List of Poor Game Design Practices

Discussion in 'Game Design' started by Erenan, May 12, 2015.

  1. Erenan

    Erenan Well-Known Member

    I am planning a project which may or may not ever see the light of day. The project involves taking common game design errors and exaggerating them a lot. But I only need five and I want to pick the five that will best suit my needs, so I'm trying to cast a wide net to find the right ones. Can you help me brainstorm ideas for purported game design errors that I could use for this project?

    Here's my list so far (obviously this list is very Dinofarm influenced):
    • Excessive non-gameplay stuff (e.g. chores, walking down hallways, etc.)
    • Too much inherent complexity
    • Hidden rules (i.e. figuring out what the rules are is part of the intended play experience)
    • CYOA style story choices
    • Excessive output randomness/RPS style guessing/prediction/reading/whatever it is
    • Fighting game asymmetry
    • Endless high score attack model with no win conditions
    I understand some of these are contentious. But right now, I don't want to debate whether these things are actually mistakes or not, so let's please not clutter this thread with such conversations if we can avoid it. Or if debate is necessary, maybe create new thread or link to relevant thread for doing so?

    Also, I searched for a thread like this because I thought there was one already, but I can't find any such thread, so...
     
    alastair, keithburgun and Lemon like this.
  2. vivafringe

    vivafringe Well-Known Member Staff Member

    - Excessive pregame customization offered as a "feature" to players
    - Not patching known flaws in your game post release
    - Extremely long arc games that allow snowballing of resources
    - Single player games with 1 difficulty level, especially survival games (Sorry Empire)
    - Mixing high scores and risk management
    - The word "Facebook"
    - Skinner box tactics (actions with real time cooldowns, randomized packs, etc.)
    - Competitive games with grind barriers to the full game
     
    deluks917, alastair and keithburgun like this.
  3. keithburgun

    keithburgun Administrator, Lead Designer Staff Member

    Yeah, there isn't going to be a thread like this anywhere but here where we believe that game design is a discipline that one can actually do well or poorly.

    This list could of course be HUGE, but if I had to suggest 5, I guess I'd say these:

    - Creating rules to support a theme/fantasy (i.e. almost all videogames)
    - Any pre-game customization at all (FG asymmetry, or worse, MTG customization)
    - No hidden information (Chess, Prismata)
    - Any output randomness or unfair input randomness (Yomi, MTG)
    - Unclear/unenforceable goals (Tetris, Rogue)

    Runner up would be "un-capped execution".
     
    Erenan and alastair like this.
  4. Erenan

    Erenan Well-Known Member

    I meant that I thought there was such a thread here already, but I couldn't find one.
     
  5. ALavaVatChild

    ALavaVatChild Well-Known Member

    Games with a goal of destroying a building (or worse, all buildings): Basically every non-Relic RTS
     
    Erenan likes this.
  6. Kdansky

    Kdansky Well-Known Member

    Depends a bit on how accessible you want it. Keith's suggestions make no sense to anyone who hasn't ever touched the subject of game design. Generally speaking I would go for things that are a bit more obvious to the general public, so my list is this:

    • Too much pregame customisation. Options that break the game, such as infinite lives, or a gun that kills everything in one shot and never misses, but is optional.
    • Balancing problems. One spell does 1 fire damage. The other spell does 90000 ice damage.
    • Idiotic moral choices. You can choose between the red and the blue option. The red option gives you the aforementioned fire spell, the blue one the ice spell. The blue choice also informs you that 10 kittens died.
    • Excessive output randomness. CYOA choice where you can go for the "risky" option, which either instantly kills you or wins the game on the spot. Yay.
    • Too much complexity. Basically copy DOTAs item shop, but only have one relevant stat in the game.
     
