Video games are as much a science as they are an art; for every subjective opinion there also exists objective fact. Much of video game criticism stems from personal taste, with different individuals liking different games based on their past experiences and their own preferences. As with any art form, beauty ultimately lies in the eye of the beholder, but there are occasions when we can look at a certain aspect of a particular game and universally agree whether it's good or bad. If an advertised mechanic doesn't work the way it was intended, it's both fair and accurate to say that mechanic is broken and hurts the game's overall quality.
Over the years, a lot of mechanics have worked their way into the games we play. A lot of them are welcome innovations for the sake of convenience and have contributed positively to games as a whole. Some mechanics, on the other hand, show up with the best of intentions and ultimately prove disappointing and underwhelming. Some of these mechanics have stuck around and become so prevalent that their presence in games has started to annoy me, while certain other longstanding tropes have really begun to wear their welcome with me. The following are, in my opinion, 16 game mechanics and tropes that need to die.
Disclaimer: most of the things I'm about to list aren't inherently bad, but they're frequently used in a bad way. I don't actually mean that these mechanics should die completely (although there are a few that probably would be better off disappearing from existence), but each of these can definitely be improved and thus require some careful reconsideration from developers. Fixing some of the problems is admittedly easier said than done but these suggestions are at least a step in the right direction, I hope.
Checkpoint save systems
The justification: checkpoints came into existence as a way to make games a little easier, allowing players to restart from somewhere further into a level when they died, rather than having to restart from the very beginning each time. Checkpoints and auto-saves are a welcome feature in games, serving as a backup in the event that you forget to save or die suddenly without warning.
The problem: checkpoint saves become an issue when they're the only way of saving your progress. They often force you to continue playing past the point when you need to stop just so you can reach the next save point -- most people don't want to replay segments of a game, but you don't always have time to push forward to the next checkpoint when real life calls. Despite their intention of cutting down on time spent replaying sections of a game, they still force you to complete the exact same mundane tasks over and over again if you reach a tough encounter and die repeatedly. They also prevent players from reverting back to prior save-states if they make a mistake or if they want to experiment with another approach to a situation.
The solution: if a game must rely on checkpoint saves, don't make each checkpoint overwrite the previous one, and allow the player the very basic courtesy of a "save and quit" option that will let them resume right where they left off if they must quit the game. Quick saves and hard saves are ultimately much more desirable options to allow players the freedom to come and go as they please while letting them experiment with the game on their own terms.
Doors that close and lock behind you
The justification: the idea, I suppose, is to prevent players from backtracking away from the story and getting lost, while pushing them along the intended path and maintaining the narrative pace of the story. They can also be used in things like boss encounters to create a more tense scenario, wherein the player has no escape and must deal with the conflict before him to advance.
The problem: often used in conjunction with checkpoint saves, doors closing behind you can seal you out of game content you had every intention to experience if you accidentally choose the correct path to advance the main game, when you meant to save the main path for later. It's infuriating leaving something that you want to come back to, only to poke your head around a corner and find you can't go back for it. Auto-locking doors prevent creative use of backtracking, such as to use a health kit you passed but didn't need at the time, or if you realize you don't like the weapon you picked up and want to go back for the one you dropped. It's very restricting, and you usually want players to have as much freedom as possible.
The solution: just don't include doors that automatically close and lock themselves behind you. They're fine when used in the proper situation (ie, boss fights and similar dramatic scenarios) but absolutely should not exist in ordinary gameplay and exploration.
The justification: regenerating health allows players to suffer the tension of near-death experiences while not slowing down the overall pace of the game by having them scrounge for health packs between encounters. It's also useful for making sure that players will never find themselves trapped in a situation where they're low on health and unable to move forward in the game, since they can just wait it out and be back on their way.
The problem: regenerating health removes resource management from games, which is an integral mechanic in genres like survival-horror. While still allowing for short-term life-or-death tension, regenerating health removes long-term tension in survival; since you either survive (effectively) without a scratch or die in an encounter, you don't have to worry about being left low on health afterwards and aren't rewarded for efficiency. It also takes the tension out of combat in shooters when you're able to hide behind cover for a few seconds and be back at full fighting strength. Static health encourages players to explore the environment thoroughly in search of healing resources and promotes a stronger sense of risk vs reward, with more player states than simply "alive" or "dead."
