Posted by: darthmeanie | July 9, 2012

Designing Your Best — Subtractive Design

A designer knows he has achieved perfection not when there is nothing left to add, but when there is nothing left to take away.
Antoine de Saint-Exupéry


I’m here to talk about two different ways to approach building a moveset: additive design and subtractive design. Additive design is what seems to be the default for most people in Make Your Move, and it’s certainly the easiest method to make a moveset. You start off with a core idea or element, and then add to it with each subsequent move to create a layered playstyle section with many options to build off of the core concept. This is probably what describes most movesets written, including most of my own historically. With subtractive design, the goal is different. Instead of making the set more complicated with each move, you eliminate as much complexity as possible to make the moveset as streamlined, and simple as you can without sacrificing the core vision. That last part is important.

Additive design is clearly what’s popular in Make Your Move right now, especially with some of the sets that are considered at the top right now; Urabrask, Pennywise, Smot, Luxord. Additive design has been a staple concept for Katapultar’s movesets for a long time, and the increase in popularity in this approach has no doubt contributed to his newfound position as a leader in Make Your Move, as well as his unshakeable work ethic. But this sort of additive design has more weaknesses than are immediately apparent. Let me start with a purely theoretical example, one that I’ve been guilty of at least twice, just off the top of my head, and that I’m certain many other members of Make Your Move have done as well. Let’s take for example a moveset which has a core vision of keeping the opponent on the ground and attacking from above. For such a character, the Up Aerial doesn’t immediately fit into the vision of a playstyle because, frankly, he doesn’t want to be attacking opponents who are above him. To get around this problem and avoid having a seemingly dead move, the Up Aerial is written to have the character throw the opponent below him, putting him back below where you want him. Problem solved, moveset improved, right?

Not necessarily. It’s true, with an Up Aerial that throws the opponent down, there’s one more element that you have to mention in a blurb when you write the playstyle section, but is the playstyle really deeper? The character’s primary weakness was having opponents in the air above him, and now you have a move that’s designed to circumvent that problem. The more interesting sub-game of trying to keep the opponent from getting above him in the air in the first place is now less rewarding for both players, as the opponent has less reason to take advantage of the core vision’s weakness when a single move will stop him, and the player can handle what would have been a troublesome problem with a single input. Certainly, there’s still some depth to landing it, spacing it, air dodges and what not, but you’ve made a more complicated move that hasn’t really made the set any deeper. And this is just a basic example. With additive design, a set’s moves seem to feel almost like patches to cover up any weaknesses the core concept the set revolves around creates, and as a result it’s practically assumed that the character you’re playing as will win if the player is remotely competent. After all, you have but to follow the flowchart each move creates to add to the set! The moveset ends up more complicated, but not any better for the complexity.

Make everything as simple as possible, but not simpler.
Albert Einstein

I freely admit I have yet to master the art of subtractive design, but I think that there’s a lot of value to practicing it, and that subtractive design can allow for movesets just as deep or deeper than additively built movesets, but without all of the bloat, forced move interactions, and awkward attacks that tend to come hand in hand with that style of movesetting. Two-Face is a moveset that ran with the concept of subtractive design. Now, I know that Two-Face was hardly a fantastic set; I made it in the course of three hours, two of which were lost when my computer crashed working on it. But I ran through many ideas for how to add to the moveset; tranquilizer darts, traps, smoke bombs, thugs, all sorts of mechanics that would have contributed to his ultimate goal of landing a successful coin flip and getting a successful KO off of it. But I eliminated almost all of those ideas because they didn’t actually give the playstyle more depth; making it easier to pull of a coin flip would in turn take away from the far more interesting risk of bluffing between fake flips and real flips and countering with the Forward Smash, and making it easier to land a kill move without losing the bonus from a whiffed coin flip boosted attack would remove the interesting game of using his attacks with follow-ups to try to fish for a chance to blow his stronger attack. Certainly, I think if I were to go back, I could make the concepts and moves flow together better and make his 50:50 mix-ups that made the core of the set smoother, easier to comprehend, and more diabolical, but I think that for what I did do, the set was ultimately better for what I subtracted rather than what I added.

