Deciding what to learn about next? Then why not visit brilliant.org/ExtraCredits/ to get 20% off the annual Premium Subscription and level up your knowledge. All while helping support the show in the process!
in the near future id like to see a series based on me my rule power wealth might and my past life in my past life of me AKA MRM AKA PREZ MRM THE GREAT I
Could you do a video exploring how/if players playing the same game or series over and over can cause diminishing enjoyment out of that game and lower the popularity of that game over time? When a game becomes unpopular it's usually balance that the community blames for it but not this. I think there's a reason people don't play one game actively for their whole lives without stopping or taking breaks. I think the Megaman/Megaman X games and World of Warcraft would be good games to look at, being an example of a single player game and a multi-player game that I think suffered from player fatigue.
A super important part of this is how a bunch of branches often merge back to a single point before splitting up again. Dynasty Tactics had this same design and that happened a bunch there.
That is honestly SUPER important to keep the scope of the game reasonable. If the game fundamentally changes after almost every single decision and doesn't link back up...It starts getting exponentially larger while not getting any closer to the resolution.
@@AegixDrakan But it can still be done, you realize that right? Especially with these big budget companies? You know not all paths have to have vastly different lengths right?
@@ramenbomberdeluxe4958 Man, I'd love to see more (big title) games give you a hard "Game Over" 5-10 minutes after making a dumb decision. Like, one where you _know_ it's dumb, but you do it anyway to see what happens and how it's gonna bite you in the arse.
Yeah, as a Narrative Designer, sometimes these decision trees can get terrifyingly huge. There's a debate at the end of a questline in a game I wrote for that was getting SO big that by the end, that just moving nodes around literally made Unity lag a little. There were so many bits of dialogue, decision splits, little bits of previous decisions to take into account, etc etc, that when I was done, I said "There's definitely not going to be one like this in the next expansion, thank you"
This video didn't really talk about the art and science of decision trees, it was mostly just about how many Mass Effect has, and not even very much about how different those are.
A key concept for taming decision trees is the idea of "orthogonality" - the idea that the effects of different decision points don't need to interact - a crude example would be if you have one character ask you what your favourite food is - lasagna, pizza, or burgers - and another character ask what your favourite flavour of ice cream is - strawberry, chocolate, or vanilla - then when a third character takes you on a date that they've secretly been gathering intel for, the whole date could be done as a single scene with 9 possible variations - lasagna followed by strawberry ice cream, lasagna and chocolate ice cream, lasagna and vanilla ice cream, pizza and strawberry ice cream, pizza and chocolate, etc. Or, because the main and dessert don't actually affect each other, the date could be broken into two scenes, each of which has three variations - lasagna, pizza or burgers followed by strawberry, chocolate or vanilla ice cream - which is 6 shorter scenes rather than 9 longer scenes, so a bit more than a third of the work of making each combination separately. Of course, breaking things down that way can weaken the illusion - each of the three mains has to fit with any of the three ice creams, and vice versa, and if people replay, there's only at most a third as much variation available. But by adding possibilities rather than multiplying, it limits the total amount of work required.
This is usually how I analyze writing quality for decision trees (in games or choose-your-adventure books). If the path through the decision tree feels natural even after combining with other branches, then it was well written. I've seen games where the merge point was written to be too vague, so it didn't feel like choices mattered at all. Or worse, I once read a published "choose your own adventure" book where the merge point clearly assumed you had taken a specific branch to get there. If you took the other path, it didn't make much sense. A common solution I've seen in games is to use "mini-braches" that are determined by a previous choice. These "mini-branches" typically just contain a single snippet of conversation (or even just one word) so they can acknowledge the previous choice without re-writing the entire scene for each choice. This can save a lot of money by reducing the spoken lines to be recorded, the actions to be animated, and the assets to be stored.
"there's only at most a third as much variation available." Dunno how you got to that. You still have 9 possible variations for the dinner, no matter how the scene is created.
@@rolfs2165 It's a matter of writing 6 scenes parts instead of 9 full scenes. If you write a full scene for each possibility, you basically have to write some choice results twice. If you just write each result once then combine them you save a bit of time. And saving bits of time everywhere drastically saves time in the long run :)
@@rolfs2165 let's say both main and dessert take 1 to build if you make the meal as a single scene you'll need 9x2, but if you make the single options as scenes and then play them in sequence you only need 6
I dunno if you guys take suggestions, but I was wondering about the game design in Factorio, and how it works in a way that causes minimal lag whilst having sooooo many entities and moving parts(literally and figuratively)
Optimization is a pretty deep rabbit hole. I think it's obvious that factario doesn't use super complex graphics on a 2D plane, but to get into the meat of it, one would need to talk to the devs. I think if they could get the devs, it could work. What some of it will come down to is just paradigm, where they bother and care to optimize, and may have from the start.
Factorio developers actually kept a weekly blog where they talked a lot about the development of the game. They called it Factorio Friday Facts, and you can learn a lot of interesting information from them.