    Erenan likes this.
  7. Fenrir

    Fenrir Well-Known Member

    Last edited: May 13, 2015
    keithburgun likes this.
  8. keithburgun

    keithburgun Administrator, Lead Designer Staff Member

    I'm still against difficulty modes, as they're seen in most videogames, wherein you're forced to "choose" which to play on. I think single player games need to scale their difficulty with the player's skill (See Auro's Play Mode). I wanted to do that for Empire but CMS wouldn't budget time for it, so the "quick fix" hack solution was to just scale up the score goals, which is what Empire does. Which I agree is not sufficient!

    But yes, my point of view does change as I learn more (I think this distinguishes me from many others in my field). When I said that games shouldn't have different difficulty settings at all, I was mistaken.
     
  9. deluks917

    deluks917 Well-Known Member

    I strongly disagree on this one: "Fighting game asymmetry"

    I still think asymmetrical games are pretty great and interesting. Though keeping the number of fractions fairly small is probably a good idea. Certainly no more than 10 imo. I think most people find games where both players are "doing different things" much more interesting. This is probably for a good reason. As if the strategies are sufficiently different its more interesting to see how they play out and interact. Of course you can get the same effect with input randomness (random advantages to each player or even random chars etc). Using an "input randomness" system is actually better imo as this forces people to learn more of the system. On the other hand "input random" introduces a huge amount of up-front RNG and idk if people would find this acceptable. In addition if a game is very complicated to play (Guilty Gear for ex) its unrealistic for people to master the whole system for dexterity reasons (This is its own problem). Still I think normal fighting game style asymmetry works pretty well even if its not ideal and should not be considered a bad practice.
     
  10. Evilagram

    Evilagram New Member

    I think it would service this thread if bullet points were given a reason they're poor design practice or in what sense they're poor game design practice, like the one directly below this preface, I'm not sure what it's supposed to apply to, and the next one which isn't applicable in all cases, it obviously applies to more limited cases, for a reason in those specific cases. Sorry to not follow the rule about disagreements.

    -Too much inherent complexity
    I think this is only problematic if the complexity doesn't translate out into depth, or if it is difficult to approach the game as a whole for a new player, or if it makes handling edge case scenarios prohibitively expensive in development.

    -Hidden rules (i.e. figuring out what the rules are is part of the intended play experience)
    Testing behaviors and creating models based on their interactions is a skill as well; One that people enjoy succeeding at in the games that cater to it. It can be poor practice for say, a fighting game, or another multiplayer competitive game, but I don't think it should be ruled out without a reason (nor included without a reason).

    -CYOA style story choices
    This is vague. Is the level select in Castlevania 3, or the decision to partner up with the other characters, CYOA style story choices? Is the level select in Shadow The Hedgehog this? Is the initial dialogue sequence in Deus Ex Human Revolution where Sarif asks you which approach you'd prefer for the mission (and consequently, which gun you'll be taking) CYOA style choices? What's the difference between a story choice and a level select? Between a story choice and a shop menu? What's a story choice at all?

    -Excessive output randomness/RPS style guessing/prediction/reading/whatever it is
    I can agree that I dislike output randomness, though I'll contend that RPS style guessing isn't strictly random. From reading around, I can see everyone is disdainful of Yomi in general, so I won't push the issue. However I will say that if there are no RPS interactions, or if everything is input randomness, then solving games becomes phenomenally simpler, as I'm sure you're aware.

    Having it be "excessive" (undefined here), with few constraints or changing circumstances to either give one player the upper hand, or alter the options both players are utilizing in that scenario, (such as knockdown, hitstun, being close, being far, high ground, low ground, short opening, wide opening, lots of resources, no resources, hanging from the ledge in smash, being denied the railgun/rocket launcher, etc), just pure guessing with no motivating factors on either player's part every time or other strategic elements is repetitive and dull.

