How To Think Like A Game Designer

Video Statistics and Information

Video
Captions Word Cloud
Reddit Comments
Captions
Okay, so during the development of Alien:   Isolation, the game had a  pretty basic save system. It would automatically cache your data  whenever you tripped invisible checkpoints,   which were dotted throughout each level. It was added because it was easy to  implement, simple to understand - and,   in truth, because that's how saving  worked in pretty much every game. Design director Gary Napper  said "like any development team,   we all play a lot of games  and each have our favourites. Often our decisions and choices  are coloured by the games we play". But... as development went along,  the team started to question   if this save system was actually the right choice. You see, because players knew that if they died  they'd only lose a couple minutes of progress,   they would just waltz around the space  station without a care in the world. Not ideal... in a horror game. *Alien scream* So while these checkpoints might  have worked in Call of Duty and   Bioshock - Napper says it was "not  the right approach when making a   game that is designed from its core  to terrify and put people on edge". Because here's the thing. It's totally fine to take  influence from other games. Many designers borrow mechanics from  their favourite titles - often evolving,   combining, and remixing them into something new. But: it’s essential to understand  why those mechanics work in one game,   before copying and pasting them into yours. How do you do that? Well, I'm Mark  Brown, and this is Game Maker's Toolkit. Alright, so... way way back in 2004, , three game designers  got together and wrote an academic paper,   featuring a framework that would prove to be  perfect for analysing the mechanics in games. It's called MDA, and it  breaks games down into three,   neat sections: mechanics,  dynamics, and aesthetics. Mechanics describes the actual workings  of the game: like the rules, the systems,   what the buttons do, and all of  the individual stats and numbers. For example, one mechanic could be the  maximum amount of ammo the player can   hold - and we could set that number really high. Then, dynamics is how the player acts and  behaves, in response to those mechanics. So because the player has tonnes of  ammo, they may storm into battle,   and shoot wildly in the general  direction of every enemy they spot. And, finally, aesthetics - not to be  confused with the graphics or the art style. Here, aesthetics is how the player  feels when acting in that way. Their emotional response. So this wild shooting could  make you feel reckless,   powerful, unstoppable - basically, like this. *Machine gun fire* To put it another way - mechanics happen in the  code, dynamics happen in the player's actions,   and aesthetics happen in the player's feelings. Now, unfortunately, game designers can't just go inside the player's head and directly alter their actions or feelings. But they can change the code - and because MDA describes a casual   relationship, these changes will cascade  down and alter the dynamics and aesthetics. For example, if we tweak that knob  the other way, and massively limit   the player's ammo stash, that's definitely  going to change the player's behaviour. Now, they will act cautiously before heading  into battle, or might not engage enemies at all. Every shot fired will be done  with careful consideration,   and more time will be spent hunting  for ammo in the environment. And this will lead to  completely different aesthetics:   like disempowerment, fear, and caution. So, this MDA framework takes mechanics -  and then puts them into a wider context,   which asks 'how do they make the player  act, and how do they make the player feel?' And we can use these questions to analyse  the mechanics in the games we play. Like, we could ask 'why do swords  break in Zelda: Breath of the Wild?'. Well - how does that change player behaviour?  Maybe players will rely less on direct attacks,   and so spend more time sneaking past enemies,  or finding creative ways to ambush foes. It will certainly encourage players   to use different weapons - and  constantly seek out new ones. It can also lead to exciting moments of  drama in the middle of a fight - (so far   I've used dynamics to refer to player  actions, but dynamics can also be used   to describe consequences that might  bubble out of the game's systems). So how does that feel? Well, it  could make you feel underpowered,   it could make you feel crafty and creative, and  it will likely make you feel like an explorer... in a world that's decaying, and falling apart. If there's one hard part of this analysis process,   it can be putting words to the subconscious  feelings that arise when playing games. We're looking for way more than just "this  game was fun", - we're looking for strong,   emotive feelings like, this game  made me feel powerful, creative,   sneaky, tense, intimidated, curious,  deceitful, cooperative, flustered. So MDA lets us see game mechanics as powerful  vessels for delivering emotions to players. And so when borrowing mechanics from  other games - or making entirely   new ones - you can pick ones with  associated dynamics and aesthetics   that will compliment the rest of your  design - and avoid ones that will clash. For instance, designer Jenova Chen says that  the game Flower once had a level-up system,   spells you could cast, resource management,   and time limitations - expected  features of other video games. But they all had to go,   because they went against the intended  emotions of relaxation, calm, and peace. Chen says "we’ve played so many games  growing up these bad habits form. A lot of the time we like to make things very fun,   but fun doesn’t always help the  emotion you want to deliver". That's why it's useful to truly understand your   game's vision - the overarching feeling  or experience you want to give the player. Because if you know you want to make the player  feel, say, scared, then you know you'll want   to pick mechanics that induce feelings of  fear, dread, isolation, and disempowerment. That vision could be a single  statement: Subnautica was based   around the phrase "thrill of the unknown",  and every mechanic had to suit this idea. Resident Evil Village was built under the  banner of "struggle to survive" - and so   when playtesters hated the game's messy combat,  the designers could use that phrase as, like,   a lodestar to rebalance and  rethink the game's mechanics. The vision could also be