gman-003

[gman-003]

Magic Experiment Log, Experiments 001-006

I love Magic: The Gathering. Shit, half the reason I’m on Tumblr is to read Maro’s blog.

One of the things I particularly enjoy is the ability to build new things. Like many players, I don’t copy popular decks, I brew my own. Usually they’re jank, sometimes they work, but I enjoy them all the same. The fun is about the process, the trying out of new ideas. And I do take a somewhat scientific approach - I don’t track numbers all that precisely, but I test ideas by experiment.

I go a bit beyond most players. I do custom card design so I can explore design space I don’t think will ever be explored. I’ve tinkered with creating entire formats - I’ve played Magic without rule 104.3b and 104.3d, just to see what that does to the metagame.

And so it should surprise nobody that, when I made a Cube, I did some weird things. I’ve been playing it for a few months now, so I’ve tried a lot of new ideas. Some of them worked really well, some didn’t. And in the spirit of scientific “publish or perish”, these are the rough results of all my sciencing the shit out of Magic.

Experiment 001: Altered Pack Structure

Most official sets consist of packs with one rare/mythic, three uncommons, ten commons, and a basic land. Some sets will swap the basic land for a common nonbasic (like a guildgate) or a themed card (a conspiracy or draft-matters card in CNS/CNS, or a contraption in UST), and some will swap a common for a double-faced card.

Hypothesis: Having only a single rare reduces drafting decision space. Having a minimum of two rares means your pack 1 pick 1 isn’t just automatically the rare.

Experiment: The pack structure is as follows: six commons, four uncommons, two rares/mythics, one dedicated multicolor slot, one dedicated “illegal” slot, and a nonbasic land. The latter three slots can be of any rarity, intended to be roughly equal between common/uncommon/rare. This gives an average of three rares, five uncommons, and seven commons per pack.

Result: Overwhelming success. The two-rare minimum has been very well received. Drafting involves a lot more tough decisions. Rares make it farther into the draft round - I’ve gotten rares as the last card in a pack! And the higher as-fan makes games less swingy. You’re much more likely to see your bombs - games have less variance. I’ve never felt that I lost because I didn’t see the one powerful card I needed - and I’ve never felt I won only because I started with the one OP card.

Experiment 002: Illegal Cards

As a fan of the more out-there cards, and as a guy who’s bought boxes of both Conspiracy and Un-sets, I obviously would want to try those cards in my Cube. But, I did not want to make it a full-on “silly” cube, rather just add a little bit of fun to an otherwise normal game.

Hypothesis: Adding a dedicated slot for whimsical cards, not legal in constructed formats, bolsters enjoyment without making the game too weird.

Experiment: A slot was added to the packs, containing silver-bordered cards, conspiracies, and draft-matters cards. Due to the low as-fan, parasitic cards were excluded - there is no host/augment, no dice-rolling-matters, no contraptions. As a singleton Cube, conspiracies needing you to note a card name were also excluded.

Result: Huge success. Players universally liked the fun cards, but keeping it checked kept the games relatively normal. Sure, sometimes you’d die to an animated library, or to a Modular Monstrosity with protection from everything, triple strike, lifelink, squirrellink and horsemanship, but it still plays like Magic.

I am slightly concerned by how rarely some cards see play. Split Screen sees play every time it’s in the pool, but I don’t think I’ve seen Gus or Extremely Slow Zombie make the cut. And Archdemon of Paliano is the only card to have effects during the draft, which I think players are overlooking.

Experiment 003: Three-Color Overlapping Archetypes

The difference between a proper Cube and a pile of cards is design. You can’t just select your favorite or most powerful cards, you need them to work together well enough that strategies exist, but not so synergistic as to eliminate strategy.

Hypothesis: A set built around enemy-color synergies within wedge-color archetypes will give a good balance of synergy and creativity.

Experiment: The Cube was constructed with wedge-color archetypes (WUR equipment, UBG poison, BRW attack triggers, RGU energy, GWB graveyard), with the intent of overlap along enemy colors (WB is aggressive but can recurse from graveyard, UR has noncreature-cast triggers, GU has proliferate, etc.). Additionally, a dedicated multicolor slot gave a very high as-fan for multicolor “signpost” cards - for each three-color archetype, one three-color card and one card of each included color pairing is included to provide guidance.

Result: Eventual success. Early iterations of the Cube were too directional, and included too many cards dedicated solely to a single archetype. But, after revisions and expansions, it seems to be working much better. Players are still able to find novel synergies (mono-green Voltron?), but they can still find direction if they need it.

Experiment 004: Multiplayer

I, like many others, am a big fan of multiplayer Magic. Most of my players come from my Commander playgroup, or my Multiplayer Modern playgroup. So it’s only natural the Cube is also multiplayer.

Hypothesis: The Conspiracy Draft model, of drafting in a group of 4-8, then if needed split into pods of 3-5, will work even with only a modest amount of multiplayer-matters cards (3 Melee, 2 Dethrone, 2 Will of the Council, 6 Council’s Dilemma, 4 “can’t attack you”, 2 goad, 1 myriad, 1 friend/foe).

Experiment: Drafts with 6+ people split into two pods for play. If we have the stamina for multiple rounds, we switch up groups each time.

Result: Expected success. It’s not really a surprise that something Wizards actually did works in a more general case. The higher variance seems to be making three-player pods play better than three-player Modern, of which I have played far too much.

Experiment 005: Non-Color-Balanced Packs

Something I have noticed I’m weak at is hate-drafting. I often end up with a surplus of cards I can make use of, and so does everyone else because I didn’t take the pick 4 bomb that wasn’t in my colors.

Prior to this experiment, I had been color-balancing individual packs. I treat colorless as a color for purposes of filling packs - the six commons are one of each color, and the uncommon slots and rare slots did not contain repeats.

Hypothesis: Producing packs with deliberate color focus will force more hate-drafting and add thoughtful decisions to the draft.

Experiment: The Cube packs were filled as follows. One pack each of each single color or colorless cards (multicolor was of that color, with an off-color ability). One pack each of each color pair. One pack each of each of each three-color wedge. Two packs of all five colors. Two packs of colorless cards.

Result: Abject failure. This led to extreme dissatisfaction. One color was present in every single player’s deck, and another color was completely absent despite having a mono-colored pack in the second round. I abandoned this experiment without even drafting the full box, emptying the half that remained to re-seed. I don’t regret the experiment - it wasn’t a “failure” as an experiment, just as a hypothesis. The experiment worked wonderfully - the data showed very strongly whether it was a good theory or not.

Experiment 006: Adding Planeswalkers

My initial version of the Cube did not have planeswalkers. Mostly because they’re hella expensive, and I was on a budget, but also because they generate a lot of value over time. You get a good planeswalker out, you dominate the game quickly.

And then I played War of the Spark limited. So clearly I had to add some, especially with proliferate being such a central mechanic in the Cube.

Hypothesis: Adding planeswalkers will be hella fun, and possibly rad.

Experiment: Twenty planeswalker cards were added. The uncommon slot got five monocolor PWs. The rare slot got five monocolor rare/mythics. The gold slot got five enemy-color hybrid uncommons and five enemy-color gold rares/mythics.

Result: Mixed, mostly positive. The hybrid planeswalkers aren’t quite a good match for the mechanics. Saheeli and Nahiri are great, Kiora isn’t on-theme but works well anyways, Kaya is a bit expensive for those aggressive colors, and Vraska just outright doesn’t fit (I have very little deathtouch, probably because of the high level of infect). The mono-color rares are very swingy, possibly stronger bombs than I wanted. But, maybe some higher variance will be good - people certainly enjoy it when they open a pack and see Gideon Blackblade.

The uncommons work the best. The monocolor ones slot in nicely to everything, and the hybrids usually fit. They don’t take over games, but they do put in the work, especially with proliferate around them.

I think that’s enough science for this post. Maybe if I leave some for later, I’ll actually post regularly.

#mtg #magic: the gathering

[gman-003]

I’m going to make a contrarian argument: Vote Bird.

What does Lightning Bolt do? Just two things: it’s removal against creatures and planeswalkers, and direct damage to players. There’s lots of nuance there - killing a four-toughness creature by hitting it after combat damage, bolting a runaway planeswalker to buy another few turns before it can ult, the best time to throw a bolt to the dome. It does both of them quite well, but it’s still ultimately two things.

Birds of Paradise does more. It’s mana acceleration - by its low cost and ability to generate any color of mana, it’s arguably the greatest mana dork ever printed, although not brokenly overpowered in the same way as acceleration like Black Lotus and the Moxen. But it’s also a creature, which gives it yet more uses. It lets green chump-block a dangerous flier - sure, you don’t want to waste a bird this way, but it’s saved me many times. And it can attack, if you can find a way to grow it. I’ve won games by swinging through an Ensnaring Bridge at zero, with a Bird I then pumped to 20/21.

That’s what makes Birds such an endearing card. There’s stuff you can do with it that doesn’t feel like it was designed. It feels like you’re finding a new and unintended way to use a card, whenever you throw a Giant Growth, Rancor or Grafted Exoskeleton onto it, or use it with an anthem or exalted triggers.

Magic, to me, is a game of finding interactions. It’s a game of building decks, not just cards. Bolt is a great card, but it’s a great card in isolation. What red decks have the choice to run it, but do not? Birds of Paradise is a great card that only gets better when you combine it with others.

The Grand Final

It’s here.  After 589 surveys and more than 16,000,000 votes, we have reach the finals of the Magic Bracket.

You can vote on the final between Lightning Bolt and Birds of Paradise right here.

It’s been a wild ride - thank you to everybody for all your support over the past two years.  Looking forward to announcing our winner on October 31st!

This is also the last stretch for our fundraiser for The Samaritans.  Please give generously to support their vital work supporting the lonely, depressed, and suicidal: https://www.justgiving.com/fundraising/magicbracket

#TeamBirds #mtg

[gman-003]

hi, i saw your comment and post about the Planeswalker set, and I'm wondering if i could pick your brain about it? I've been trying to make a custom card cube with a higher than normal Planeswalker count and I've been having trouble making Planeswalkers that aren't too goodstuffy and samey, what kind of triggers and such were you playing around with?

So there were a couple different things I tried with the triggers. I was particularly focused on trying to get Planeswalkers working at lower rarities (rare/uncommon), which needs them to stay on the board for a shorter time. Board state pretty quickly gets out of hand if planeswalkers aren't dying.

For low-rarity planeswalkers, the trigger usually replaced the plus ability - something would happen and the ability would put a loyalty counter on them. The minus and ultimate abilities stayed about the same, although you need to cost appropriately. I only experimented with fairly straightforward triggers - "Whenever you draw a card, put a loyalty counter on CARDNAME" or "Whenever a land enters the battlefield under your control" or "Whenever you cast an instant or sorcery". There's dozens that you could do for this - battalion triggers, exalted triggers, constellation. Conditional upkeep triggers might also work - "At the beginning of your upkeep, if you have no cards in hand, put a loyalty counter on CARDNAME" for a Hellbent-like trigger would be worth investigating.

I found some problems while trying to replace the minus or ultimate abilities with triggers. Replacing the minus leads to the plus ability being the only thing to do until you can fire off the ult, which is kind of boring. Replacing the ult is more viable but players might hold off on playing the card until they can trigger it immediately, which makes it harder to respond to, which in turn makes it problematic to do anything really powerful. Finding good conditions is also hard - although I did come up with some. "Whenever you cast your tenth spell of a turn", "Whenever you attack with ten or more creatures", and "Whenever you cast a spell with converted mana cost ten or more" might work as starting points for you.

I had better luck replacing the ult with a static ability, and making the minus effect stronger and costlier while still not being quite an ultimate. "Zombie creatures you control get +1/+1; +1L: return target Zombie card from your graveyard to your hand; -5L: create a 2/2 black Zombie token for each card in your hand" was one example (note the synergy between the two loyalty abilities).