I just replayed Dragon Age 2 and was really impressed when two characters talked about how my character was in a relationship with Merrill but had also had a less serious relationship with Isabela before that. The game had a recorded dialogue specifically with those characters, about having a relationship with one specific character while before that having a relationship with different specific character. That just makes me wonder how many dialogues the games has for all other possible scenarios. xD
Lots! Some people don't give that game enough credit because the graphics were dated for when it released and it didn't have quite enough gameplay varation, BUT the writing is fantastic! The interpersonal interactions are really where it shines. I've played it through a few times and it takes a lot of hours before you run out of smatterings of new dialogue, depending on your choices. Same for inquisition, which I've logged even more hours on (like, hundreds because in THAT game the gameplay also slaps! XD)
@@purpleghost106 I have a couple hundred hours in DA:I, too. Because I did a single completionist run (including DLC). Racking up hours in that game is not hard, not getting bored and burnt out while doing the same stuff over and over again in the forest area, then the desert area, then the seashore area, … _that_ is the hard part.
@@purpleghost106 If you manage your expectations, I'd actually argue DA2 is the strongest written one of the bunch. On the whole it is my favorite. The whole rival/friend dichotomy appealed to me a lot more than the general disposition in the other ones. Having a rivalry romance with Anders gave me one of my favorite moments in the game (YMMV, I personally like emotional gutpunches that connect, because it also meant I really started to care about the characters) almost talking Anders out of something big. Also: This may be my particular weirdness, but I appreciate the attempt to make the elves look really distinct and less 'human with pointy ears'. It didn't always work, but I prefer my elves somewhat alien.
My favorite example of decision trees in the Mass Effect series is Port Hanshan in the first game, where you need to obtain a pass in order to leave the port and continue to the next area. The game guides you naturally along one path, but there are actually five main ways to do it plus room for variation within each path. These paths are... 1) Uncover evidence of the director's corruption and give it to an investigator (which is the path most people take) 2) Give the evidence to the director so he can cover it up 3) Give the evidence to a businessman so he can blackmail the director 4) Tip off the director to the identity of the investigator, at which point they kill each other and you pick the pass off the director's corpse 5) Completely skip all of this and just rat out a smuggler who was part of a random side quest which the game in no way indicates is related to your search for the pass.
I think I ended up helping out Lorik Qui'in? I think that's the businessman one you're referring to. I may done both that and helped out the investigator; hard to remember.
Yes, Gianna asks you to convince Lorik to tesitify which leads to Anoleis being arrested but you can just give Lorik the evidence and he'll give you a pass. Fun fact, if you do this then in ME3 you'll get an email from Lorik where it turns out he took over as director after ousting Anoleis.
I kinda feel sad. There's a conversation, a very interesting and important conversation, to be had about actual dialogue trees. A simplified overview of how they're programmed, how games might note them with triggers. The effect of recalling certain decisions to players and how that betters the experience. The fact we can't actually program every computational outcome. I feel like that conversation is the something, and I hate to say it, that older Extra Credits would've looked at. And the reason I grumble about this is that there have been a lot of episodes of Extra Credits about dialogue design, and this video does nothing new. In fact, Choices vs Consequences discussed how numbers of choices didn't matter. And the actual decision trees game developers map out look plenty different from the visual used for this episode. The episode The Illusion of Choice is all about the practicality behind the complex part of making choices. The video essentially says "decision trees determine what can and can't happen" and that's it. No difficulty in programming, no 'you should remember this not this', or even talk of how games might change fundamentally or in small ways. Do better.
They don't have an actual game designer working for them like they did originally, so they have to rely on people like a "Star Trek consultant". Or on the rare occasion they do have a real game designer to write an episode, it ends up being an ad for that person's game.
Agreed, most of the newer Extra Credits episodes are rather superficial, especially compared with the older ones. There's so much potential to dive deeper into most of the topics they talk about, because they have done more general overview video's of many topics already. Instead, they just rehash and even simplify their older videos. There's still so much to learn about game design, but I don't get the feeling I will find that here anymore.
I had absolutely no interest in playing mass effect, regardless of the accolades, until it was described to me as a dating sim disguised as an action RPG. Gimme that space beefcake!
@@edoardoprevelato6577 OH, come on, you can't just assume just because they're alien muscles, they're not muscles at all! ... Ignore the identical model composition.
@@saxor96 i mean, i suppose Turian biology, being dextro based, and then avian evolved, is so absurdly different that Garrus could be the birds' version of the Incredible Hulk
the fact that there are trillions of branches does not necessarily make it the case that every mass effect play through is likely to be unique, because not all choices are going to be equally likely. To compare the total number of theoretically possible playthroughs to the number of playthroughs that have actually happened, you would have to assume that any of those theoretically possible playthroughs is equally likely, and that would require not only that each option be equally likely to be chosen, but for your decisions to be independent, that is the choice you make in one place having no effect on any other choices you make down the line; both of those assumptions are fairly obviously not the case here. For example, it's fairly likely that there are quite a few people who played "pure paragon" or "pure renegade" runs, or as close as they could come to it, and all of those runs may well end up identical.
Not a Mass Effect example, but I loved how the original Gold Box games back in the 80's and 90's let you export and import characters between games. While this didn't have any story implications, as the plot and events were different game to game, this *did* allow keeping your spellbooks and gave a significant boost to character progression if you maxed out your levels in the previous game - I like to think of it as branching spell selection. Curse of the Azure Bonds' character creator started at level 4, but you could have level 6 characters with over 2 levels worth of experience from Pool of Radiance imported in. You could train them to level 7 (capping your experience to just shy of 8th level), get in a bar fight for a handful of xp per character, then train them again to level 8. You'd be short on equipment to start since you spent a chunk of the money you were found with on training, but this was before the days of level scaling in games, so the increase in power from being higher level was worth it.
It's really weird seeing youtube links in our school homework. We're going through some game develpment, and our book literally cites your videos. Thanks so much for making game design more approachable!