The solution: if you're going to incorporate regenerating health, make sure there are still long-term consequences for taking damage. Perhaps only allow partial health regeneration, so that sustained damage progressively lowers your maximum health and thus the maximum amount that can regenerate, with opportunities along the main game path to restore yourself back to full capacity and/or allowing the player limited-use items to remove the cap reduction. Perhaps restrict health regeneration to occur only during calm downtime between encounters, or make it so slow that it can't be exploited.
GPS quest markers/waypoints
The justification: quest markers and waypoints show the player where they have to go so that they're not stuck wandering around aimlessly, thus making tasks more streamlined and efficient.
The problem: quest markers and waypoints take away a lot of the satisfaction that comes from problem-solving and figuring things out for yourself. They often streamline the gameplay to such a degree that you're not expected to use any of your own input in solving a quest or navigating the game's world; instead you just shut your brain off and follow the waypoint. Compounding this is the fact that in some games the worlds are just so large and unmanageable to explore that they absolutely require waypoints, otherwise locations would be virtually impossible for players to find. They emphasize the final destination more than the path to the destination, making their worlds feels transient and fleeting as you get stuck with tunnel vision homing in on the waypoint.
The solution: design worlds that are navigable, with distinct landmarks and layouts, and have characters describe directions and locations to the player. Use a journal system that keeps track of these directions. When a marker is absolutely necessary, put it on a static map that the player has to consult and study, rather than automatically showing him where to go constantly.
The justification: games are often played by a wide variety of audiences, and developers don't want to preclude anyone from enjoying their games by making them too difficult or too easy. Including difficulty settings which make the game easier or harder, or which actively tailor the difficulty to match the player's skill level, allows casual and "hardcore" gamers alike to enjoy the same game.
The problem: games are fundamentally designed around a default difficulty, and more often than not different difficulty settings only adjust statistical sliders -- making enemies have more health and deal more damage while reducing the player's health. If a game feels too easy on normal mode, bumping it up to hard won't necessarily make the game more satisfying, since it'll provide the exact same gameplay experience but just make it take you longer to achieve your goals. These changes feel cheap and superficial. Games that feature level-scaling enemies often reduce the level of challenge one experiences in a game since everything is designed to be appropriate for your level wherever you go, and you therefore feel less satisfaction from getting stronger and leveling up.
The solution: design the core gameplay experience around a certain level of challenge with fixed statistics, and empower the player to choose their own difficulty through their in-game actions such as where they go, how they play, and so on. If you must include difficulty settings, make sure there are gameplay rewards and incentives for playing on higher difficulties, not just achievements/trophies.
The justification: scaled loot ensures that no matter where the player chooses to go in the game, he'll find gear appropriate for his level and be able to feel the constant progression of getting better gear as he levels up.
The problem: like with level-scaling enemies, scaled loot takes a lot of the psychological reward out of RPGs. It doesn't really matter where you go or what you do, since everything will be tailored for you. There's no satisfaction from getting a rare bit of high-level loot or the subsequent anticipation of getting the right stat requirements to use it. There's no reward for meta-gaming in replays, the concept of using prior knowledge of gear to min/max. In some cases you're actually punished for the way you explore game content because you get the "epic" quest rewards at a low level, which stay fixed at the level you acquired them and quickly become obsolete. There's no real depth to such loot systems since the item threshold remains a flat line as you progress; it's difficult to feel any sense of progression when everything is only ever marginally better than what you already have.
The solution: don't scale loot to the player's level. Make items have fixed statistics and place them in the game in such a way as to encourage and allow players to carve their own path, going after easier or greater rewards with proportionate challenge, relative to their level.
Stealth take-down cutscenes
The justification: in stealth games, the take-down cutscene animation provides a more visually interesting appeal, making the gameplay look more cinematic and exciting.
The problem: in practice, it gets to be quite boring and repetitive when all you do in a stealth game is sneak up to the general area near an enemy and press E to watch a three-second cutscene. Whether it's an actual cut-away scene where the camera leaves your perspective or just a super-elaborate attack animation, the bottom line in these situations is that they remove a certain level of skill of and interactivity from the game, since they quite literally take all control away from you, making the action much simpler and therefore less engaging.
The solution: employ a stealth take-down system that requires the player to manually aim at a target's weakpoints, like the back of their head or between their shoulder blades. Such a system requires the player to be positioned more precisely and aim more precisely, making stealth take-downs require more skill and therefore feel more engaging.
Invisible walls and impassible barriers
The justification: video games require boundaries; developers can't render an infinite amount of space, and so need to find ways to confine players to a set area. Invisible walls and impassible barriers allow the player to see the world beyond their own confines, giving the illusion that the world is bigger than the playable area.