An important realization I had about pushing planeswalkers as a card type was that they need to act more like support cards. You make a planeswalker to support a specific strategy - make a Storm 'walker or a Control 'walker or an Affinity 'walker. It isn't the card that wins you the game, it's the thing that helps the rest of your deck win the game. This goes a long way towards making planeswalkers less samey - assuming you have enough different archetypes in your format.

[gman-003]

Why are Google’s algorithms so bad?

As I write this, the top two stories in the “Science” category on the News app on my phone are a moon landing truther piece and a Nibiru doomsday prophecy piece.

Those are literally the opposite of science. They’re bullshit. I don’t think they should be shown to anyone, but fine, Google doesn’t care about not being evil anymore. If someone is an idiot and would click on it to hear the writings of another idiot, Google would probably show it to them. But they’ve got access to basically all of my data. They know I have the SpaceFlightNow app and do Google searches about rockets and have followed Dr. Plait’s astronomy blog through three different sites. They have my browsing history to know that I clicked on stories about the eclipse or solar storms or new rocket technology. They have my Youtube history to know I watch every rocket launch livestream that happens while I’m awake.

Why the hell do they think I’d click on some conspiracy-theory science-denial bullshit?

It isn’t at all uncommon for me to see something like that from Google. I’ve seen stuff about UFOs or chemtrails or horoscopes or alien contact crap pretty much weekly. And never once have I clicked on it. Why do they still keep popping it up? Can they just not distinguish between “interested in astronomy and spaceflight” and “raving nutter who thinks Hillary has a pedo ring on Mars”?

Their algorithms elsewhere are ass, too. Youtube recommendations? They’re basically just more vids from people I’ve already watched, plus some of whatever is trending globally - and they somehow managed to forget some videos I’ve watched, and want to recommend them to me again. Google Search? Loves to give me crappy sources (why are W3Schools’ inaccurate docs ranked higher than MDN when Google is literally a contributor to MDN? Why is Google Search so bad at pulling up Google’s own Android documentation?) and the autocorrect loves to correct things that I was specifically looking for (why would you auto-correct one ten-character model number into another?). Google Maps? Passable as a “get from one known address to another known address” tool, but just try asking it for recommendations.

If this were the 90s, and Google was just a tiny startup, this would be tolerable. Cute, even. But Google has approximately all of the money, and literal armies of developers. These problems aren’t easy but they aren’t that hard. They’re solvable problems.

#angry rant #google #computers #bullshit #algorithms

[gman-003]

Cleverness - The Legend of Zelda: The Minish Cap Analyzed, Part One

Finally. After what feels like a solid year talking about Skyward Sword, I'm moving onto the second game on the list: The Legend of Zelda: The Minish Cap.

The game is, overall, pretty unremarkable. It makes no big changes to the classic formula every 2D Zelda game since Link to the Past has used. But it makes a lot of small changes, and a lot of them try to be, for lack of a better word, "clever". We'll look at two of them today.

Recontextualization

The Minish Cap re-uses elements all the time. Enemies get palette swaps to go with new abilities or increased stats - but that's a very common trick. The Minish Cap also likes to use early-game minibosses as late-game regular enemies - but this, too, is a standard trick to ramp up difficulty. What I found clever was how they used enemy designs in such a different context that they become massively more dangerous.

Many 2D Zelda games have a parallel-map system of some sort, a "dark world" or something. Minish Cap uses the same map, but changes the player's size, shrinking them to allow them to fit through nooks and crannies, explore unexplorable corners, but also limiting your ability to climb or fight. While normal-sized, you dispatch moblins and darknuts with ease, but while small, even harmless insects turn into enemies. This is exploited to turn low-tier enemies into full-on boss fights.

Chu-Chus, at normal and Minish size

Early in the game, before the first dungeon even, the slime-like Chu-chu enemies are not very dangerous. They're just cannon fodder, something to keep the player engaged while traversing the overworld. But the first dungeon takes place in a tree stump, entirely while shrunk - and the boss of that dungeon is a chu-chu. A short cutscene shows it to be normal-sized, but compared to the player, it is now enormous. It isn't a palette-swap fight at all - the graphics are completely different, to fit the appropriate scale, and it is fought in a completely different way. From an abstract design perspective, it's a completely new fight, but the re-use of an identifiable in-story enemy type serves not just to borrow some of the player's experience fighting them at normal size, but emphasizes to the player the difference in the two scales of the world.

Octoroks, at normal and Minish size

This exact trick is used again, later - this time with an Octorok, in the Temple of Droplets. It uses this trick exactly the same way. The boss uses completely new art assets, is fought in an almost completely different way, and in the abstract, can be treated as a completely new enemy... except for the visual relation. The player "knows" it's the same enemy, that it's them who has changed.

This dungeon also features a reversal - a "giant" Chu-chu serves as the miniboss, but this one is a new variety (blue) with a new ability (electricity). Only after the miniboss do you begin to encounter normal-sized blue Chu-chus, and can draw on your experience with the miniboss battle to defeat them easily. I'm not entirely sure that reversal works well - it forces the player to experiment with items in a rather tense setting, and then robs the later introduction of the normal-sized ones of its impact.

Attack All Resources

While the concept of game-designer-as-antagonist has rightfully fallen out of style, there is some value to thinking that way. "How do we beat the player?" is a useful mode of thought, because a challenge loses its edge if victory is not in doubt.

The Minish Cap has a useful trick up its sleeve - while enemies in most Zelda games are content merely to damage a player's health, some enemies in The Minish Cap will deal damage in the form of rupees.

Half the enemies on this screen will steal from you

In the late-game, players are rarely fazed by damage. You've found enough heart pieces that you can just tank hits. The classic response is merely to increase the damage values, to negate the increase in health, but when done too often, that wipes out the sense of accomplishment. The Minish Cap throws the player the occasional curveball, with several enemies that steal your hard-earned rupees. A quick kill will return them, but running away will cost you.

That is what this lesson boils down to: every asset the player has, is an opening to attack. Whatever the game gives, can be taken away.

In fact, The Minish Cap could have gone much further with this. No enemies drained your magic meter, or ammunition. The only item to be stolen is your shield (something the series has done since the beginning). A single cutscene-only enemy steals a key, which came across as more tedious than dangerous.

Seriously, what was the point of this?

Would the game have benefited from enemies stealing your other weapons? Would that key-stealing crow have worked as a repeated enemy? I think so - the game did drag during the last few dungeons. Some extra variety would certainly have helped. But too much of this trick undercuts the importance of the health system, so maybe it's for the best.

Next time: more small changes.

#the legend of zelda #everyzeldaproject #loz #minish cap #game design

[gman-003]

Skyward Sword play log excerpts

Oh hey, guess who’s still alive? I kind of got sucked into a vortex of freelance projects and moving, and I had completely no time to work on this. I didn’t even queue up the stuff I’d pre-written, things were so hectic. But it seems to be stabilizing now, more or less. I’m not going to try to commit to a schedule yet but I’ll still update on a Tuesday when I do.

While playing, I kept a log of thoughts and experiences, to aid the writing process. Since I'm in the middle of playing, I tend to write in an abbreviated, stream-of-consciousness style. Sometimes it turns out kind of funny. I thought I'd go ahead and share some of the better bits.

And then I swear we are DONE with this fucking game and I’ll be posting about Minish Cap next.

Side note: start with six hearts? wtf worst zelda ever

Six notes in and I'm already declaring it the worst game in the series. I am the *best* at games criticism.

Do the Loftwings not get names? Very bad writing to say "you have a close bond" with a nameless bird. Even just "Crimson" would have been something.

I continued to refer to Link's Loftwing as "Crimson" throughout my notes, because seriously, why don't they have names, if they're supposed to be so important? And why couldn't you just let the player pick a name, the way you could name your horse in earlier games. I judged this complaint too trivial to waste time expounding on, but still, it bugged me.

The idea of a sentient spirit embedded in the Master Sword (this isn't the Master Sword yet but I'll be fucked sideways if the game doesn't end up with that happening) is cool

I also don't filter out my profanity when jotting down notes. "Fucked sideways" is an abbreviated form of "fucked with a sideways rake", which doesn't actually make much more sense, come to think of it. I'm not sure what axis you rotate it on to go sideways...

An invisible voice (a new one this time) tells me to hit it with a skyward strike. I initially thought this was a blatant trap to release some ancient evil, and tried to find ways to bypass it but no, that's actually what you're supposed to do.

Seriously, I thought hammering the sealing spike home was a trap to release an unspeakable horror that would form the rest of the game, and was looking for a sneaky way around it for a secret ending. Nope. You're supposed to listen to the invisible voices. That's practically the moral of *Skyward Sword*, since Zelda also triumphs through the power of listening to invisible voices.

minesweeper in a zelda game? I was not ready. Or actually I was. Perfect beginner game on third try, moved on.

Do not ask how I got to be so good at Minesweeper. It would ruin my hardcore-gamer cred. I'm just naturally good at logic puzzles - that's my story and I'm sticking to it.

game stubbornly continues to give me rupees (full wallet) instead of hearts (1/2 heart left)

You'd think eleven thousand words complaining about a single Zelda game would be enough, but no, I had to cut a lot.

technoblins are constantly blocking me. checking strategy guide for how to not get blocked. Strategy guide useless

This would not be the last time the strategy guide lied to me.

woo vent crawling. expect headcrabs?

wtf groose you live in the sky how do you not know what BIRDS are

 so impa's an angel now. and also gandalf

fucking water temple

This will likely not be the last time this phrase makes an appearance.

Tentalus is supposed to look scary? It looks like a plastic toy teenage girls attach to their cell phones

hey ghirahim, all you had to do to find the second time gate was follow Link stealth-like and wait for Fi to remind him

A lot of game villains would do well to spy on the hero and overhear their advisor characters telling them exactly what to do.

wasn't hyrule polytheistic?

Seriously, can someone make sense of Hyrulian theology? Is Hylia a higher or lower goddess than Farore/Nayru/Din? How is she supposed to fit into the creation myth? If the Hyrulians have a race-specific guardian deity, do the other races have the same? THESE ARE THE QUESTIONS WE WANT ANSWERED.

obligatory zelda tennis boss

Again, we'll probably see this note in pretty much every play log.

tadtone (eugh) search goddammit it's not even a search, it's a bunch of "solve my maze" bullshit

Superman 64 is my go-to reference for this particularly stupid and annoying minigame archetype. Does anyone like it when a game puts the fun on hold and makes you literally jump through hoops?

Oh sure THIS GUY recognizes me as a hero right off the bat, no stupid PROVE THY WORTH quest

I don't remember exactly who THAT GUY was. One of the dragons, probably.

LD-Link-16 start your theories

In reference to a robot in the mines section, near the dragon. But I never actually came up with a good theory for this, although it *is* evidence of prior Links in the timeline, which implies *Skyward Sword* isn't the first one.

#the legend of zelda #everyzeldaproject #loz #skyward sword #game design #isanyoneactuallyreadingthis

[gman-003]

I felt like changing things up for this week’s post.

A few months back, I was inspired to design a custom Magic: The Gathering set based around a simple concept: what happens when you push planeswalker cards as hard as possible, get them to an absolutely insane as-fan for any noncreature card type? We’ve had artifact sets and enchantment sets and creature sets - why not a planeswalker set? What other design tech do you need to pull that off, and what design space does that open up?

I didn’t finish the set. Not even close. The file has sixty cards in it and half of them are garbage. But I treat custom sets as closer to “exploratory design” than “design” - a place to try out new ideas, see what the problems are. There’s some designs in there good enough to share.

This is intended to be the set’s chase rare - the uber-splashy, iconic card that makes people buy booster boxes. It’s the kind of card Wizards would spoil first, so I decided to make it my own first spoiler.