What's crazy to me is when you actually look up how many variables there are in Bioware games, it's often way more than you think or notice when just playing the game. They make sure so many different options and variables work. And yet, many of them are never even seen or noticed by most players because most players play a certain way.
I prefer to think of branching narrative as a branching river rather than a branching tree. You don't often see a branch bending back into a tree or into another branch, which is what happens when plots converge. If you ever take a look at a branching dialogue tree you can always see every path reconverging at different points so things don't spiral out of control. Take the Outer Worlds or Divinity 2. If you fail or succeed a skill check you might get one line of unique dialogue relating to what you just said or did but then they all reconverge on a single node for if you are progressing a failed path or another node if you're progressing the successful path.
“Think of it like a series of ‘if then’ statments.” As a programmer, this is probably how they’re implemented if a specialized plugin isn’t being used…
Really it's more of a rhizome than a tree, except if you want to never finish your game. You rarely get a whole narrative arc sprouting to never join up with a main branch. In the end it's much less different outcomes than just small details changing to acknowledge player's choices. The problem with BioWare is them always advertising a diversity of outcomes when it's rather different flavours of the same set-in-stone story beats, and that's why they didn't see a problem in wrapping the entire ME trilogy around a central, final path, whereas the players were expecting much more. They make up for it by making high-spectacle action-RPGs with dating sim elements for mass gamer appeal. Compare and contrast with Encased, Tyranny, Underrail... smaller budget, smaller scope, but better handled choices-and-consequences.
this is why i have such a hard time finishing my interactive fiction games - get too caught up making new branches rather than finishing the story! gotta work on keeping myself on track lol
While it was interesting to learn about the concept, I'm more interested in learning about the nitty gritty of its implementation in games. Any hope of you guys doing an episode on that?
Actualy is not only prety hard to make games with lots of meaniful branches is also a risk. You can create one good history for your game but even if you spend the time to make several some of them will probably be better than others which means you have players having a not so great experience with your game. That is why most decisions in games are meaningless making a litle dialog change here and there but with few changing the plot. More significant changes use to be at the ending as well. This solves another problem of this type of games that is having to replay it a lot of times if you want to see all content. So designers use to fake branches
@@massimookissed1023 Yes with few branches, but if there are like a lot of branches you endup replayng a lot of parts that don't generate fun. Like havin to re-watch half of a movie 10 times
I bet there is a lot more overlap in play throughs considering the paragon renegade progression system encourages following one or the other exclusively. Something like the Witcher 3 probably has more variety in the decision tree between plays since it doesn't lable its decisions with a specific playstyle.
For Dragon Age Inquisition, you can go to the Dragon Age Keep website and make different World States to upload to Inquisition (rather than having to have the save files all in one place). Makes it a lot easier to see how different decisions in earlier games affect things and really neat to play with in general!
The image about the three ends is just amazingly descriptive on why nobody likes them. But considering having billions of possibilities makes it impossible to finish it nicely for everyone.
For a while i found the bioware romances cute and nice to have actual choices with my character but ... what bothered me was that i was never in a position where the other side did the first steep. And usually every other steep too. It was always my character who had to proclaim interest and try to who the love interest never them trying to who for my characters attention. I think it was in the pathfinder game where i finally had an npc who did initiate the first steep and did occasionally do some wooing of their own but not certain ^^ Still a good healthy written romance should show interest from both sides and not have just only one pursuer, however the vast majority of rpg romance plots tend to be rather one sided. (I am not talking about VN here because well ... they are all over the place ;) and have a more book feeling then RPGs do)
Playing backwards through the Bioware catalogue, I've found that in the older games it actually is your companions who will initiate to some degree. The game starts you off in an 'active' romance state, your companion will flirt with you, and you get to decide whether or not to reciprocate, at which point the romance is continued or concluded. Being used to the newer system where you have to choose flirt dialogue options to start a romance, the older games had me googling which stealth romance option I had chosen to have all these dudes declaring their undying affection lol
Yo, EC, can you make a video on how devs start developing a game? Is it from an interesting mechanic they were playing around, a story they want to flesh out in a game world, something else?
My favorite mass effect decision in the whole fucking series!!!! I never make it, and only discovered it watching my brothers playthrough, but my god is it funny! And I think they even make references to it in each game with a similar choice.
I found out that my journey with Commander Shepard was unique when I discovered that Wrex didn't have to die on Virmire. And for that sin, Ashley was killed.
"Best Quarian pick-up lines," when the animation was FemShepxGarrus. I'm not sure if this means the writer was BroShepxTali and the animator was FemShepxGarrus, or a subtle reference to the fact that GarrusxTali is canon if Shep doesn't take one of them. And yes, I'm assuming it was intentional and not a screw up, because I like to think positively once in a while. Edit: Watching for several minutes makes me think it's the latter.
I realized my journey was unique when my girlfriend Tali killed herself and I only had four members in my team at the end of Mass Effect 3 = (. Lonliest experience ever
The game gets so much better when you don't just try to make a Paragon or Renegade run. Also I like the ending. I feel like it is a grandiose ending for a grandiose epic. Go on! You can put my spaceship in lockdown now!
I could sit and listen to people gush about Mass Effect all day. Best trilogy ever. I've seen 10 different journeys. Looks like I have a few hundred billion to go.