    -Fighting game asymmetry
    - Any pre-game customization at all (FG asymmetry, or worse, MTG customization)

    This one is short-sighted. No individual character can be all the other characters. Even if a single character, or army, or what have you, can be diverse in its play styles, it can't encompass all the playstyles of all the other characters. Sure, there's a lot of ways to play Terran, but you'll never zerg rush or reaver drop. Protoss and Zerg will never be able to scan opponents unblocked by their anti-air defenses. If you combined all 3 starcraft armies together, sharing resource pools and buildings, then the resulting army would become overcomplicated, many units and play-styles would be obsoleted by the merger, and entirely different synergies between the units would take effect. In essence, the constituents of the merger wouldn't retain all their individual properties post-merger, the resulting army would play differently from all 3 of the original ones, even if technically it can perform any operation any of the 3 can perform individually. A friend and I discussed the hypothetical merger of all the best traits of every smash bros character, like if one character had the best version of every other character's moves, like the down B shine, high gravity, high land speed, low friction, forward air is falcon's knee, falco's down air, etc. The trouble is, even if this character avoided being overly complex like the prior example, they still wouldn't have the synergy between moves that other characters have, like jigglypuff's sing into rest (arguably nothing has a stronger synergy with sing in the game), Marth's forward air into down air, or his up throw into up tilt and up air, Falco's shine with his down air and high gravity, jigglypuff's weak knockback fast recovery moves with her low gravity, high air speed and multiple jumps.

    This is shortsighted like saying that you shouldn't make different games, just make one game that can be played so differently you can emulate the playstyles of all the others. No individual can be everything. Fighting game style asymmetry usually increases the depth of the game by making it so there are more viable and relevant playstyles, at the cost of invoking imbalance, on either a matchup level or a cast-wide level. Not to mention that Ryu versus Ryu has always been considered one of the most perpetually interesting mirror matches. It's obviously not necessary, but no component of a game is except in context with the other components. Many games don't include the bare minimum of components necessary for them to work (having more than one map in CS isn't necessary, but people play on maps other than de_dust2 sometimes) and different matchups tend to be played very differently, meaning players learn and master different skills depending on how they need to play each matchup, and this isn't simply a matter of cheaply trying out all the things one by one, different playstyles and strategies in of themselves exist in different matchups, leading to say jigglypuff versus peach playing very differently from jigglypuff versus young link, which is very different from young link versus fox (kudos if you can detect the pattern here). Simply put, jigglypuff versus jigglypuff plays differently from jigglypuff versus bowser. One match is based on lots of turtley play, generally trying to whittle down the opponent and go for a timeout, where the other is based on crazy aggression, where bowser attempts to combo jigglypuff into oblivion, while evading jigglypuff's rest attempts that are super effective against bowser's larger hurtboxes. Both of these matchups are unique across the entire game, with no other combination of characters motivating similar behaviors.

    You can't make a single character, single army, single component, that can capture all the different possibility space of asymmetric components.

    - Not patching known flaws in your game post release
    I would like to add that I similarly disdain the patching of things that do not harm the majority of consumers, but provide value for a subset of the consumers, such as exploits, glitches, etc. I'd like to see preservation of those when they actually add something to the game. It's especially disheartening to see this happen in single player games where your experience doesn't affect anyone else's.

    - Single player games with 1 difficulty level, especially survival games (Sorry Empire)
    Disagreed, and disagreed with Keith's statement on scaling difficulty levels. Many players want the difficulty level to stay consistent. For example, I'm endlessly frustrated by the enemy handicap system in DMC4, which makes enemies weaker if you die multiple times on the same level. I have to restart the entire mission just to turn it off, which I suppose is a penalty in its own right.

    - Unclear/unenforceable goals (Tetris, Rogue)
    twitch DOT tv/kevinddr/c/5983832
    "Yo, I just beat Tetris."

    - Creating rules to support a theme/fantasy (i.e. almost all videogames)
    I don't think this is good design or bad design, it's neutral strictly speaking. It depends on the rules you implement and whether they're good or bad. The problem is really that people create rules to support a theme/fantasy without actually considering whether that makes the game better (Iron Sights, randomized bullet spread).

    Idiotic moral choices. You can choose between the red and the blue option. The red option gives you the aforementioned fire spell, the blue one the ice spell. The blue choice also informs you that 10 kittens died.
    Moral choices are just feedback. The more pressing issue is that designers are using points built up across play sessions to lock you into either one scenario or another as a cheap bid to make you replay the entire game to see the bit that is different rather than building a game that is a significantly different experience on replay. The red and blue spells aren't a moral choice, they're just a weak spell and a strong one, until the strong one actually affects something else, then it's a strong spell that consumes a resource.