Obviously, this card is heavily inspired by the infamous Tarmogoyf. But he has a different feel to him. He can get bigger - a LOT bigger. With the current list of planeswalker types, he can grow to a 35/36, and I expect that would roughly triple if the set was completed. Yet in a normal metagame, he’d rarely reach 2/3, bad stats for a 3-drop green creature. He also grows more slowly, since planeswalkers tend to stick around longer than creatures. But that’s the kind of weirdness that happens when you take the rarest, most powerful card type in the game, and make it your theme.

The cost was the hardest part of this card. The concept was simple, the templating was already done for me, the name is just Latin for “spark eater”, and the other stats were intrinsic to the template. But I still don’t know whether 1G or 1GG is the right call. At 1G, he might see Modern/Frontier play (assuming at least some of the ‘walkers break into the meta), but he’d be banned instantly in Standard, which would probably want him at 2GG or higher. 1GG is, I feel, a good compromise that still calls back to earlier lhurgoyfs.

(I am aware the art is twenty types of awful. I am not, in any way, a visual artist. But I can never find art that fits the card, and I think it’s better than leaving it blank, like most custom-card designers do.)

#magic: the gathering #custom magic card #custom mtg #programmer assets

[gman-003]

Refusing the Call: The Monomyth in Interactive Media

While this was inspired by my Skyward Sword playthrough, it's really more of a general question.

The monomyth is literally the archetype of a legend - a recurring pattern, seen through mythology and legends and epic stories across every culture. We see it in the myths of Osiris and Gilgamesh, we see it in the Eddas and the Odyssey, we see it in J.R.R. Tolkien and C.S. Lewis, we see it in Aladdin and Harry Potter. Star Wars and The Weatherlight Saga and Journey were deliberately written to that pattern. It’s all over epic fiction.

Not all components of the monomyth are required. "Temptation" is frequently skipped, as is "Crossing the Return Threshold". It's not a strict formula, the way the modern three-act structure is.

Most Zelda games incorporate at least a few steps of the monomyth. There's always a "Call to Adventure", a "First Threshold", a "Road of Trials", and an "Ultimate Boon". There's usually a "Meeting with the Goddess" and a "Magic Flight".

Skyward Sword hit a lot of the steps more often skipped, as part of its attempt at an epic, legendary feel. But one of them fell flat - even to the point of outright frustration. "The Refusal of the Call" - upon receiving the call to adventure, the hero at first refuses it, often out of a sense of obligation or inadequacy. Luke Skywalker at first refuses to join Obi-Wan on the adventure, out of duty to his family; Bilbo at first refuses Gandalf's call, preferring to stay home and live a quiet life; Peter Parker at first uses his powers for his own enrichment, not to fight evil, until stirred to adventure by his uncle's death.

Link, in Skyward Sword, tries to refuse the call twice, both times in cutscenes. When he first meets Fi, below the Goddess Statue, he tries to shirk the responsibility of hero, needing Headmaster Gaepora to convince him otherwise. Then again, when meeting the wise old crone at the Sealed Grounds, she repeats the Call to Adventure at Link, who tries to shrug it off. I remember shouting at the screen, "Just shut up and let me go on the adventure already!". I - the player - was champing at the bit to answer the Call to Adventure, but Link - the player character - was holding me back.

Much has been written about the relationship between players and player characters. Sometimes a disconnect is forced, to further the aims of the story. The player can be given information the player character does not have, to patch up a plot hole. We might be given a different reason to desire the same goals - the player character may want to liberate his homeland, but the player just wants to fight more bad guys.

This doesn't feel like a deliberate disconnect. Nowhere else is a similar player/avatar disconnect used. Never again does Link show this lack of confidence. Pretty soon, he's the one guiding others into this strange new world.

Part of the problem is the sheer length of the game's intro section. This cutscene takes place over an hour into the game, before any of the gameplay that one expects from a Zelda game has started. If you're anything like me, you'd have been exasperated by this point, frustrated by the lack of adventuring and exploring you expected to be doing.

But I don't think that's what really killed it. It was the lack of interactivity - taking agency away from the player - that allowed for that player/avatar disconnect.

A true Refusal of the Call seems to be pretty rare in the gaming medium. I can recall offhand only a handful of games that did - mostly RPGs. It's problematic in games because it is intrinsically incompatible with the core game mechanics - whatever gameplay your "adventuring" consists of must be abandoned in order to make a Refusal of the Call seem like an actual refusal. You could, I suppose, create entire new gameplay just for it, but that's a lot of work to do something that probably ought not be done at all. After all, if the gameplay of the Refusal is more fun than adventuring, players will be let down by the vast bulk of the game; if it's less fun, then we're just wasting even more of their time.

So when a game does do it, it takes the form of non-gameplay. If it's not a pure, cinematic cutscene, it's an RPG-style dialog, perhaps giving you the "choice" to not go out on an adventure, but not letting the scene end until you choose properly. These are actually more effective at ending the Call to Adventure stage of the monomyth, since they force the player-character to choose to leave their old world behind.

Is it possible for a game to successfully spend a lot of time on the Refusal of the Call? I think not - the player, simply by virtue of choosing to play the game, has already chosen adventure, and the Refusal of the Call can only delay getting the player to what they want.

Maybe there is a way, though. Perhaps it could be built into the tutorials, another early-game section that keeps the player from the core game, but a more obligatory one. Merging the two would obviously give you more leeway to do a Refusal bit, since it wouldn't be adding to the barrier time, but I can't see a workable narrative for it. That doesn't mean there isn't one, though.

The existence of one possible way to do it well suggests the existence of others. Can you think of a way to do the Refusal of the Call with gameplay instead of cutscenes? Or even have examples of it being done well in an existing game?

#monomyth #writing #games #zelda

[gman-003]

Trust - The Legend of Zelda: Skyward Sword Analyzed, Part Six

(This is part of my ongoing series analyzing The Legend of Zelda: Skyward Sword, part of a bigger series where I play through and then talk excessively about every game in the Zelda series. If you want to catch up, earlier installments can be found with the #everyzeldaproject tag)

Every game has to trust the player to be able to play the game. At some point, after some amount of tutorials and hints and explanations, if a player still doesn't get it, that's their fault, not the game's. As a designer, you need to give the player everything they need, and then trust them to be able to do what you want. A puzzle isn't a puzzle if you hand them the solution. A battle isn't a battle if you make it impossible to lose. A dungeon without exploration isn't a dungeon.

Skyward Sword fails at this a lot, and I have suspicions about why. But I'll start with the places they did it well.

Good: Bokoblin Base

There's a section late in the game where, while skydiving into an erupting volcano, you're hit by a sudden updraft and end up captured by Moblins and stripped of your vast array of weapons and items. You escape, but without weapons you have to sneak around enemies that you're used to easily killing, and without tools, you can't easily move about the area you've thoroughly explored (but which has been built over by enemy fortifications).

Item by item, you reassemble your arsenal. First the Mogma Mitts, then the Gust Bellows, then the Clawshots, and so on until you get back your trusty Sword. And while the game does point you at the next item-bearing chest, it stops at that. It doesn't waste time telling you how to use each item - you've already proven you know that. It doesn't point you at the obstacle your new item can get you through - you can remember passing by it, or if you've forgotten, you can go explore a bit.

It's not a very lengthy section of the game, and it can get a bit excessive about pointing out where enemies are, but overall, I think it's one of the best sections of the game. Everything just clicks together. The area gets new life by giving a very different context, the low-level enemies become dangerous again, and you get a fresh set of puzzles. It's a moment of game-design cleverness in a game otherwise starved of it.

Okay: Sky Keep

The final dungeon of the game also makes an attempt at clever design. And it feels like the designers realizes that, in the final dungeon, it would be appropriate to cut back on the handholding, since if the player doesn't know how the game works by now, they never will.

But the final dungeon is a gimmick dungeon - it's a three-by-three grid, with the rooms being the pieces of a slider puzzle, which you control from special pedestals. The rooms have limited connection points - you can't just wander freely between rooms - and they even require you to enter from the "correct" entrance in order to make it through, thanks to liberal use of one-way doors.

That gimmick is explained - but what was not explained was that there is more than one location from which you can control the dungeon layout. My natural inclination when faced with a slider puzzle is to attempt to solve the entire thing. But that's actually impossible here - I actually sat down and mathematically proved it impossible before I resorted to the strategy guide.

However, lots of players would have ended up reaching for that guide anyway. While the dungeon presents the illusion of control, being built around a standard puzzle, it really forces you down a single, very specific path. You don't even seem to be able to go after the three Triforce pieces in whatever order you want - it's completely fixed. The slider puzzle is less a single puzzle and more three separate slider puzzles, where you have to guess at what the intended result is. It's functionally identical to three separate, unrelated puzzles, in a dungeon with a fixed, linear layout. The controlling pedestals are always positioned in ways that limit your possible moves.

The Sky Keep doesn't trust the player. It merely pretends to.

Bad: Hand-Holding

Everywhere else, though, Skyward Sword practically rips the controller out of your hands. The designers were deathly afraid that a player might, at some point, not know what to do - and so they made sure that you were always, always, being told what to do and where to go.

The normal Zelda routine is for NPCs to make references, in cutscenes and dialog, to your next main objective. You get a cutscene after each objective, where some exposition-spouter points you in the right direction, and if you forget it, you can talk to NPCs and they'll generally get you pointed again.

Skyward Sword does that, and then wallops you over the head, in case saying "there's something weird going on at [place], someone should go check that out" is too subtle. After every single of those post-objective pointer cutscenes, Fi pops up and tells you, to your face, where you need to go and what to do. There isn't even a delay - she just pops up, during the same cutscene, regurgitates the information you just received, and practically orders you onward.

Modern Zelda games have had somewhat illusory exploration. You always get a good push in the right direction, to keep you moving forward. But they maintained the illusion - the pretense that it was really you, the player, who was deciding what the best place to explore first is. It allows you to trick yourself into thinking that it was all your idea. Fi's repetition isn't just repetitive, but it shatters that pretense. It's no longer a game of the player exploring, guided by friendly NPCs, but the player merely following the path the NPCs have set for you.

This isn't even a full rundown of the ways Skyward Sword hobbles you with training wheels. There's the gossip stone, which gives video tutorials on request. There's the tag on your save file, reminding you of your general task when you come back to the game. There's a few sections (mostly sidequests) where it outright rips the choice out of your hand, putting the right item into your hands automatically and preventing you from switching. There's dowsing, which largely negates the need to explore. I'm sure I managed to forget some, since I'm writing this well after finishing the game.

The Wii Audience Problem

I have a theory about why Skyward Sword is so overloaded with hand-holding. It's because of the success of the Wii with casual gamers.

Twilight Princess, the first Wii Zelda game, had the normal amount of hand-holding for a recent Zelda game. More than I'd like, but it was still a very opaque, complex game. Other than a tedious pile of tutorials at the start, it generally lets the player do as they wish.

But it came out as a launch title on the Wii, before anyone knew just how popular the Wii would be with casual gamers. With it's dozens of mechanics, deep references to past games, and great length, Twilight Princess is a very poor game to get a casual gamer hooked on the Zelda series.

So is Skyward Sword, but I can see many things were done to try to make it a good casual player's introduction to the series. It has dungeons not much simpler than Twilight Princess (which isn't saying much), but it gives you plenty of save points and pointers on where to go. It strips away some complexity, getting rid of tunic-swapping and automating warping. And there's the handholding, constantly making sure that the novice player never gets lost or confused.

Constantly showing the controls on-screen? That's a very Wii Sports-like move, something you'd do for a game you might hand to someone who's never played it before at a party. Constantly telling the player what to do? People whose gaming up to now consisted of Bejeweled and Wii Sports aren't used to keeping track of multiple goals, like you have to do in any Zelda game.