So while there are a ton of potential scenes and things to do, that's just looking at the potential routes, which is a feat of design for Mass Effect, no doubt, but how many of those scenes are reused? Game development, especially something like mass effect, is expensive, and game designers like shortcuts and reusing assets where they can, especially for a game made for a console, so file size of the end product especially matters, don't want to make something bigger than a disc can burn to
Mass Effect: Nearly a full Quadrillion branching paths for the player to navigate full of intricacies, romance, adventure, and tradgedy In Space With Markiplier: *FIX IT FROM OUTSIDE*
Kinda missing out the fact that most of the decision tree branches will rejoin. and net you the same results. You may have 59 branches bout 20 of them point to them hating you, 10 have them remain neutral, and the rest have them biased towards you. The key point is to simply remember the last significant node.
All those variations come down to a binary choice at the end that every player gets. Mass Effect isn't even close to being the best example of Decision Trees.
Dragon Age, also done by Bioware, also imports in saves from prior games. I believe even Witcher and other games that share the same protagonist over multiple games do the same.
It wouldn't be too hard, I'd guess. It checks for a save file on the drive, scans for the relevant nodes on the decision tree, then adjusts a 'default' tree based on which nodes it reads from the data.
It's actually stupidly easy. Built into the new game there is a "reader" program which reads your save file and identifies "flags", which are a set of information written as numbers. In the case of mass effect, for example: who is councilor at the end of me1? If "0" then = andreson, if "1" then = udina. If you download the save editor mod you'll realize how easy it is.
@@cristianbernal8009 The Witcher one is actually interesting. It couldn't scan a save file like how Mass Effect did (each witcher game was on a different console), so instead each game starts with an in-universe questionnaire about what choices you made in the previous games.
@@Dramatic_Gaming well mass effect doesn't scan either. You transfer a physical save file when creating a Shepard at character creation but yea I am now remembering that Witcher asks the big questions first.
On the topic of importing previous saves: did they fix Conrad Verner in the Legendary Edition? Or does he still react like you blew him off in 1, no matter what actually happened?
No, there is still a small and limited number of outcomes, only a few are mapped. 9 trillion variations isn't that large of a tree anyway (in term of number of nodes)
Most of the decisions you make don't actually end up creating a unique path. Using a tree is a bad metaphor; a better one would be a stream. You're floating down the stream, and there's an obstacle you have to decide to go left or right around. But after the decision you pass the obstacle and see that the two forks merge back into the same stream.
Well that's because the real end of the ME3 is the end of the Citadel DLC. Anytime I wanna play ME, I start with the first, play through all 3 until I hit Citadel and call it a day
decision trees and stories are great, too bad so much development time is spent on "romance" instead of actually interesting choices more linked to the main story and actual gameplay.
4.57 : the screen clearly shows 978,447,237,120. Isn't it 978 BILLION. And yet the voiceover says 978 TRILLION. Is it me who counts wrong or is it them.
in the near future id like to see a series based on me my rule power wealth might and my past life in my past life of me AKA MRM AKA PREZ MRM THE GREAT I
Best decision I made playing the trilogy was to romance Jack. I've talked to a lot of people about the game and I've never found another person who chose her during their first run. It's a shame the end suuuuuuuucks
@@AegixDrakan Ain’t that the truth. That’s why I made a fanfiction about her and Shepard. Called Shepard and Tali Parallel universe on Archive of our own. And why whenever I play through again, I always go with Tali every time.
![YoungBoy Never Broke Again - Missing Everything [Official Video]](http://i.ytimg.com/vi/dTYeMZGzOzw/mqdefault.jpg)
Deciding what to learn about next? Then why not visit brilliant.org/ExtraCredits/ to get 20% off the annual Premium Subscription and level up your knowledge. All while helping support the show in the process!
Man I love watching these videos alot my favorite is history of coffee and Poland fights in ww2
in the near future id like to see a series based on me my rule power wealth might and my past life in my past life of me AKA MRM AKA PREZ MRM THE GREAT I
Could you do a video exploring how/if players playing the same game or series over and over can cause diminishing enjoyment out of that game and lower the popularity of that game over time? When a game becomes unpopular it's usually balance that the community blames for it but not this.
I think there's a reason people don't play one game actively for their whole lives without stopping or taking breaks.
I think the Megaman/Megaman X games and World of Warcraft would be good games to look at, being an example of a single player game and a multi-player game that I think suffered from player fatigue.
A super important part of this is how a bunch of branches often merge back to a single point before splitting up again. Dynasty Tactics had this same design and that happened a bunch there.
That is honestly SUPER important to keep the scope of the game reasonable. If the game fundamentally changes after almost every single decision and doesn't link back up...It starts getting exponentially larger while not getting any closer to the resolution.
@@AegixDrakan But it can still be done, you realize that right? Especially with these big budget companies? You know not all paths have to have vastly different lengths right?
@@ramenbomberdeluxe4958 Agreed! I think Detriot Become Human does that well
Another example would be in space with markiplier where you kind of get 1 ending and then continue to get a second
@@ramenbomberdeluxe4958 Man, I'd love to see more (big title) games give you a hard "Game Over" 5-10 minutes after making a dumb decision. Like, one where you _know_ it's dumb, but you do it anyway to see what happens and how it's gonna bite you in the arse.
Yeah, as a Narrative Designer, sometimes these decision trees can get terrifyingly huge.
There's a debate at the end of a questline in a game I wrote for that was getting SO big that by the end, that just moving nodes around literally made Unity lag a little. There were so many bits of dialogue, decision splits, little bits of previous decisions to take into account, etc etc, that when I was done, I said "There's definitely not going to be one like this in the next expansion, thank you"
Any chance we can know what game?