    Balancing problems. One spell does 1 fire damage. The other spell does 90000 ice damage.
    Could be more accurately phrased, options don't have individual niches, option redundancy, or obsolescence. If two options are identical, but one is better, it obsoletes the substandard option. Adjoining problems are useless or ultra-low utility options, and opposing options or counter options from an adversary that invalidate some options.
     
    Laukku likes this.
  11. vivafringe

    vivafringe Well-Known Member Staff Member

    BTW @Erenan I should have mentioned this before, but I think your ambition to make a thread that just involves making lists but has no discussion is DOOOOOOOOOOMED.
     
  12. Erenan

    Erenan Well-Known Member

    LOL. Yup. No disagreement there.
     
  13. Kdansky

    Kdansky Well-Known Member

    @Evilagram: The things you say are not entirely wrong, but the issue is that we've discussed them at length already, and you're now just catching up.
    e.g. Asymmetry in fighting games: http://keithburgun.net/debunking_asymmetry/ - And then there are at least four threads with fifteen pages each on the topic around, probably more.

    It boils down to this: If you have 10 characters, then you have 100 match-ups (including mirror) total in your game, or 10 for every character you choose. That isn't depth, that's just content in the form of breadth. To put it in different words: If you have pre-game asymmetric choices like character selection, you basically designed an engine with 100 different games in it, because every match-up is just its own game. Some of these matchups are way more interesting than the others, so just get rid of the boring ones. Nobody wants to watch MorriDoom fireballs. If your game needs that to be interesting, then maybe your game isn't good enough. Chess or Go don't even have two "characters", just one.

    Most of us used somewhat unclear phrasing here, because we know each other fairly well, and don't need to spell everything out.
     
  14. Lemon

    Lemon Well-Known Member

    I do think 'fighting game asymmetry', in the form currently implemented, is a bad practice - simply because it breaks the most fundamental condition of a contest: fairness. We all know and agree that it's impossible to perfectly balance all match-ups in a fighting game. However, I think some small tweaks can make things a lot fairer. For example - if a player could switch character every time they lose a stock in a smash-bros game. Then counter-picking becomes a part of the game proper and not the nebulous pre-game 'metagame'. Alternatively, you could change character at any point - provided you can charge up the move to do so. This could even have ripple effects on the problem of camping. If you camp, the other player might switch character to a (somewhat) favorable match-up. And I'm sure there are other ways to fix this problem.
     
  15. serebei251

    serebei251 Well-Known Member

    Very minor nitpick, I think Chess and Go actually have two "characters" because of the first turn advantage and komi.
     
  16. Waterd

    Waterd Well-Known Member

    i think its fair to say that black and white play considerable different, so, I agree wtih you there.
     
  17. Lemon

    Lemon Well-Known Member

    There may be 2 characters, but there's only a single matchup.
     
    Kdansky likes this.
  18. Erenan

    Erenan Well-Known Member

    @Evilagram I probably didn't clearly express why I was making the list. It's perhaps selfish of me to do this, but my primary purpose here was actually not to make a list of things designers ought to avoid. It was really just to brainstorm ideas for things I could intentionally do completely wrong in a project I'm planning. I kind of figured that even if that's selfish it wouldn't matter a lot because a list of things purported to be problems could potentially be useful to others anyway. And I kind of expected that despite what I said about not cluttering the thread with debate, people would debate these things anyway, and now that I think about it, I guess that's fine. It kind of helps with my brainstorming anyway. And I suppose it helps clarify why they are or are not problems. Maybe I should have linked to articles or threads where we discussed these things. Come to think of it, maybe I'll spend some time editing the OP to do that.

    RE: Black and White in Chess: I don't think that counts as two different characters in the "fighting game asymmetry" sense because they both have all the same resources and abilities available to them. That's why Chess 2 has six armies and not twelve.
     