The problem: nothing breaks immersion more than attempting to go somewhere you logically and apparently should be able to go, only to encounter an invisible wall preventing you from going that way. Invisible walls discourage exploration and make the world feel much less realistic, exemplifying the shortcomings of video game settings' inability to fully simulate the real world. Impassible barriers are a more logical solution to the problem, providing a contextual reason for why you can't just go anywhere you want, but these sometimes feel just as contrived and unrealistic as invisible walls, especially when your character in unable to hop over a two-foot ledge or climb over some waist-high boxes.
The solution: do not include invisible walls in a video game, anywhere, ever. Physical walls and obstacles are preferred, but even these must be logically placed and designed; if a person could get past your walls and barriers with relative ease in real life, it needs to be rethought.
The justification: audio logs allow developers to tell some of the game's backstory in more or less overt ways, with spoken narrators telling their own story while you continue exploring, instead of having to stop to read a written journal.
The problem: the occasional audio log is fine when it pertains to the main story, but common logs scattered throughout the entire game often feel completely out of place and detached from everything else. It doesn't always make sense when or where audio logs turn up, and it seems really implausible that all these people would be documenting their thoughts and their stories just to leave the tapes idly lying around. It also feels like a really cheap storytelling tactic, since the developers can just have some voice tell you the story without them having to animate cutscenes or design meaningful interaction between the player and the narrators in the audio logs. It doesn't help that it's all too easy to tune out the content of audio logs while you focus on exploring and looting rooms.
The solution: don't rely so heavily on audio logs to tell the story. Use real dialogue with real characters whenever possible, and use the environment to show and imply the effects of the backstory. If you need a disembodied voice talking to the player to get the point across, make sure it's in a believable framework.
The justification: cutscenes create a more cinematic storytelling experience, with the developers controlling the camera angles and the player's focus, and allowing more dramatic scripted action to happen that might not be possible in the regular engine. They make a game more visually exciting.
The problem: cutscenes inherently take control away from the player, making the game feel like less of a game and more like a movie. Video games are supposed to be interactive, but it's difficult to feel a part of the game when you're not in control of your character. It's difficult to feel tension in a situation where your character is on-screen panicking, since you're quite literally invulnerable during the cutscene, forced to sit there and watch whatever events are scripted to happen. Excitement and emotion that occur during a cutscene are merely shown to the player since you don't get to experience the actions yourself. They can also create a dissonance between the player and the on-screen character in the case of games that allow role-playing options but which feature a set voice and personality for the protagonist.
The solution: try to leave the player in control of the character as much as possible; even if there's nothing to do and the player has no choice but to be forced to watch or listen to the events around him, he'll feel more immersed in the setting and be anxious of what to expect up ahead since anything can happen to him while he's in control.
The justification: quick-time events allow developers to use complex cutscenes while still retaining some element of interactivity, blending the exciting action and cinematic control of cutscenes with user input and interactivity.
The problem: the action that happens on-screen rarely ever translates to genuine player control, and the input prompts used in these quick-time events rarely ever match the rest of the game's controls. You're often expected to push buttons seemingly at random with no rhyme or reason, making it feel kind of like you're playing a more visually stimulating version of the electronic game Simon. The "gameplay" in these segments often feel cheap and shallow, since you're not actually controlling what's happening on screen -- you're just pressing buttons to advance a pre-rendered cutscene.
The solution: just don't use quick-time events. If some kind of quick-time event must be used in a game, make sure the control prompts and input translate to the action that happens on screen, like clicking on plausible spaces of the game screen or using movement controls that are consistent in and out of the cutscene.
Errand-boy fetch quests
The justification: I don't even know how developers justify this one. It's a simple quest structure that's easy to design and gives the player chores to do in the game.
The problem: these quests are just that -- chores. Mundane tasks such as delivering a crate of goods across town or venturing to the nearby cave system to retrieve a stolen item for an NPC in town are acceptable enough when you're just starting out an adventure, but they start to feel like tedious wastes of time once you've established yourself within the game's setting. If you're the legendary hero destined to save the world from total destruction, you should have better things to do with your time than pick plants for some guy in town. These quests often involve very little user input to solve, being the most totally unengaging and straightforward types of quests in existence, especially when combined with GPS quest markers and waypoints.
The solution: try to minimize the amount of mundane fetch quests that are included in the game. If the errand-boy fetch quest structure must be used, make sure the quest is developed in a rich and interesting way so that the player has genuine incentive to complete it.