@@fakjbf3129 Outward. The debate in question is in the Soroboreans Expansion. :)
@@AegixDrakan heyheyppl watch sseths video
@@HI-kb2cg Saw it when it came out, was pretty hilarious. XD
Imho developing Mass Effect probably destroyed Bioware. They bit off more than they could chew.
This video didn't really talk about the art and science of decision trees, it was mostly just about how many Mass Effect has, and not even very much about how different those are.
A key concept for taming decision trees is the idea of "orthogonality" - the idea that the effects of different decision points don't need to interact - a crude example would be if you have one character ask you what your favourite food is - lasagna, pizza, or burgers - and another character ask what your favourite flavour of ice cream is - strawberry, chocolate, or vanilla - then when a third character takes you on a date that they've secretly been gathering intel for, the whole date could be done as a single scene with 9 possible variations - lasagna followed by strawberry ice cream, lasagna and chocolate ice cream, lasagna and vanilla ice cream, pizza and strawberry ice cream, pizza and chocolate, etc. Or, because the main and dessert don't actually affect each other, the date could be broken into two scenes, each of which has three variations - lasagna, pizza or burgers followed by strawberry, chocolate or vanilla ice cream - which is 6 shorter scenes rather than 9 longer scenes, so a bit more than a third of the work of making each combination separately.
Of course, breaking things down that way can weaken the illusion - each of the three mains has to fit with any of the three ice creams, and vice versa, and if people replay, there's only at most a third as much variation available. But by adding possibilities rather than multiplying, it limits the total amount of work required.
This is usually how I analyze writing quality for decision trees (in games or choose-your-adventure books). If the path through the decision tree feels natural even after combining with other branches, then it was well written.
I've seen games where the merge point was written to be too vague, so it didn't feel like choices mattered at all. Or worse, I once read a published "choose your own adventure" book where the merge point clearly assumed you had taken a specific branch to get there. If you took the other path, it didn't make much sense.
A common solution I've seen in games is to use "mini-braches" that are determined by a previous choice. These "mini-branches" typically just contain a single snippet of conversation (or even just one word) so they can acknowledge the previous choice without re-writing the entire scene for each choice. This can save a lot of money by reducing the spoken lines to be recorded, the actions to be animated, and the assets to be stored.
"there's only at most a third as much variation available." Dunno how you got to that. You still have 9 possible variations for the dinner, no matter how the scene is created.
@@rolfs2165 It's a matter of writing 6 scenes parts instead of 9 full scenes. If you write a full scene for each possibility, you basically have to write some choice results twice. If you just write each result once then combine them you save a bit of time. And saving bits of time everywhere drastically saves time in the long run :)
@@rolfs2165 let's say both main and dessert take 1 to build
if you make the meal as a single scene you'll need 9x2, but if you make the single options as scenes and then play them in sequence you only need 6
I dunno if you guys take suggestions, but I was wondering about the game design in Factorio, and how it works in a way that causes minimal lag whilst having sooooo many entities and moving parts(literally and figuratively)
Optimization is a pretty deep rabbit hole. I think it's obvious that factario doesn't use super complex graphics on a 2D plane, but to get into the meat of it, one would need to talk to the devs. I think if they could get the devs, it could work. What some of it will come down to is just paradigm, where they bother and care to optimize, and may have from the start.
Factorio developers actually kept a weekly blog where they talked a lot about the development of the game. They called it Factorio Friday Facts, and you can learn a lot of interesting information from them.
@@theplaneshifter803 thank you, I’ll be sure to check that out!
I don't have any answers, but as a fellow fan I think you'll enjoy Trupen's videos for funny Factorio stuff
@@Ahrpigi of course, I even wasted an afternoon building a diagonal smelter
Al-Jalani punching femShep and stealing Garrus is both the funniest thing I’ve seen today, and the most disheartening. GarrusXfemShep is life.
I just replayed Dragon Age 2 and was really impressed when two characters talked about how my character was in a relationship with Merrill but had also had a less serious relationship with Isabela before that.
The game had a recorded dialogue specifically with those characters, about having a relationship with one specific character while before that having a relationship with different specific character. That just makes me wonder how many dialogues the games has for all other possible scenarios. xD
Lots! Some people don't give that game enough credit because the graphics were dated for when it released and it didn't have quite enough gameplay varation, BUT the writing is fantastic! The interpersonal interactions are really where it shines.
I've played it through a few times and it takes a lot of hours before you run out of smatterings of new dialogue, depending on your choices. Same for inquisition, which I've logged even more hours on (like, hundreds because in THAT game the gameplay also slaps! XD)
@@purpleghost106 I have a couple hundred hours in DA:I, too. Because I did a single completionist run (including DLC). Racking up hours in that game is not hard, not getting bored and burnt out while doing the same stuff over and over again in the forest area, then the desert area, then the seashore area, … _that_ is the hard part.
@@purpleghost106 Heck yeah! xD
I have to admit that I prefer the gameplay of DA2 to the others but that's just cause I'm wierd. x)
@@purpleghost106 If you manage your expectations, I'd actually argue DA2 is the strongest written one of the bunch. On the whole it is my favorite. The whole rival/friend dichotomy appealed to me a lot more than the general disposition in the other ones.
