away from Tales of Mediocrity IV-Z: Hyper Legend of Browsing the JRPG Tag on Steam (Waifu Edition).

Remake is the first Final Fantasy game I’ve played in… golly, two decades where none of the game systems felt confusing or vestigial. I understood immediately how equipment and materia worked, and how to upgrade them. The controls all made sense, there were no extraneous meters or counters, I didn’t have to keep looking up what combination of trigger buttons activated each of my six combat modes. Enemy data was clearly presented. I wanted to use all the characters. I actually cared about the treasures I found because they were useful consumables or pieces of equipment, and not vendor trash or digipoop cards or whatever.

(It might actually be that the last Final Fantasy game whose gameplay was this tightly designed was the original Final Fantasy VII. FFVIII had that pointless GF compatibility system no player ever noticed or cared about, and FFIX had limit breaks that only triggered when there was a single wounded goblin left alive, and never in tricky boss fights.)

Square-Enix has been noodling with this idea that you can blend action and menu-based RPG mechanics together since long before there was a Square-Enix. They’ve had… varying levels of success. The primary disconnect in these systems is positioning. In an action game you push the sword button, and your guy swings his sword, and if there’s a dude standing there the dude takes damage. In an RPG you have a sword command on your menu, and you select it, and the game rolls some invisible dice, and if the dice are good, the dude takes damage even if he’s on the other side of the screen. Both conventions work in proper context, but the push-and-pull introduced by trying to blend them makes for frustrating situations. Remember missing monsters in Secret of Mana even though you watched the pixels of your weapon touch the pixels on the monster? Or getting blasted by an AOE super attack in Final Fantasy XIII because your idiots refused to spread out?

Remake largely avoids these problems. There is the occasional positioning or camera snafu inherent to all 3D games. Sometimes you’ll push the gun button and Barret will take a step leftward for no reason and fire a full clip into a wall. But for the most part what we have is a well-grounded action combat system with good gamefeel and clear feedback. Running behind cover really does block the enemy’s laser, even if they roll a crit. Holding the block button really does reduce the damage you take. Picking a melee attack off the menu will cause you to spend resources to punch the air unless there’s actually an enemy nearby. It really feels like you and the bosses are playing by the same set of rules, at least as far as connecting attacks goes.

There are four player characters, and some thought went into making their playstyles unique without having to re-learn how the combat system works. In fact, the difference between characters boils down to “what does the triangle button do?” Cloud uses it to switch between two useful and easily-understood attack modes, Barret uses it to fire a super shot that recharges automatically over time, Tifa unleashes karate combos she can power up manually using meter, and Aerith gets a Mega Buster. Outside of these commands, and the standard attack/block/dodgeroll gameplay that’s so familiar nowadays, everything is selected manually. Your meter fills up as you hit enemies or block damage, and you spend it on casting spells, using items, or doing special weapon attacks.

You control one character and the AI controls the other two. Your AI partners play considerably more slowly than you probably are, so it can feel like their meter isn’t filling as fast. It also feels like enemies will prioritize attacking your controlled character, if possible. This is how the game gently nudges you into changing characters a lot, keeping track of who has meter and who doesn’t, threading commands for one character into the attack animations of another. It feels a bit like how they probably wanted Lightning Returns to play, back in the day.

Stagger is back. Stagger is a mechanic whereby you have to put the enemy into a vulnerable state before you can actually damage it appreciably. This is another system Square-Enix has fiddled with for a while now, and it finally feels like they landed in the sweet spot. Scanning an enemy tells you if an enemy can be staggered and, if so, how to do it. You get a clear visual indicator near the enemy’s health bar telling you when they can be staggered quickly — and here’s the key bit — the stagger bar actually does fill up quickly. A marked improvement over the fiddly mechanics of the FFXIII series, where it could often feel like you were nickle-and-dime-ing enemies to death because they were impossible to knock over properly.

The combat system felt great while I was succeeding, and when I was failing, it provided clear feedback as to why. High marks.

Aesthetically, the game is a dream. Its environments are gorgeous, the UI is clean and efficiently laid out, animations are snappy and fluid. The voice acting sounds very conversational; they took a light touch with the awkward pauses and anime grunts. Many (if not most) of the individual story sequences are quite enjoyable. This game has a lot of cutscenes and I was rarely bored watching them. I loved watching the characters interact, socially and visually.

I loved seeing the materia actually slotted into each character’s weapon during cutscenes.

Remake‘s soundtrack is my favorite OST in years. Modern Final Fantasy soundtracks tend to get a little operatic for my tastes, too much wailing chorus and not enough electric guitar. More than that, though, I appreciated how the music was used. It’s not just a 1:1 remaster of each individual song; rather, the tracks are remixed and blended together as the scene demands. You get lots of played-straight versions of Anxious Heart and Under the Rotting Pizza, sure, but you also get sublime mash-ups like Turks Theme/Let the Battles Begin! as boss music. Plus we get some new earworms, like Hollow Skies, which I’d honestly love to hear a 1997 PlayStation version of.

I had a lot more nice things to say about Remake than I did about the last game I hateliked this much. Let’s slide on down into the dregs.

Just Thought I’d Mention

This game has a lot of cutscenes. I think I said that already. But you may not realize that by “a lot of cutscenes” I meant both “this game has a large number of individual cutscenes” and “a large amount of this game’s runtime is cutscene, by volume”. In a 34 hour game I probably spent a good seven or eight hours just holding the controller and watching the screen.

This usually didn’t bother me. The long story sequences were often very good, and I’m an old hat at Metal Gear Solid, so I have the fortitude for this kind of thing. But judging by the flak MGS gets for that, a lot of players don’t. If you’re familiar with the original FFVII, imagine going through something the length of the unplayable parts of the Kalm flashback sequence maybe ten or twelve times over the course of the game.

What did start to annoy me were the sheer quantity of, let’s call them “banter scenes” which infested every gameplay area. See, Remake doesn’t trust you to explore the map, or even to follow the bright “GO HERE” icon. Instead, every time you flip a switch or enter a new room in a dungeon or encounter area, your characters have to stop and talk about it. Sometimes at great length. You know you have to flip a switch. The game wrangles the controller out of your hand, close-up on Tifa, she says “Hey Cloud, I think I see a switch over there!” Slow pan across the map to the switch. Back to Tifa. “Hmm, looks like we can get there by using that bridge.” Slow pan over to the bridge. Brief rumination about how it looks dangerous/hmm how do we get there/we have no other choice. Slow pan back to the camera-behind-Cloud “ok you can move now” position.

Over and over. Six or seven times per area. For the entire game.

This kind of banter scene is crucial for character development. One of the things everyone remembers fondly about FFXV is how the bros are constantly joking or bickering or exchanging quips. But the game doesn’t stop and force you to listen. Prompto can be whining about how he saw a bug or whatever while Noctis is climbing a ladder or exploring the corners of a dingy room. It’s annoying to see them get this so wrong after they’d already gotten it so right.

All the Gameplay is Ruined, Though

They made two lousy choices when finalizing this combat system that kind of ruin the whole experience.

At any time during combat, you can pause the game and rewind time to just before you triggered the combat. This sounds like a nice quality-of-life feature — it means you can never get a solid, demoralizing Game Over. The other side of that coin, though, is that it gives rise to degenerate strategies. See, you can’t change equipment or materia during combat, and for much of the game you don’t have enough slots to equip all of your possible attack magic. (You can do that, if you want to commit to an almost purely magic build, but I didn’t want to and I imagine most players won’t.)

Each of these three mechanics is fine in isolation, but when you combine them, you get to make a really dumb decision any time you face a reasonably tough monster whose stagger mechanic involves hitting them with a particular element. To wit: you scan the monster and see what it’s weak to. If it likes materia you’re not wearing, it’s almost always faster to just retry the battle, swap out your magic, then re-engage. You even get a rung up on the action economy if you do this, since you don’t need to waste meter on a scan, now.

Decision points are what create depth in games like this, but decisions between two bad options should probably be avoided. I spent a lot of time near the beginning of each area hitting Retry for exactly this reason, and it felt like a dumb exploit every time I did it. But the alternative was slowly chipping away a monster I couldn’t stagger, because I had Wind and not Lightning magic, or whatever. A more elegant solution may have been allowing the player to spend meter to swap one piece of materia. They have to do this anyway, from the main menu, either after they chip the target down or after they select Retry, so we might as well move that menuing into the combat. If I could have stayed in the fight and spent a renewable resource to course-correct I’d have done that instead, and my scan log would not now have so many holes in it.

The other big dumb thing they did, like almost every action game in recent years, is completely flub their difficulty settings. This is particularly frustrating to me because they almost got it very right.

So we have this hybrid combat system. The action elements test your awareness and reflexes; you have to know where you’re standing, where the enemy’s standing, when you should block or dodge, and when it’s safe to run in and strike. I am bad at these kinds of games, and my preference is to scale this sort of difficulty back, if I can. (If it’s a really well-designed game, sometimes I’m motivated to practice the lower difficulties and eventually move up to higher ones.) The RPG elements test your game knowledge and resource management skills; you have this many bobbins, each thing costs some number of bobbins, you want to end the fight with more bobbins than you entered it. I’m pretty good at these kinds of games, and I’m fine with having this sort of difficulty ramped up from the start.

Remake has three difficulty settings available from power on. “Normal” is the hardest one, and is the intended play experience. “Easy” is like Normal except the monsters do less damage (and I believe act less frequently). In fact, Easy is so easy that it essentially trivializes the resource management challenge of the game. If you pick it, you will probably never have to heal during combat, and will end up with a million of every kind of potion. Obviously I don’t want that.

But then there’s “Classic”, which causes your active character to move, block, and dodge on their own. It essentially applies the same AI to your guy as it does to your partners. This tones down (but doesn’t actually trivialize) the reflex challenges of the game, since the AI is pretty smart about blocking small attacks and rolling out of the way of big ones. And they were smart enough to still let you keep control of your hero in Classic mode; you can override the AI at any time by just playing the game normally.

However, Classic forces the enemy difficulty down to Easy levels. Which makes the whole game boring.

I went back and forth between Classic and Normal a lot as I played. I really preferred Classic, and I learned a lot about how the game expects me to engage with monsters by watching how the AI does it, and also because I’m just bad at dodge rolling in general. But I really did want the other half of the combat to have some teeth. That they recognized there were two elements to the system is to their credit, but for some reason they don’t allow me to tweak the difficulty of each element individually. In the areas where I played on Normal I constantly got punished by attacks I didn’t see coming until a fraction of a second too late, and in the areas I played on Classic I just bulldozed everything.

I cannot fathom why they didn’t have Classic/Modern and Easy/Normal toggles I could set independently. The end result is, as good as this combat was, I never felt completely at home in it. The areas I played on Classic were still fun, but it was a very passive kind of fun, similar to zoned-out level grinding in oldschool JRPGs.

This would be such an easy thing to fix in a patch. I hope they do.

What Is This Even

Final Fantasy remakes have run the full scale of quality, but generally they trend toward “excellent”. Dawn of Souls is my preferred way to play Final Fantasy I. The Advance versions of FFV and FFVI are both terrific, and the latter even has some great extra content that feels like it could have been there all along. (We, ah, won’t talk about the Steam versions of those games.) The DS versions of FFIII and FFIV are so good and so different that I consider them to be completely new games. And then of course there’s FFXII: The Zodiac Age, which just may be the crown standard of how to remake an old game you don’t want to re-invent from the ground up.

(I’m really looking forward to seeing how they remaster FFXIII on the PS5, when the time comes. Maybe for my birthday next year, Square-Enix?)

We knew for years that the FFVII remake was going to be different. This is the company’s flagship title, one of the most popular video games of all time, the most singular defining point of this series and of its genre as a whole. It’s understandable they wanted the remake to be the highest possible fidelity, and to feel like something truly unique. Every game in this series has been remade at some point, but FFVII Remake is the first one that was positioned in the market as a mainline entry.

But, uh oh, it’s not actually possible to make a cutting edge super HD game that encompasses the entire scope of a globetrotting adventure, and hasn’t been since the PS2 era. FFXII and FFXV use a lot of shortcuts to try and evoke that same feeling, but don’t actually come close to succeeding, and Remake wasn’t going to, either. So they dropped the bombshell: Remake wasn’t going to be a remake of the whole of FFVII, just the game’s first act: Avalanche’s struggles in Midgar.

We’re going to get all spoiler-y now. Like, really super spoil-y, both for Final Fantasy VII Remake and the original Final Fantasy VII. Don’t scroll past this picture of Aerith riding a giant robot hand if you don’t want to know what happens in these two games.

Hold △: Make a Fist

FFVII‘s first act involves a terrorist group and an authoritarian government going to war with each other. We play as the terrorists, and blow up a bunch of their stuff. Eventually the government concocts a plan to cut us off at the knees, and they succeed, and end up kidnapping one of our own. We break into their HQ to rescue them, and while inside discover some insidious secrets that put our actions so far into a broader context. Everything culminates in a thrilling chase down one of Midgar’s highways, a giant robot explodes, and our heroes regroup just outside the walls. They have here a chance to breathe, to re-assess, and to plan for the future.

This is an excellent place to end Act I of a five act story. It is a godawful place to end a 34-hour video game I payed $60 for.

I was highly critical of this decision long, long before we even knew how Remake was going to play out. Just the announcement was enough to convince me it was a terrible idea. Before the game’s release I was mostly told to “wait and see”, and given vague assurances about how it would probably be fine as long as I shifted my expectations. In the months since its release, most of my criticisms have been brushed aside by folks who have played the game, again, by telling me to adjust my expectations.

My issue, before playing the game, was that no matter how I set my expectations, it seemed like a “Midgar only” game was a bad idea. If I criticized it for not actually being a remake of FFVII, I was told Remake is trying to be its own stand-alone game. If I criticized it as a stand-alone game, I was assured that later episodes would tie it all together. If I criticized it as being a single episode in a larger game, I was told to not really think of them as episodes but rather individual entries in a larger series, as is more common in video games.