My theory is that partway through development, it was decreed that this game be a good "gateway" Zelda - something that the Wii's casual audience might buy in effort of making their $300 purchase do something besides bowling, which would put them on a path of maybe buying more hardcore games.

This even explains certain other oddities. Why is Skyward Sword canonically the first in the series, despite not really fitting as an origin myth? Because that way, series neophytes aren't scared off by the thought of jumping into a story midway, or having too much nerdy lore to absorb. Why were motion controls so heavily integrated? To make it more appealing to the Wii Sports audience.

I don't have any evidence for this theory, besides the game itself, but it makes a lot of sense.

While this, finally, concludes my analysis of Skyward Sword, there's one tangent I want to go on before moving forward into The Minish Cap. Join me again next week!

#the legend of zelda #everyzeldaproject #loz #skyward sword #game design

[gman-003]

From the Archives: The Nintendo Niino and Super Smash Bros. N

I have a rather large folder of ideas I'd come up with, liked enough to bother recording, but didn't or couldn't actually pursue. Some of them are worth looking at again, particularly when intervening events cast it in a new light. (And if this post is received well, it might become a regular feature.)

In this case, it's an idea for a Nintendo console, which turned out to be surprisingly similar to the Switch, even though, with a timestamp of September 27, 2014, it predates any public knowledge of the Switch, and possibly even predates the Switch development project. While I rewrote it into a more readable format, I have not changed any of the actual details - preserving the state of the idea as it was when I conceived it. It's not a detailed plan, more an executive-summary proposal.

2014 was a lackluster year for Nintendo - sales were plummeting, the Wii U had fizzled, and the New 3DS was flailing. Even fans could tell the company was struggling.

I made a key insight, though I don't claim it was a unique one: the gaming console market is under pressure from two directions. Going back to the 90s, the cheaper handheld consoles served as the entry point for new gamers, with home consoles offering a superior experience at higher cost, and gaming PCs being yet more expensive. But the rise of smartphones has devoured the low end - an iPhone or Galaxy or Nexus is an objectively worse gaming platform than any handheld console, but since they're essentially free as gaming devices (since consumers will be buying a smartphone anyways for communication), they offer the cheapest entry point. The rise of cheap gaming PC hardware (itself an effect of slowing desktop sales) has pressured the high end, driving home consoles to lower price points. This prediction turned out to be fairly true, although I also thought "nanoconsoles" like the Ouya would contribute to the demise of the two-tier console market, which completely failed to happen.

The logical conclusion was that handheld and home consoles will need to merge or displace each other. Sony simply gave up on their portable line, focusing their in-house developers on the home console. Nintendo could easily have given up on their home console line, throw everything onto the 3DS or its successor, and hope to compete with Sony and Microsoft on the merits of portability and game design rather than technical specifications.

But that's only the obvious way to do things. There are problems with that approach - Nintendo has a much longer history than Sony of maintaining both form factors, and there are many Nintendo fans who would be angered if Nintendo chose to abandon one or the other. A non-obvious solution is needed - and a non-obvious solution is what we got, with the Switch. But let's take a look at the idea I came up with, because it took a third approach.

The console I came up with was to be named the "Niino", punning off "Nintendo", "Nano", and the double-i pattern used in the Wii, Mii and Amiibo. The name is kind of dorky, I'll admit, but it's better than "Wii" and that thing sold like hotcakes.

The core principle was that you couldn't make a handheld that worked as a home console as well as a dedicated home console would, and vice versa. The two need to be fully software-compatible, but even just making an ergonomic controller fights against being able to put it in your pocket. So I didn't go as far as Nintendo ultimately chose to - I still had different hardware for home and handheld use. But they were to be different SKUs of the same console, not separate consoles - all games would run on both, down to using the same cartridges.

I called the two versions the "Niino Home" and "Niino Pocket". I specified that both were to use the same architecture. An eight-core ARM-64 CPU was specified - either K12 or Denver, depending on whether AMD or Nvidia was offering a better deal. I further specified a fully-unified memory architecture, with 16GB of GDDR6 memory. (I will note that I also wrote that it would be released in 2020, so my use of stuff that doesn't even exist three years later isn't completely groundless). Games would be stored on internal flash memory (512GB on the Home, 128GB on the Pocket), or on removable cartridges. As a minor twist, the cartridges would be partially-writable - patches would get downloaded and stored, and your game saves would be on the cartridge itself, in a special R/W memory segment.

The Niino Pocket was spec'd with a 4.5" 1920x1080 screen, featuring capacitive touchscreen capabilities (aka multitouch). The Niino Home was specified to target 3840x2160 output resolution (fed over Mini-DisplayPort or HDMI), with an expectation that it would normally downscale to 1920x1080 (getting some free antialiasing in the process). To give the Home enough processing horsepower to render the same game at quadruple resolution, I gave it twice as many GPU compute units, running at twice the clockspeed, and bumped the CPU clocks up by 50% (while keeping the CPU core count the same). That math checks out if the CPU on the Pocket spends 25% of its time on controlling the GPU, and game logic does not scale with resolution which are generally reasonable assumptions.

For controls, I didn't do much unusual. Two analog sticks - full thumbsticks on the Home controllers, smaller PSP-style "nubs" on the Pocket. A D-Pad. Four main face buttons, presumably A/B/X/Y. Four secondary face buttons, two per side - Plus, Minus, Home and Share. I had two analog triggers, with a digital "click" at full travel, like the Gamecube did, as well as two bumpers. And to finish it off, I listed dual accelerometers - they aren't all that useful, but they're so cheap now, why not include them?

Games would be required to run well on both the Pocket and Home, as part of certification. Thanks to the identical architecture, OS, and even screen aspect ratio, the programming to support both would be minimal. Assets could be authored for the Home, and downscaled for the Pocket, or authored for the Pocket and simply reused, not taking advantage of the extra power. Some UI elements might need to be redesigned to work better on the Niino Pocket's physically smaller screen, or more particles might spawn on the more powerful Niino Home, but I was aiming for it to be easier than developing a smartphone app that runs well on multiple screen sizes. I even mandated that the cartridges themselves be the same - if you for some reason buy both a Home and a Pocket, you would still only buy the game once, and play it on both. With savefiles being stored on the cartridge itself, that could be a very useful way to play - almost as seamless as the Switch ended up being, although significantly more expensive if you want that capability.

That unified architecture would allow Nintendo to stop splitting their development resources across two consoles, which would in turn allow Nintendo to develop a more robust library of first-party games. As Nintendo would offer the only viable handheld console, it would make them a more attractive target for third-party devs.

But Nintendo consoles sell because of Nintendo games, so that's what I had plenty of. I listed two pack-in titles, four additional launch games, another four titles within the first year, and five in the second year, as well as as many third-party titles as possible. My third-parties list is amusing from today's viewpoint - I correctly predicted that there would still be annual Call of Duty, Assassin's Creed, and Monster Hunter titles, but I seriously missed the mark by listing "Sonic Boom 3".

The Niino would have two pack-in titles. Nintendo Sports was supposed to be a deeper follow-up to Wii Sports, adding campaign modes, character customization and even some map-making (on the Golf game). But that was really just there for the sake of having a "complete" pack-in game. In retrospect, this was a horrible idea - it would either miss entirely what made Wii Sports a system-seller, or would require compatibility with Wiimotes, which makes it a horrible way to show off the new system.

The other pack-in game was "Super Smash Bros. N", which was a free-to-play Smash... kind of. The pack-in version would include only a minimum of characters, items, and levels (I listed Mario, Link, Donkey Kong and Pikachu, along with a Mii-based custom fighter). But any Niino game could add more - first-party or third-party. That game's developers would be responsible for all the assets and initial programming, although balance patches would come from Nintendo's Smash team, and certification would make sure it worked right and at least came close to being balanced.

So when you bought the obligatory Mario launch title, you automatically get (for example) Peach and Luigi, along with a stage and some items, added to Smash. Spla2oon (I am still surprised that's not what Nintendo's actually calling it) would add Inkling. Metroid: Paralysis would add Samus (don't ask for details on the games themselves, I was just making up titles and one-sentence concepts). Call of Duty would presumably add one of the Captains Price. Even Virtual Console games could get in on it - some people would totally spend another $15 to buy Final Fantasy VI again if it gave you a Terra assist trophy (I think full playable characters are too much to expect for a VC game).

As additional ways to get that content (it is kind of scummy to lock it behind a game you may not want, if you're just a hardcore Smash player), they could also be sold separately, as normal DLC, or bundled with an Amiibo, assuming those continue to sell.

I really, really like this idea. It solves two problems with the Smash series - first, it makes it possible for characters from games released after Smash to appear in the current version, instead of waiting for the next console, and second, it allows the game to eventually have the kind of mammoth size that made Brawl such a wonder. I remember the Brawl spoiler season - the hype was unimaginable. Just when you thought they were done, they dropped more on you. And it will even have more good effects - it acts as a portal to game discovery, and gives a bump to the Niino version of multiplatform games. If you're playing Smash, either at a party or online, and you encounter a character you've never seen before, that might spark an interest in the game they came from. And if you're a hardcore shooter player looking at which new console to buy, maybe the free Smash characters would be enough to tip you towards the Nintendo platform.

Would the Niino have been a better console than the Switch? Maybe. It probably wouldn't have sold as well, because the messaging would be more complicated. And is one do-it-all console better than two single-purpose consoles that do their job better? That would depend on how much better the specialized hardware works, which could only be determined by actually building the things.

Would Super Smash Bros. N have worked well as a system-seller? I think it would, although it would require a permanent support team at Nintendo, something they don't seem to do. It might anger the hardcore Smash fans at first, but it would make for an overall larger game (for free!), and half of them are still playing Melee anyways.

What are your thoughts?

#nintendo #game consoles #from the archives

[gman-003]

Aesthetics - The Legend of Zelda: Skyward Sword analyzed, Part Five

(This is part of my ongoing series analyzing The Legend of Zelda: Skyward Sword, part of a bigger series where I play through and then talk excessively about every game in the Zelda series. If you want to catch up, earlier installments discussed the use of patterns, the story and how it fits into the chronology, the controls, and the flow and friction)

Video games are composite art. They combine dozens of artistic (and non-artistic) disciplines. You have game designers, writers, musicians, and an entire array of visual artists, specializing in textures, models, animation, lighting, and all kinds of other things. I've been focusing a lot on the game design, since that's an understudied field, but all the other elements deserve analysis too. We've already examined the story, so today, we'll take a look at the look of Skyward Sword.

Good: Overall art direction, graphics

When I review games, I make a distinction between "art" and "graphics". The former, as I use the term, is the timeless quality of the art direction, divorced from that specific implementation in the game; the latter is the technical work of making the software turn the artistic vision into reality, judged by what the hardware can theoretically handle. "Art" is that which shows up even in the manual and advertising, that which is unchanged in remakes and ports. Skyward Sword, on the whole, does an excellent job on both fronts.

The graphics are nothing to write home about by 2017 standards, but for a Wii game, it looks excellent. It's important to remember just how weak the Wii was, even compared to the Xbox 360 or Playstation 3. It also has some real limitations on what you can do with it, beyond just strength - since the processor design was identical to the Gamecube, it had very little flexibility, requiring a lot of trickery from programmers to get many special effects.

I'm not really equipped to dig into the technical functioning of it, but just from playing, it seems to hold a mostly-stable 30fps, rendering the game at 360p and upscaling to 480p before rendering UI elements. This is a common trick for console games - this hides the jaggies you get with no antialiasing, and fonts benefit far more from the increased resolution than the 3D environment does, while not costing nearly so much processing time to render. There was some consistent framerate drops during certain cutscenes - most scenes of a Timeshift Stone have some really nasty reduction in framerate - but during gameplay, it seems to be solid, which is when it really matters.