"You need the blue key"
The justification: locked doors are obstacles that give the player concrete goals with small problems to solve, making it feel a little more rewarding to advance through the game since you have to earn the key to advance to the next area.
The problem: locked doors are not inherently bad, but straightforward "lock and key" puzzles that require one specific key to open one specific door aren't very sophisticated objectives. If you're not looking for a key, you're often looking for a generator somewhere, or a control console consisting entirely of one button to press. More often than not they feel like banal, dreary tasks meant simply to sidetrack the player, to make you spend more time on arbitrary side-goals than actually progressing forward. It gets quite enervating constantly being presented with a locked door or a blocked path and being expected to travel to Egypt and back, especially when you notice the repeating pattern. These are the errand-boy fetch quests of FPSs, and they feel just as shallow and straightforward.
The solution: if you must include repeated instances of "lock and key" puzzles, make sure there's some mechanical variety in the solutions and the player's course of action, whether that's using physics to manipulate the environment, solving a more traditional puzzle, or combining inventory items -- don't just make it a matter of "press the action button" in the correct hotspot for each one.
Infinite respawn and backspawning enemies
The justification: infinite respawn can be used to place the player in more dire situations as they face a relentless barrage of enemies, forced to consider long-term survival more with no slack in the tension. Backspawning enemies keep the player on his toes.
The problem: these types of enemy behaviors are infuriating in most instances because they're often used to artificially inflate the difficulty of a given counter. If you're playing a linear corridor crawler and you kill everything in your path, leaving nothing but corpses in your wake, you'd figure your backside should be safe, and so it feels cheap to have enemies spawn behind you just so they can get a few cheapshots on you while your guard is down. Games rarely make it clear from the start of an encounter that you're dealing with infinite respawn, leading to the all-too-frequent limbo phase wondering if they'll continue respawning so long as one is left alive, or if they'll stop after killing a certain number. There's usually no reinforcement for any of these lines of reasoning, so you're often stuck wasting time and resources.
The solution: if you're going to feature a scenario with infinitely spawning enemies, make it clear to the player, somehow, that they're dealing with infinite respawn; don't just leave them guessing. If you want enemies to come from areas that catch the player off-guard, have them spawn from flanks within the player's current zone, not enter from areas he's already cleared.
The justification: people talk to each other in real life, so voicing all of the dialogue in a video game adds to the realism and immersion of that setting, and can even add immersive flair like accents, which are lacking in text-only dialogue.
The problem: having each character deliver fully-voiced lines works alright in games with smaller casts of characters, but in larger games with a much bigger ensemble of active and ambient characters, it actually breaks the immersion hearing the same voices repeated throughout an entire continent. From a development standpoint, having a ton of voice actors inflates the production costs and lengthens the production time. If things change later in development, it can be a problem getting the same voice actors back to record new lines, and having to record all those lines of dialogue puts a restraint on writing dialogue choices and branching conversation paths, ultimately making for a more straightforward and shallow dialogue system. The spoken dialogue is kind of a waste, anyway, when most people will skip through it since they can read the subtitles faster.
The solution: in games with large casts of characters, considering using only partial voice acting -- hire voice actors to voice the main, prominent characters in the story and leave all other conversations to text boxes. Consider having voice actors deliver the first line or two in a dialogue, to introduce the character's voice to the player, and then switch to text-only lines, letting the player internalize the character's voice.
The justification: borne out of the days of pen-and-paper RPGs, random encounters were meant to simulate statistically probable encounters adventurers might experience while spending a few "in-game" days traversing the world map.
The problem: random encounters make sense to me when used in the context of a reduced world map, but otherwise they come and go without reason or context. It's really jarring to be doing one thing and suddenly find yourself transported to another plane of existence to fight some enemies that weren't there before, and it's really frustrating trying to do something as simple as walk across a room and find yourself constantly bombarded with combat. Random encounters feel like a constant impediment to your progress, and in fact the frequency with which they occur often makes them feel more like statistically improbable encounters -- if you're lost and wandering around an elaborate cave system fighting a ton of enemies, you'd think everything would eventually be dead and you'd finally be left alone, but the encounters often persist long past the point when they reasonably should have stopped.
The solution: only use random encounters when using a reduced world map to simulate greater travel. In more local, realized spaces, use on-screen cues like visible enemies (that possibly come from somewhere off-screen, or are already on-screen waiting to aggro or be aggro'd) to trigger the battle plane so that encounters feel more predictable and sensible.