I eventually stopped criticizing it because, you know, I hadn’t played the game. It was possible they knew what they were talking about and I didn’t.

But now I’ve played it, and my opinion hasn’t changed. I mean, it changed a little. I enjoyed my time playing, at least. Still, Remake is at best a competent implementation of a really, really garbo idea.

Here’s a hot take: Midgar is not the best part of FFVII. It’s a great start to the story, but the story and gameplay both get much more interesting after you leave. There’s this breathtaking moment outside of Midgar’s walls, some six hours into the game, where you’re placed on the world map for the first time. Midgar is this big black splotch next to you, but it’s only a speck compared to the three enormous continents you now see on the full map. Of course no other area of the game is as large as Midgar was, but that’s not really the point. That six hours was just the first step of your enormous adventure, which is going to span cities and caves and a theme park and a huge icy wasteland and an Escher maze and outer space before it’s all over.

If you really want to take one-sixth of FFVII and get the most bang for your buck, I think the most interesting section is the Huge Materia hunt from disc two. Open in Junon, where Tifa and Barret are held captive, and Sephiroth has just summoned Meteor. Cloud is missing. The world is about to end and two bitter enemies — the Avalanche remnants and the Shinra company — each have their own mutually-exclusive ideas on how to attack the problem. During this slice of the game you get to visit a lot of different locations (one of which, I must emphasize, is outer frikkin’ space), fight the massive WEAPON monsters, score some real victories over Shinra, and solve the mysteries of Cloud’s tragic backstory.

There are reasons to not do this, of course. There’s too much connective tissue hanging off either side of the story if you pull a chunk right out of the middle. Who is Cloud and why do we care why he’s missing? Shinra’s evil, but they seem to have a good idea on how to destroy this catastrophic meteor, so why are we fighting against them? Wait, we’re hinging all our bets on some lady I don’t care about who died in a flashback?

These aren’t insurmountable problems, but they are much harder to write around than just pulling out the first one-sixth instead. And doing so would only barely appease me, and piss off pretty much everyone else. And it still wouldn’t be an actual remake of actual FFVII.

But, okay, they didn’t do that and were never going to. Remake is just the most boring part of FFVII, stretched out to 34 hours and sold to me at full price. Fine. But what actually is it? If my enjoyment of it hinges so much on calibrating my expectations going in, what do those expectations need to be?

FFVII Remake is the first episode in what will be a much larger game.

This was the most popular consensus for most of the time Remake was in development. The first episode of the game was going to take you to the edge of Midgar, and each later episode would have some portion of the game, and the full game would be all the episodes together. The kind of thing Telltale Games used to do before they imploded.

I think this sort of structure would pose a lot of pacing challenges, both in terms of gameplay and in story, but nothing a bit of smart work couldn’t solve. You’d probably have to reduce the game down into a series of stand-alone areas. Maybe episode 2 is the gang figuring out how to infiltrate Junon. Episode 3 could be a contained open world experience, sort of like FFXV, encompassing the important areas of the big continent and ending at the Temple of the Ancients. Carrying on in this way you might finish the game in five or six episodes, assuming you can keep the playerbase interested that long. (And, ah, you can’t. That’s part of the reason Telltale Games imploded.)

I think it’s clear Square-Enix didn’t do this. It might have been reasonable to think this was their intent, while the game was in development, but Remake‘s release day killed the notion entirely. For one, you can’t have a 34-hour “episode”. If we’re thinking six episodes, at 34 hours a piece, and I haven’t mishandled calc.exe recently, that’s like 63 thousand hours. Some poor slob is going to buy the Collector’s Edition Platinum Box at the end of that and nobody will ever see him again.

For another, a multi-episode work implies a strong connection between each entry. Telltale didn’t have to think much about this because, mechanically, your character at the end of one episode is the same as that same character at the start of the next. They just have to import a few booleans to know what sorts of key decisions you made, and that’s good enough.

But in Remake, I ended the game with a nearly full set of maxed materia, a closet full of armaments, 90 mega potions and more than 6000 hit points. If they balanced subsequent episodes around importing characters from the older ones, the numbers would have to be ramped up to absurd levels to keep pace. And if they aren’t, then anyone who imports anything just steamrolls the rest of the content. Old Ultima and Bard’s Tale games did these sorts of imports but it’s folly to think a modern RPG series could.

No, Remake is not just the first chunk of a larger game. You will not start Remake 2 with 6000 hit points.

FFVII Remake is the first game in a much larger series.

The first, much stupider reason this fails is, we already have an FFVII series. It’s already a game which spawned a greater universe of spin-offs and side content. Living in a world where we have an entry in a larger series that is itself two separate series makes my brain hurt.

But, it’s not an unreasonable way to present a video game. In this scenario, we’re to look at Remake as something more along the lines of a modern Tomb Raider or Assassin’s Creed game: it’s the first game in a series of games which, when taken together, complete the overarching narrative of a particular character or setting. In this case, the story of Cloud and Sephiroth and a big-ass meteor and some green glowing rivers.

In a longer series like this, it’s reasonable to expect that my hero will begin each new entry at a beginner power level. I’m not going to play Horizon: Forbidden West and be all butthurt that Aloy is starting over with just a bow and her magic iPod. There are certain contrivances we accept as part of the necessary reality of sequels. I’m not excited for Forbidden West because I want to wtfpwn all the new robot dinosaurs with stuff I earned in Zero Dawn.

No, the reason I’m excited for Forbidden West is because Zero Dawn staged an intriguing setting, introduced a likable character, and told one of my favorite sci-fi stories in a long time. Aloy experiences a full character arc, and so does the world where she lives. The story has a beginning, a middle, and an end. It is complete. The story is about the things Aloy accomplishes, and the ending implies there is much still left to accomplish — but doesn’t say what. Not yet. Because that’s a story for another time, and another game. There’s a post-credits sequel hook, sure. But the story I played was done and it was satisfying.

I can’t say any of the same things about Remake. If you really squint you might get away with claiming that Barret starts his character arc, but certainly nobody else has one. And Avalanche accomplishes nothing — absolutely nothing — during the game. It’s established early on that the things they think are victories are just traps set by their enemies, and then their enemies win, and then their enemies dick them around, and then their enemies kill 50,000 people, and then their enemies get away.

And then some new enemy shows up out of nowhere, with no proper introduction, and you have no idea who he is or what he wants besides some vague unexplained connection to the hero. He shows up from so far out of left field that Knowledgable Magic Girl has to literally step forward and insist he’s the real threat you should be fighting, and there’s no time to explain why that’s the case, just take her word for it. So you fight him and you don’t win, and he says some forboding stuff, and then he vanishes. The heroes regroup in their take-a-breath scene, roll credits.

I liked Remake‘s gameplay, but not really enough to justify playing future games in this series, based on what I played in this first entry. And while the story is made up of some really great individual scenes, the whole of it profoundly sucks. I feel like these heroes didn’t resolve anything, didn’t change or learn or grow at all, and don’t really deserve a chance to expound on all the little story teasers strewn throughout the game. If I didn’t already know where this story was going, and who these characters are, I would have zero interest in continuing on their journey with them.

A game needs to have its own identity, and its own message, even if it’s being positioned as the first step in a much larger story. The only identity this game has is, “Be sure to buy the sequel, coming soon!” And from there we spiral down into a discussion about exploitative practices game companies use to vaccuum dollars out of cunsomers and I end up very depressed.

There are two moments in particular I think really betray future titles in the Remake series, if that’s what we’re calling it. The first happens early on when Cloud starts seeing Sephiroth. First of all, blowing the Sephiroth wad so early is a mistake from a pure storytelling standpoint. Part of why the Midgar section of FFVII is so effective is because you can really take it as a story about terrorists vs. government, before the other shoe drops and the stakes get raised. But okay, Sephiroth is perhaps the number one bankable character in this entire franchise, and we’re not going to save him for later. Fine.

But they don’t just sprinkle Sephiroth into the story way too early. They show him ominously turning and walking through a wall of fire. This is one of the most iconic visuals in any video game ever released, but it’s also a key moment in this actual story. The fire is Nibelheim, Cloud and Tifa’s hometown, which Sephiroth has set ablaze. It’s the moment that galvanizes Cloud’s hatred for Sephiroth. Showing it in Midgar, one hour into the game, robs all the emotional impact of its eventual reveal in context much later. It doesn’t serve the story at all, it serves someone on Square-Enix’s board of directors, who probably played FFVII in college, who sent up a memo saying “hey you gotta show sephie walking thru flames cuz that’s cool”.

The other moment happens when Shinra drops the Sector 7 plate in an attempt to squash Avalanche. I was actually very impressed with how Remake handled this scene. Part of the FFVII experience is having a brutally violent event right in the middle which changes the stakes. They’re obviously not going to have Sephiroth kill Aerith in the Shinra Building, though, so the writers tweaked some things so the plate drop would have a similar impact. The build-up is much larger now, for one. Cloud and his gal pals can see the plate engulfed in gunfire and explosions the entire time they’re working their way through the train graveyard. Their cause is urgent and they’re stuck and you can feel their desperation as you play.

Then, the entire sequence is expanded tremendously. In FFVII you just climbed the pillar and fought some Turks, and established that your Avalanche buddies died along the way. In Remake you fight your way up the pillar, then play an interlude as Aerith moving through the scrambling, terrified populace to save Barret’s daughter. The scenes with the Avalanche kids are more drawn-out, and pack more of a punch because of their expanded roles in the game’s plot. (There’s a whole away mission where they drag you off to meet Jessie’s mom and get into a motorcycle chase and fight some Metal Gears that wasn’t in the original.) There’s an interjection into the Turks’ helicopter where they have a brief crisis of conscience about murdering thousands.

Then the dumb thing I hate happens.

After the dumb thing I hate, Cloud and his friends decide to go back into Sector 7 to see if anyone survived, and you experience the aftermath of the plate drop first hand. It’s a really effective scene that segues into a new dungeon area which, in turn, reinforces all the reasons you already have to want to take down Shinra. The plate drop really is the fulcrum of the plot, with a lot of unpacking to do on either side of it.

The dumb thing I hate is a maybe ten second cutaway shot, just after the plate falls and causes obscene destruction. Standing above the city, watching the devastation below, hammering his fists against the ground in rage and grief is… a stuffed cartoon cat.

Who we never see again.

Players familiar with FFVII will know this cat is Cait Sith, a robot operated by Reeve, the only member of Shinra’s upper management with a conscience. In both versions of the story we see Reeve begging the president not to go forward with mass-murder. For these players, this is a cheeky winky-nudge moment. “Oh cool! Cait Sith’s here!” A moment which is tonally inappropriate given the scenes we just saw and are about to see.

For those unfamiliar, it’s a confusing trip to the wiki or one of the subreddits to find out if anyone knows wtf is up with the weird cat.

In FFVII, Cait Sith sort of hops into your party at random upon arrival at the Gold Saucer theme park. He introduces himself as a fortunetelling machine and then insists on barging into the story when the fortune he gives Cloud turns out to be too intriguing to pass up. He’s a goofy character who attacks with dice and megaphones, and unlike every other character you’ve met up to this point, there’s no “Cait Sith required” setpiece in Gold Saucer where you learn more about him.

The twist comes later, when Cait Sith steals an item from the party and delivers it to Shinra, and then reveals he’s been a Shinra spy all along. He also reveals the party can’t dispose of him. If they don’t agree to let the robot spy continue traveling with them, the robot’s operator will be forced to retaliate against Barret’s daughter, who is within their reach.

This is an effective moment of storytelling. The reason it’s such a gut-punch is, after leaving Midgar, it’s easy for the player to start feeling like they’re always a step ahead of Shinra. You infiltrate their parade, stow away on their boat, swipe some materia from one of their defunct reactors, and fly off in a seaplane they wanted to commandeer. Most of the second act of the game is introducing and reinforcing individual character stories, and this “oh crap, Shinra had my number all along” moment is the abrupt end to that. It’s the point in the story where the lollygagging is over and the heroes have to get back on task and actually start handling some heavy stuff.

Of course, the twist only works if you don’t know Cait Sith is from Midgar, and now every player does. The second, smaller twist, when you learn it’s Reeve at the controls, is also spoiled, because there’s only one guy at Shinra who spoke out against the plate drop, and we’ve seen his robot cat crying about it.

Remake is full of little moments like this, where it robs later series entries of their story beats to fluff up the first installment. It’s the over-eager author who has a big basket of worldbuilding notes and he just can’t wait to breadcrumb it all out for you. As the introduction to a series, it’s lousy because it lacks its own identity and tries to make up the difference by touching its nose and giggling, “but just you wait!”

FFVII Remake is a standalone title.

The game already doesn’t work as a standalone title, for reasons I’ve gone into re: the character’s don’t achieve anything. But I think it’s interesting to consider the game from this angle, because if you really wanted to condense the full story of FFVII down into just Midgar, you very easily could. All of the aspects of the original which made it a globetrotting adventure can be boiled down into Midgar if you really wanted to.

You wouldn’t even have to cut any of the characters. In this hypothetical game you’d rescue Red XIII from a research facility much earlier in the plot. Yuffie and her dad are running a materia-thieving operation out of the Sector 3 slums and sending the proceeds back to Wutai, and Yuffie joins your party because she wants to stop the “Wutai is behind the terrorists” propaganda. (Also she wants to steal all your materia lol.) Rocket Town and Gold Saucer can be moved into Midgar’s borders without incident, so Cid and Cait Sith’s storylines are in tact. Shinra activates Vincent as a secret Turks operative after you beat up Reno and Rude too many times, but he switches sides because he’d rather pursue his personal vendetta against Hojo.

The major plot points we need to keep the plot coherent are: Cloud explains the Nibelheim incident; Cloud receives the black materia at the Temple of the Ancients; Sephiroth kills Aerith to stop her from activating Holy; Reunion occurs, causing Cloud to hand the black materia over to Sephiroth; Tifa finds Cloud and helps put him back together; Cloud and co. discover Aerith was successful in activating Holy after all; Cloud and co. topple Shinra and Sephiroth and save the Planet.