Some very interesting effect is done to create a depth-of-field-like effect, blurring the far background in a painting-like way. Again, I don't have the tools (or, quite frankly, the skills and time) to dig into exactly how this is done, but the effect is really well-done. This serves a gameplay purpose too, by making foreground objects (like enemies or items) stand out more from the background.

The design of most of the characters is also spot-on. Everything is stylized, anatomy is often exaggerated, and colors are bright and vibrant. There's visual depth to everything, lots of layers, which balances out the flatter colors. Faces are expressive. Costumes are iconic without being overly simple.

Even lots of the fully background characters look pretty good. Wander through Skyloft, and you'll see lots of interesting characters. There are very few background characters with a bland or boring design.

Okay: The Heroes

Which makes it so weird that the two central characters of the game are so dull. Skyward Sword's Link and Zelda aren't bad, really. They're just... completely unremarkable. 

Link is bland. He doesn't have any of that exaggerated styling that made other characters fun. He just looks like the Link from Twilight Princess with new textures. There's very little depth or layering - the dominant feature is just that blank green tunic. Given how much green there is in the environment, Link often fades into the background. This never reached the level of a gameplay issue, at least for me, but it's still not great when the player character isn't visually distinct from the game world, especially when the story is about him entering a strange new world.

Link also just doesn't fit in. His proportions are far more realistic than the rest, making him look out of place next to caricatures like Beedle or Strich or Gondo. It's not consistent - he looks fine beside Pipit or Kina or Peatrice - but that inconsistency is itself a minor problem.

Zelda's first design in the game is fine. It's got contrast, it's got depth, it's not the best thing ever but it's pretty good. And she's very well animated - I won't call it Pixar quality but it's leagues ahead of where they were even a few years prior. She's emotionally expressive and responsive. But then she has a costume change, and it's just... nothing.

It's a white dress. No other color. No layers. No depth. No frills or poofs or anything, besides a bit of lace designs that you can only see in the concept art. They still have her moving well but it's just... so... wasted. Could they really think of nothing more visually interesting than a featureless white dress to symbolize the purification stuff? You don't see her in this outfit all that much, but it's the outfit she's in for the majority of the game, so it's still very important. It wouldn't have taken much to make her design better. Give her any splash of color - a red rope belt (red string has a suitable symbolic meaning in east-asian culture), a purple feather from her Loftwing in her hair, anything really that isn't plain featureless white. Even making that lace rendition of the Hyrulian Royal Crest more visible would have worked. Even if you absolutely insisted on flat white, you could layer the clothing to give depth instead of a featureless white void. Or give her more to wear - a wimple or veil over her head even adds to that "purification" intent (and it's not like Zelda games haven't borrowed religious imagery before). I can't really point to anything that was actually done wrong with Link or Zelda. They just aren't nearly as good as many of the other characters.

Bad: The Villains

Some stuff was just outright bad, though.

Tentalus, the boss of the Sandship dungeon, is a bizarrely cartoonish enemy design. If its battle was intentionally funny, that could have worked - but the fight is framed as one of the most dramatic in the game. The ship shudders under its attack from the moment it appears, twisting in a section reminiscent of the Call of Duty 4 level "Crew Expendable", of all things. The skies are dark under the pouring rain, lit by distant lightning... the music builds to an ominous tension... as this Baby's First Lovecraftian Horror shows up. It's purple Cthulhu with tentacle dreadlocks. As more than one fan artist has noted, Tentalus looks like it came from a Monsters Inc. knockoff, not a Zelda game with a French Impressionist aesthetic.

And the single eye as a weak point, in this late stage of the game, is frankly unnecessary. Any player who's made it this far knows how boss weak points work. Absent a gameplay justification, it just further serves to make the boss look child-like.

For a boss that features so prominently in the game, The Imprisoned is also quite poorly designed. His core problem is his lack of features, and how the "gameplay-oriented" features look so out of place.

On the face of it, a dark behemoth, featureless except for a gaping maw, seems like a perfectly fine idea. The early concept art certainly looks good. But now add some white blobs as "toes" that serve as weak points - first, it distracts from that singular focus on the mouth, and second it undercuts the intimidating look of the boss, because seriously, you kill the boss by stabbing its toes? The same thing happens when, in the second and third forms, they added some ungainly long arms, tipped with white blobs as "fingers". They were certainly necessary to make the boss fight playable, but there had to have been another way to do it. It just reeks of a boss that was designed by an artist, and then had to have gameplay elements tacked on.

The Imprisoned also fits very poorly with the painterly aesthetic of the rest of the game. It, almost alone in the game, is harshly geometric, with his hundreds of identical scales. The concept art is quite a bit more abstract, so maybe it's a matter of the Wii not having the horsepower for all those particle effects. But I have to judge the game they released, not the game they wanted to make, and this game has a central villain that doesn't fit.

Had that felt more deliberate, it could have worked. The Twili had that sort of effect in Twilight Princess, with a techno-ish sound and harsh digital effects to play up their alien-ness to Hyrule. But in every way except its appearance, The Imprisoned fits the game normally - normal music, normal gameplay, normal sound effects.

Overall, the game's aesthetic sensibilities are on point. It really is a lovely game. But those few mistakes are all the more glaring when they're surrounded by excellence.

Next time: the greatest flaw of Skyward Sword

#the legend of zelda #everyzeldaproject #loz #skyward sword #game design #character design

[gman-003]

Quick thoughts on game engines

I’ve tended to roll my own game engines. Growing up, the video games I made mostly were built from scratch, using plain old SDL, OpenGL and OpenAL. Not entirely coincidentally, none of them were ever finished, but I think that has more to do with the games themselves not being good enough to finish than the extra work of coding an engine. And yes, it actually is a ton of work - work I, and many other programmers, tend to enjoy - but work nonetheless. Every hour spent on a physics system or a renderer is an hour not spent on making a fun game.

I tried modding (I think there’s an alpha build of Bitwar out there on the interblags somewhere), which gave me a decent bit of experience with Source, and a bit of experience with UnrealEngine 2. As a level designer, I loved Source - the Hammer editor is very, very effective at building level geometry, and the I/O system is fast for building simple scripted events, and with effort can do some quite complicated things. But as a programmer, I hated it. Builds failed mysteriously, it relied on arcane, never-explained preprocessor macros, and iterating on code was slow and tedious.

UnrealEngine 2 was almost a flip of that. Coding in UnrealScript was simple and straightforward, although I remember the documentation being a bit lacking, and iteration was still slow. Level creation, though, was awful, at least in my experience. Full-blown constructive set geometry might make sense for CAD but not for video games. Making even simple rooms and corridors was painful. Maybe it just never “clicked” for me, but I’m surprised anyone was ever able to do anything with it.

All of that was back in high school, or early in college. After that, I pretty much coded everything from scratch. I was pursuing ideas that didn’t lend themselves well to mods, and I didn’t think game engines were really worth it - particularly since, at the time, getting a full license was expensive.

I still take advantage of modding - it’s a great way to build a quick-and-dirty prototype. Find a game that’s similar to what you’re making, either open-source or with good mod tools, then mod in your core game mechanics. I did this with my current game - I took Red Eclipse, and modded in the things that made my idea distinct from all the other shooters out there. The code was better than Source, though not by much, and I didn’t need the level editor. I had a playable prototype in a few days, and it was enough to prove that yes, my idea was viable.

When I started making the actual game, though, I debated whether to build an engine from scratch, or try one of the new game engines. UnrealEngine 4 was out, and is free “until you make a million bucks”, which may as well be free for indies. I gave it a try. (I considered Unity and CryEngine, but went with UE4 because it’s aimed more at PC/console, while Unity is more mobile-focused, although both reportedly work well everywhere)

It’s night and day compared to how things used to be. The code is clean and fairly well-documented. The Blueprints system that replaced UnrealScript is unbelievably simple, but still powerful enough to do pretty much anything (as someone nigh-allergic to drag-and-drop flowchart coding, that’s high praise indeed). The level editor is... well, I’m still faster in Hammer, but I can at least see how people can master it, and it’s not bad at all. Most importantly, it iterates fast. Anything done in Blueprints can be changed and tested again in seconds. Compiling the code takes a quarter-minute. Just being able to play in an editor window instead of starting a game instance changes everything.

I got it to the level the mod prototype was at in about five days of work - including going through the “First Person Shooter” tutorial. Now, I’m working through all the other things that need to be done to make a complete game. And I am consistently impressed by how fast and easy things are. Adding screenshake? That’s like one line of code and a minute in the blueprint editor, and it makes an unbelievable amount of difference when it comes to game feel.

I’m probably still going to code my own engines for some games. I think it makes sense in certain contexts - games with lots of procedural content, or collections of minigames, or simulations with a high focus on realism. But I’ve flipped my general stance - you should use a prebuilt game engine, unless you have an overridingly good reason not to.

#game programming #game engines #unreal engine 4 #source engine

[gman-003]

Blog Update

I am still alive.

I had quite a few adventures last month, a pile of life crises (crisises?) to deal with, but nothing that did permanent damage.

It did, however, eat up a lot of time and energy. I ended up about a week and a half behind schedule on the video game I’m developing (details of which are forthcoming), and I focused on getting caught up on that rather than blogging. So once the pre-written, pre-scheduled blogs ran out, things went kind of quiet here.

Well, as of today, I’m back on schedule with the game, so I want to get back to this little thing. I’m going to try to alternate between the “Play Every Zelda Game” posts and miscellaneous other subjects, because those posts are a lot harder to write, and I don’t want this to be just a Zelda fan-blog. (I mean, it is, but I don’t want it to just be about Zelda).

Regular posts should resume next week, once I put out this last little fire and catch my breath a bit.

(Oh, and no, I haven’t actually played Breath of the Wild yet. Everything I’ve heard sounds great, which perversely makes me want to hear less about it. If they’ve finally returned to being a game about exploration, I want to be the one doing the exploring, with as little outside information as possible.)

[gman-003]

Flow and Friction - Legend of Zelda: Skyward Sword analyzed, part four

(This is part of my ongoing series analyzing The Legend of Zelda: Skyward Sword, part of a bigger series where I play through and then talk excessively about every game in the Zelda series. If you want to catch up, earlier installments discussed the use of patterns, the story and how it fits into the chronology, and the controls)

Game design, like all artistic fields, has its own jargon. "Friction" and "flow" are related terms: "flow" is the state games are trying to achieve, where the player is sucked in, with each action they do naturally leading into another. Without anything to disrupt their experience, players will just continue to play. Many things can disrupt flow - difficulty is the one that usually gets all the focus, because a game that is too challenging is frustrating, while a game that isn't challenging enough is boring, both states that take the player out of flow. But there are other things - other internal disruptions are often from "friction", where the game makes something not more challenging, but more difficult or more time-consuming. A cumbersome interface is friction. Loading screens are friction. Even just forced dead time is friction - my favorite example is shimmying along a ledge, which, regardless of the game, always seems to take forever but rarely requires anything more than just holding left or right.

The Zelda series has always had pretty good flow, despite having quite a bit of friction. The structure of the game keeps the player advancing - when they get a new item, there's a natural urge to go find where it can be used, and using it in the right place gives them another item to repeat the experience with. Combine that with constant new places to explore, and hints of places you'll eventually get to but currently can't, and it takes a pretty big disruption to break the player out of flow.

Good: Priming

There is one particularly genius bit of flow-promoting design I noticed. You probably didn't even notice it if you played, because it's a very subtle thing.

Games are composed of nouns and verbs - objects, and their interactions. In Zelda games, the player has a variety of tools to interact with objects, and every time the game introduces a new noun, whether it be a new tool, enemy or environmental object, it needs to teach what verbs will work on it.