  19. Evilagram

    Evilagram New Member

    I've read the article in question already, my post should make that obvious. I'm aware I'm two years late, but I don't like to leave nonsense unchecked. I think the distinction between breadth and depth here is tenuous. I think it's fairly tenuous in general. What comprises breadth? If a character has a lot of moves, is the character deep or broad? If those moves have a lot of possible outcomes/interactions is that depth or breadth? How is character selection more or less than another option in playing (and winning) the game? Also that's 55 matchups with 10 characters, not 100. 45 of those are the same matchup flipped. Not to mention that each individual matchup in the majority of fighters is deep unto itself, so it's not like fighting games are having a crisis of depth that is offset by how much content there is (like league of legends).

    Every matchup isn't it's own game, otherwise we might well argue that Dark Souls 2 with two daggers is a different game from dark souls 2 with a greatsword and shield, or dark souls 2 with a pyromancy/sorcery build. The characters in your typical fighting game all use the same number of buttons, they have access to meters of similar dimensions that build at similar rates, they all block the same way, they all jump the same way using the same model of gravity, in the same arcs, if you're in king of fighters they can all hop, hyper hop, and super jump. If you're in third strike they can all parry by pressing forward 7 frames before being hit, they all take chip damage from special attacks but not normal attacks, they all treat the same attacks as high and low as they're marked for that particular game. We can say that if you're a guilty gear character, you'll have a double jump, super jump, airdash, backdash, run, faultless defense, instant block, tension meter that allows overdrives at half meter along with roman cancels, and YRCs or FRCs at a quarter meter, instant kill, RISC. meter, 5 normal attack buttons which chain up through each other, typically in a similar order, a dust that launches enemies into space, and a few more things. KoF characters have their 4 jumps, 4 normal buttons, directional throws on C and D with a fast close contextual attack as an alternate, a run, a backdash, a roll that you can throw people out of, a guard cancel roll, and depending on the game, 3 or 5 stocks of meter and an extra meter that can cancel special attacks into each other. Yomi characters will have one attack/block/dodge/throw for every card in a poker deck, they'll all have a speed rating and a combo point count. Divekick characters all use only 2 buttons, pressing both will do a special, one jumps, one kicks, they can kickback on the ground. Characters feel like they're parts of a set, in the same game, which people get an affinity for as a whole, which doesn't carry over into a game with different identifying characteristics for their characters. Not to mention, if you have one matchup and only switch out one character, it's kind of silly to say you're playing an entirely different game, though I guess that's a bit of a Ship of Theseus type of thing.

    As for Morridoom, isn't that more that the character has an issue, not the matchup? I don't think the players would necessarily say that playing that way is lame, even if spectators consider it that way, and regardless, few people play Morridoom as successfully as Chris G in the first place.
    youtube watch?v=4et5xOKj9tQ

    It's absurd to say off the bat that asymmetry is automatically a cardinal sin, even games like TF2 which have one matchup have imbalance depending on the map. A ton of games that don't even have varying maps have imbalance. It's double absurd to assume that either fighting games need it to be interesting (and therefore are using asymmetry to compensate for depth the matchups themselves lack, which if you've played street fighter or any other fighting game to any significant extend you can clearly see they don't) or that if the matches themselves are deep that they should remove the asymmetric components, only allowing mirror matches. Many board games with a single matchup are less deep than even the least deep of street fighter matchups, and no more balanced. You can't make a rule that will prevent boring matchups, only make sure the characters don't have tendencies that will lead to matches being boring.

    Making a game with Asymmetrical Components that are selected at the start of a match and remain in place for the duration of the match is a style of design with benefits and drawbacks, the benefits obviously being that there are more distinct situations to play out, that more disparate playstyles can be expressed mechanically in the same overall design space, the drawbacks being a higher commit to player memory up front, and difficulties in balancing. It would be smarter with rules like this to acknowledge there are advantages and disadvantages to many of them, which we can draw upon when we feel it fits what we're designing, instead of disregarding them out of hand like guns with randomly spread bullets (or other such high output randomness things with no means of compensation), which in my opinion there's absolutely no redeeming factors to.