This is all stuff that can happen in or around Midgar. The Temple of the Ancients is actually a ruin in a sectioned-off area of Sector 2, and its conduit directly to the Lifestream is why this geographical point was chosen by Shinra to build Midgar upon. A second site accessible from somewhere near the Sector 5 church contains the ruined city of the Ancients, which is why Aerith always felt drawn to the place, and ends up being where she dies. Instead of raiding Shinra sites around the world looking for Huge Materia, we can raid some of the other six reactors instead. Northern Crater is significant to Reunion because it’s where Sephiroth’s body washed up after being thrown into the Lifestream. Instead, let’s have the suction of the eight mako reactors draw him to a point deep beneath the Shinra building, which Shinra accidentally makes accessible when the Sector 7 plate shears a hole into the abandoned cave system there (complete with icicles and whirlwind maze!).

It’s very possible to tell the complete story, in a satisfying way, with all nine characters, and only have to leave Midgar once for the Nibelheim flashback. And I think you could do it inside of 40 hours. (The original did, after all!) A remake along these lines would make The Zodiac Age look like a slapdash effort.

Alas, for what might have been.

FFVII Remake is a companion piece to the original FFVII.

Now we come to it, the only actual reasonable way to consider Remake‘s story at all: it exists alongside the original and soft-requires you to have already played it. As it happens, it’s impossible to actually discuss what happens in Remake without referencing the original FFVII or speculating what future events from FFVII they will or won’t have to change.

The first clue is the biggest one: nobody actually knows what the hell happens at the end of this game. Long before I played it, I read the spoilers and watched the video essays, and clicked the black boxes in my own Discord that people thought I wouldn’t click. There are lots of theories and potential explanations for the last hour or so, but nobody actually knows what happened, and cannot know until more games come out.

Remake pretty faithfully follows all the story beats from FFVII‘s first act. Indeed, I referenced this earlier as a major problem, because the characters don’t actually accomplish anything during the first act. There’s another reason Square-Enix chose to do this, though: they wanted to write in the whispers.

The whispers (colloquially known as “time ghosts” across the Internet) are supernatural beings whose job is to ensure that everything happens the way it’s destined to. Their primary concern seems to be making sure people don’t die before they’re supposed to, or live after they’re supposed to have died. Remake sticks to the established plot so closely in order to make it really stand out when the whispers show up to alter reality if it looks like something out-of-place might happen.

Of course the only way you can know what’s out-of-place is if you’d played FFVII before. Otherwise they’re just random time ghosts.

Whispers show up to make sure Aerith and Cloud don’t get shot by Turks in the church. They show up again to miraculously heal Barret from a fatal stab wound. (Ask me how I know it’s fatal.) They show up again to prevent Wedge from escaping the Shinra Building, after he was supposed to die in the plate catastrophe hours earlier. They’re neither good guys nor bad guys, they’re enforcers.

The final area of the game, which explodes out of Sephiroth’s butt with a gooey metaphysical splat, reveals to us that he wants to destroy the whispers in order to defy destiny. The inferrence is that he wants to do this because he knows his master plan re: summon Meteor is destined to get foiled by Spikehead and Gun-arm and their dumb robot cat. And we do see that the whispers are destroyed; the Remake epilogue shows Biggs waking up in bed, having survived injuries at the plate support tower that were supposed to have killed him.

This does seem to be the only interpretation of Remake that makes any sense, and it only makes sense with the context of already knowing what’s supposed to happen in FFVII. I really shouldn’t have to explain why this is an absolutely awful way to approach a remake.

The speculation from here stretches out in every direction, as fanboys grapple for any explanation their fevered fanfic-writing minds can concoct, but I find them all underwhelming. Did FFVII already happen in a different timeline, and Sephiroth somehow traveled from that world to this one via the Lifestream, and that’s why he knows the future (and maybe also Aerith kinda does too)? I don’t care, because that story is terrible. I don’t want to play the multiverse clones of the guys who already did the adventure. I don’t care about fighting Sephiroth now that I know he can sidestep timelines or whatever. I don’t care if Meteor destroys this reality if there are already other realities out there where that was averted.

The writers would have to be geniuses to resolve this story, and its unmistakable connections to its ancient forebear, in a way that will satisfy me. And I just don’t think they are. I think each successive Remake sequel is going to get more and more confused as they try to detangle themselves even as they run out of clever foreshadow nuggets to drop in.

I won’t even get into this.

Remake is going to inspire a lot of players to go back and experience the original FFVII. I’ve already stated I think that is fantastic. FFVII is one of the all-time greats, and I love thinking of a new generation of people who are going to stamp their feet when Yuffie strands them with no materia, desperately watch the seconds fall off the Emerald WEAPON timer as Knights of Round drags on forever, or play Aerith’s Theme at their wedding.

Some of those people, though, are going to re-visit Remake after their inauguration, and see the Cait Sith scene, and go, “Oh, that’s why the cat king was there. It’s Cait Sith. Huh, kind of dumb that he’s there, actually.” And they’ll be in about the place I am now. And they’ll still have two or three of these games to, er, look forward to.

Thanks for reading the longest thing ever!

Outer Wilds: Thoughts, Spoilers, and Unfair Comparisons

I bought Outer Wilds because people recommended it to me. These were not normal “hey you might like this” recommendations. More than one person who knows my tastes pretty well mentioned they thought this could be one of my all-time favorites. It didn’t happen in conjunction with some other game they knew I’d played, either. Dudes were not all “Oh hey, Brick said he liked Death Stranding, if that’s the case he’ll love this other thing!” I checked online to price the game on Steam and noticed that Limited Run Games were going to be doing a PS4 pressing, and I figured if the game really were as special as all that, it might be nice to have a hard copy. So I pre-ordered then waited like eight months.

It was worth the wait, and I’m glad I will have Outer Wilds‘s physical spine staring at me from my shelf, because this is one I’d like to be reminded to go back to from time to time. A kind of warm, familiar reminder you can only get by glancing at a shelf, and not from, say, a massive digital list of titles in tiny sans-serif font.

Outer Wilds was fantastic, and exactly the sort of game I needed right now. I went into it completely blind, and that’s how you ought to go into it, too. That said, it is not the life-moving incomparable experience I was maybe expecting after close to a year of self-imposed radio silence. I think it would have been okay for someone to tell me what genre the game was in, so that’s what I’m going to tell you: it is an adventure game, in the tradition of Myst, Riven, and Obduction.

I think this is important to say up front, because often when I encounter these “you have to play it completely blind!” games, the advice is trying to preserve some genre-defying new gameplay secret, or is one of those wink-nudge meta storytelling jobbers that’s all the rage these days. Outer Wilds is neither (thank goodness). The reason to go into this one blind is, while it is a fairly conventional adventure game, the breadth and direction of its exploration-based gameplay is quite unique. There are games I can point to and say, “Outer Wilds is kind of like that!” but you’ll have more fun if you don’t know anything about the specifics of what you’re getting into.

One thing I will say, by way of review, is that this game’s controls are very bad. They start out feeling quite cumbersome and a lot of the early game is spent getting used to them. The game is pretty short, so there’s not enough time to really master them over the course of your adventure. Even if there were, this isn’t the type of game that really rewards mechanical mastery. There are arguments to be made that bad controls have a point, and there may even be a smudge of merit to those arguments. I really dislike bad controls, though, and one of the things I love about adventure games is they’re usually designed without reflex challenges. People can play them who are bad at Mario or Doom or whatever. This game departs from the Myst template in that you have to be considerably more “in the moment” than just walking around and engaging with things intellectually; sometimes you have to also be good at playing the video game, and if you’re not, you could end up losing a lot of state.

SPOILERS FOR OUTER WILDS BEGIN NOW

I don’t feel like I want to break down the story in Outer Wilds, although it really was quite good, and the game was quite good at telling it. I also don’t have much to say in the vein of pointing out gameplay problems and possible fixes for those problems, like I usually do. What I want to do instead is contrast Outer Wilds against a few other popular adventure game archetypes to examine 1) what makes Outer Wilds so unique in this field, and 2) why Outer Wilds kind of almost broke as a gameplay concept.

I like to get stuck in adventure games. I love to solve puzzles, and I am incredibly stubborn, and I have a reputation as someone who will stay stuck on a puzzle far, far longer than what is usually considered reasonable. There are heights of euphoria to be had here, in that moment of finally figuring something out you’ve been stuck on forever, that cannot be reached outside this genre. It’s not like banging your head against a boss fight for six hours in Dark Souls, because that’s just a disconnect between designer intent and player skill. And it’s not like that moment where some badly-conveyed gamaplay quirk clicks with you, as happens sometimes in combat-heavy games with lots of fiddly controls. In a well-crafted adventure game, the designer knows you’ll be stuck and probably has a good idea of where, and engages you in a battle of wits. (And observation, and experimentation, and maybe a bit of trudgery. But mostly wits.) When pushing through these kinds of barriers in other games it often feels like a relief. An end to frustration. And I think that’s the difference: in a really well-designed adventure game I don’t get frustrated. I have faith the thing I’m missing is out there, somewhere, and the designer has pointed me at it, if only I can see the sign.

(It’s possible, of course, to lose faith in this process, but Outer Wilds didn’t do that, at least not directly.)

I completed Outer Wilds over the course of five play sessions. Three quite long ones, then one short one to wrap up the story, then one even shorter one to try out a few weird things I thought could get responses. (And did! Much to my delight!) The second of these sessions was actually pretty awful as a gameplay experience. It was seven hours long, and during that time I did not discover anything at all. I didn’t make any progress. The state of my game knowledge at the end of that seven hours was the same as at the start of it. I eventually did think my way through one of my barriers, but the thing I found on the other side also didn’t advance my knowledge of the world at all, so it didn’t feel rewarding.

So I shut the game off that night and spent some time laying in bed thinking about the comparisons I’m about to make.

An adventure game is, in the most abstract, a set of doors you have to unlock. There’s a big door at the end which completes the game, and it’s locked, and the path between your starting location and the big end door is frought with smaller doors with smaller locks. The various subgenres of adventure game differentiate themselves by changing the actual shape and configuration of these doors and locks.

At one extreme you have something like Monkey Island. The doors and locks in this style of game (which sits outside of the general concept of Mystlikes, but is still useful for illustration) are linear and tangible. You can’t open Door A without finding the corresponding Key A, and you can’t reach Door B without first going through Door A. You advance in this game by using the correct object from your inventory on the correct object in the game world. The puzzles usually “make sense” in terms of Monkey Island being a comedy story, which is to say, they kind of suck from a logical real-world interaction point of view. When you get stuck the game is often reduced to just trying every inventory item on every game object. The game tries to prevent this by giving you a lot of inventory objects, but there are never so many of them that this method becomes impossible; the possibility space is never bigger than maybe 30 things. This is boring and stupid but it eventually does work, just by pure chance, and you make progress.

Monkey Island: getting unstuck is BORING but making progress is GUARANTEED.

As you might expect, Outer Wilds is at the opposite end of this spectrum of “how do I get unstuck” and “does getting unstuck help me progress” from Monkey Island, but let’s look at two more data points along the way.

The next step is something like RHEM. The doors and locks in RHEM are usually very literal; they are physical doors locked by physical objects, and you have to move those physical objects in some manner in order to remove the door. There is no physical inventory in RHEM (at least, not in the first game), but you still do collect “keys” for the locks you encounter. Many of the physical locks involve entering combinations, and you learn what the combinations are by finding diagrams or working machines elsewhere in the game world. Once you figure out the proper way to use a machine and you write down what it tells you, identifying the corresponding lock and entering the combination is often trivial. The keys, therefore, are often intangible. If you have good notes from a previous playthrough of RHEM, you can skip a lot of the game because a lot of those “keys” are already “collected”. You can’t skip all of it, though, because there are still actual physical doors, and their movement changes what you can do and where you can go. Getting unstuck in RHEM involves trying every physical combination of pieces you can reach and making sure you see them from all angles. You’re working with the knowledge of where you’re stuck, though, so you have a generally good idea of where you should be applying your efforts. Like Monkey Island, though, getting unstuck always involves moving forward, because it means you can reach a new area, or you’ve gained some new knowledge you can now apply elsewhere, or both.

RHEM: getting unstuck is REASONABLE and making progress is GUARANTEED.

The original Myst would be the next step along this line. Myst has many puzzles but very few of them involve physical motion; they’re almost all knowledge-based doors and locks. The possibility space is too enormous to experiment with, here. Two of the first puzzles in the game involve picking three individual star charts out of a library of millions, or playing the proper sequence of notes on a 36-key piano. Like RHEM, opening any door is trivial once you know how to unlock it. Unlike RHEM, actually reaching a given door is also trivial, since the only things in your way are other doors you’ve already trivially opened. Famously, the big last door in Myst is reachable from the beginning of the game, and you can complete the whole adventure in minutes if you know precisely what to do. Which you will, after a single playthrough. The puzzle areas of Myst are also small and self-contained. Getting unstuck involves interacting with things in that specific area until you figure out what they’re doing and what that means for you. Once you know that, you’re off and running again to the next thing.

Myst: getting unstuck is QUITE REASONABLE and making progress is GUARANTEED.

And now we land at Outer Wilds. By necessity, the game is every bit as open as Myst is. Every inch of the game world has to be reachable inside of just a minute or two, because of the 22-minute cycle that drives the game’s central conflict. There’s no inventory and only two examples of what I’d call a physical lock. If you have your notes from the previous run you can not only go win right away, you can do anything in the game right away. Also, you don’t even have to take the notes yourself; your spaceship does that for you. These notes end up being the keys you collect. Bits of crucial information are scattered across the solar system, and as you begin compiling them, knowledge about how to find new areas or traverse certain obstacles is gained. Once you know how to reach the ocean planet’s core you can travel there whenever you want, but you’ll (probably) never figure it out without helpful clues from elsewhere in the system, several of which are themselves locked behind other knowledge keys from elsewhere.