Subtractive design is not a call for forced simplicity, unimaginative concepts, or even necessarily in-smash movesets. It simply means that simply adding something to a moveset, even something that seemingly increases the character’s options and playstyle, can ultimately simply make it shallower and more complex. It’s a plea that sets not let their original vision for how they should play be clouded by extraneous ideas that don’t contribute to what makes the moveset truly unique and engaging. Design like this isn’t as flashy as interconnected mechanics and is more low-key than movesets that are more liberally constructed, but if we analyze movesets carefully, and see them as more than simply the sum of their parts, we can see how a well designed moveset can consider all of the different options and clearly present a well-written moveset.




    • This also, ironically, summarizes my feelings about this article. I would actually take it seriously if, as MW said, you were able to find a set using this philosophy other than your’s.

  2. I can see your point far better when you’re using that example with the uair, and that point is actually good. Flow should not cover up a set’s weaknesses en mass, and the intended weaknesses should actually be weaknesses.

    Still, that does not mean that the character should have absolutely no way to deal with their weaknesses. I think the most blatant example of my own sets would be Black Hole Bowser, whom hates people getting behind him due to having few attacks that hit behind him and having horrible turning lag. I still do give him attacks that hit behind him, but they’re bad, and much worse than what he has to start with. Characters still need to typically acknowledge their weaknesses in most cases, unless it’s just something like a bad recovery. I do like where the article is going in regards to that, though, and it’s certainly something to think about. Just adding in my own two cents.

    Using Two-Face as an example was not among your wisest choices. I know you don’t read sets anymore and say you’ve only recently gone to this approach, but surely there’s something you could’ve thought of from the past, some Junahu set or something, that used this mentality better? Because even you aren’t exactly wild about Two-Face.

    • My point isn’t that movesets shouldn’t cover for their weakness either. Though I can understand hoe you thought that. The point is that adding more to a set can actually hamper the playstyle, and that streamlined focus on the core concepts you want to get across can make a better, deeper set.

  3. This is actually a perfectly legitimately good article(I was expecting it to be a lot more condemning) and frankly David you suck at making jokes. Maybe you should actually read the article and not try to deter people from something that has a good point with your horrific sense of humor.

    • Jokes? In case you haven’t bloody noticed, FA, the only goddamn example he uses for this ideology is his own set. As said, I would actually take this seriously if he was able to name more examples that didn’t come from him.

  4. Using your own set as an example automatically makes you terrible now? I’ll keep that one in mind.

    • As I said to you in private chat: when the only example for something he is trying get people to adopt is his own set, it’s pretty damn bad. He’s able to name multiple not-him examples for the ideology in which he is attempting to point out flaws in.

  5. Now admittedly, that is a problem and I wish you had brought that up earlier. At any rate though, in his defense not a lot of people seem to have even tried that style. It’d seem more usable if there was more to it’s name, I suppose.

  6. Tutankoopa and Baron are the two sets that brought playing on your own weaknesses to the forefront, both play on that paradigm and get plenty creative. I wouldn’t be so quick to group together practically every moveset that comes out because they have similar focuses.

    It comes down to a set-by-set comparison, and I feel this may be your lack of reading at show. True, most sets now try their hand at this kind of playstyle, but they have varying levels of success and there is value in sets that do it well.

    Nonetheless, different types of movesets are always good, so I’m glad you presented the argument. It does come across as generalising more so when you can’t find any good examples of ‘subtractive design,’ however, but can find many of ‘additive design.’ And I don’t think Kat came back to seize the opportunity of the current set climate, that’s some tinfoil hat stuff.

  7. I think there are hardly differences between Additive and subtractive movesetting. regardless of whether or not every idea you have for a set makes it to the final cut, you are adding before you ever subtract. There are incredibly focused sets and (the more popular type at the moment) multi-layered sets with a vast amount of options for playing them, but all sets are ultimately conceived in a similar manner, and it is up to the movesetter as to how far they take their character and ideas.