    Oh, and I'd personally like clarifications on some of the unclear phrasing, like the CYOA style choices one, which only really makes total sense in CYOA games (because they suck). In other games you could describe them as choices with poor feedback, or which don't affect much. I think the moral choices bulletpoint is misplaced completely, because moral choices are just choices.

    That feature is in Project M (mod of Brawl, google it), called All-Star Versus mode. Or King of Fighters (google that too). Also Marvel and Skullgirls (also google that). Also, counterpicking is part of the game proper. The metagame is just the body of information outside the game that affects player choices. If you pick the wrong character, that's as much a wrong decision as choosing to shield instead of dodge, or take the ledge instead of centerstage versus Marth, or try to outcamp Samus on the far side of a stage with low platforms she can missile cancel off of, like pokemon stadium, or allowing falco to start up his laser spam. The other trouble is, if you give people chances to change their character, mid-match then you change the nature of the game, much as with merging the 3 starcraft armies as mentioned previously. If you want to cut down on camping, then have the camper have to pay something in order to camp, such as negative penalty in guilty gear, the degrading shields in smash, or the limited ledge invincibility grabs in project M, so they can only keep it up for so long.

    You can get pretty damn close to balancing them all, and some games certainly do a better process of that than others (like Project M, Guilty Gear, Yomi, and Skullgirls). Otherwise characters naturally fall out of use, as in the case of Smash Bros, Third Strike, and most other games, and people compete over a more limited, and balanced, selection. It's a regret that sections of the game that someone worked to design fall into disuse, though I'd argue it's not a problem as long as what remains is deep/interesting.

    The division between the game and pre-game, or game and metagame, how much of the game counts as the game, seems like a matter of perspective/framing here.

    And I'm sorry for blatantly ignoring the rules you were trying to set and digging up a 2 year old topic. I'll contribute don'ts next time I visit. Busy at the moment.

    • If there is a way to bypass a challenge to progress, make bypassing it as difficult or more difficult than taking the challenge head on.
    • Don't strictly outlaw unintended behaviors, find ways to twist them into interesting tradeoffs or challenges, or consider keeping them if they are that already.
    • Don't introduce user experience issues in order to service your story, setting, or theme. (unskippable cutscenes are my persistent favorite, followed by save systems that don't allow you to suspend play on the dot)
    • Don't make an ingame timer that records load times, cutscenes, or end credits. (the Souls series (and bloodborne) has excellent timers with the exception of Dark Souls 2, which records longer times than you actually play for in realtime)
    • Don't make "interactive cutscenes" or scripted sequences in engine that play out in realtime that you can't skip or bypass.
     
    Last edited: May 15, 2015
  20. vivafringe

    vivafringe Well-Known Member Staff Member

    To set the record straight, I don't think every game should have dynamic difficulty adjustment. What you're describing is... a pretty common negative side effect of DDA, and it's one that's pretty challenging to try to address. I think DDA is doable in theory, but hard to balance correctly. Right now Auro has annoying "ping pong" issues where a player has to "grind" through one of the difficulties to play on the difficulty level he prefers.

    I *do* think it's an absolute requirement to have multiple difficulty levels, unless your game is some sort of content driven thing that's meant to be played once and then discarded. The basic issue is that, even ignoring differing skill levels across your target audience, my ideal difficulty for a game when I start playing pretty much *can't* be the same as after I have played for 2 hours.

    I mean, we've all played some survival game where the game plays fine at first, but then you learn more and more of it, and after a while you're grinding through tons of tedious gameplay just to get to the part where the game can actually threaten to kill you.

    The best model for difficulty that I've seen so far has been Super Hexagon's: you play until you do something that shows utter mastery of the difficulty level (in Super Hexagon's case, survive for 1 minute), and then you've "won" and can advance to the next level.
     
    Last edited: May 15, 2015

Share This Page