This ends up being the gameplay loop of Outer Wilds: go explore some new place, stack up all the clues now available to you, sift through the ones that seem to point to other places, go get the clues from there, and then repeat this process until you have the small handful of clues which enable you go open the big door at the end. (Which, again, is something you can do from power on, if you know the exact sequence of actions.)

In my first play session this went really smoothly. There are so many “easy” places to explore that you will spend many hours just stacking up clues and chasing down leads and generally blowing through locks and doors without thinking much about it. At first it seems like there are a dizzying amount of loose story threads, but as you work these start to connect to each other and dwindle down to a single common goal.

The problem with having an “inventory” made up of intangible clues became apparent during my second play session, that gross one in the middle where I batted the game around for seven hours without getting anywhere. See, physically reaching and reading a clue is one way to get through a door in Outer Wilds, but another way is to reach the door first and then experiment with ways to open it. Sometimes the door yeilds and you have an amazing moment of discovery. Sometimes the door remains steadfast and you decide to revisit it later, when you’re really stuck in all directions. I had unlocked many doors without finding their keys, but those keys were still out there, often hidden behind other doors I hadn’t unlocked yet.

When you’re stuck in Outer Wilds, it often can help to continue experimenting with whatever you’re stuck with. But it’s just as likely that you won’t grok the solution without finding the key. There’s no strong correlation between where you find a key and where you find the door it unlocks, either; the nature of the game involves clues being left in every corner of All Over The Lovin’ Place. And when you do get stuck, and push against all the locked doors you have left, it becomes increasingly likely that the key you find inside unlocks a door you’e already opened.

Outer Wilds: getting unstuck is ESSENTIALLY UP TO CHANCE and making progress is NOT GUARANTEED.

This happened to me with the anglerfish.

I visited Anglertown before I had seen any of the clues that related to how to get through it. Of course, on my first visit, I got eaten. But I had work to do there, so I flew directly back, and this time I decided to look for clues on the anglerfish themselves. They have big glowy dingle-dangles, and enormous fanged maws, and cold dead eyes. You can actually get a pretty good look at them, since they’re larger than your ship and don’t aggro right away, even when you’re close. Since proximity didn’t trigger getting eaten, and it appeared that they were blind as well, the only other thing that made sense was sound. The only sound you can really make in this game is by blasting your thrusters. I tested this out by pointing my ship directly at an angler’s face while far away, firing slightly, then drifting toward it. The thing didn’t aggro until I’d literally bumped into its butt, and it immediately spun around and ate me, but this time I had a solid idea: if you knew where you were going you could just drift toward your destination without alarming any nearby fish. This got me to where I needed to be, but it felt like a kind of kludge-y solution. It felt like the kind of gameplay you might cobble together when you don’t know the real solution. There was one location in Anglertown I hadn’t reached yet, and it didn’t seem possible to reach it without using my thrusters near a fish, so I reasoned that I had stumbled upon about half of the solution and that I’d come back later once I learned the rest.

In a city elsewhere in the game, some NPCs left a helpful note about Anglertown. As children, in another part of the solar system entirely, they used to play on a huge dead anglerfish skeleton, and they thought the rules of the game they used to play could help them reach the Anglertown location I was trying to get to. So now I had a lead: go back to that area and find the skeleton. The skeleton is in another NPC town and reaching it involved actually learning the layout of the town (which was a non-trivial task!), solving a minor puzzle, then traversing some platforming challenges in a dark maze on a time limit. It took three tries to reach the skeleton (see “bad controls”, above), and when I finally did I learned…

…that anglerfish are blind.

Which I’d already intuited.

The actual solution I was “missing” is that the physics in Anglertown just so happen to “always work out” so you drift past the main gauntlet of them in the final area without firing any thrusters. It doesn’t seem like it should work, but it does, and if your destination in that area were somewhere else, or if the anglers had a different starting configuration, it wouldn’t work at all. So that’s lucky! Once I did it correctly the first time I realized the skeleton clues were saying a little more than just “lol they blind tho”, but decided I still didn’t like the solution because drifting slowly through a large area is dull. (And there’s even a reason this portion of the game needs to be time-consuming, which made sense in retrospect, which also didn’t improve my opinion of the solution.)

There were other areas in the game where I invested a great deal of time and elbow grease just to happen upon a piece of choice info I’d already figured out. As a matter of fact, as the game’s various plot threads and daisy chains begin converging, there seems to be maybe three or four Really Big Clues you need. One of these is a Myst-style combination that unlocks the big door at the end. Another is an incredibly specific clue I imagine completely defies intuition, but the possibility space around the door that clue unlocks is… well, not small but small enough that I could have stumbled upon the rules after an hour or two of playing with them. (And I spent longer than “an hour or two” reaching the Really Big Clue in question.)

The other Really Big Clues were things I figured out by experimenting with the game, sometimes quite early on. The doors that held the clues to these last few things were harder to get through than the doors the Really Big Clues opened, which meant I spent a lot of the second half of my time working on things that… ended up not mattering. So they didn’t feel good to discover, and didn’t help me progress.

This has an effect on the state of euophoria I mentioned. When you take down a tough puzzle in Monkey Island, it’s usually because you brute forced it, but it feels good because you immediately get some funny dialogue and then you progress on to the next thing. In Myst and RHEM the tough puzzles all open new areas, or reveal new information you can’t progress without. That doesn’t always happen in Outer Wilds. Sometimes I cracked a tough puzzle only to keep standing still. The list of stuff I could work on now was the same as the list of stuff I already had, minus the thing I’d just cracked.

I think there’s a lot of merit to this style of information-only progress gating, without constantly falling back on the old standby of writing down combinations to punch into button pads. (Although, there are a few locked one-way doors in Outer Wilds I kind of wish just gave me a combination to open them from the outside, the first time I reached them.) If there are other games like this, I imagine they have the same issue. And it’s a large issue. I mean, no matter how much I praise this game, or how much I enjoyed all the other parts of it — including all the parts outside the scope of this article — there’s a seven-hour black hole in the middle of it where I didn’t learn anything and wasn’t able to apply what I already knew.

It’s the first adventure game I can remember with well-designed puzzles where I did not enjoy getting unstuck. If nothing else, it was a new sensation for me.

I think I may have one more spoileriffic Outer Wilds post in me, where I break down how effective the story was, but we’ll wait and see if Next Week Brickroad wants to write it.

Thank you for reading!

My Custom Villainous Decks (and the rules they break)

Disney Villainous is an asymmetric board game. That means the game contains a variety of characters who play by different sets of rules all at the same time. This might sound insane if you’re not used to the idea, but it’s actually pretty common across all form of competitive games. Consider Street Fighter: one of the rules is “this button does a kick”, and that applies to all characters. However, another rule is “this button sequence makes Ryu do a fireball”, and another is “this other button sequence makes T. Hawk do a command grab.” There are no button sequences that make T. Hawk do a fireball or Ryu do a command grab. The two characters play by different rules, but that’s okay, because the central framework of the game — e.g. the set of core rules that applies to all characters — is robust enough that the game still works.

Another way to think of this: which of the above characters is breaking a rule? The game has rules about the various range of attacks; Ryu’s fireball breaks them by being a projectile that travels across the screen. The game also has rules about avoiding or teching throws; T. Hawk’s command grab breaks them as well. In fact no character in the game follows all the rules. It’s the part of the game’s design that make each character unique. This is something I immediately noticed about Villainous, and after a couple of games I knew I would be playing and thinking about it for a long, long time.

The asymmetry creates another property about the game I really like: it is endlessly expandable. Three expansions have been released so far, with a fourth on the way, each one introducing new villains that break the rules in fun and interesting ways. Even sticking with just the villains in the main box, though, I identified what must have been part of the game’s design doc: “there is a core rule set, but each villain breaks it in one way.”

I mean, each villain breaks it in lots of ways. That’s what cards are in board games; specific instructions on how you’re allowed, for that one play, to break the game’s rules. But each villain also has a special gimmick, something unique to just them, which breaks one of the game’s major rules. Other players are responsible for playing Heroes to your Realm, except for Captain Hook, who can (and probably must) do it himself. Realms have one locked location that becomes unlocked with a certain card, except for Ursula, who toggles back and forth between two locked locations. These deviations became more pronounced with each expansion. The first had a villain building a new kind of deck with cards from his own one. The next had a villain splitting her Fate deck into separate piles. All three in the expansion after that have new kinds of tokens and feelies to track various mechanics.

When I started designing my own custom decks for villains from the Final Fantasy series, I made “each villain breaks one rule” my own cornerstone. This post is a brief summary of the decks I’ve made and the rule each of them breaks. Buckle in for a long read, I’ve been at this for well over a year now, and there’s fourteen of them.

Garland (FFI)

“In Villainous, all cards remain in play, either in a Realm, or a player’s hand, or a deck, or a discard pile.”

Garland breaks this rule by having cards that remove Heroes from the game entirely. As one of the first decks I designed, I didn’t yet have a good handle on the ins and outs of how a well-constructed Villainous deck should look at various levels of development. The first round of feedback for Garland was that the deck felt absurdly powerful, and had a lot of interesting moving parts, but he wasn’t actually winning games. A lot of discussion, testing, observation and many rounds of changes went into getting the Garland deck into a position where it was fun and playable and an actual threat to win.

Part of the puzzle was giving Garland several cards that not only defeat Heroes in his Realm, but removes them from the game entirely. Garland’s Fate deck is as punishing as his villain deck is powerful — it helps give the deck a brutal, unflinching nature, kind of like the mean old NES game it’s based on. When Heroes kept cycling through his Fate discard pile and back into his Realm, Garland had no clean shot at victory. Giving him the ability to strike a fatal blow to the opposition makes him feel even more powerful but gives him an actual chance at the win.

Playing as Garland, it feels incredibly cool to send a Hero packing for good. Playing against Garland, there’s a feeling of horror at the same moment. It’s dark, it’s mean, and it’s exactly what Garland would do.

Sho-gun (FF Legend II)

“In Villainous, every deck breaks a rule.”

Sho-gun’s deck doesn’t break any rules, so that’s the rule it breaks.

I made this more or less as a joke deck, because I thought having a bunch of cards with all Game Boy graphics, based entirely around a quirk in the translation meant to avoid drug references, would be hilarious. The deck is still intended to be viable and I’m sure with some playtesting it will be.

Golbez (FFIV)

“In Villainous, you may only play one card at a time.”

Or two cards, if you’re on a space with two Play a Card buttons. Unless you’re Golbez, then you can play as many cards as you want. Indeed, you have to, on your final turn to win the game. You need to have four special cards in your game, and the Power to play them, and not have your play blocked by one of the Heroes in your Fate deck that can do that. This can be a hefty restriction, since you usually only have a hand of four, and every card you’re holding on to until your last turn is a spot not being put to use for something more useful you can play in the meantime.

Most villains have to discard lots of things they’d rather play, in order to get to the cards that are really important. Golbez still discards a lot of cards, but he also gets to play a lot more if he wants. This has always been one of the easier FF decks to learn and to win with, and this simple rule-break gives Golbez the flavor of a master manipulator who is, indeed, holding all the cards.

Dark King (FF Mystic Quest)

“In Villainous, you can only take one Fate action per turn.”

Fate is very, very strong. For most villains, it’s your only opportunity to complicate another player’s progress. It’s something every villain can do, and it’s something the game is very careful not to restrict. However, it’s also something the game is very careful to police. You can only do it once per turn, because you never have access to more than one Fate button at a time. Lots of villains have cards that grant extra actions, but these cards are always carefully worded to ensure the extra actions aren’t Fate actions.

Dark King begins his first turn by Fating someone. He has to. And he starts every turn that way, for the early game. Eventually this swarm of Fate stops and Dark King moves on to bigger and better things, but he can turn the early game into a real train wreck. In fact, this is the point. In the very early game, players tend to work towards their own goals, only using Fate when the opportunity arises. Nerts to that. Dark King throws monkey wrenches at everyone, throws up speedbumps all over the place, and gives everyone something to think about while he climbs the long ladder up his win condition.

I don’t know whether this breaks the game too much, or not. It’s a relatively new deck, and anything could happen. There’s a weird edge case where, with a potentially self-destructive sequence of actions, Dark King can perform three Fate actions at once. That might need reigning in, or maybe it’s fine as a once-per-game oddity. We’re keeping an eye on it, but it’s a characteristic I’d like the deck to keep into its final state.

Gilgamesh (FFV etc.)

“Rules are dumb, let’s just break all the rules, what’s the worst that could happen.”

Well, the worst that could happen is the deck becomes totally unplayable, or breaks the game in fundamental ways to the point where play cannot continue, or a winner cannot be definitively known, or bees swarm in through all the windows and sting everyone to death. I don’t think Gilgamesh (aka Greg, aka Glorgamorsh, aka Gilgafred, aka @FF5ForFutures) is at that level of tomfoolery, but he might be close.

After his initial introduction in Final Fantasy V, Gilgamesh just started showing up in whatever games he wanted, with no respect for the cosmological rules of whatever universe he happened to pop into. Indeed, he even started showing up retroactively when pre-V games got remade. Sometimes he’s a joke boss, sometimes he’s a sidequest, sometimes he’s a $4.99 DLC adventure, I think at least once he was a summon spell. Gilgamesh doesn’t know the meaning of the word “rules”.

When building this deck, I threw every stupid idea I possibly could into it, with no regard to coherence or reason or swarms of bees. Gilgamesh plays cards to other players’ Realms. His Conditions interrupt their actions. He can perform actions on your board. He can expand his own Realm to five locations. And then he can summon Enkidu to do it all again.

It’s nice to just cut loose and see how many shreds you can reduce everything to, sometimes, and Gilgamesh is exactly the right character to explore those ideas with. I haven’t played this deck myself, and I don’t think it’s gotten a win at my table, so it clearly needs some work. Maybe a lot of work. But it’s important that, at the end of this arduous project, I have one deck that’s just “everything exploded and then we just scraped off the stuff that splattered against the wall”.