There's a section in chapter 5 (as I count it), where you're climbing up some rock towers in the Lanayru Sand Sea to reach an old house. This area introduces a new creature, a bird called the Furnix. To defeat them, your best weapon is the whip, which allows you to drag them in close to kill them with your sword. How does the game teach it to you?

Immediately before the first Furnix, you encounter some Peahats, which you already know have interactions with the whip (pulls them out of the ground) and the hookshot (once airborne, you can use them as grapple targets). The level has you do this once, as a refresher, then puts you on a platform with a few planted Peahats. You'll probably take your whip out to pop the Peahats, which seems to trigger the Furnix to spawn. You'll probably try the sword first, and quickly find that it doesn't work. What will you try next? Well, you already have the whip ready, maybe try that. It's the path of least resistance for players, and while they won't always follow that path, a good number of them will. The Peahats, a known noun to the player, primed them with the proper verb for the Furnix, an unknown noun. It isn't the only hint as to what tool to use - the Furnix's tail looks like some of the previous whip targets - but it helps.

I spotted this "priming" technique again in Chapter 6. You're exploring the Fire Sanctuary, and you encounter some piles of dust. Previously, you'd learned that items are often hidden in dust piles, and can be obtained by using the Gust Bellows. Shortly after, your path is blocked by puddles of lava, a new challenge. The Gust Bellows may still be on your item button, and "blow on thing to cool it off" is an intuitive interaction, right? And indeed, that's how you have to solve this puzzle.

I have more examples of priming, and I probably missed a bunch of other uses, but two is probably enough.

This is clever. This lets the player be the one to figure out what tool to use, but it makes it so that they're more likely to get it right with their first guess. The game isn't telling you what to do, it isn't holding your hand, it's just making you have the right tool in your hand. It keeps the game flowing.

Okay: Loftwing travel

Pretty much every 3D Zelda game uses some sort of overworld faster-travel mechanism to make the world seem larger while still having acceptable travel times (walking back and forth across a large overworld to get to the place you need to be is very flow-disrupting). Ocarina of Time let you ride a horse. Wind Waker had a boat. Skyward Sword's transportation is based on the Loftwing, a large, rideable bird.

By themselves, the riding mechanics are good enough. The lack of camera control can make sharp turns messy, but it feels good, and while I'm not 100% sure if it makes any actual difference, it seems to give the player stuff to do to get from point A to point B faster. Overall, it feels like a better, three-dimensional version of the ship in Wind Waker.

The problem comes from the context it's placed in. The sky world itself is barren, with only a few destinations worth visiting on their own, and the small islands with the Goddess Cubes are unneeded until you've done something on the surface. The two most worthwhile destinations nominally in the sky, Skyloft and the Lumpy Pumpkin, are behind loading screens. Traveling to the surface is done by singular points - you can't skim the treetops in your Loftwing, you never dodge ejecta to dive into the volcano, you just fly to a point in the sky, jump off, then select the right spot on the surface-world map to appear there.

Reducing the overworld to a set of loading-screen transition points makes the overworld feel tedious. It doesn't invite exploration, it doesn't present a meaningful challenge, it just exists and you have to go through it.

The three-dimensionality of flight may have been part of the problem. Enemies do exist, but they seem very scarce because they can be avoided effortlessly by flying higher or lower. It also forced a more empty overworld to avoid the complexity of real three-dimensional navigation, which quickly becomes maze-like.

The clustering of important destinations together in Skyloft, behind a loading screen, seems to have been primarily a technical problem. Having both the huge sky overworld and the populated, detail-rich town be in memory at the same time would have overtaxed the 88MiB of memory the Wii had. If the choice was "have one big, rich sky-island with everything on it" or "have the town scattered across dozens of islands, but they're all boring and identical", they clearly made the right choice... but there were other options. There was no reason Skyward Sword needed to use the sky as a setting, or use giant bird mounts for travel. Perhaps they only discovered this after the artists had produced too many assets for the sky to throw it all away? Maybe they thought they could optimize enough to fit a rich sky world? This is a flaw they had to have been aware of, even though it's not that painful to the player. I actually found it kind of relaxing. But I also enjoy flight simulators, so I might be an anomaly here.

Bad: Horrible UX flow

We come now to what I consider the second-greatest flaw of the game. The game constantly disrupts its own flow at a very low level. Little things constantly drag you out of the experience.

For example, every time you pick up a crafting-trash item for the first time, it pauses to pull up the inventory screen, show the new thing appearing in its spot, and show the item count going up by one. The whole process takes about seven seconds, by my rough count. That seems like a minor quibble, but two things make it worse.

About a third of these crafting items are dropped by enemies. Lizard tails? Jelly blobs? You pick these up by walking over them after killing an enemy, and these enemies almost always come in groups. So you kill one, and are running towards another, sword drawn... and the game yanks you out of that combat flow to tell you that you picked up a thing. By the time you get back to the fight, your timing is completely thrown off, and you might even take a hit from an enemy you were about to block, but now have no time to react to.

As an extra insult, it resets the state of "have you picked up this item" every time you load your save. It doesn't matter if you already have thirty of them, the game considers it absolutely vital for you to know that, when you step over a monster horn, it's putting it into your inventory. And to rub salt in the wound, it appears that the game's save file actually tracks whether you've picked up the item before, even if you no longer have any! The save file knows "this player has picked up Ornamental Skulls before, we should let them look at the item description even though they used them all in crafting", but it doesn't bother to check that when deciding whether to ruin a fight for you by wasting several seconds on an inventory screen pop-up. I'm a programmer, I know how easy it would have been to check the "have they ever had this item?" variable instead of the "have we shown the pickup sequence for this item this play session?" variable. Fixing this should have taken five minutes, tops. It might actually have been faster to do it this way to begin with.

The item pickups are not the only instance. This game's companion, Fi, repeatedly pops up to explain things - most infuriatingly, in my playthrough she popped up during the final battle to tell me that my health bar was low and I might want to drink a potion. But she's constantly popping up, relaying some information you've probably already heard and definitely don't need to hear again.

Cutscenes, too, are a constant interruption. Not the big ones, but the small ones. The short close shots of an major feature of a room that plays when you enter it. The reverse shots of the door closing and locking behind you. These are useful ways to keep the player informed, but Skyward Sword uses them so often it becomes grating.

Games suffer greatly when the player isn't allowed to play it without interruption, and Skyward Sword filled itself with interruptions.

#the legend of zelda #everyzeldaproject #loz #skyward sword #game design

[gman-003]

Control -The Legend of Zelda: Skyward Sword Analyzed, Part Three

(This is part three of my post-play analysis of The Legend of Zelda: Skyward Sword, part of an thing where I play and then talk about every game in theZelda series. Check out the earlier installments, where I talked about the story or the use of patterns.)

Games differ from all other fiction by being interactive, and video games differ from their analog brethren by being software, capable of being far more interactive. How you play a game is an absolutely crucial part of how good or bad it is, and as with all things, Skyward Sword is a bit of a mixed bag.

Okay: Motion controls enforce delay

Skyward Sword uses motion controls. That was what everyone read and heard in the lead-up to launch. This was the game that was going to finally make good on the promise of the Wii: a game that was far more than just a collection of minigames, but still made full use of the Wiimote's capabilities.

Well, for better or worse, it did. The game actually uses motion controls in a game for the hardcore. And it immediately ran into the fatal flaw of motion controls: they enforce a delay.

Latency is absolutely critical to gameplay, in all but the most strictly turn-based games. Mere frames of latency between the player pressing a button and the corresponding action happening in the game can be ruinous. Motion controls, though, have forced latency because they require larger physical motion before the sensors can pick up on it. Even the deepest mechanical keyboards have only a few millimeters of physical travel before they trigger. Most game pads have even less. But with an accelerometer, you must make large gestures - even the smallest flick is still several centimeters of travel. And if a precise angle is needed, you may as well expect several seconds to pass as the feedback cycles between the player and the computer.

That latency impacts game design. If not adapted for, you have enemies moving faster than the player can hope to respond to them. To its credit,Skyward Sword adapts the 3D Zelda formula to the appropriate pacing. It slows things down to the point that the player rarely feels like the controller latency is holding them back. It wasn't even that much of a slowdown, because past Zelda games have been slower than they strictly needed to be, since they inherit from a game that ran at 20 frames per second.

But it's still there. You can feel it in the more precise swordfighting battles, such as against Scervo and Dreadfuse. Often an enemy will change its block pattern while you have physically begun to swing the Wiimote, but before the controller has registered that movement. This gets quite aggravating, and pushes the player to avoid using the sword on enemies that can punish you heavily for being blocked.

Good: Slower gameplay became deeper gameplay

Which is quite a shame, because the sword combat is actually pretty good otherwise. The slower pace called for greater depth, and Nintendo gave it to us.

Previous Zelda games had attacks from different angles, but they never really mattered. You had slashes or stabs for engaging a single enemy, spin attacks for attacking multiple enemies. Some attacks were tied up with movement - the jumping overhead swing, or that rolling backstab from Twilight Princess. But enemies didn't care how you hit them, just what you hit them with and when.

About half the enemies, and almost all of the bosses, in Skyward Sword care about the angle of your attack. Bokoblins will try to block. Deku Babas need to be cut from a certain angle. Skulltulas can be flipped onto their backs by an uppercut. Moblin shields can be cut apart. The controller may have forced the game to slow down, but the enemies give you plenty to think about during that newfound time you have. Which is only slivers of a second, but it's not that much to think about, anyways.

Many of the other weapons take advantage of motion controls. The whip has a natural feel to its flick. The beetle is steered entirely by motion controls. The bow, slingshot and hookshots aim quite well. Bombs get a nice extra interaction - while you could always drop or throw them, now you can also roll them, a minor but appreciated addition.

Overall, it's a welcome set of changes to the formula. The combat in 3D Zelda games has always been a bit lacking, and a small decrease in pace is more than worth it to get some much-needed depth.

Bad: Lack of buttons on Wiimote

The controls were somewhat forced by the console. Nintendo couldn't exactly say "eh, this game plays poorly with motion controls, we'll just put it out on Xbox and Playstation instead of the Wii". So even though the use of motion controls was perhaps sub-optimal, they did a great job at adapting the game design to the motion controls.

Surprisingly, what hurt the game more was actually the traditional control elements. The Wiimote was pretty good, for a groundbreaking motion controller, but it works pretty horribly as a regular controller.

Let's do a comparison. The Wiimote, with the Nunchuk and MotionPlus accessories, gives the player a single analog stick and a single D-pad. Four buttons are immediately under the player's fingers: the A, B, C and Z buttons. Four more buttons are available to games, but are further away from the player's fingers: +, -, 1 and 2. (The Home button isn't usable for game controls, so we won't count it).

Other Gen-7 consoles had more inputs available to the player. Both the Xbox 360 gamepad and the PS3's Dualshock gave two analog sticks, a D-pad, four easily-reached face buttons, four triggers, and a pair of out-of-the-way buttons. That's a gain of a stick and two buttons overall, but more important is the four extra buttons that can be used for fast gameplay actions.

(For what it's worth, Nintendo clearly recognized this as a mistake - the Wii U Gamepad has the same complement of inputs as the traditional consoles, plus the touchscreen. And the preceding Nintendo console, the Gamecube, had nearly the full complement of inputs, missing only a fourth trigger and a select button.)

Zelda games give the player avatar lots of ways to interact with the world. This must be mirrored by giving the player lots of ways to interact with the game console, or else you must resort to piling multiple actions onto each input. Since the Wii physically lacked the buttons to map each action to a different button, they were forced to make them ambiguous. A long-press of B brings up the item select menu. The A button has a different meaning in almost every context and can change depending on other buttons being held.

