Thursday, June 25, 2015

How I would do it: Splatoon

Splatoon is revolutionary and freaking adorable, but it could be better.

Before we get to what I'd do differently, we have to talk about the game itself.

When I realized that doing so was really post length itself I cut it out and put it in it's own post.


The first thing that needs to be understood is that Splatoon, as it currently exists, needs multiplayer bots.  I'm just going to foonote that [1] because it's more a patch on existing things than the fun how I would do it stuff.

Table of contents:


Ok, so, single player is awesome.  More of it, thank you very much.

Also more integration of it.  This could be purely aesthetic.  When you start the game you're told on the local newscast that the power source for the city is gone, and since the power source is a giant electric catfish (creativity points for inventing the "zapfish" electric catfish instead of making it an eel, by the way) that's kept in public, you can see it's no longer there.  But, other than it not being there as part of the scenery, there's nothing mentioned about it ever again until you complete the single player campaign and return it.

Local newscast says its back, now you can see that it's home.

The local news cast is used to announce stage rotation which happens multiple times a day and also announce any new features added to the game (new weapons, new stages.)  Mixed in with these could be things about the power shortage.  Lights could flicker.  Things could look like a city that has power problems.

As you returned the little zapfish stopgap measures could be implemented to deal with the problems and thus you could see what you were accomplishing.

Like I said, purely aesthetic.

That said, it would also be nice if you could have things that do affect gameplay.  Like bringing some of the gear you buy in the plaza on single player missions.


Even if the game thinks it's the first time someone has seen this, it should be skippable.  Cutscene?  Be able to skip it.  Newscast?  Be able to skip it.

Results of the match you just played?  Be able to skip it.

It should be always, always, always be possible to skip non-interactive content.

The sole exception is when the content serves the same purpose as, "On your mark.  Get set.  Go!" and that is something that should never take a long time.

Non-combat inter-player interaction.

When you're walking around the plaza you can encounter other players in a miiverse way.  That means you can see them, as they have chosen to appear (skin color, eye color, clothing) and read any message they've shouted to the world, but you're not really encountering them because they're basically playing the role of NPCs.

I'd be in favor of making this an interactive non-combat multiplayer.  Have it be a chance to talk to your fellow players between matches and such.  Talk could in fact be text so voice chat wouldn't have to be implemented; that said, people want voice chat and for good reason.

Multiplayer is a team of four vs. another team of four.  Working together with your team in a way such that you complement each other is essential.  And yet you can't talk to them.  You can't say, "I can hold this position so no one else needs to wait here," or, "Cover me," or, "I've got you covered," or, "Don't go that way there's a sniper," or, "Everyone attack on 3," or, "If you just tell me what you're trying to accomplish, maybe I can help you."  Or anything like that.

But we'll get to multiplayer.

For now the plaza.

The thing is, you could still do the mii-verse thing that's being done here so that the area looks like it's populated people people doing people things while doing what I'm advocating.  The difference would be that the NPC population would be made of people who aren't in a position to interact here (say people who are logged off, instead of people who are logged on.  Or people currently in matches, or something.)


The Plaza

In the plaza is a cafe you can't go into (because it's not like your character needs to eat or drink) the shops, an arcade game (not an arcade that you can play arcade games in, a single arcade game machine in the open) the place you go for multiplayer, the place you go for same-system multiplayer, and the Amiibo station.

First the shops.  I was disappointed when I learned that headgear, which includes hats, helmets, glasses, goggles, headphones and probably other stuff (oh: gas mask, head band) is one item only given that I want my character to be able to wear glasses and a hat.

Simple solution: split off eye-wear into a different shop.  The game involves paintball (well ... inkball) it would make sense for everyone to wear goggles by default and the eye-wear shop could be divided between things like cooler goggles, normal glasses, sunglasses, Geordi's visor (but not called that so you don't get sued) and such.

Headphones would also be good to separate out since they're there already but are somehow prevent you from wearing glasses or a headband.

Now, since gear comes with abilities there could be balance issues, but I'm sure they could be worked out.  Before I do quick math on that, consider this:

You can pick your headgear, you can pick your shirt, you can pick your shoes, you can't buy pants.  (We are the squids who wear no pants.)  Instead everyone wears the same featureless, black, form-fitting athletic shorts.