Kefka (FFVI)

“In Villainous, each deck has two copies each of two Conditions.”

Conditions are special cards you play on another player’s turn, interrupting for a moment their action to take one of your own. In the Disney decks these Conditions are always named after personality traits the villain possesses — Cowardice, Rage, Opportunist, etc. — there are always two of them, and there are always two copies of each. Card games with lots of turn interrupts can get pretty complicated, so I can definitely see why the designers wanted to keep this aspect pretty simple. With the FF decks I didn’t slavishly adhere to the personality trait naming convention, but I did stick with two Conditions per deck, two copies per Condition.

Except Kefka. Kefka has ten Conditions in his deck. Two of them trigger off other players’ Condition cards. His final card to win the game is a Condition. There’s a sense of lurking dread playing against Kefka. Since much of what he does takes place outside of his turn, you can never quite tell what he’s up to.

If you know my decks, you might think this isn’t the weirdest thing about Kefka’s gameplay. He also splits his Villain decks and discard piles in half, which no other villain does. This sounds like a second broken rule, but it’s actually a logical emergence from the first one. Kefka doesn’t win on his own turn, he wins on yours. He has to put his Realm into a particular state before he can do that, though, so he needs unique ways to hide or play with information. Face down double discard piles is a way Kefka can keep track of his own business while making him tough for everyone else to follow.

I actually really hate playing as this deck, and hopefully it doesn’t need much more refinement. It’s kind of hard to tell if it does. That’s probably how Kefka would want it, all told.

Magus (FF Time Travel Gaiden Superventure)

“In Villainous, each villain has one Objective.”

Magus has two. They’re each of the form “Begin your turn at X location while y action avaiable and z card is played there,” but each of the variables has two possible mutually exclusive values.

At the time I started designing Magus’s deck, with its two Objectives, his gimmick was unique. That was before the second expansion, and the Ratigan deck. Ratigan (from The Great Mouse Detective) also has two Objectives, but they’re handled a little differently from Magus’s. Ratigan always starts with Objective A and then switches to Objective B if a specific thing happens during his game. Magus chooses an Objective at the start, and both he and other players have opportunities to flip it. Because his Objective can turn on a dime, and the two Objectives are mutually exclusive, his deck is packed with lots of odd abilities to help him pivot course.

The balance of this deck feels okay so far, but the intent is for the two Objectives to be equally obtainable from start of game. If one is easier than the other, the deck might become overly predictable. Or maybe that’s okay, because a table’s meta will change as players try to guess and second-guess whether this Magus is going for the easier Objective, or the harder one, or picked the harder one because he assumes you’ll assume he picked the easy one and then flip it on him, or vice-versa. (Assuming one is harder than the other in the first place, which I don’t have good data for yet.)

Magus is a conflicted character who does evil things in dogged pursuit of vengeance brought on by grief and loss. I hope the deck captures that duality.

And yes, Chrono Trigger is a Final Fantasy game as far as I’m concerned.

The Turks (FFVII)

“In Villainous, there is one villain mover.”

I was done making Villainous decks, for real I was. I had designed the eight core decks to the point they were table-playable (e.g. the cards were written and imported into Tabletop Simulator, albeit some without art assets, and in something like a playable state) when I decided to create two more “expansion decks” — Gilgamesh and Magus — rounding everything out at a nice even ten. And that’s where I should have left it, until we started brainstorming one night about what a deck with multiple villain movers might look like. The idea was so tantalizing I had to make another deck for it, and The Turks was a natural fit.

(For the record, each deck started with an idea rather than an actual villain, usually the idea that coalesces from some form of “what if we break this rule?” With the exception of Garland and Golbez, villains I knew I wanted in the collection for personal reasons, I never started with a villain and then went in search of deck mechanics that would fit that villian’s themes. This is why there are no Sephiroth or Kuja decks; none of the ideas I had presented themselves as good fits for those characters.)

The Turks is by far the most complicated deck I’ve made, and all those complications stem from the idea that while most villains are solitary creatures, moving about their Realms and performing actions, The Turks is a group of four. On your turn you select one to move and perform actions. All of the deck’s resulting weirdness stems from that core idea. Each Turk has Personal Actions they can perform in conjunction with the Realm Actions printed on the board, allowing for somewhat customizable action sequences. And they have an extra Company Action, too, which they take collectively, and depends on how many Turks are present in a single location. (This “extra action” concept is explored in a different way in the Dark King deck, which I decided to make after The Turks because I didn’t want an odd number of decks.)

This deck has gotten some heavy play and, while it is the most complex of all the decks, it’s also incredibly fun to work with. There’s so many moving parts and so many permutations of action sequences that it really does feel like you’re playing a group of people cooperating with each other, rather than a single villain pursuing his own goals. Someday Elena will be introduced in Final Fantasy VII Remake 2 or 3 and we’ll have a nice HD render of her character to include on this board.

Ultimecia (FFVIII)

“In Villainous, Realms have at most one locked location.”

Ultimecia has three. She has to unlock them, do some stuff, then lock them back up to win.

This deck first hit the table in a literally unplayable state. I mean it hard locked to the point where the player could not do anything. With a lot of work and testing the deck has landed in a state that is not only playable, and not only winnable, but also somewhat elegant. There were times early in Ultimecia’s development where we played around with changing a lot of the rules for locked locations just for her board, but in the end the only necessary change was this: in addition to the cards that specifically unlock locations, Ultimecia can unlock a location by playing an Ally to it.

Everyone who plays Ultimecia gets into big trouble on their first try. One of the deck’s main characteristics is locking stuff up, but the more you do that the fewer places you have to take actions. The actual order you want to do things is determined by the action symbols in the Realm (whether you need access to Power or Play or Fate buttons might change depending on what villains you’re playing against) and what Heroes you end up seeing, and in which order they appear. There is one powerful “lock EVERYTHING down” card, but it can be pretty obvious you’re planning to play it, if your opponents also know the deck and are paying attention.

Queen Brahne (FFIX)

“In Villainous, Heroes are bad for you.”

The core game already has one villain — Captain Hook — who breaks this rule. And in the wider world of custom Villainous decks, “defeat such-and-such Hero” is a fairly common Objective. (And that’s another blog post unto itself, maybe!) I really like Hook. He’s probably my favorite of the core decks. So I definitely wanted to try making a deck in the same style. In Hook’s case, there’s a Hero he needs to play and several ways to get that Hero out. Other players can Fate Hook and bring the Hero out, too, and in my experience that’s almost always bad for Hook. What I wanted to try was a deck where the Hero starts in play, and can never be removed, but needs to be semi-controlled in order to win.

(The third Villainous expansion introduced Mother Gothel, who works on a similar idea. Between that, and Magus’s double Objectives, and Gigamesh’s Quest structure, I feel at least a little smug about independently designing deck ideas that went on to be good enough to explore in the main game.)

The board ends up feeling like a chase, and that’s intentional. The Queen has to chase the Princess around in order to get certain cards played at certain times. Sometimes you can manipulate the Princess to where you need her, sometimes you have to work with wherever she ended up. My suspicion is that the deck is too hard to win with, but I actually really like it in its current state. Any refinements on the basic Brahne concept are going to be light touches.

Sin (FFX)

“In Villainous, you always have access to the lower row of action symbols.”

Sin has to destroy his own action symbols to win.

Sin is a slow and steady deck. It’s going to win, and no amount of Fate can prevent that, it can only deter the inevitable somewhat. Playing against Sin isn’t about disrupting its Objective, it’s about pushing it back long enough that you can meet your Objective first. (If you’ve played the base game, Prince John has a similar feel.)

There’s really not much more to go into here. The deck has the same “problem” as Ultimecia, in that you can box yourself in by blocking off actions too early, but the solution is to just learn the deck and avoid doing that. Very few refinements were necessary for Sin. It’s a simple, straightforward deck that’s another good candidate to introduce a new player to the expansion.

Shantotto (FFXI ~ A Shantotto Ascension)

“In Villainous, you can hold a card  in your hand as long as you want.”

Shantotto can’t. Some of her cards have a Channeling cost, which is an amount of Power she needs to pay to keep the card in her hand at the end of her turn. This led to a major emergent property of the board that is necessary for honest play: Shantotto always plays with her hand revealed. Other players know what you’re holding. This concept of having no hidden information, not even your own hand, is thematically appropriate for Shantotto. In her game, specifically in the expansion pack this deck is based on, Shantotto tries to be deceptive and to hide things from the player, but is so obviously up to no good that the ultimate betrayal is more or less played for laughs.

This deck currently exists on a razor’s edge. The card Shantotto wants to play to win costs 24 Power. That’s a lot of dosh, so the deck is mostly geared around getting infrastructure in place to get lots of Power very quickly. The card has a Channeling cost of 2, so if Shantotto is holding it but can’t gain at least 3 Power per turn, she’s just spinning her wheels. There do seem to be decent strategies for playing the deck depending on whether you see the winning card in the early, mid, or late game, and I suspect we’re close to the deck’s final state.

It was this time last year I was playing Final Fantasy XI to get art assets for this deck. It was hell. I have scars. (But I still love Shantotto.)

Doctor Cid (FFXII)

“In Villainous, every Realm has multiple Gain Power actions.”

Not Doctor Cid. He gains Power by taking a special action to reshuffle his discard pile back into his deck, and his gameplay is designed around starting with a small deck and making it larger over time. If you’ve played Final Fantasy XII it should be obvious why this is good theming.

I’m glad I eventually had a good idea for a Doctor Cid deck. FFXII was a big hole in the collection for a long time. It’s one of my favorite games in the series, and its primary antagonists Vayne Solidor and Doctor Cid are amongst my favorite villains, so it kind of sucked having decks for the likes of Sin and Queen Brahne, but not one of them. Well, now I do. (And then, to even things back out, I whipped up the joke-y Sho-Gun deck to top it all off.)

This deck is so new it hasn’t been played yet. There are some obvious problems I want to look at and who knows how many non-obvious ones. This is why I like to playtest decks quite a lot before releasing them to the public table.

Ardyn (FFXV)

“In Villainous, you defeat Heroes with Allies.”

Ardyn has no Allies. He must defeat Heroes himself.

This wasn’t a new idea, when I sat down to design Ardyn. Ursula (from the base game) and Evil Queen (from the first expansion) both have alternate ways of dealing with Heroes in their Realm, and it’s a popular design idea across many of the custom decks I’ve looked at. The twist with Ardyn is, instead of simply having alternate Hero-slaying mechanics, he uses the standard Vanquish action to do it with his villain mover, rather than relying on Allies.

To ensure that Heroes were a central mechanic to the deck, this property emerged: Ardyn needs reasons to go into his own Fate deck and play his own Heroes. Once that was in place, his final gimmick — cards that have extra abilities if he possesses the Fate Token — was obvious.

It’s dangerous giving villains the incentive and ability to go into their own Fate decks. If Fate is necessary to advance a villain’s Objective at all, other players may not want to Fate them. There is some debate whether you should fate Captain Hook, for example. (He’s the only villain in the base game that needs to play a certain Hero to win.) We worked very, very hard to carefully word each of Ardyn’s Fate cards so that playing Fate against him is always damaging to him. There should never be a turn where Ardyn gets Fated and then says thank you.

I may or may not do more in-depth looks at these or other Villainous decks on this blog. I’ve definitely been playing and thinking about it a lot over the past year-and-a-half. I do maintain a YouTube playlist where I go in-depth into each deck as it’s released, but even some of that information is obsolete now with further playtesting. (Garland and Golbez both need to be revisited in future videos.) You can check out that playlist, or the Tabletop Simulator mod itself, at these links:

Steam Workshop: https://steamcommunity.com/sharedfiles/filedetails/?id=1901065673
YouTube Playlist: https://www.youtube.com/playlist?list=PLJaCM0fQ67Zaf3q34bWo5EKho4mNZ9FhA

Thanks for reading!

Shantae and the Seven Sirens

I have a sorta-kinda reputation amongst my viewers for being a huge Shantae fan.

And I am a fan, unquestionably. But not as big a one as people probably expect. I like Shantae. I like the characters, I like the dance magic, I like the corny sense of humor. I like that the games are breezy fun, and don’t aspire to be much beyond that. I like the gameplay, too, despite how sloppy it can be. I grew up in the ’90s, so I especially like how each entry to the series feels like it could be an episode of its own Saturday morning cartoon.

Do you zoomers even know what the phrase “Saturday morning cartoon” means, or why it has different connotations than if I’d just said “cartoon”? If so, you understand a small piece of why the games are special.

But I didn’t like Shantae and the Seven Sirens much. I was bored for most of my time with it, and so I had a lot of time to think. About Shantae as a character, about Shantae as a series. And I’ve concluded that while Seven Sirens is the only one I didn’t really enjoy playing, if I’m being honest, the series as a whole isn’t… great. Good, certainly. Flirting with greatness, at times. But almost always falling short.

A Brief History

I’m going to have to do a lot of work to make that statement make sense. I’ll start with a bit of history. I own a copy of the original Shantae GBC cart. A good friend of mine is a pixel artist by trade, who has a great admiration for the craft that eventually grew to be his career. Way way back in the long-long ago, around the first time Cave Story was the new indie hotness — that’s 2005, not 2010, for you zoomers — this friend recommended the game as being quirky and having really neat pixel animation. I do not know whether he had played it or not. Copies of the game were widely available at the time. I bought mine for $20 and played it on my GameCube’s Game Boy Player and then did not think about it for years.

Back when my YouTube channel was even smaller than it is now, I had a pretty good idea I wanted to be doing gaming commentary videos for a very long time but also realized the stable of NES titles I was actually good at would run thin very quickly. And I wanted to record something that wasn’t already covered to death on YouTube. Shantae seemed like a perfect fit, and it was. I got to showcase a game most people didn’t know existed, and had a crisis of conscience about whether anyone would actually watch fifty-nine videos of me getting lost in swamps and snowbanks. I created two Shantae playlists: one that included every video in the series, and one which only included videos in which I made actual progress, which I dubbed the “Platinum Edition”. (I eventually deleted that nonsense.)