This leads to a noticeable increase in how often you have to wait for a second because you're long-pressing a button to trigger the alternate mode. You can't rapidly switch between items, the way you could in every other game (even the Wii version of Twilight Princess) - you have to hold the B button. Climbing a block has you walking into it until the game is sure you're not going to press A to try to push it instead. It's a tiny thing, true, but over the course of the game, it adds up. And if it leads to the player getting killed because they were waiting for a long-press to finish... well, players absolutely hate it whenever they are penalized because of something the game did wrong.

While the limitations of the hardware make it so no game of this complexity would control flawlessly, Skyward Sword doesn't even do the best it could with the tools it had. Let's compare it to Twilight Princess, which utilized the buttons of the Wiimote better than its successor.

Calling your companion was moved from D-Up to D-Down. A small change, but a noticeable one, because D-Down is a much more accessible input on the Wiimote. The harp was stuck onto D-Up at all times - wasteful, since you only ever need it in a few known places, and it could have just been a contextual action as howling was in Twilight Princess.

Gone is the ability to set different items to quick-use slots, something seen since very early in the series. D-Left and D-Right are completely unused, forcing you to enter the equipment menu for every single non-sword item.

The 2 button is dedicated to a "help" action that I doubt any player has ever tried to use. I seriously can't even find a screenshot of what it does, and neither the game manual nore the official strategy guide bother to explain it.

The +, - and 1 buttons all bring up different menus - items, map and equipment. But each of those can be reached from any other. At least one of them could have been given up without harming the user experience.

So that's two buttons literally unused, another that could have been removed entirely, a fourth that could have been made contextual, and if necessary, a fifth that could be folded into another.

What other uses could those buttons have been put to? A quick-select item or three, for starters. Let me quickly pull out items other than the one I literally just used. A dedicated button to get to the item-select wheel might also be in order, instead of the long-press on B. A dedicated "run" button would be nice, while we're at it.

This seems like an incredibly minor complaint, but while playing, this was in my top five most frequent complaints, and it really does make the gameplay feel worse. It even had a knock-on effect on game complexity - since selecting items is so tedious, they refused to make items that weren't strict upgrades. They solved the problem Ocarina of Time had with its long, annoying sequence to switch between boots or tunics, not by making the controls better, but by making the game simpler so you never have to do that.

The button limits and the bad choice of mappings is something I'm sure they'll fix in the inevitable HD remake. I notice the Switch has a more reasonable amount of traditional buttons... perhaps the touchscreen can be a suitable replacement for the motion controls?

Worst: Calibrations

(I refuse to apologize for this stupid joke)

The motion controls work fine when you're just making relative motions - swinging a sword around. It falls apart when you try to use it as a pointer, because the calibration drifts like crazy. You pull out the bow, and suddenly you're spinning around because the pointer is off the screen to the left. Same for dowsing, hookshots, slingshots, the beetle, or menu navigation.

The designers must have realized this. Every time you use the Wiimote as a pointer, the D-down button is available to "center" it. You will be using thatconstantly. Every time I pulled out one of the aimed items, went into first-person view, or pulled up a menu, I had to center it before I could use it. It got to the point of muscle memory - I pulled out a new item, better calibrate it before I end up spinning around like a top.

I don't even know why they bother making you calibrate the Wiimote when you start the game. You're going to re-calibrate it every few minutes anyways.

What adds insult to this bit of tedium is that the game completely ignores the Sensor Bar. Something it could have used to calibrate itself, even just for the pointing sections, gets ignored. So instead it tries to reconstruct the relative position and angle of the Wiimote based on reconstructed acceleration. It seems almost pointless to ignore this.

Overall, the controls did more harm than good. Swinging the Wiimote around feels great, and it forced a much-needed update to the combat system, but it was overall a frustration and a hindrance. I look forward to the HD remake, whatever console it's on, which should fix the more glaring issues.

#the legend of zelda #everyzeldaproject #loz #skyward sword #game design

[gman-003]

Legends - The Legend of Zelda: Skyward Sword Analyzed, Part Two

Levias is an even worse fit. He's supposedly a guardian spirit for Skyloft, but he's never mentioned until the part of the game where you have to go get a MacGuffin from him. After that, he never appears in the story again. I think he gets more total screentime in Hyrule Warriors than in Skyward Sword. He feels like a Link's Awakening reference that was tossed in at the last minute, when it was decided that Skyward Sword was to be the origin myth for the Legend of Zelda series. Games have become a very narrative-focused experience. When the Legend of Zelda series began, all you had was "you're the hero, go rescue the princess from the monster by getting all the magic things". You're still the hero, there's still a princess and there's still monsters and magic things, but the series has gotten a lot more story-focused as time has passed.

That's not, in and of itself, a bad thing. Early games skimped on the story because they didn't have the technical capability to tell it well. The NES basically didn't have the memory capacity for books' worth of text or minutes-long animations. And a story done well is always worthwhile.

But it has to be done well. Is the story of Skyward Sword good? Well... kind of. It has great bits, it has okay bits, and it has some bad bits.

Good: Zelda has agency, Groose has arc

Certain tropes have gotten a lot of flak for their prevalence. The "damsel in distress" trope, the helpless female character who you must rescue, gets some deserved criticism, and the Zelda series has used that trope pretty extensively in past games.

Not here.

Skyward Sword gives us a Zelda who takes action. She does things. In the earliest parts of the game, she sticks up to a bully to protect you. When she gets thrown to the surface world by a tornado, you go to rescue her, only to discover a) she's taking care of herself, b) she's on her own quest, and c) she's actually kind of busy right now, let's try to meet up later, k?

Even when she becomes trapped in a crystal thing... she was the one who put herself there. She had to do some weird magic stuff to keep The Imprisoned... imprisoned, and until you kill the thing for good, she's too busy to come have sloppy makeouts so get off your ass and finish the quest already, Link. Get it together, because she's got hers.

The complaint about the trope is often misused. Tropes are tools, and tools exist to be used (although yes, the extent to which that particular one is used is indicative of problems with our culture's gender roles). Using the damsel-in-distress trope to, say, give the broad strokes of a story because you're on the NES and putting a more detailed story in the game would require cutting out half your gameplay, is perfectly valid. Nintendo could have told a good story even while using that trope... but instead, they gave us a Zelda who felt like a real character, who did real-character things and had a real story of her own.

In Skyward Sword, Zelda has an arc. She has character development. She has growth. She feels like a real person, with internal conflict and personal change. She's an actual good character.

She's not the only one. Groose - who I had pegged at first as the annoying comic-relief character - has an arc. He starts off as a bully - childishly cruel to those weaker, but a coward to those who he can not or will not harm. He sobs inconsolably when Zelda goes missing... until, eventually, his grief is overwhelmed by his envy at Link's exploits, and he tries to usurp him as protagonist, trying to be the one to save Zelda and, he hopes, win her favor. But visiting the strange surface world changes him. He learns to respect others. He finds his courage, helping Link fight a demon. He uses his brain, and not just his brawn. And, eventually, he realizes that Zelda is her own person, that he's doing both of them a disservice by trying to win her.

That's a story right there. That's an arc. It's not one that's well-suited to game form, honestly, but it's great character development. The low-level writing, the words and sentences of dialogue, is never particularly great in Skyward Sword, but the mid-level character crafting is on point.

For everyone except Link.

Okay: Proving your worth to be a hero, not being a hero

Link as a character has one big problem in *Skyward Sword*. He doesn't have agency. He never does things without another character, or several characters, pushing and prodding and guiding and dragging him through the story. His supporting cast is so supportive that he loses his protagonist status - because he gets told to do everything that he does, he becomes merely the pawn for other character's successes, rather than a hero in his own right. He's a bullet fired at Demise by Zelda. He's the dumb meat swinging Fi and the Goddess Sword around. He's the errand-boy of Impa, in two ways.

Link's character arc is barely an arc at all. It's all buildup for almost no payoff. Link spends the first three dungeons trying to prove to Impa that he's good enough to keep up, to not be a liability to her and Zelda. Then he spends three dungeons proving himself worthy of the sacred fires needed to forge the Master Sword. Then he has to go prove himself a hero to the three dragons. Link only "becomes a hero" at the very, very end, when he defeats Ghirahim and Demise - specifically, he becomes a hero when Demise gives him a chance to run away, and Link chases after him instead. So, about fifteen minutes before the credits roll, depending on how much trouble the final battle gave you. (It gave me a lot of trouble, by the way). Other games in the Legend of Zelda series did not make you wait nearly so long. In *Ocarina of Time*, Link gets his first hero moment when he first faces Ganondorf, and draws his sword against the king of evil. In *Wind Waker*, Link shows signs of heroism from the beginning, when he ventures into the forest to rescue a fallen stranger. Perhaps the writers of *Skyward Sword* intended that early moment where Link ventures to the surface in search of Zelda to be that moment where he shows even a sign of heroism... but they sapped it of any meaning by having him be so forcefully prodded into it, by not one but two characters. I was yelling at the screen for them to shut up and let me go on an adventure already, but the cutscene dragged on of Fi and Gaepora telling me how important it is to go on an adventure. (There may be a blog entry entirely about this scene, as well as the role of certain steps of the Monomyth in interactive media, if I feel like it's worthwhile). Link isn't really changed by his adventures. He's learned skills and acquired powers he didn't have before, but he's fundamentally the same character, as far as I can tell. Maybe it's just hard to show a player character developing self-reliance and confidence, because their actions are controlled by the player? But plenty of other games have pulled this off - the 2013 Tomb Raider reboot springs to mind.

That brings up an interesting aside: does the protagonist of the story necessarily have to be the player character of the game? I would argue that Zelda is the protagonist of Skyward Sword's story, even though she's never controlled by the player. Link doesn't take initiative. He starts the game doing things because Zelda told him. When Zelda's gone, he does things because Fi told him, or because a dragon told him, or because someone just asked nicely. Zelda has far more agency than Link, and honestly has a better character arc.

In the end, it was Zelda that defeated Demise, not Link. Link might have been the one swinging the sword, but the victory was arranged by Zelda. She (as Hylia) placed all the tools Link needed in the places he would get them, and kept The Imprisoned from escaping until Link had assembled all the pieces necessary to kill it.

And even though I'm not sure the Legend of Zelda gameplay formula would be the best fit for such a story... I kind of wish I had been playing as Zelda instead. Or maybe playing as Impa, since she had lots of action going on, but was more closely involved with the story's protagonist. Either one seems like it could have been a good game.

Bad: Elements out of place

While Link is a bad fit for the protagonist role, he's at least a good fit for the player character role, and his story isn't jarringly bad. But there are elements that seem much more out of place.

The story of Skyward Sword has an unfortunate tendency to bring new characters in without foreshadowing or justification - characters that, logically, ought to have been known to the player earlier, but were not.

Demise, the game's final boss, sort of comes out of nowhere. You never see him except in the bestial, non-sapient form of The Imprisoned. You never see art depicting him as a humanoid or writings referencing him as such. This isn't completely without foreshadowing, as you do know that The Imprisoned is not in its original form, and it was once a fearsome demon king, but it's still quite abrupt to see him standing before you, wielding a sword. It was clearly intended to be a surprise, but it comes off more as a "we made the final boss fight without telling the writers" than a "clever twist ending".

Levias is an even worse fit. He's supposedly a guardian spirit for Skyloft, but he's never mentioned until the part of the game where you have to go get a MacGuffin from him. After that, he never appears in the story again. I think he gets more total screentime in Hyrule Warriors than in Skyward Sword. He feels like a Link's Awakening reference that was tossed in at the last minute, when it was decided that Skyward Sword was to be the origin myth for the Legend of Zelda series.

And the game's very status as an origin seems to contradict its internal story. Skyward Sword has an obsession with the past. You're fighting an ancient evil that was sealed away in an ancient battle by ancient peoples under the guidance of an ancient goddess. Time travel is a theme of one of the dungeons and is a major component of the story - and you always go backward to the past, never forward to the future.