    And unless you’re some massive 3v1 boss character like Ameno-sagari, you should in fact have some weaknesses or at least specific styles of play you have trouble countering – though obviously we have come to value balance less and less in this contest and many, many sets end up with 0/100 and 100/0 MUs against certain others.

  8. My position as of now doesn’t necessarily have anything to do with the current design trend – it’s just that now I actually know what I’m doing and how to get it after a good deal of inner struggling in both MYM and real life. Also the fact that I’m done with school (that doesn’t mean I don’t work irl though).

    I can only guess that additive design, in practical terms, would have at least one weakness in that the moves do not have as much freedom as a Brawl character’s given how focused they are for a particular purpose, though there are moves that attempt to play like normal Brawl ones anyway. Even I don’t really know all that much what you mean by subtractive design, possibly because it doesn’t feel all that well described here, but I can only believe it’s referring to recent sets like Shana and Sayaka who try to be simple but still have some cool stuff going on in them which makes them well-liked anyway. As a rule I do try to make sure my sets have their melee attacks feel like practical Brawl moves yet at the same time serve some purpose to the playstyle so they wouldn’t feel restricted or underpowering in a Brawling environment, though with certain sets I just cannot do that or at times end up getting carried away. Speaking of getting carried away, this article is supposed to deal with the implementation of a character’s weakness, is it not? Kind of maybe.

    It’s nice to see an article from you like in the old days though and showcasing your seriousness for the situation with quotes from famous people. At the very least we haven’t fully lost you.

  9. I don’t presume to grok everything meanie says – and like Warlord, I’m no doubt misrepresenting his original argument – but these delineations are always more useful as ways to think about our movesets differently than literally different methods, so here’s what I’m seeing:

    Additive design: Take a basic playstyle concept (for instance, deceive the foe into not knowing where you are at any given time) and then build up on it, twist it off in various directions, give the character more playstyle options and different ways to flow in any given game. You have Gengar.

    Subtractive design: Take a basic playstyle concept (keep the foe trapped on the ground between two poles) and try to get it across and then work DOWN, trying to get it across in as few and as simple moves as possible. Can you make a complex playstyle function without having complex attacks? Can you create a deep, fun playstyle without letting the character branch out in as many directions as your imagination can come up with? No? Now you have Wolf Man.

    I think HR’s sets are very good examples of what DM is talking about, and although I work in half-measures as ever, this is also largely the sort of thing I was going for with Skeleton, Tutankoopa, Gastly, Mouse Man, Wolf Man, and even Fulci (on all of which – exceptions of Mouse Man and Wolf Man – I cracked at some point and gave in to my more additive leanings). Fulci is a set about minions that can kill you as well as the foe, and with the exception of silly stuff like the burial and the ooze I did my very best to put that playstyle forward in an interesting way in as few leaps as possible.

  10. A problem of additive design in MYM, is that by everyone’s own admission, nobody plans their movesets enough to make this approach truly shine. If you create a set linearly, and make a 50-idea pile up, you end up with something that people will like in the short term, but that ages horribly in the long term.

    As Warlord says, an example from my back catalogue probably would have served your point a little more. Airman and Zigzagoon spring immediately to mind, or Billy Hatcher, if the example needs to be more recent.
    Actually, just my two Ditto sets side by side would have demonstrated the difference between Additive and Subtractive design.

    • Air Man is actually a wonderful example and one I wish that I had used. It’s one projectile being used, but it uses it smartly. It’s a shame the set’s subtractive direction and confusing organization (by design) made it mostly glossed over.

      That and I’m getting old and crochety and can’t keep up with every newfangles set that comes out.

  11. Zigzagoon cheats the approach by simply not having inputs to subtract from.

    And yes, obv nobody read/remembered Air Man. Such a glossed over set.

What do you think?

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


%d bloggers like this: