In-Game Asymmetry

Discussion in 'Game Design' started by keithburgun, Jun 15, 2013.

  1. EnDevero

    EnDevero Well-Known Member

    Actually, what I disagreed with was the idea that there isn't any cognitive dissonance. I am fine with what you said in the above comment though.

    Edit: For clarification, I'm talking about cognitive dissonance in the common emotional sense. I don't mean that I'll ever choose to play non-optimally in a competitive scenario, I mean that the clash between desire and objective will irritate me. I'll play to win, but I won't enjoy it as much as I could.
  2. Leartes

    Leartes Well-Known Member

    I have to say, I like the article a lot. You have changed your writing since you published your articles on what games are and the result turns out very well.

    Additionally I agree with a lot of points you make. Only I come to a different conclusion :D

    I think it can be impossible for a superior symmetric game to exist in a limited genre. Again going back to the dune-style rts, I still find the claim reasonable that you can't make a symmetric single-race game with equal replay-value. If this is the case then you simply have to pay the prize. You may be forced to make it asymmetric because you run into a wall otherwise. Furthermore if this holds true for dune-style rts, then it is reasonable to assume it holds true for more genres as well.
    For instance in fighting games, I don't think you can make them symmetric either and keep them interesting for the same fighting game players that play them right now. With making symmetric I don't mean cutting an existing game until it fits, but making a new game from scratch using mechanics close enough to retain the appeal to the target group.

    Small nitpick, I think weapon selection in CS and selection of the first hero in WC3 is both the same as pre-game race selection. You usually make the choice before you have any information on your opponent thus it totally doesn't matter if you picked the hero in the lobby, on the load screen or actually ingame. I think we could think of WC3 heroes as sub-matchups with the hero choice of similar importance as the race choice in AoE.
  3. pkt-zer0

    pkt-zer0 Active Member

    Well, if you have to handicap yourself for others to pose any sort of a challenge, even despite doing their best to win, then your problem is with the tiny player base, not necessarily with the game itself. Not an issue that most folks are going to run into, I'd say.

    It doesn't seem like that's quite what you were getting at, though. Hmmm.
  4. Nachtfischer

    Nachtfischer Well-Known Member

    That's what Keith does. And that's how he has obviously been able to derice very useful guidelines for practically designing games. If you don't like the approach, then don't read the articles. Just by realizing what the overall approach is, you can't criticise it. That's stupid.

    First, the article explicitly states that asymmetry is not necessarily "bad", but comes with certain costs. Second, the part you're referring to just explains what the whole piece is about. It hasn't even started to criticise or value anything. It's all in your imaginative head.

    I mean, you repeatedly state how difficult and complex of a problem designing a game is. And now you think mere players who have probably no idea of the field should happily take on the role of the designer? Yeah, that'll surely work extremely well!

    Only if there is no dissonance between the options. I mean, you're "playing to win" statement is exactly the point. When you "pick a race" you're not even playing yet.

    Sounds pretty much like Sirlin's "you have to learn less to play the game" argument. Which was debunked by the article as well. Only because you have 30 characters, doesn't mean you don't have to know them all. Especially if you're going to play competitively and don't know who you'll be facing.

    If you make such a game, then it is already balanced. You can't simply balance a game as an "afterthought", if the whole system is screwed beyond (balancing-wise) repair already. That's like saying: "First make a ball, then make sure it rolls!"

    That wasn't even the point. You as a player have less options of how you influence the emerging dynamics, because you're already kind of "half-way into the game" after picking a side.

    I have no idea how you make the connection between matchups and random maps? It was repeatedly mentioned in this thread that input randomness is NOT the problem. And neither "asymmetry" as a part of that.

    Not entirely. But if they're weird, quality-reducing and just induced by the industry to make money. Why not?

    As if anyone had ever started out to make that.

    Guess you just had a trollgasm, huh?
    keithburgun likes this.
  5. Dasick

    Dasick Well-Known Member

    You seem to have a little false dichotomy going on here. I'm referring to a concept that I have no idea how to phrase it in any other way. Take the development of the A-bomb. It's a leap because a lot of resources were spent into creating the infrastructure needed to manufacture A-bombs, it's faith because you've only got hypothesis and circumstantial evidence, certainly no-one's seen or used an A-bomb before, and it's rational because a group of people got together, and after weighing pros and cons decided to go ahead with this thing. The phrase isn't really as ambiguous as you make it out to be.

    Moving on...

    Keith, I'm not sure your article addresses some of the points I made pro-asymmetry. To make things easier, here they are point-by point:

    1. Breadth
    You kind-of addressed this with the variety/replay value point, but here is the main point I don't see addressed: gameplay is created by placing limitations on the players, it's the difference between what is possible and what isn't. If you take Blake's Terran based game where Terrans can create bio-labs that spawn zerglings, when playing this game this is something you have to consider. When you play a straight up TvT or TvP - no zerglings are possible, and you can use that information in your strategy. You don't have to fear being ambushed in an open area (which favors a swarm of zerglings), you don't have to fear an early 6-pool rush, you can do earlier air raids because there will be no overseers to detect your cloaked wraiths etc etc. It's not that you're exploring the same web just a different part of it, you're exploring a different web altogether.

    2. Combating Fatigue
    Playing the same web over and over is fatiguing (I think this is what some people mean when they say symmetric games get boring). Switching things up, just a little bit, helps avoid that, and allows people to play a game they like for longer. Note, that I'm not talking about 'being done with the system' and prolonging that, I'm talking about game sessions, how much you want to play at a time. As much as I like playing Go, I can only play one or two games in an evening, but I have no problem playing other games afterwards.

    3. Elegance
    Each individual match-up can be more elegant. When you consider match-ups as part of the game, and you have the whole meta-game on top, yeah, no, fuck that noise. But I'm considering MUs to be like individual gamemodes and stuff, there should be some other way to implement those without metagame and counter-picking and design-your-own-game problems.

    4. Cutting floor
    Not everything left on the cutting floor is a bad idea - sometimes there just wasn't room for it in the actual game. For example, there was this scoring system in one Auro version, where your combo points don't decay if you picked up a power tile last turn. Yeah, it was completely broken, and re balancing around that would mean throwing out all of the current balance, but why can't it be it's own character gamemode? Or what about iceblock-based ice tree... that never worked as intended, but what if the entire system was rebalanced around that? etc etc.
    keithburgun and Nachtfischer like this.
  6. Nachtfischer

    Nachtfischer Well-Known Member

    Dasick, I think all of your points pretty much come down to each matchup being kind of its own game. So, instead of a "design choice", the choice of race (or character or whatever) comes down something like choosing a game mode. I think nobody (Keith included?) would have a problem with that if all the matchups were equally interesting and equally balanced. (At least then the "design choice" becomes a "false choice" and so it's really more a choice of flavor.)
    It's probably still kind of weird to not have a mode that is a specified "main mode" by the designer. But, if you really made like 10 equally valid games, they probably should all be "main". I guess Keith's problem then derives from the fact how difficult it is to even get the balancing (and interestingness) of one single game right. So, can you make 10 equally valid ones at a time (many asymmetrical games don't even stop there)? Maybe. But it's probably incredibly hard, so it'll most likely come with compromises for almost each single "game in the game". The article also says, that it's not inherently bad to take the asymmetric approach, it's not impossible to achieve something with it, but it comes with certain "costs".

    EDIT: Oh, and I totally agree with this:
    I'm just not sure if asymmetry is the way to go against that, if it comes with compromises you have to make in terms of the overall quality of the game? Maybe just make a variant that's clearly defined as such. Or another whole new game, so people have multiple systems of yours to take into their "playing rotation". And as long as it's great it's likely to stay "on rotation" for a long time. I don't think your game has to be playable with the same amount of "fun" involved, without ever stopping to play something else... in fact, you probably will not achieve that without some devious exploitation of the human brain (see Diablo, WoW... I played them for many hours straight, if you take out the real-life parts probably even for months without ever playing a different game. Now I play Outwitters once or twice a day for half an hour or something, or Puerto Rico twice a week on average... and that's totally fine, and still I think they're a million times better).
    Dasick and Leartes like this.
  7. Dasick

    Dasick Well-Known Member

    I didn't say anything about not having a main mode... just because modern videogame asymmetry has plenty of sucky attributes, doesn't mean they're all inherent to all sorts of asymmetry.
    Nachtfischer likes this.
  8. keithburgun

    keithburgun Administrator, Lead Designer Staff Member

    Ok. I think I agree with everything you wrote there. Was there a specific question for me about that?

    I actually think that if you're getting bored of a game, it's not because it lacks asymmetry. Go gets boring because of design problems, namely, it takes a long time to play overall and it's very end-game heavy, meaning that like the first 100 moves or something all feel kind of like "setup" and then the game only begins really late. So while asymmetry CAN combat fatigue for a short time, I think it's kind of a cheap fake solution. To come back to Puerto Rico, I've played like dozens of games of that in a row. Same with Chicago Express and Lost Cities, none of which have asymmetry.

    Also it's worth noting that some degree of fatigue is to be expected in anything, and it's not something to be too terrified about. Even if asymmetry can stave that off a bit, again, it's kind of a cheap effect and comes with all of the negative aspects I described in the article.

    I'm certainly not opposed to game modes. AURO might eventually do something like this, where there's a specific mode where you play as a specific character that supports a different kind of gameplay mechanism. This covers your last bullet point, too I think.
  9. EnDevero

    EnDevero Well-Known Member

    :p Not exactly what I was getting at, no. What I was getting at is a bit more abstract than that and maybe won't seem practical to you. Say there is a game with a tactical layer and a strategic layer. Normally, the strategic layer would revolve a lot around making instances in the tactical layer easier for you and harder for your opponent. Well, I'd prefer a game in which the strategic layer provided incentives to actually sometimes make tactical instances more difficult for yourself. Like, both players can handicap themselves in some way for scoring advantages, stuff like that. So the game sort of becomes a game of chicken; two players pushing themselves and each other to the limit seeing just how much they can take. Ideally, a big part of the skill in the game would be understanding your own skill level and being realistic about what you can take on, while also not underselling yourself. My ideal game would be largely about showing off, basically. But that's a personal thing mostly. I don't know that anyone else is as into that sort of idea as I am. But, yeah, so it's not so much about me not being challenged, that's not the issue, it's that games usually provide incentives for lowering difficulty more than they do for raising it and adding pressure. Whether my desires are realistic or not, that's a different story.

    I guess you could sum it up as:
    I'd like a game in which players have to challenge themselves as much as they have to challenge their opponents.

    So, bringing this back to asymmetry-- My ideal asymmetric game would provide extra rewards for doing well in a difficult match-up.
  10. pkt-zer0

    pkt-zer0 Active Member

    So, winning with Ryu gets you 1 point, winning with Dan gets you 5 points, winning with ST Akuma gets you 0.000000000001 points. Something like that? That's just simple risk-reward balancing, I think. Or accurate skill measurement? Maybe way too meta for me to comprehend (and I've been reading about recursive algorithms recently): a game about estimating how good your are at the game. A tournament where you bet on what place you're going to finish in? Dunno, I'm too tired.

    As for asymmetry, the ideal game would also be balanced, meaning all matchups are equally difficult.
  11. EnDevero

    EnDevero Well-Known Member


    I mean, it's nothing crazy, it really does come down to risk-reward. I just like the risk-reward elements to be emphasized in as much of the game as possible. From my point of view, most games fall into the trap of having interesting risk-reward balancing for one section of the game, but leaving other parts of the game to become about simply minimizing risk. I'd rather not have too much of the larger strategy of a game revolve solely around risk aversion. I want the majority of risk-reward interactions to be dynamic and two-way, so you always have to consider the possibility of adding more risk to the game.
  12. pkt-zer0

    pkt-zer0 Active Member

    Now that my brain-juices have been refreshed, there's one issue I noticed: from the system's point of view, intentionally handicapping yourself and poor play are indistinguishable. That means you'd end up with a system that rewards making sub-optimal choices, which seems like it'd be the opposite of accurate skill measurement.
  13. EnDevero

    EnDevero Well-Known Member

    The idea is that it wouldn't actually be a handicap, because there would be sufficient rewards to make it a sane choice. I'm talking about risk-reward structures at different layers of strategy, if that makes sense. So, at some tactical lower level of the game, you might only be making safe survival choices, but at some higher level of strategy you'd be making decisions about whether or not to take higher risks. Games already have these risk-reward structures in them, I'm not trying to pretend they don't exist, I'm just thinking about how far we can take it.

    For example, take some fighting game with difficult combos. That combo system is kind of like a tactical, lower layer of the game. What if you could press a button to multiply the damage of your next combo, but also have some sort of negative effect if you drop it before a certain point? Like, your opponent gets multiplied damage on a follow-up hit or something. And then, to add challenge along with the risk, maybe when you press that button your next combo will also have stricter chaining/linking timing. That's the sort of thing I'm talking about.
  14. pkt-zer0

    pkt-zer0 Active Member

    Execution-based push-your-luck elements have the problem that the reward structures can change drastically just based on how much time you spend in boring 1P training mode, basically. Anyway, seems that's veering pretty off-topic, might make more sense to continue such discussion in the "randomness and skill caps" thread.

    As for on-topic: I noted lack of practical applications as a flaw, so some case studies on symmetry vs asymmetry would be nice. Sirlin's games make for good candidates, since "Game X with asymmetry" has been a design goal with many of his projects: Dominion vs Puzzle Strike, En Garde vs Flash Duel, Chess vs Chess 2.
    link6616 likes this.
  15. EnDevero

    EnDevero Well-Known Member

    You sort of missed the point of my example. Yeah, the focus of the game would shift to getting good at combos, but pretend getting good at combos is the point of this particular fake game. The reward structure would allow you to show off how good your combos are. But, yeah, "execution-based push-your-luck" elements is what I'm suggesting, except the "execution" part would be whatever mechanic is at the core of the game, not an extraneous mechanic like comboing in a fighting game. Off-topic, though, yeah, so conversation over.

    On-topic: Case studies would be real nice, yeah. Not much of a pool to draw from though. Those you listed aren't exactly fair.
  16. Leartes

    Leartes Well-Known Member

    To a certain degree Ido what you describe when I play Go. There is this save move destroying bad aji in my corner, there is the save extension grabbing some territory, then there is the ambitious extension that can easily be attacked and there is the crazy aggressive move starting a fight - if your opponent doesn't chicken out and makes a weak trade. You can go all out aggressive turning the game into a kill-or-get-killed or you can play a smoth counting game. Pros might approach the game a bit differently, but I think you can become very very strong before some style becomes infeasible.
  17. pkt-zer0

    pkt-zer0 Active Member

    Well, those are the designs that are closest to each other, except for asymmetry, that I'm aware of. If you have something in mind where this symmetric/asymmetric difference can be more clearly examined, you could go with that. At least these comparisons make more sense than saying "Street Fighter is asymmetric Chess" or "Go is symmetric StarCraft".

    Also, these games actually exist, so they have practical concerns built into them - unlike some theoretical symmetric game that may not even exist. That's not something to be discounted, since even if the ideal game is symmetrical, in practice, who's to say it's not an asymmetric game that best approximates it?

    You may say the unfairness comes from not comparing like to like, that these asymmetric games expand the symmetric originals such that they're better. What they should instead be compared to is a symmetric game, constructed such that it encompasses the asymmetric one, being a strict improvement. However, even if you suppose that this theoretical game exists, why would you say it could not possibly be improved further? As the original games show, you can add asymmetry to improve a symmetric game, then why not this one? Of course, then you could further continue this infinite chain of symmetric-asymmetric-symmetric improvements, but why would this chain necessarily end with a symmetric game?
  18. EnDevero

    EnDevero Well-Known Member

    I'm not trying to say we should be using theoretical comparisons, I've actually been arguing for a more scientific approach to this. I'm just saying that there are unfortunately not very many case studies to be made right now. Not a lot of data to examine, so I wouldn't know how to go about it.

    That's cool :). I love when games allow for different playstyles throughout the course of a match, and when they let you raise the tension.
  19. keithburgun

    keithburgun Administrator, Lead Designer Staff Member

    I think that by PKT-Zer0's "we can only establish guidelines based on what already exists" suggestion, we necessarily won't be able to advance.

    And yes, there's a huge bias towards asymmetrical games because of the fact that many modern games (which are generally more advanced designs in many ways) happen to be asymmetrical.
    Nachtfischer likes this.
  20. pkt-zer0

    pkt-zer0 Active Member

    "Theories have to fit the already existing evidence at the very least" is more like it. Guidelines that don't apply to the universe we happen to live in are not going to be too useful. Never mind the flaws I just pointed out that exist even if you restrict yourself to the purely theoretical and run with some of your assumptions for no reason.

    Of course, you're welcome to continue avoiding saying anything that'd be practically applicable, if you feel it's more important for your theories to be unsullied by inconvenient realities, than it is to actually make better games.

Share This Page