Having a rivalry romance with Anders gave me one of my favorite moments in the game (YMMV, I personally like emotional gutpunches that connect, because it also meant I really started to care about the characters) almost talking Anders out of something big.
Also: This may be my particular weirdness, but I appreciate the attempt to make the elves look really distinct and less 'human with pointy ears'. It didn't always work, but I prefer my elves somewhat alien.
My favorite example of decision trees in the Mass Effect series is Port Hanshan in the first game, where you need to obtain a pass in order to leave the port and continue to the next area. The game guides you naturally along one path, but there are actually five main ways to do it plus room for variation within each path. These paths are...
1) Uncover evidence of the director's corruption and give it to an investigator (which is the path most people take)
2) Give the evidence to the director so he can cover it up
3) Give the evidence to a businessman so he can blackmail the director
4) Tip off the director to the identity of the investigator, at which point they kill each other and you pick the pass off the director's corpse
5) Completely skip all of this and just rat out a smuggler who was part of a random side quest which the game in no way indicates is related to your search for the pass.
I think I ended up helping out Lorik Qui'in? I think that's the businessman one you're referring to. I may done both that and helped out the investigator; hard to remember.
Yes, Gianna asks you to convince Lorik to tesitify which leads to Anoleis being arrested but you can just give Lorik the evidence and he'll give you a pass. Fun fact, if you do this then in ME3 you'll get an email from Lorik where it turns out he took over as director after ousting Anoleis.
Happy to see more Extra Credits lately. I love Extra History but this stuff is so interesting too
eh, I mean did it really tell us anything that we hadn't already figured out ourselves after playing through mass effect?
The best way for commander Shepherd to avoid all this conflict is to say.
"I should go"
"Goodbye, Commander"
"I'm going to have to let you go."
"Let me go? Do humans view conversation as a form of imprisonment?"
In ME1 and 2, I tend to punch Al-Jilani out...and then in 3, I play nice...nothing confuses her more. XD
I'll be here if you need me.
[RENEGADE] You should go.
I kinda feel sad. There's a conversation, a very interesting and important conversation, to be had about actual dialogue trees. A simplified overview of how they're programmed, how games might note them with triggers. The effect of recalling certain decisions to players and how that betters the experience. The fact we can't actually program every computational outcome. I feel like that conversation is the something, and I hate to say it, that older Extra Credits would've looked at.
And the reason I grumble about this is that there have been a lot of episodes of Extra Credits about dialogue design, and this video does nothing new. In fact, Choices vs Consequences discussed how numbers of choices didn't matter. And the actual decision trees game developers map out look plenty different from the visual used for this episode. The episode The Illusion of Choice is all about the practicality behind the complex part of making choices.
The video essentially says "decision trees determine what can and can't happen" and that's it. No difficulty in programming, no 'you should remember this not this', or even talk of how games might change fundamentally or in small ways.
Do better.
They don't have an actual game designer working for them like they did originally, so they have to rely on people like a "Star Trek consultant". Or on the rare occasion they do have a real game designer to write an episode, it ends up being an ad for that person's game.
Agreed, most of the newer Extra Credits episodes are rather superficial, especially compared with the older ones. There's so much potential to dive deeper into most of the topics they talk about, because they have done more general overview video's of many topics already. Instead, they just rehash and even simplify their older videos. There's still so much to learn about game design, but I don't get the feeling I will find that here anymore.
I had absolutely no interest in playing mass effect, regardless of the accolades, until it was described to me as a dating sim disguised as an action RPG. Gimme that space beefcake!
Fun fact: the only muscly people you can date are humans
@@edoardoprevelato6577 OH, come on, you can't just assume just because they're alien muscles, they're not muscles at all!
... Ignore the identical model composition.
@@saxor96 i mean, i suppose Turian biology, being dextro based, and then avian evolved, is so absurdly different that Garrus could be the birds' version of the Incredible Hulk
the fact that there are trillions of branches does not necessarily make it the case that every mass effect play through is likely to be unique, because not all choices are going to be equally likely.
To compare the total number of theoretically possible playthroughs to the number of playthroughs that have actually happened, you would have to assume that any of those theoretically possible playthroughs is equally likely, and that would require not only that each option be equally likely to be chosen, but for your decisions to be independent, that is the choice you make in one place having no effect on any other choices you make down the line; both of those assumptions are fairly obviously not the case here.
For example, it's fairly likely that there are quite a few people who played "pure paragon" or "pure renegade" runs, or as close as they could come to it, and all of those runs may well end up identical.
Not a Mass Effect example, but I loved how the original Gold Box games back in the 80's and 90's let you export and import characters between games. While this didn't have any story implications, as the plot and events were different game to game, this *did* allow keeping your spellbooks and gave a significant boost to character progression if you maxed out your levels in the previous game - I like to think of it as branching spell selection. Curse of the Azure Bonds' character creator started at level 4, but you could have level 6 characters with over 2 levels worth of experience from Pool of Radiance imported in. You could train them to level 7 (capping your experience to just shy of 8th level), get in a bar fight for a handful of xp per character, then train them again to level 8. You'd be short on equipment to start since you spent a chunk of the money you were found with on training, but this was before the days of level scaling in games, so the increase in power from being higher level was worth it.
Someone pointed out that the journey to the ending for Mass Effect 3 was more important than the single choice ending.
It's really weird seeing youtube links in our school homework. We're going through some game develpment, and our book literally cites your videos. Thanks so much for making game design more approachable!