The folks at WayForward watched that series, including Matt Bozon, who co-created the character with his wife Erin. I didn’t know this until several years later when the sequel, Risky’s Revenge, was greenlit for the Nintendo DSi. They plugged my YouTube series on the Risky’s Revenge development blog, which caused a huge influx of viewers to my channel. Matt reached out to me to offer me a download code for the game, but I didn’t have a DSi at the time, and so the company sent me one. It’s in a drawer right now with like 40 other DSes I’ve accumulated since.

Risky’s Revenge includes three corny references to my original video series, which still make me grin every time I pick up the game and replay it.

I really, really like Risky’s Revenge. It’s the only game I’ve ever attempted to seriously speedrun; at one point I managed to climb as high as #5 on the leaderboards. The game is smaller than the other games in the series, by like a lot, a fact which I’ve seen attributed to early DSiWare limitations. But the smaller size makes it feel punchier than the original. Each area is filled with more purpose. One of my most ironclad beliefs about game design is that limitations and restrictions drive creativity rather than stifle it. I realize the game wasn’t WayForward’s ideal Shantae vision, but for me, it was just the right mix of all the components to really shine. The game gets in, does its thing, then gets out without really having space to drip and droop.

Drippiness and droopiness were some of my complaints with the following title, Pirate’s Curse, which WayForward also provided me a copy of. I got to play it a few days before everyone else did, as long as I promised to not spoil the final boss for anyone. That final boss is so terrific I won’t even spoil it here. Really one of the all-time greats. (I did a full write up of the game in an old blog post here: Shantae and the Pirate’s Curse.) Teal deer, the game is fine but not a smash like its predecessor, and it has one really godawful level that I still see sometimes in my darkest and most fevered nightmares. I played through it a second time when the Steam version dropped, but not again since.

Even before Pirate’s Curse came out, WayForward organized a Kickstarter campaign to fund the next game, Half-Genie Hero. I backed the campaign and was not disappointed one bit upon finally getting my hands on it. I’m always wary of HD graphics in platformers, as many developers use the extra pixels as an excuse to cram lots of unnecessary detail on each screen, making them harder to read. Half-Genie Hero did not do this. The game looks fantastic and showcases some truly creative level design. The biggest compliment I can pay the game is that it eschews the Metroidvania structure of the previous titles for a more Mega Man X style level select. This was the perfect move for a series that had never quite gotten the hang of Metroidvania-style level design, but having now played Seven Sirens, I also suspect it again wasn’t WayForward’s ideal vision. That maybe there was some technical limitation why Half-Genie Hero couldn’t be a big connected world. If this is the case, it again just supports my theory that restriction drives tighter, more creative game design. Since it came out, Half-Genie Hero has been my go-to for replaying games in this series.

So now that Seven Sirens is out, that leaves me with two games in the series I really like, one I like with several asterisks, and one I probably won’t play again. And the original, of course, which I haven’t played since I recorded that YouTube series, and which I definitely will never play again. Because it’s not good, and I don’t like it.

Why the Original Shantae is Bad (and Why That Doesn’t Matter)

The original Shantae is a labor of love like no other I know. It’s big and ambitious, but also clearly held together with spackle and stardust. The team that made it was driven and talented, but also clearly working with thin resources. It barely clawed its way onto a dead handheld at the tail end of its lifespan, and even that must have been a herculean effort.

It’s not a terrible game, but it’s far from a good one. My friend who recommended it to me was right about its absolutely charming art style and the wonderful animations, but as time went on I also realized he was willing to overlook large gameplay flaws in his search for beautiful pixel art. Shantae has a lot of flaws, which are on clear display if you watch any of those videos I made which I originally deemed unworthy of the “Platinum Edition”. The world is vast, but it’s also sprawling and hard to navigate. The screen size is tiny and there’s no map, and large areas all made out of the same grey blocks make getting lost inevitable. Shantae moves faster than the screen can keep up with her, but the view doesn’t pan in her direction of travel, so running into monsters you couldn’t possibly see or dropping into pits you couldn’t possibly know were there are just everyday occurrences.

So the game is unpolished and clumsy, and kind of a real chore to play, if you hunker down and go through it. It starts out as good fun but becomes wearying quickly as you realize all the little niggles and irritants are going to be with you for the whole rest of the playthrough. I know it’s getting released on Switch soon, so if that’s going to be your first time playing I guess this is your fair warning. It’s old and stingy and definitely not a showcase of the best the series has to offer. All the charm is there, so if that’s what you like the games for you probably won’t be disappointed, but you are going to die a lot and it won’t be your fault and if you’re anything like me that sort of thing makes you sour.

A year or so ago I recorded a few short serieses I called the “old mean NES games”. This included Castlevania II, Faxanadu, Conquest of the Crystal Palace, and Battle of Olympus. This is about where on the 2D platformer spectrum I’d put the original Shantae. They’re decent, and it’s certainly possible to enjoy them for what they are, but they’re very hard to recommend nowadays because of their jank and rough edges.

But.

Shantae might still be one of the most important games ever made. The late ’90s and early ’00s weren’t a kind time to 2D platformers. The big companies weren’t making them anymore and neither were the small companies. Metroid and Castlevania kept rolling on sheer inertia, but outside of that and a couple of Mega Man X spin-offs the genre was dead, dead, dead. There wasn’t a proper mainline Mario platformer on either the Game Boy Color or Game Boy Advance that wasn’t a port. There wasn’t much of an indie scene to speak of either. If you liked platformers a lot, as I did, Shantae was important because it was bearing a torch no one else wanted to carry. This is what I meant when I said the game was a labor of love “like no other”. Someone in the world, for some reason, stood up and said, you know, we’re gonna make an oldschool platform game. Even though it won’t sell. Even though nobody wants it. Even though we’re not Metroid. We’re gonna make it because we know, in our heart of hearts, that classic 2D platformers are one of the truly timeless video game genres, even if these turdbois with their PlugStations and their Q-Boxes have temporarily forgotten that fact.

Someone said this even before Cave Story. And I can do nothing but salute that sort of dedication.

We are awash with 2D platformers in this golden modern era, more than a man can ever possibly play. And there probably always will be. And to be clear, Shantae was not a herald. It didn’t cause the genre’s return to glory. But in the Dark Ages it was there, alone in the wasteland, carrying the standard.

So Okay, Seven Sirens, Then

The original game was a great release for 2002, for reasons I’ve mentioned, but I have serious doubts it would have cut the mustard if it had been, say, a 1994 Super Nintendo release. Standing alongside Metroid Fusion and Castlevania: Whatever of the Whocares I think the game did fine; if it had stood alongside Mega Man X and Donkey Kong Country and Yoshi’s Island and Sonic & Knuckles and fifty other titles, I think it would have been washed away like so many other long-forgotten platformers of the era.

And this is the problem I kept having with Shantae and the Seven Sirens. In 2020 it’s simply not enough to be a 2D platformer, or even a Metroidvania. Nowadays you have to compete with, just as a sampling, Axiom Verge, Micro Mages, Bloodstained, Guacamelee!, Ori and the Blind Forest, Hollow Knight, Mark of the Ninja, and a thousand others. To even register on the scale, your game has to do something really special.

And Seven Sirens doesn’t. As I reached the end of the game, I realized that I had only bought it because I liked Shantae, and that was really the only reason to play it. Other Shantae fans are the only people I would recommend the game to, but that’s pointless, because other Shantae fans already know about it.

Again, all the charm is on display here, and if that’s what you like most about the series, you probably won’t be disappointed. Alas, charm alone isn’t enough for me anymore. I need the gameplay to sink its talons in deep. This means different things for different genres; for Metroidvania titles, though, it means the first and most consistent thing I’m going to notice is the moveset.

In Seven Sirens Shantae can wall climb. She can air dash. She can ground pound. She can double jump. And that’s… it, as far as general use movement options go. Her moveset is rounded out with two “move through special types of ground” transformations that are functionally identical to each other, and four full-screen “trigger the special onscreen thing” magic spells that are functionally identical to each other. Oh, and her usual repertoire of attack items and potions, which are barely worth mentioning, because the game is so easy anyway that combat just doesn’t matter.

Here’s a hot take: I don’t want combat to matter in my Metroidvania titles. I had to give up on Hollow Knight and, while the original Guacamelee! won me over I wasn’t quite as enamored with Guacamelee! 2: ¿No hemos hecho esto antes? The explore-y bits are what I’m here for. I’m the stubborn jacknose that hundo’d Axiom Verge without a walkthrough, even though it came down to painstakingly drilling individual blocks one at a time and mapping the whole world out by hand on graph paper. I think X-Ray Scopes are for weiners. I want a bunch of trivially-cleared enemies that I don’t have to think about as I plow back and forth through areas I’ve seen a hundred times. I got stuck for hours in Bloodstained and refused to look up the solution even as my Twitch chat was watching and slowly going insane.

(And the solution was dumb, I decided. If it weren’t dumb, I’d have figured it out sooner. Of course.)

So believe me, I don’t want Shantae to nerf the pike ball and limit my potions. Toothless, softball monsters and infinite healing are not why I was bored.

The game world of Seven Sirens is closest to the original Shantae in terms of navigating the world. The screen size is much bigger, and you have a map now, so you won’t get lost or fall into a cheap hole. But you will spend a lot of time in very same-y looking environments, screen after screen of perfectly uniform blue or orange blocks, with just not a lot of interesting things to do.

It really does come back around to that boring moveset. In a Metroidvania game I want to unlock a move that’s so fun I just can’t help but do it everywhere. I want to chain Juan’s uppercuts or fling Ori around like a pinball. All Shantae can do here is jump to the next platform or, if it’s a little too high up, turn into a lizard and climb up to the next platform. Or, if the screen has one of the Special Things on it, you push the matching Special Thing button to unlock it and get some goodies. It’s up to you if you want to stop and fight the monsters or not, but I certainly didn’t.

This has always been at least a minor issue with this series. The first game gets a pass, for being the first, and coming out during the Dark Ages before the pulsating ooze of the Igavanias seeped into the genre. And the second does too, for sticking to punchier, nicer-looking versions of the first game’s abilities. The third tried to innovate by replacing the standard dancing moveset with pirate gear, but the pirate gear was just the bog standard “double jump air dash wall grap slow fall” kit. It worked well enough and some of the gear combined in fun ways, but there were no surprises. Seven Sirens not only has no surprises, it tries to make do with fewer actual movement options than any of its predecessors.

(This is another area where Half-Genie Hero’s level design really wins the day. Because it wasn’t made out of interconnected levels, it could afford to lean way harder on its gimmicks and platforming challenges, giving every location a unique feel. You were hanging from fish hooks, or leaping between magic carpets, or racing a volcanic eruption. That’s not something a Metroidvania can easily do, and one of the reasons I felt moving away from that archetype was good for Shantae.)

There actually was one surprise waiting for me, as I traversed the map: the load times between each area. I have to say I really wasn’t expecting that. I couldn’t swear they’re longer than the ones in Guacamelee! or Bloodstained, but I definitely noticed them a lot more. Maybe the load screens just pack a bigger punch when all you’re doing is switching from blue hallways to brown ones.

Most of what you’re doing, on a moment-to-moment basis, is running fetch quests for NPCs. So-and-so wants the such-and-such, or maybe four or five thingamajigs, and wouldn’t you just be a doll and go gather them all up? Risky’s Revenge did a little of this to squeeze some extra use out of the game’s limited areas. Pirate’s Curse did it to sort of introduce you to each new area as you reached it. But neither of those games were what you’d really call pure big-boy Metroidvanias like Seven Sirens is. And the convention just doesn’t work here. The NPC sends you off to some part of the map, yeah? But I mean, I was going to go there anyway, just as part of my baseline exploration of the world. When I reached a new area I didn’t want to explore at all, because I knew it’d just be wasted time and I’d just have to cover all that ground again once I found the relevant NPC. For reasons I hopefully don’t have to explain in much detail, “I don’t want to explore this new area” isn’t a sensation a player should be having in this genre of game.

The dungeon content from the first three games is back. These are the areas where the level design really comes into sharp focus; small-ish but tightly-designed areas built around a common theme and, later, whatever your new dance move is. Each dungeon has a mini-boss which cheekily advances the game’s story, which is a cute idea I really liked. There is some very light “match three shapes” style puzzle matter to contend with, and then a forgettable boss fight you’ll plow through because you have the super pike ball and infinite healing.

(I think I would really enjoy a batch of Shantae dungeons that went all in on the puzzle design. There was one squid in one of Seven Sirens‘s dungeons that had me stumped for a couple minutes, because I had misunderstood one of my new dance’s interactions. My preference would be for every dungeon to stump me repeatedly, because they were deliberately designed to. I realize this isn’t part of WayForward’s toolbox, but I can’t think of another series that’s making sideview Zelda-style dungeons, so this is a niche Shantae could really occupy, if it wanted.)

What’s missing from the dungeons, much like the moveset, is creativity. You’ll shoot through a cannon maze. Okay, that was kind of fun in previous games, where it was already dropped in from Donkey Kong Country. But we’ve done it before, and Seven Sirens doesn’t do anything new with it. There are red/blue toggle switches that make red/blue platforms appear. There are breakable walls and whippable blocks. There’s… uh… Pac-Man rooms, for some reason, which at first I thought was a great gag, but then the game threw a second and later a third one at me, and I started to wonder if maybe it wasn’t a gag.

There was one dungeon I quite enjoyed, where you’re put on a time limit and given what seems like way more tasks than you can reasonably complete in that limit. But even this was an idea explored already in Risky’s Revenge, so while it was enjoyable, it wasn’t new or surprising.

There is a badge system now, where sometimes monsters will drop equippable doodads you can put on for perks like “climb ropes faster” or “fireballs deal more damage.” But you can ignore this system entirely, because there are purchasable items in the shop that replicate the few badges that are actually useful, and because the only really good badges are the boss ones you have to purchase from NPCs anyway.