They need a pants, shorts, and skirts shop.

Thus expanded, the shops would be:
Eyewear, hats, headphones, shirts, pants/shorts/skirts, shoes.

Headphones kind of stands out as "Wha?" but they're already in game as headgear.  It's canonical that inklings are known for their liking of music, so each of them having some kind of music thing (doesn't have to be headphones, could be an nondescript MP3 player clipped to a belt) isn't really a stretch.

That's six shops.  If the new things all have the same one primary ability and up-to-four secondary abilities thing going on then you could have every ability as a secondary as opposed to the half you could have before (if the random rolls went just right in either case) in addition to the six primaries (compared to 3) in a single outfit.

I think it's not really desirable for there to be the potential to have everything, but it's not like there was some law of nature that necessitated there be precisely 24 abilities to begin with.

If we double the shops we could double the abilities.  Some of them are already pretty esoteric.  "Haunting" is the one that comes to my mind.  It marks the person whop killed you so that everyone on your side can see where they are (yourself included when you respawn.)  I'm not precisely sure what controls how long this lasts, but it's probably the same as the point sensor and echolocator which do similar things.  If that can be an ability you can imagine how many possibilites there are.

Run speed and swim speed being increased are both special abilities, but nothing about jumps.  Jump length and jump height are both things that could be increased.

I'd like there to be an anti-gravity-in-ink-ability so that when you're in ink on a wall and not moving you stay where you are rather than slide down.

Being able to submerge in enemy ink could be useful, even though enemy ink is harmful to you and has a severe movement penalty.

The point is, the abilities out there to abilities you can have ratio could be kept the same because I'm sure more ones could be thought up.  (Or, alternatively, just don't have as many slots.)

For that matter, you could have a feature where the maximum number of abilities is capped but the amount of gear is variable.  As it stands everyone always has three bits of gear.  You can't go barefoot or shirtless, and if you want there to be nothing on your head the best you can do is buy contact lenses for your head gear.

I'd still want the inklings to be wearing clothes, but things could be made optional.  The shorts currently contribute nothing, have them be the default and still contribute nothing.  Have a similarly bland shirt be the default and contribute nothing.  Have all other gear be optional.

Then put caps on the abilities.  If you think the game is perfect as it is, you can only have three primary abilities (if you have four or more things on some of the primaries, you should be able to choose which, get demoted to secondary) you can have at most 12 secondary abilities, if you've got six things on all of them maxed out, only the first 12 secondaries count (this is easy to figure out because they have a set order) so from the things with non-demoted primaries you get two secondaries each and discard two secondaries each.  From each thing with a demoted primary you have the primary as secondary, the first secondary, and discard the three other secondaries.

My point here is that there are plenty of ways to have more gear without unbalancing the game, so I have no fears about adding gear types and would do just that.


I wondered, to some degree, if people would find these ideas absurd, the game isn't dress up, after all.  It turns out that there's an online movement to add a pants shop.  People also want a hair salon.

Inklings don't have hair.  The top of their head has something that looks hair-ish from a distance but it's actually a remnant of your squid-form color, the most hair-like part are the still-long tentacles.  You can't get a trim.  It would have to be surgery and I'm definitely opposed to that being in the game.

However the tentacles can be styled in a way.  They can be tied.  The boys all bind theirs toward the top back of their heads, the girls all wear them loose.  A couple of NPCs who are older than the players and have extra long tentacles have theirs bound in ways that are more complex and clearly feminine.

I'd definitely support a tentacle styling shop.  Especially because if we assume that loose tentacles are how you say to the world, "I'm a girl," and bound at the back ones are how you say, "I'm a boy," (which seems to be the case) then someone with long but bound at the back tentacles would openly be a trans boy, someone with short but unbound ones would be a trans girl, and we'd be able to have instant representation just by letting people decide how to do their "hair".

(We probably wouldn't get it, though I'm all for NPCs that look like other players to populate the larger world I'm going to be describing, thus we could ensure that we got it.)

Other styles would be a way of saying that either you're neither gender or you care more about looking how you want to look than people getting your gender right at first glance.


People also want pets which I don't really see the point of, but ok.  Some specifically want ink-pets that they can use in multiplayer which, I guess, would take the form of a special and ... there's nothing wrong with that, some of the specials are weird.  Or maybe it would be a secondary weapon (throw out your pet) which, again, some of the secondaries were kind of weird so ... sure, why not?

People also want apartments/houses.  This actually makes sense because the loading screen when you start the game is of an inkling's room and if you look at the gamepad you see a map of said room.

In fact, having a home could be really useful for being able to sort your gear.  Right now all your weapons are on one unsorted screen and if they take up more than one screen you have to flip right or left to see more.  There's not rhyme or reason the order they're in (perhaps the order you acquired them?) and it's a bit of a mess to find what you're looking for.

If you actually had a 3D space in which to put your stuff you could sort it in a way that makes sense to you.

Additionally it opens the door to having swag you could collect.

X-Wing Alliance gave you a room (complete with a growing souvenir collection) in spite of the game being a flight simulator.


People want to see more of Inkopolis than just one plaza and the multiplayer stages.  I'm with them on that.  It would be cool to be able to walk around and look at the place you save in single player and live in regardless.  The multiplayer stages are explicitly normal places that they've cordoned off to make into a match spaces temporarily.  It would be neat to be able to walk around the non-cordoned off version when it's not in use as a match.

(It might also be neat to go to one of the places currently in use for matches and spectate.  Obviously it could only show one match at a time even though many are going on, but: who cares?  Be a spectator.)

For that matter, since I'm basically talking about turning a simple match based multiplayer game into an MMO, it would be cool to arrange unofficial matches where you told people, "Show up at X and bring your ink," and the fight didn't really have rules or time limit and didn't affect your standing or ranking at all.

And being able to wander the city comes back to something about the merchants.  It would make more sense to me if the shops were fully loaded with stuff all the time and it were just the case that certain sections weren't open to you when you hadn't reached a sufficient level.  Furthermore, that would could lead to decisions about various things because if multiple companies are offering more or less the same product, you've got to decide which to get.

And what about if you're choosing how to dress entirely based on fashion and care not about abilities?  We know there's a mall in Inkopolis, maybe the stuff there is cheaper but doesn't have abilities or doesn't have as many.  (Could be everything there is primary only without a single secondary slot.)  The mall could also have stuff you could use to decorate your room.

I'm going to get into travel around this expanded city, but first one more thing on clothes.

What if you really like something in terms of stats, but not so much in terms of how it looks.  There ought to be an alteration shop that will help you change how things look or, failing that, the ability to let some piece of clothing soak in dye of another color until it's something you like.  (Boring white tee?  That's what tie-dye is for.)

Ok, so, bigger city.  The super-jump from multiplayer, or the squid launchers from single player, would justify fast travel from one location to another so that you wouldn't be stuck slowly making your way from point A to point B.  That said, there are other means of travel and the plaza really feels too much like a human design and not enough like an inkling design.

Inklings move faster in squid-form so you'd expect the design of their cities and buildings to reflect this.  The plaza is a no-splat area so you can't lay down a trail of ink to swim through, but there could be channels of some sort of opalesque universal ink that everyone could swim through to move faster than on foot.  Also, in single player a mode of fast travel to difficult places was shown, ink-rails, which are sort of like ... um ... beams of ink projected in straight lines until they reach a joint that redirects them in another direction or their end point.

Both of these things could be ways to travel around larger areas without getting bogged down in travel time.


Multiplayer game modes.

Right now the multiplayer is a sort of balance between laying down ink and fighting your opponents to hinder them from laying down ink while preventing them from hindering you.  That's fine, I'd add two things though.  One would be a mode in which it's just laying down ink.  You cannot die, you cannot kill.

There are still downsides to someone shooting at you --there's a severe movement penalty on the other team's ink, you need your own ink to refill your ink tank-- but in general the game is primary objective only: ink more territory than the other side.

The other would be the opposite.  It doesn't matter how much ink you lay down, the object is to splat the enemy more than they splat you.  The way squid-swimming through ink changes movement and what it does with respect to stealth means that this would be very different from what goes on in a traditional shooter, but the objective would be the same.

Right now all matches are four on four.  I'd open it up to more possibilities.  I don't think one more person would kill the system and cause horrible things, so three teams of three seems an option.  Four teams of two definitely.  Free for all: eight colors of ink being shot around.

Also, instead of 4 on 4, 3 on 3, 2 on 2, 1 on 1.

And this is part of where the footnote, apparently the only footnote (at least the only one so far), comes in.  Bots really are needed to deal with cases where someone's connection dies, and they could be very helpful when the matchmaking software can only find seven people to put in a match and thus no one gets to play, but if there were significantly more multiplayer options (I just described 42 options of which only one is in the game) then it is absolutely essential that the system has a way to deal with things when there aren't enough human players.

The more options I have, the fewer people playing in each option, which means the more important it becomes that a match can be had when there aren't enough people who want that type of match at that time, which means the greater the need for really, really good multiplayer bots.

Not really, really good in terms of ability to win, really, really good in terms of acting in a humanish way.  Beginners don't need their butts constantly kicked, if they're going to have artificial opponents those opponents need to be on their level.

Further complicating existing multiplayer is that since it's four on four all the time, you're always on a team with three other people and you didn't get to pick their weapons.

Maybe you're great with a blaster provided there are other people on your team with rollers to lay down ink for you.  If you equip a blaster and get stuck on a team where everyone else has blasters ... sucks to be you.

There are four primary weapon types.  I think that leads to 35 possible combinations on a 4 person team.  Basically any level of control over what kind of team you're on will require the matchmaking software to consider all 35.  (I suppose the exception is if you could name only one that you wanted to be a on a team with, then things could be paired off and pairs stuck together, that's two sets of ten combinations so I'm not sure that's any simpler.)

And, again, you start to see how it would be vital to have a supply of players in cans (really good bots) to choose from to fill out any matches.

But it would at least mean that you don't start a match with a feeling of, "Fuck, I suck with a team armed this way."



Levels are just earning your way into being able to buy better stuff.  Rankings are really important.  There are official rankings which happen as a result of "Ranked Battles" (good name there, I wonder how they came up with it) and I'd like to think that there's some kind of unofficial behind the scenes ranking system that tries to keep a bunch of newbies from squaring off against the most experienced veterans of all time in the non-ranked matches.

Too often, in both types of matches, one side never stands a chance.  This is a serious problem.  It doesn't help, it isn't fun, it's just no good.  There needs to be a better system for making sure that the teams which face each other both at least have a chance of winning.

Moreover, ranking needs to take into account something that the official visible ranking does not: there are four kinds of weapon.

If you're the best there is with a roller you'll probably be able to get a high rank, but if you then decide to start practicing your charger skills there's no way in hell it makes sense to throw you into a match against top ranked players.

It needs to be divided by weapon so that you're playing at the right level for you no matter which weapon you have equipped.



The places you play are called stages.  I'm not totally sure why.  Maybe it's nomenclature I'm just not familiar with, maybe it's because the game already uses the terms "level" and "map" to mean different things.

Regardless, there need to be more of them (there's at least one more on the way.)

In addition to needing more, though, there's also a need for different ones.

The single player campaign made heavy use of an interactive environment.  The ink rails I mentioned were part of that.  You can see how it would work, and offer a risk/reward combination in multiplayer.  Shoot the start thingy to turn the rail on in your color, then use it, but watch out because the enemy can shoot the start thingy too.  I'm not sure if doing so should turn it off (with a second shot turning it on in their color), or immediately switch it to their color.  Either way you'd be dropped right out of it.  (If it did immediately switch to their color you'd be inked up close to being splatted when you fell.  Or possibly splatted outright, either way.)

There was invisible stuff that you had to ink up to see.  On the one hand, this could be brought in the same way, or it could be made so that not only does there have to be ink on it for you to see it, but it has to be your ink.  Enemy ink doesn't show it to you.

The sponges were a truly interesting thing.  Shoot them with your ink and they grew huge and you could interact with them as you could with any inked surface.  But if the enemy shoots them they shrink back down.  Two possibilities.  One is that all sponges are neutral.  In their uninked state they're just waiting for someone to ink them so they can grow up huge and take the ink's color.  When shot with the other color they shrink back down to the uninked state and then, from there, can grow up huge in the new color.  The other is that some sponges are one color, the others are the other color.  Shooting with the color of the sponge can only make it grow, shooting with the other color can only make it shrink.  You can make an enemy sponge small and useless, but you can't convert it to one of your own.

There were things powered by propellers that you'd shoot ink into to make them go.  generally elevators or things that moved side to side.

Single player also had stages with dynamic parts from simple spinning things to blocks that rearranged themselves.

My point in bringing all of this up is that it's in the game.  The code is already there and it's just waiting for someone to take it and make multiplayer stages using these features.

I'm not advocating turning multiplayer into a platformer like single player, but I do think that the unique features from single player could be introduced to multiplayer.  Should be introduced, in fact.


Random other thing: I don't know how hard it is to make a very basic skateboarding simulator, but one of the multiplayer stages is a skate park.  It would be kind of cool to be able to try to skateboard there when it's not being used as a multiplayer stage.  It is, after all, a skate park.

I really feel like I should be able to pull at least one other random bit of random out.

Oh, post single player victory it would be pretty cool, in my opinion, to be able to have non-violent interactions with the octalings.  The war is over, you won, they're probably bitter and pissed off, but maybe you could do something to help their situation which seems to, kind of, suck.

[1] Historically this has proven to be an incredibly difficult thing to do because while making an opponent that will be difficult for a player to beat is fairly straightforward, making an opponent that acts like a human being is less so.  Great strides have been made in this arena, but since Splatoon is not a traditional game, it is likely that a lot of what is needed to make a faux player needs to be built from scratch.

In addition to all of the stuff that goes into trying to make a shooter bot, which has traditionally been the problem approached, a faux player would need to be able to decide when to shoot vs when to lay down ink, where to lay down ink, when to switch between the two tactics, when it's best to lay down ink in an area that doesn't matter in order to charge their special, how to use the special, the best way to use things like beacons, when it pays to drop all reason and go for a Hail Mary because you're about to lose, what the Hail Mary would look like, how to best support and complement allies, and so forth.

The reason that Splatoon needs multiplayer bots is that, since the object is not to rack up the most kills, losing a player is devastating.  In a traditional shooter if you lose a player because their connection was lost, they were called away from the game, or whatever it's bad, but not catastrophic.  The reason is that while you have one fewer person on your side trying to shoot the enemy, you also have on fewer target for the enemy to shoot at.  It doesn't make up for a lost player, but it does mitigate the effect.

In Splatoon the object is to lay down ink.  Four on Four and you lose a player means that there are four of them inking the stage and only three of you which means that all things being equal the end result should be that they cover 57% of the stage.  What would be a tie is a loss.  What would be a win is a loss, only what would be an impressive win still constitutes a win.

I wonder if vast strides could be made in multiplayer bot making simply by collecting an analyzing the data that's being produced every time people play.  All games are run through nintendo online so in theory, if they wanted to, the could record every single game, use that to gather data for how human players actually play, run the data through pattern matching software to simplify it into a finite set of machine-understandable tactics and strategies, and then attempt to simulate those tactics/strategies.  (Still using the real life data to create very complex, but basically if-then based, ways to make the decision of which tack to take and when to change tack.)

We're talking about massive amounts of data to sift and analyze, which means ideally you'd be doing this shit on supercomputers (which, more often than not, are being put to better uses), but it does make for an interesting thought.  Every multiplayer match is a trove of data on how actual human beings approach actual gaming.  As far as I know that data is never saved, but it could be and analysis of it could do some very interesting things.

1 comment:

  1. These are some really neat suggestions - they're definitely things the designers should be looking at if they can sink more money into the game.

    I think the cheapest is probably matchmaking improvements, and your suggestion of having weapon-specific rankings is the way to go. I don't know how much your global ranking should affect your weapon-specific ranking (someone who's played the game a lot but never used a charger is going to be better with it than someone who's never played at all), but that's a problem that can be solved by experiment, looking at how matches turn out.

    Multiplayer bots is the most important, but I expect it's also the most expensive. It's doable, though - I remember Left 4 Dead and Left 4 Dead 2 had decent AI for your buddies.