What's crazy to me is when you actually look up how many variables there are in Bioware games, it's often way more than you think or notice when just playing the game. They make sure so many different options and variables work. And yet, many of them are never even seen or noticed by most players because most players play a certain way.
If I have a hard time making choices in real life, what on Earth makes video games think THEIRS will be any easier for me? 😅
Mid-game: trillions of possibilities
Ending: what flavor ice cream does the galaxy get
I'd like to thank you for teaching me so much about video game design
I prefer to think of branching narrative as a branching river rather than a branching tree. You don't often see a branch bending back into a tree or into another branch, which is what happens when plots converge. If you ever take a look at a branching dialogue tree you can always see every path reconverging at different points so things don't spiral out of control. Take the Outer Worlds or Divinity 2. If you fail or succeed a skill check you might get one line of unique dialogue relating to what you just said or did but then they all reconverge on a single node for if you are progressing a failed path or another node if you're progressing the successful path.
“Think of it like a series of ‘if then’ statments.”
As a programmer, this is probably how they’re implemented if a specialized plugin isn’t being used…
Firewatch seemed like a linear experience, but there was a huge decision tree that affected all the dialog in lots of large and small ways.
Silent Firewatch protagonist is best Firewatch protagonist.
PANR has tuned in.
Really it's more of a rhizome than a tree, except if you want to never finish your game. You rarely get a whole narrative arc sprouting to never join up with a main branch. In the end it's much less different outcomes than just small details changing to acknowledge player's choices. The problem with BioWare is them always advertising a diversity of outcomes when it's rather different flavours of the same set-in-stone story beats, and that's why they didn't see a problem in wrapping the entire ME trilogy around a central, final path, whereas the players were expecting much more. They make up for it by making high-spectacle action-RPGs with dating sim elements for mass gamer appeal. Compare and contrast with Encased, Tyranny, Underrail... smaller budget, smaller scope, but better handled choices-and-consequences.
Mass Effect really made its mark on gaming history with how it's *still* talked about today. Is a good thing!
I believe the first time I encountered a decision tree, was in Oregon Trail. I died from it. :(
🎵 State machine, a really terrible, state machine 🎵
this is why i have such a hard time finishing my interactive fiction games - get too caught up making new branches rather than finishing the story! gotta work on keeping myself on track lol
This is perfect timing, because I'm literally casually working on a game concept where this is a core aspect.
While it was interesting to learn about the concept, I'm more interested in learning about the nitty gritty of its implementation in games. Any hope of you guys doing an episode on that?
look all I'm saying is, programming is basically If Then with extra steps
When you get down to the assembly code, yeah.
Actualy is not only prety hard to make games with lots of meaniful branches is also a risk. You can create one good history for your game but even if you spend the time to make several some of them will probably be better than others which means you have players having a not so great experience with your game. That is why most decisions in games are meaningless making a litle dialog change here and there but with few changing the plot. More significant changes use to be at the ending as well. This solves another problem of this type of games that is having to replay it a lot of times if you want to see all content. So designers use to fake branches
It does give a reason to play through a game again, for a different experience.
@@massimookissed1023 Yes with few branches, but if there are like a lot of branches you endup replayng a lot of parts that don't generate fun. Like havin to re-watch half of a movie 10 times
I bet there is a lot more overlap in play throughs considering the paragon renegade progression system encourages following one or the other exclusively. Something like the Witcher 3 probably has more variety in the decision tree between plays since it doesn't lable its decisions with a specific playstyle.
I was going to say that I'd like to see a game let you take decisions from game to game, I had no idea that mass effect already let you do that! Cool!
For Dragon Age Inquisition, you can go to the Dragon Age Keep website and make different World States to upload to Inquisition (rather than having to have the save files all in one place). Makes it a lot easier to see how different decisions in earlier games affect things and really neat to play with in general!
@@livelaughstuck neat!
4:00 We got a friggin' Alien Lover ovah here!
Quick little side note: I thinks it’s funnier whenever the _Reporter punches Shepard_ in retaliation.
Could you do one about skill trees?
The image about the three ends is just amazingly descriptive on why nobody likes them. But considering having billions of possibilities makes it impossible to finish it nicely for everyone.
For a while i found the bioware romances cute and nice to have actual choices with my character but ... what bothered me was that i was never in a position where the other side did the first steep. And usually every other steep too. It was always my character who had to proclaim interest and try to who the love interest never them trying to who for my characters attention. I think it was in the pathfinder game where i finally had an npc who did initiate the first steep and did occasionally do some wooing of their own but not certain ^^ Still a good healthy written romance should show interest from both sides and not have just only one pursuer, however the vast majority of rpg romance plots tend to be rather one sided. (I am not talking about VN here because well ... they are all over the place ;) and have a more book feeling then RPGs do)
Playing backwards through the Bioware catalogue, I've found that in the older games it actually is your companions who will initiate to some degree. The game starts you off in an 'active' romance state, your companion will flirt with you, and you get to decide whether or not to reciprocate, at which point the romance is continued or concluded. Being used to the newer system where you have to choose flirt dialogue options to start a romance, the older games had me googling which stealth romance option I had chosen to have all these dudes declaring their undying affection lol
Yesssss, narrative systems design! Give me *more!*
And this is why virtually all the ME2 Squadmates were basically written out of ME3, too many people can die.
I love that you guys play Mass Effect.
It might be impossible to talk about mass effect without talking about the ending.
Ahhh, Bioware games were really something special. I hope they return to focusing on this sort of thing.