Since all the badges are monster drops and all the upgrades and spells are for sale in the shop, that doesn’t leave much to hunt for out in the game world. This is really the meaty center of any Metroidvania: worthwhile upgrades hidden in clever spots. In Seven Sirens these are all either heart pieces, gem caches, or gold nuggets used for buying the rare boss badges. This is a complaint I had about Pirate’s Curse too, where the actually good upgrades were gated only by money, and so the emphasis was on grinding up gems rather than hunting for secrets. It’s just as well, since most of the secrets in Seven Sirens are of the “use every spell on every screen and see what pops out” variety. At a certain point in the game the world becomes populated with tinkerbats poking at walls or digging in the ground. This is a clever way to hide secrets that otherwise wouldn’t be signposted. However, sometimes I couldn’t find a way to interact with them to reveal the secret. This is a pretty strong clue to come back later with a different upgrade, but tinkerbats that are poking at a secret don’t get cleared away once you reveal it. So the post-game clean-up would just be going to every screen with tinkerbats, trying each ability and spell in turn, reloading the screen in between tries, with no way of knowing whether I’d already gotten the thing from that screen already. Needless to say, I didn’t bother with the hundo.

(Sometimes the game does pop a permanent “you found this!” indicator out, but as far as I can tell these are just for an achievement and serve no other purpose.)

I think that about covers everything. Oh, wait, the final boss was really bad. That covers everything.

Who is this game for?

It’s for Shantae fans, duh. The series built up a pretty strong little following with Risky’s Revenge, and WayForward earned every inch of them. Pirate’s Curse was a more modern take on the original game, which I didn’t like as much, but which I’m sure many fans appreciated since the original was so impossible to find by that point. And Half-Genie Hero strikes out in a new direction and shows what WayForward can really do when the creative engine is burning.

What I mean to say is, the first four games all have strong reasons to bring new players into the fold. Seven Sirens feels like the fanbase is now in place, and we know how many units this thing is going to move, and we can kind of just connect the dots and color in the lines. It grabs some good ideas from the older titles and puts them back together in a way that’s new only in the sense that they haven’t been placed in this exact configuration before. You want some of that quality, in a series with five entries, for sure. You don’t want each new sequel to be unrecognizable. (Looking at you, Risk of Rain 2!) But you do want to see at least a few new pieces.

Seven Sirens is a competently-made by-the-numbers Metroidvania title. In 2020, that’s just not enough anymore. What is Shantae’s unique voice in what is no longer a niche genre? I know how Juan, or Ori, or Miriam, or the Salt and Sanctuary guy would answer that question. Maybe in Shantae 6 we’ll get an answer here, too.

Thank you for reading this long post about hair-whip girl!

Super Metroid Rando > Symphony of the Night Rando

Super Metroid is a nonlinear exploration-based platformer from the ’90s. It is one of my favorite games of all time. It has a randomizer which helps to keep the game fresh and interesting, and even after dozens of seeds I’m thirsty for more.

Castlevania: Symphony of the Night is a nonlinear exploration-based platformer from the ’90s. It is one of my favorite games of all time. It has a randomizer which helps to keep the game fresh and interesting, but after only a couple of runs I’ve seen everything it can do and am tired of playing it.

Wait, that doesn’t make any sense.

Randomizers and Gating

One of the ways game randomizers keep things fresh and interesting is by using gating. In the vanilla game you need to kill the Dog Boss to get the Dog Key and open the Dog Door, or whatever. Any progression beyond the Dog Door is locked off until you do this. What a randomizer does is take the Dog Key, Lizard Key, and Ferret Key and puts them into a big sack, then shakes the sack up and redistributes the keys to the Dog Boss, Lizard Boss, and Ferret Boss. The vanilla game usually has you fighting Dog -> Lizard -> Ferret, and there are implications for exploring the game in that intended order. The randomizer has you fighting these three bosses in some arbitrary order, though, depending on what blocks your path to them, so you have to explore things quite a bit differently.

If a game really does only have three gates (Dog/Lizard/Ferret) the rando options aren’t going to be very robust. You go fight whatever boss is available from the start, use the key you get to reach whatever boss it allows you to reach, then use the next key you get to reach whatever boss it allows you to reach, then go win the game. Six total paths and you’ve kind of seen everything the rando can throw at you.

The more gates a game has, the more robust the randomizer can be, and the more potential paths you’ll have to expore as you play through seed after seed. The nuts and bolts of the concept can get a lot more complicated than this, but it serves as a good starting point for these two games. What a rando typically does is looks at all the gates in the game and builds a path from the start (where you have nothing) to the end (where you win) by placing one item at a time, keeping in mind what its already placed and what can be reached after each step. An item placed on your path in this way is said to be “in logic”.

Gates in Super Metroid

What’s in logic in Super Metroid? Let’s break it down to just the mission-critical stuff you absolutely need to beat the game. To get almost anywhere in Super Metroid you need to use different kinds of ammunition to open doors and break blocks. Missiles, Super Missiles, and Power Bombs are required to make any progress in any direction. You also need the Morphing Ball to traverse certain areas and to lay Power Bombs. Even the wackiest seeds are pretty reserved until you have this basic set. For the rest, let’s start at the end and work backwards.

(I’m using a pretty loose definition of “need” here. With glitches you can go “out of logic” and skip gates in a variety of ways, and you can usually configure your rando so this sort of thing is required. I’ll be ignoring glitches for purposes of this text.)

To win the game you must destroy Motherbrain. This requires three Energy Tanks to survive her candy laser. In addition, to kill any boss in Super Metroid, you need either enough ammunition (Missiles/Super Missiles/Power Bombs) to deplete their HP, or the Charge Beam.

To reach Motherbrain you need to destroy the other four major bosses: Kraid, Phantoon, Draygon, and Ridley.

Reaching Kraid and Phantoon requires nothing special beyond the basic set.

Reaching Draygon requires the basic set plus the Gravity Suit to move through Maridia’s water, and the Speed Booster, as there is a hallway filled with speed blocks along the path. Also, the Space Jump is required to get out of Draygon’s room once you’ve killed her.

Reaching Ridley requires all of the above plus the Varia to keep Norfair’s heated rooms from damaging you.

The final list is:

  • 3 Energy Tanks
  • Morphing Ball
  • at least a few Missiles/Super Missiles/Power Bombs
  • either lots of ammo or the Charge Beam
  • Speed Booster
  • Gravity Suit
  • Space Jump
  • Varia

In a rando, once you have everything you need to complete the game, you’re in “Go Mode” because now you can go win.

This is only about half of the possible items in Super Metroid, and depending on your seed other items might be required. For example, there’s one item hidden behind a block destroyed with the Grapple Beam; if one of the above items is located there, the Grapple Beam is in logic and you’ll have to find it. In addition, even items that are out of logic are almost always helpful in your seed; the Hi-Jump Boots are never required, but they make moving around the game world a lot easier.

The keys to making these gates work are: 1) there are lots of items in Super Metroid, and any of them could be something you need; 2) every major item you find expands the list of places you can now check; and 3) once you’re in Go Mode there’s very little left to do except kill the bosses and win.

Super Metroid randos are great exercises in routing, problem solving, and application of game knowledge. It’s not really possible to be bored because, even if you’re in Go Mode, the boss fights don’t take very long and aren’t trivial to win with low item counts. It’s not uncommon for a player in Go Mode to keep looking for items anyway, for more energy and ammo.

Gates in Symphony of the Night

Symphony of the Night is a much larger game than Super Metroid in terms of physical space to explore, number of bosses to face, and the hero’s range of action. Why then is the experience so boring? Let’s break down the game’s logic.

To win the game you must destroy Shaft and Dracula. To open the room where they dwell you need the five Vlad relics: Eye, Ring, Tooth, Rib, and Heart. You also need to be able to physically reach and move around in the Inverted Castle.

To reach the Inverted Castle you must defeat Richter with the Holy Glasses equipped.

To reach Richter you must have some form of flight: Soul of Bat or Gravity Boots plus some “midair item”.

The final list is:

  • Eye of Vlad
  • Ring of Vlad
  • Tooth of Vlad
  • Rib of Vlad
  • Heart of Vlad
  • Holy Glasses
  • some way to fly, either:
    • Soul of Bat or
    • Gravity Boots + a “midair item”

(A “midair item” is anything Alucard can do in the air, including the Leap Stone’s double jump, any weapon with a thrust move usable in the air, or any transformation. Doing any of these things puts him into a midair state that lets him jump repeatedly with the Gravity Boots.)

This is a really tiny list of relics. Depending on your seed and settings, other items may be required. For example, there are some item checks behind a switch in a cave only the Demon Familiar can open. If any of the above items are there, Demon Familiar is in your logic.

But now we start seeing problems.

Unlike Super Metroid, very few of the items in Symphony’s logic actually expand the player’s moveset. The Vlad relics all raise your stats, but nothing is ever gated behind having a high stat. Most relics that do open gates only open one specific gate. The Holy Glasses only get you through the Richter fight; they have no implication on gameplay other than that. Other items, like the Jewel of Open, Merman Statue, and Spike Breaker similarly only serve to open certain rooms but not actually alter the hero’s moveset at all.

Once you can fly, you can go just about anywhere in the castle you want. And once you can get passed Richter, you’re in Go Mode… except you’re not. You’re in “Almost Go Mode”. Flight is the only meaningful way Symphony’s map ever opens up. Once you have it, you really just need to find the last couple Vlad relics and then go win.

Being in “Almost Go Mode” is boring. It’d be boring enough if it were just five arbitrary items you had to seek out that could be anywhere, but it’s actually worse than that. Since there are still lots of “key-and-lock” style relics the game likes to daisy chain them into long strings requiring a lot of backtracking. Your last Vlad item might be behind a grate, which requires Form of Mist, which is behind a spike room, which requires Spike Breaker, which is behind the floodgate, which requires Merman Statue, which is behind some other door locked by some other item that exists only to open that one door.

Early Symphony seeds are pretty fun, since you’re usually fighting through the first few levels with weird equipment you’d never use in the normal game, but even this little bit of randomness is devalued because you have access to all your spells as soon as you have MP to cast them. You very quickly end up with flight, since about 80% of the map is inaccessible without it in some form, and from there it’s Almost Go Mode. You full-clear each area of the castle to identify the areas you’ll need to daisy chain later, then you daisy chain, then you finally go win.

In general…

I’ve played lots of randomizers for lots of games. In general, you reach a point where you just need one more item for Go Mode, and it can be in any of a couple dozen places. In Symphony you reach a point where you need three or five uninteresting items, and they can be in any of a hundred places.

Restricting Checks

Someone will point this out if I don’t, so I guess I’d better. Most of this post has been written under the assumption you’re playing Adventure Mode in the randomizer. This is a mode which means any item (including in-logic progression items) can spawn in any spot where the base game has any kind of equippable item. There’s another mode that limits your in-logic progression to just those locations in the base game which had relics. This reduces the total number of checks by a great deal.

It doesn’t fix the problem, though, because there are so many “useless” relics in Symphony. Most of the relics you can find just aren’t that interesting, by which I mean they don’t open up any new checks. Your checks are still spread out across the entire castle, many of them are still gated behind boring key-and-lock pairs, and many of them only serve as keys to open singular locks.

Not-Adventure Mode also has huge locations in the castle which are always useless. If there’s not a relic in a location, there’s no reason to ever check there. The result is you tend to always check the same areas in the same order, ignoring the same pointless areas each time. You don’t get cool/weird moments like “Oh wow, I actually have to go to Reverse Library this seed!”

What can be done?

Most randos I’ve played launch in a pretty boring state, but get better over time. I think there’s potential for Symphony too. In fact I hope there is, because I love Symphony and a good rando of it will be on my plate for a very long time.

Like everything in life, I think the solution is to steal. Other randos I play regularly have ideas I think would be great additions to Symphony, providing they’re possible to implement and the developer actually has the will to implement them.

Door Rando

VARIA, the Super Metroid randomizer I use, has an excellent option to randomize areas. The game world is broken down into nine large chunks, and the entrances and exits of those chunks lead to random places. For example, the door that usually takes you from Crateria to Tourian might lead to Lower Norfair instead.

What this adds to the rando is a sense of needing to first place all the locations and the connections between them, then actually work your way through the levels in an order you’re probably not used to. It also adds to the list of interesting checks: doors are now on the menu.

In Symphony, red doors lead either to new areas or to warp rooms. There are two castles, each with twelve sections, and each with five warp rooms. There are lots of red doors, some of which are one-way until you pass through them once. Include options to scramble each castle individually, or both together. Now instead of the tired Entrance -> Alchemy Lab -> Marble Gallery -> Outer Wall -> Library path literally every rando seed goes through, you might have to fight through part of a tougher level to get anywhere useful.

The first area of the game (the Entrance) has two red doors accessible immediately (in a door rando, we should remove the one-way wall next to the area warp room), one red door accessible if you can gravity jump or fly, and one red door that’s a one-way exit from another area. With some logic that ensures all areas are accessible and all doors get used, this option alone could make every run unique.

This sort of randomization does devalue the Holy Glasses a little bit. If you scramble the two castles together, it would be possible to reach the Reverse Keep through some other door, and now Richter isn’t a required fight anymore. And it doesn’t do much to solve the problem of some areas not being useful; the Reverse Library is still a dead end and there’s still nothing there.

Incentivized Locations

This was similar to a problem the Final Fantasy rando had early on. In that game, some locations just aren’t useful to visit and people never went to them. You could reasonably finish the rando by going to the same list of locations every time, and avoiding the ones you knew didn’t have anything. They fixed it by slightly expanding the pool of Good Items (which includes your in-logic progression items) and then hand-picking a couple of interesting but seldom-visited locations and flagging them as “there might be a Good Item here”.