a fantasy  that you're trying to deliver. Perhaps you want the player to feel like  Batman, or an assassin, or a world leader. It could be a feeling - like how  Flower tried to make you feel relaxed. Or it could be a specific experience  - when making FTL, the devs said their   starting point was "we wanted to recreate  the atmosphere of commanding a starship". With a strong vision in mind,  it becomes easier to evaluate   whether game mechanics are the right  choice for the game you're making. And, of course, it goes beyond game mechanics. All the other elements in the game, like  visual style, music, animation, story,   colour, camera framing,  and so on - those all create aesthetics too. (Just, without the dynamics bit). For example, Dead Space composer Jason  Graves said that EA originally requested   a predictable sci-fi soundtrack for the  game - full of electronics and drums. "After we did that first little  piece of gameplay," says Graves,   "EA came back and said that  it wasn’t scary enough. The music made players feel heroic,  and they wanted them [to feel] scared." And ultimately I think games are most clear and   coherent when absolutely everything  is pointing in the same direction. Jenova Chen says "all the elements have to  sing the same notes to make the impact strong". Perhaps the best example of this  - for me - is DOOM, from 2016. It also had a strong vision statement  - in this case, "push forward combat". And pretty much everything in  the game supports this statement. You can see it in the game mechanics  - from the rapid movement speed,   to the way you pounce on enemies to get health,  to the way certain demons are told to run away   from you - every feature pushes you to race  towards enemies like an unstoppable predator. And that's supported by the non-game  elements, like the heavy metal soundtrack,   the violent animations, and the Doom  Slayer's visual design and personality. Of course, none of this as easy as  I'm making it sound - and there are   so many other considerations  to make when designing a game. For starters, a game isn't just one mechanic - but   hundreds, perhaps thousands if  you want to go super granular. My interactive video essay, Platformer Toolkit,   shows how a single character can be defined by  dozens of distinct stats - which could change   the game from feeling like Inside,  to feeling like Super Meat Boy. And these mechanics interact, overlap,  and can even undermine each other. In The Callisto Protocol,  you have very limited ammo,   which is intended to make the player  feel underpowered and fearful. Buuut, you also have a one-button,  insta-kill stealth attack with a   slick animation - which can make  you feel powerful and unstoppable. Also, while you can make best guesses about how  a player will act in response to a mechanic,   you can't know for sure until  you do a lot of play-testing. Players might misunderstand or ignore a mechanic  - or do something completely unexpected. Sometimes that's a fun, emergent behaviour  - like the flying cars in Rocket League,   which suited the vision just fine,  and so could be kept in the game. But at other times it can create degenerate  strategies that undermine your emotional goals. Going back to Alien: Isolation, the  automatic checkpoints meant that if   you got jumped by the Xenomorph, a  smart strategy was to just beeline   it towards to the next checkpoint,  get killed, and respawn in safety. Furthermore, sometimes you want to have different  emotions, at different times in the game. And so will need to change  the mechanics accordingly. If the story depicts a character who  starts out weak and naive - but ends the   game feeling competent and powerful, how could  you reflect that growth in the game mechanics? And finally - and perhaps most importantly  - aesthetics are, ultimately, subjective. A scoring system might make one player feel  competitive - and eager to replay every level. But it could make another player feel  judged - and eager to get a Steam refund. Time pressure might feel fun and exhilarating  to one person - but anxiety-inducing to another. And this is often related to skill and ability. Take the new Tango game, Hi-Fi Rush:  it clearly wants you to feel like a   rockstar as you brawl along to the beat -  but as someone with zero musical ability,   I just felt like a bumbling idiot. At the end, I’ll recommend a video about  balancing a fantasy with differing player skills. So, going back to Alien: Isolation - the team  eventually tried a different save system. Something closer to older games. Now the game features a small number of save  points that you have to operate manually. And - to make things worse - it takes  a while for the save point to turn on,   and for your game to actually save... giving the alien a chance  to sneak up and kill you. And this new mechanic changed  everything - completely altering   the player's behaviours and emotions. "We were afraid," says Napper. "If we didn’t make it to the save point and  successfully save, we would lose our progress. Saving became tense. Looking for a save became tense. The simple act of   saving had become supportive to the game's  driving factors of terror and isolation". So - it's fine to borrow game mechanics -  but you need to understand why they work. MDA is a powerful tool for this because  it helps us predict how a game mechanic   will make players act, and make players feel. If those feelings compliment the  vision for your game - then awesome. If they clash, then they probably need to go. Because, when it comes to picking, pinching, or  producing game mechanics - you should listen to   Shovel Knight studio Yacht Club Games, who says  "it depends on the game you are trying to create,   the emotions you’re trying to evoke, and the  experience you want your players to have". Hey, thanks so much for watching! This  video was originally given as a lecture   at universities in London,  Breda, Boden, and Skellefteå. If you'd like me to visit your school,   please get in touch - you can find  my email in YouTube's about section. I'm also going to GDC next month, so if you  see me - say hi! For now, though, click here   to watch that video about delivering a specific  fantasy, to players of different abilities.
Info
Channel: Game Maker's Toolkit
Views: 665,951
Rating: undefined out of 5
Keywords:
Id: iIOIT3dCy5w
Channel Id: undefined
Length: 13min 6sec (786 seconds)
Published: Fri Feb 17 2023
Related Videos
Note
Please note that this website is currently a work in progress! Lots of interesting data and statistics to come.