The game does not feel primitive. You have plenty of advanced technology - rotorcraft and electricity are in active use on Skyloft, and the are friggen robots on the surface. That doesn't jive at all with the supposed placement at the head of the timeline. The technology of the series jumps all over the place - Skyward Sword has all this advanced, modern tech, but the next few games per the official timeline, Minish Cap, Four Swords and Ocarina of Time feature very little tech beyond that of medieval Europe. You can't even argue that all the technology we see in Skyward Sword is ancient, pre-apocalyptic tech, which will break down and be lost over the next century or so. Some of it was indigenous! Beedle had electrical generators and human-propelled aircraft, stuff we emphatically do not see in our visits to the past. He didn't salvage that from an ancient ruin, he invented it. So why isn't that seen in every other game, since they all take place after Skyward Sword?

I suspect, contrary to the claims of Aonuma, that Skyward Sword was not intended to be placed at the head of the series from the start of development. It feels like it was created either to be at a vague future point in the timeline, or with no concept of being a sequel or prequel at all. Before the three-branched timeline was made official, it was commonly argued that the series had no strict chronology, that it was literally a legend, in the sense of being the same story told over and over by different storytellers.

Under that paradigm, Skyward Sword fits well. It has the magic sword of power, it has the princess and her protector, it has the mentor, it has the companion, it has the demon king, it has the whale, it has all the stock locations. It has all the recurring nouns of the Zelda franchise, it just puts them together differently, as every game did.

With its focus on the past, I could even buy it being a distant sequel to another game, perhaps on the Wind Waker branch of the timeline. I have not yet played the games billed as direct or indirect sequels to Wind Waker, though, so that may have its own contradictions. It could even fit around the rest of the timeline - have the "past" of Skyward Sword come before every other game, and the "present" come after. There's tons of theories that could be made to fit - and all of them fit better than the actual way they chose to present it.

Overall, I just don't feel like Skyward Sword is a perfect, natural starting place for the Zelda series. I have a theory as to why it was made the Official First Zelda Story, but that will have to wait for another time.

#everyzeldaproject #the legend of zelda #loz #skyward sword #game writing

[gman-003]

Patterns - The Legend of Zelda: Skyward Sword Analyzed, Part One

Any good work of art has a certain rhythm to it. Music has melodies, a series of notes repeated to build the song, and on a higher level, there are verses and refrains, a repeated chorus. Visual art has proportions, a pattern not in time but in space. Poems have scansion and rhyming schemes. Humans love patterns - we see them everywhere, and we put them everywhere. I'm even writing these analyses by a pattern - each will cover a good bit, an okay bit, and a bad bit, linked together by a common theme. Patterns are part of creation, and video games are no different.

Good: Theme & Variations fights

The Legend of Zelda as a series loves its patterns more than most. Skyward Sword establishes patterns on multiple levels. The main section of the game consists of a repeated series of Overworld Challenge, Dungeon, Miniboss Fight, Item, More Dungeon, Boss Fight. This is a classic Zelda formula, but the thing about patterns is changing them up. Nobody listens to a song that is the same four notes over and over, and nobody plays a game that is the same exact pattern over and over. In music theory, there's a concept called "theme and variations" - you establish a "theme", then repeat it with "variations" - changing the instrumentation, changing the key, altering the rhythm or bits of the melody, but leaving it always recognizable. Game design, as it so often does, lacks an equivalent term, so I'll just be stealing music's. You can see this concept perfectly in the Lake Floria and Ancient Cistern chapter of the game.

First, you must defeat the first Silent Realm Trial, a sort of stealth/racing section, which the first three dungeons did not have before them, but which this and the next two do. Already, you're beginning with something new. This Trial gives you an item allowing you to swim underwater, which lets you into a large tree that you have seen before but could not enter. The tree is full of enemies to defeat, complete with its own sort-of miniboss with the first appearance of a moblin. Reaching the top starts you on a bit of an investigation to find the entrance to Lake Floria. Finding the clue lets you into Lake Floria proper - but you're not done yet. Now you have a long water section, making sure you have a good grasp of the swimming controls, and ending with an underwater fight, which you've never before faced.

You're almost at the actual dungeon, but one more thing remains - you're given a fetch quest, and the trinket to fetch... is in a previous dungeon, Skyview Temple. But the Temple has been modified since your last visit - it is populated with more dangerous enemies, and many doors have been re-locked, with keys hidden in areas inaccessible without the items you've acquired since your last visit. This fetch quest culminates in a battle against a trio of Stalfos, in the same room where you had first fought a single Stalfos. 

At last, you reach the dungeon proper, the Ancient Cistern. The game actually lets you explore for a bit, but your exploration is limited by a door. Normally, you'd just wander the available areas until you found a key, but this door instead has a combination lock, and the combination is given to you through in-world clues. There's a stone tablet letting you know what the order is, and then the actual inputs (it's directions instead of numbers, so you can enter it with sword slashes) are hidden in the environment. It's not a real stumper, but it's a more complex variation on the concept "locked door -> key -> open door". 

The miniboss is a Stalmaster, a four-armed skeleton warrior, who has two phases. His first phase is just a harder version of the Stalfos fight you first experienced several dungeons ago - three of them were needed just to be a miniboss earlier in this chapter. But his second phase is substantially different, with new, much harder patterns. A good iteration on something you've experienced. This gives you a new item, the whip.

Shortly after, you're sucked into a new section of the dungeon. While the higher floors, where you began, were bright and cheerful, these lower floors are dark and foreboding. My notes while playing read simply "spooky area is spooky". You have zombie variants of old enemies, thick fog, a purple-black color scheme, and skulls everywhere. Poisonous water damages you and leaves you vulnerable, unable to attack until it wears off. Eventually you make your way back up, but you have to bounce back and forth between these two sections to solve the rest of the dungeon, making heavy use of the whip along the way, as in every formulaic Zelda dungeon.

Finally, you reach the boss, Koloktos. A six-armed automaton, it fulfills the pattern set by the Stalfos and Stalmaster fights. Two, four, the next number has to be six, and so it is. The fight works differently, however, relying heavily on the whip. I found the battle extremely challenging - partly due to it being a real test of your skill, and partly because of a poorly-placed lock-on target making me think I was supposed to attack a place I wasn't supposed to. But it ends the chapter either way, and then it's off to the next section of the game.

We can see several layers of pattern here. There is the rising pattern of the number of arms of the climactic battles - the two-armed Stalfos, the four-armed Stalmaster, and the six-armed Koloktos. There is the thematic pattern of skeletal enemies in Skyview Temple, as well as your entire visit there being a variation on the original dungeon experience. The Silent Realm trial that started this chapter is the first statement of the theme that will run through the next two chapters. There's the contrasting patterns of the bright, clean upper floors and the dark, toxic lower floors. All of these patterns are recognizable but never come off as formulaic.

Okay: Timing of repeated fights

The flow and rhythm of the dungeons is generally pretty good. But some of the patterns established at a higher level have problems. At a higher level, the timing of some of the repeated fights is off, namely the three fights each with The Imprisoned and Ghirahim.

If we treat each dungeon as the end of a chapter, you fight The Imprisoned at the very start of Chapter 4 (the Faron Woods/Lake Floria/Ancient Cistern I spoke of just above). Then the second fight happens at the very start of Chapter 7, right after the Fire Sanctuary... and the third fight is shortly before reaching the dungeon of Chapter 7, separated from the second fight by just some questing up in Skyloft, an aerial boss fight with the Bilocyte, and a decent amount of plot dump.

Several hours of gameplay exist between the first and second battles, but there's barely any time at all between the second and third. In the official strategy guide, the two fights are separated by only eight pages. The evolution between the second and third fights is also minimal - while the second encounter introduced the Groosenator, gave the Imprisoned arms and the ability to climb walls, the third introduced only the ability for it to slowly fly over the walls, and capped with a very short (but dramatic) final phase. They come across as almost identical battles.

I know that the story is trying to emphasize how The Imprisoned keeps breaking out easier and easier, leading up to the ultimate conflict of the game, but... there's a bigger gap between the third Imprisoned fight and the end of the game. If I had directed this game, I would have pushed the third fight further into the game (making the three fights about evenly spaced), and either taken away some element of the second fight, or added newer elements to the third fight, to make it more of a variation and less a repetition.

Your three fights with Ghirahim also have some uneven pacing. You first encounter him as the boss of the first dungeon, Skyview Temple. Then, after many cutscenes of him showing up but not fighting you, he serves as the boss of the sixth dungeon, the Fire Sanctuary. And then, although he doesn't appear in the dungeon itself, he is defeated immediately after the seventh and final dungeon, Sky Keep, and can be considered its boss (since it otherwise lacks one). The timing isn't quite as bad, but it does come close enough afterwards to feel a bit sudden.

Ghirahim's battles do have much better variation than The Imprisoned's, it must be said. His appearance changes radically, you encounter him in different areas each time, and his moveset and attack patterns evolve very nicely. The timing of the fights still seems unbalanced.

Other repeated fights don't have this problem. Your two fights with the Dreadfuse miniboss have a nice evolution - he gains an electrified sword, and is fought in a tighter, less forgiving location, the second time around - and there's enough space between them that it feels like a callback, not a retread. The Stalfos/Stalmaster fights had a good evolution - first you fight one, then when you come back, you face three, then you face a far more dangerous version one-on-one, and then later, you fight the dangerous one while also fighting waves of weaker enemies. They're tied to a region, so there's even an in-world justification for seeing them again, but each time, something new gets added to the experience. But the timing on the Imprisoned fights in particular make me suspect that either some entire dungeons were cut, or that one of the fights was rushed in at the last minute.

Bad: Lack of location variety

The Legend of Zelda series is often said to be about "exploration". And yet, looking back, I find that much of Skyward Sword is spent re-treading the same areas.

There are only four broad areas of the game: Skyloft and the surrounding skies, Faron Woods, Eldin Volcano, and the Lanayru Desert. Each of these has a few sub-regions - Lake Floria is different enough from Faron Woods to feel like a new area, and the Lanayru Sand Sea feels somewhat novel. But you really have only five or six terrain types, very little for a Zelda game.

Ocarina of Time

 had forests, fields, volcanic mountains, deserts, and icy lakes. 

Wind Waker

 had tropical islands, volcanoes, forests, castles, and of course the ocean. 

Twilight Princess

 had forests, fields, mountains, volcanoes, cities, lakes, deserts, the sky, a weird twilight realm and even a western-themed village. 

Skyward Sword

's relative lack of variation makes it seem a lot smaller than it is. Contributing to this problem is the constant reuse of the same locations. On an optimal path, you'll visit the major areas three times each, so you're spending a pretty good chunk of the game in the same handful of areas, either as a goal itself or on your way towards a new location. Even with the variations they introduce over time, flooding the Faron Woods or adding an escort mission to Eldin Volcano, it makes Link's epic journey seem more like putzing around the same three points for a while.

Also adding to the feeling of sameness is the lack of any towns besides Skyloft. Even Ocarina of Time had Kokiri village, Kakariko, Castle Town, the Goron city, and the Gerudo camp. So there certainly isn't a technical restriction behind this. If they were trying to create a feeling of a wild, uncivilized surface realm, they also failed because there's plenty of people living down there in organized groups, between the Kikwis, the Mogma, the Parellas, and the Goron. The surface feels inhabited, you just happen to never visit their villages while they're there.

Finally, the lack of ways to walk between the three areas makes it feel a lot smaller because you never feel the distance between the three surface locations. Having fast travel is a good thing, don't get me wrong - but because the only travel is "fast travel", it doesn't feel fast, it feels like the world is just small.

A game about exploration shouldn't feel small.

#the legend of zelda #game design #everyzeldaproject #loz #skyward sword