Symphony has lots of good candidates for this. Leave in the current Adventure Mode (where anything could be in basically any room) for people who want it, but include this middle ground “Relics and Incentivized Locations” option as an alternative. Good places to incentivize would be behind both gear puzzles, the underwater (or overwater, I guess) cave, the secret elevator in the Outer Wall, the wolf/bat puzzles in the Merman room, the stopwatch clock room in the inverted castle, the Grapes that drop in the Chapel’s confessional, and Richter Belmont’s room. (This isn’t an exhaustive list, just a few examples.)

That last one is important. Because he wouldn’t block progression if you randomize locations, we can keep him in logic by maybe locking a Good Item behind him.

Picking memorable out-of-the-way locations is key, but we should also try to pick locations that are in-character for Symphony as a game. One of the relic locations is in the game’s shop, which I think is a good idea. Keep that as an incentivized location. Also, let’s pick a monster with a really good drop everyone knows about, like the Shmoo, and make it’s common drop an incentivized check. This is well in the spirit of Symphony‘s drop system, shouldn’t be too obnoxious (it only takes a few tries to get Shmoo to drop Ramen in the base game), and makes LCK-up items valuable for this one check.

Other very Symphony-ish spots with incentive potential are the first item that drops out of the Librarian’s chair when you jump into him, or the item you get when the Sword familiar levels up enough. Maybe the second or third item you spawn with a Meal Ticket is hard-coded to be an incentivized location. An excellent idea might be to put an incentivized spot inside pool of items that drop out of candles, which turns the Cube of Zoe into a progression item itself.

(I don’t mean pick one random candle somewhere in the castle, I mean pick one particular candle drop, say Rebound Rock, and replace all of them with the progression item. First time you break a candle with one of these items you get your progression and know you can stop whacking candles. After you get it, future candles will only drop Rebound Rocks.)

These are fun things — or maybe annoying but still distinctly Symphony things — that can open up the possibility space without just making the player run across the whole castle again.

More Item Pools

Right now the Symphony rando only has two pools of items: Relics and Progression Items, and Everything Else. (Actually for the sake of Adventure Mode I think the rando recognizes equippable and non-equippable items as separate things, but for all practical purposes they just make up one big pool of basically trash.) We can leave that in for people who want to keep the searching space huge, but we can also add an option to tighten it up.

I propose three tiers: Good Items (anything that can block progression, minus the Vlad relics, plus a few good pieces of equipment like Duplicator or Dracula Tunic), Boss Items (all non-progression familiars and transformation power-ups, the five Vlad relics, then enough Life Up and Heart Up items to make up the difference), and Everything Else (everything else).

Relic spots and/or incentivized locations as outlined above contain Good Items. Boss Items drop only off of bosses. Everything else is just laying around the castle, including most of the equipment and Life Up items you’ll use.

Fighting bosses in Symphony is already seen as the de-facto major goal of the game, outside of “win the game”. I know when I play vanilla Symphony I like to fight and kill all the bosses at the very least. They already block progression in the sense that they block off certain areas of the castle, so they don’t need to drop progression items. They’re a good spot to hide other power-up items you probably won’t need, but might be nice to have.

The real smart bit here is, once you’re in “Almost Go Mode”, you can stop hunting through every corner of the castle and just start slaughtering bosses. By the time you reach “Almost Go Mode” you’ll probably have half the bosses down anyway, and reducing the search space to just the rest of them makes the boring back half of seeds go more quickly. It also presents a potentially fun routing challenge: “Oh, that’s Almost Go Mode! What’s the best path between here and Dracula that takes me through as many boss rooms as possible to find my last two Vlad items?”

Two bosses should drop Good Items instead: Richter Belmont and the Succubus. The former requires a Good Item to fight properly, so should give a Good Item in return, and the latter gives a Good Item both in the base game and the current rando.

More Doors

If all of the above is implemented, we’ve inadvertantly added another problem. (Turns out randomizers are complicated. Who knew!) If we remove Richter from progression by doing some sort of door shuffle, then you’re in “Almost Go Mode” as soon as you find a door leading to the Black Marble Gallery. This reduces the entire rando to just a boss hunt. This would still be better than the current rando, but most seeds would probably wind up that way.

Currently in the vanilla game there are a few kinds of blockades requiring specific items:

  • bookshelves you can open just by pushing on them,
  • blue doors requiring the Jewel of Open,
  • grates requiring Form of Mist,
  • tight locations too small for Alucard to fight through, such as the hallway outside of Orlok, which require some form of transformation to squeeze through,
  • switches the Demon Familiar can open, and
  • spiked areas requiring the Spike Breaker.

(The latter would have to be solid walls with spikes on both sides, so you couldn’t mist or i-frame through them like you can in the Chapel hallway, and ensure you really do spike-break through them.)

Anywhere in the game that currently has one of these blockades could be randomized to be any of these six, instead. So you wouldn’t know what’s blocking off Orlok’s room (or Akmodan’s room) until you visited there and looked. Now we have a risk/reward decision: if you have lots of key items, you should go look. If you only have one or two, you might get lucky, or it might only give you information so you can backtrack later.

We could arbitrarily add as many of these blockades to the game as we want, but that would be really annoying. A few key locations might be appropriate though. The entrance tunnel to both Succubus areas (especially if the inverted version is incentivized) would be good spots. Maybe the path in the Catacombs leading to Granfaloon/Galamoth. And then the three entrances into the inverted clock room: the far side of each Guardian hallway, and just outside the door leading to Black Wing’s Lair.

Now you’re guaranteed to need at least flight and one other door-opening item from the castle to reach the clock room, instead of only flight. Or at least to find an alternate path through the castle so you can access a door you do have the key for. Or maybe get lucky and just push a bookshelf for two seconds! (You’d still need the Vlad items, but you’re already out boss hunting to find flight and your key, right?)

Enemizer

This all turns Symphony into a better treasure hunt with more interesting gates and more complex logic to explore, but the game is still probably not threatening to most players familiar with it. I like my randos nice and troll-y. What else can we do?

Take a page from Link to the Past and randomize enemy locations. Everywhere the game has an enemy, put a potentially different enemy. This would dramatically change the layout of many rooms in the castle, especially early in the seed before you have Wing Smash. This works really well to spice up Zelda dungeons you’ve cleared a hundred times. It’d work just as well in the Clock Tower. Maybe even better, considering how many Symphony locations follow some template of “just plop a few enemies in the way”.

There should be three levels of this:

1) Shuffle Enemy Types. Each enemy type in the castle is replaced with some other enemy type, and each is used exactly once. The mass of Dark Octupuses in the Reverse Caverns turns into a mass of Stone Roses instead.

2) Shuffle Enemies. Each individual enemy in the castle is replaced by one enemy from elsewhere, preserving their populations. One of the Dark Octupuses can be a Stone Rose, but they can’t all be, because there aren’t enough Stone Roses in the castle to fill the spots.

3) Full Randomization. Each individual enemy in the castle is replaced by one other enemy, with no regard to population. The Dark Octupuses can turn into literally anything, including all Guardians or Blue Venus Weeds or whatever other horrible thing. (This might be sufficiently troll-y that we have to incentivize the location, now!)

A few enemies can’t be randomized in this way, like the gorilla skeleton that breaks the wooden bridge, and probably anything that infinitely spawns (Mermen and Medusa Heads). And we should probably leave Yorick alone too. Homeboy has enough to deal with.

If the Shmoo has an incentivized drop, #3 might be a problem. Maybe run the randomization as normal, then randomly pick five individual enemies in the world and turn them into Shmoos.

Boss Randomization and Scaling

Final Fantasy IV: Free Enterprise does something incredible with bosses: it randomizes their locations and then scales their stats to match the location where you find them. If Odin, an endgame boss that is usually very difficult, rolls into the position where you normally fight the Mist Dragon, the game’s first boss, you get to fight a very weak version of Odin. This makes for some interesting fights, in both directions.

Obviously this would be harder to implement in Symphony. You’d probably need to build a few possible boss pools based on the size of the rooms where you fight them, then shuffle the ones in similarly-sized rooms between each other. This works well on paper at least, because almost every boss room in the game is used twice; Karusamon and Darkwing Bat are both fought in the 1×1 screen room at the end of the Clock Tower.

The scaling could probably play it pretty straight; if Doppelganger10 rolls into Hippogriff’s room, it gets Hippogriff’s stats but keeps Doppel10’s move set. Maybe Galamoth rolls into the Slogra/Gaibon room and you get a silly super-easy version, then later Granfaloon rolls into Galamoth’s room and you get a super grueling match against a pile of zombies.

In actual practice, most of the bosses in Symphony are jokes, except the first two you fight: Slogra/Gaibon and Doppelganger10. These bosses are still jokes (they’re early game bosses) but in some seeds you have to fight them with bad equipment, so you’re required to at least be awake. Fighting a low-level version of Orlok or Medusa might be interesting if only because you actually have to watch their moveset for a change.

We should leave Richter and Succubus out of this, and maybe have a special option to leave Galamoth out as well. I suspect even a low-stat version of Galamoth would be impenetrable if he’s literally blocking all your progression and all you have is a Basilard.

Make Spells Items

The other reason boss fights in most seeds are trivial is Alucard very quickly has enough MP to cast high-damage magic spells which auto-target enemies. Most bosses just have no answer for this. In the vanilla game these were soft-gated behind button combinations a new player wouldn’t know, but everyone playing rando certainly knows them.

Instead of the magic scroll items just being there to show the button combo, let’s have an option to make them unlock the spells instead. You can’t cast Tetra Spirit until you find its scroll, even if you have the MP and know the combo.

I’m thinking three options: Default Spells (you can use them as soon as you have the MP), Boss Items (replace Life-Ups in the Boss Item pool with spell scrolls until they’re all placed), or Everything Else (scrolls are just in the castle somewhere, like most other equipment.)

Three spells in particular — Tetra Spirit, Soul Steal, and Wing Smash — are probably good enough we could mix them in with the Good Items instead. Or maybe that could be a separate option.

Faerie Scroll and Spirit Orb

In the vanilla game, these two relics just display information when you hit enemies, and have no other function. They are mechanically useless. Is there a way to turn them into progression without just arbitrarily making them some new dumb thing? I think the easiest way is to just let the player start with them. Seeing enemy names and damage is pretty basic functionality.

Or maybe we can get creative. If we added in the random blockades above, here’s a seventh: a type of door that only opens when hit by a certain elemental attack. What kind of attack? We don’t know until we hit the door and see what it’s called. (It’d have to be implemented as a solid enemy immune to all damage, like the Stone Skull, or something.) If you find such a door and have the Faerie Scroll, you can attack it and see the phrase “will open if burned”. Then you know to hit it with a fire weapon or attack. If you’re late in the seed and have amassed quite a lot of equipment, you could also just try a bunch of stuff until something works. This kind of check would also reward some esoteric game knowledge about elemental weaknesses that usually isn’t tested at all, and also provides at least a little value to the boatlod of junk the player picks up all over the place.

There might be a way to make one of these doors open in such a way that knowing how much damage you deal per swing is useful. I can’t think of a workable one right now that isn’t stupid, though. Maybe it’s okay we leave one mechanically useless relic in the Good Items pool.

In-game Hints

Many randos offer in-game hints to help point you to important checks. Symphony is such a huge game, with such a big search space, and it’s full of NPC cutscenes people just skip. Let’s have an option that removes those cutscenes and their terrible voice acting, and replaces them with some potentially useful text.

“There’s a Vlad relic above the Entrance.”
“The Floating Catacombs contains a power for your transformation.”
“Nova Skeleton has a powerful sword.”

Lots of these cutscenes are already gated behind progression. The boss fight in the Coliseum, for example, requires some way to get up through the passage into Orlok’s Quarters. Gaining info about where to go — or where not to — would be an especially helpful addition to Adventure Mode, where the player has to otherwise check every room.

In my experience, randos with hint systems tend to involve checking every room anyway, but hey, we have these cutscenes, we might as well do something with them.

Alt Win Condition: Death’s Collection

At the beginning of Symphony, Death steals all your cool items and runs off. (In vanilla this is all the Alucard gear, but in rando it’s some random set of equipment.) Let’s have an alternate win condition where all five Vlad relics are guaranteed to be hidden behind Death’s room in the Cave, but he refuses to fight you unless you’re wearing the exact equipment set he stole at the beginning. If you’re wearing anything else he just scolds you, tells you what you’re missing (but not where it is!) then teleports you out.

If you show up wearing the exact gear he stole, he’ll fight you as normal, but he disables your pause button so you can’t swap out for better gear during the fight.

Since all the Vlad relics are back there, you’re hunting for six random things instead of five. (Weapon, shield, armor, hat, cloak, accessory.) And it guarantees at least one boss fight where you can’t use broken gear or item spam. (Unless you happened to start with broken gear, of course.)

For the truly masochistic, let’s add yet another type of door — Death’s Door — that can roll into any of the above blockade spots, and only opens if you’ve completed Death’s collection. If that door is in your logic, you have to hunt for eleven arbitrary items instead of just five or six. That’s at least the equivalent of a Link to the Past pedestal seed!

Or maybe I’m dumb!

Any or all of these ideas might be terrible, unworkable, or just plain impossible to implement. I have no idea. I know they work pretty well in other randos, but as with the comparison to Super Metroid what works in other randos might not translate well into Symphony. I do hope the rando continues development, and I do hope one day they get it into a state where I find it fun to play. As it stands, Symphony of the Night Randomizer is going to be a Sometimes Game for me. Even at its worst, it’s still Symphony, which means it’s still pretty good to just dust off and play.

But man, I would love it to be something I play three times a week.

Useful Links

You can try Symphony of the Night Randomizer here: https://sotn.io/

You can try Super Metroid VARIA Randomizer here: https://randommetroidsolver.pythonanywhere.com/

You can try Link to the Past Randomizer here: https://alttpr.com/en

You can try Final Fantasy IV: Free Enterprise here: http://ff4fe.com/

You can try Final Fantasy Randomizer here: http://finalfantasyrandomizer.com/

I will not provide you with ROMs, please don’t ask. Thanks for reading!