_The Cave of Time!_ That takes me back.
I was playing Deus Ex again the other day... released in 2000 and still does all this better than most new games!
Yo, EC, can you make a video on how devs start developing a game? Is it from an interesting mechanic they were playing around, a story they want to flesh out in a game world, something else?
Half of this episode is just simping for Garrus and I am HERE for it.
Well, they need to do something to pad out the time, since there's so little actual content.
4:56 you said trillion instead of billion.
An excellent example of a well crafted decision tree is in space with markiplier which is pretty much just a decision tree and nothing else
Thank you for the video.
There should have been a non zero chance to be able to romance Sovereign. His speech still makes me weak in my knees.
It's a _ship._
You can't just get inside a--
Oh, wait.
Oh I ABSOLUTELY have played through again just to do that, consequences be damned
@@GGreenHeart hahaha tried to replace Saren as Sovereign’s SO too I see.
My favorite mass effect decision in the whole fucking series!!!! I never make it, and only discovered it watching my brothers playthrough, but my god is it funny! And I think they even make references to it in each game with a similar choice.
I cannot condone a player getting in the way of the relationship between Garrus and Tali.
Thanks for the video 👍
I found out that my journey with Commander Shepard was unique when I discovered that Wrex didn't have to die on Virmire. And for that sin, Ashley was killed.
how did you do that count? how did those number's get crunched? Is there a tool that you run or did you have to ask the devs?
"Best Quarian pick-up lines," when the animation was FemShepxGarrus. I'm not sure if this means the writer was BroShepxTali and the animator was FemShepxGarrus, or a subtle reference to the fact that GarrusxTali is canon if Shep doesn't take one of them. And yes, I'm assuming it was intentional and not a screw up, because I like to think positively once in a while.
Edit: Watching for several minutes makes me think it's the latter.
You can follow up with a discussion on Automata.
I realized my journey was unique when my girlfriend Tali killed herself and I only had four members in my team at the end of Mass Effect 3 = (. Lonliest experience ever
The game gets so much better when you don't just try to make a Paragon or Renegade run.
Also I like the ending. I feel like it is a grandiose ending for a grandiose epic.
Go on! You can put my spaceship in lockdown now!
And I did the math for the amount of endings possible in Mass Effect. 4 is the answer
I like decision tree games. Makes things interesting.
*Sees the video stan Garrus*
Ah, content creators of class I see.
The first second of this video gave me ptsd
"Decision trees" is a poor term. They're directed graphs, meaning you can probably prune a lot of those possibilities.
I thin "Punch in the face" should be an option for anything in any game.
I could sit and listen to people gush about Mass Effect all day. Best trilogy ever. I've seen 10 different journeys. Looks like I have a few hundred billion to go.
nice what a great video I love videos that feature erin they are always way good
This should be common knowledge to anyone that plays rpgs.
They could have at least discussed some of the details and problems with decision trees, but they just talked about combination possibilities.
So while there are a ton of potential scenes and things to do, that's just looking at the potential routes, which is a feat of design for Mass Effect, no doubt, but how many of those scenes are reused? Game development, especially something like mass effect, is expensive, and game designers like shortcuts and reusing assets where they can, especially for a game made for a console, so file size of the end product especially matters, don't want to make something bigger than a disc can burn to
Mass Effect: Nearly a full Quadrillion branching paths for the player to navigate full of intricacies, romance, adventure, and tradgedy
In Space With Markiplier: *FIX IT FROM OUTSIDE*
Kinda missing out the fact that most of the decision tree branches will rejoin. and net you the same results. You may have 59 branches bout 20 of them point to them hating you, 10 have them remain neutral, and the rest have them biased towards you. The key point is to simply remember the last significant node.
All those variations come down to a binary choice at the end that every player gets. Mass Effect isn't even close to being the best example of Decision Trees.
Listen Garrus being in the beginning is such a callout I am upset.
Am I the only one who doesn't like multiple endings?
"Congratulations. You finished the game. Now play it all over again in a slightly different way."
I love your videos
Hold up. Mass effect seriously came up with a way to import the previous games data? How did they end up doing this
Dragon Age, also done by Bioware, also imports in saves from prior games. I believe even Witcher and other games that share the same protagonist over multiple games do the same.
It wouldn't be too hard, I'd guess. It checks for a save file on the drive, scans for the relevant nodes on the decision tree, then adjusts a 'default' tree based on which nodes it reads from the data.
It's actually stupidly easy. Built into the new game there is a "reader" program which reads your save file and identifies "flags", which are a set of information written as numbers.
In the case of mass effect, for example: who is councilor at the end of me1? If "0" then = andreson, if "1" then = udina.
If you download the save editor mod you'll realize how easy it is.
@@cristianbernal8009 The Witcher one is actually interesting. It couldn't scan a save file like how Mass Effect did (each witcher game was on a different console), so instead each game starts with an in-universe questionnaire about what choices you made in the previous games.
@@Dramatic_Gaming well mass effect doesn't scan either. You transfer a physical save file when creating a Shepard at character creation but yea I am now remembering that Witcher asks the big questions first.
On the topic of importing previous saves: did they fix Conrad Verner in the Legendary Edition? Or does he still react like you blew him off in 1, no matter what actually happened?
On console, No But PC has the fix.
What my question is how do you write and map complex decision tree. Like how can a team make 9 trillion variations of the same game?
No, there is still a small and limited number of outcomes, only a few are mapped. 9 trillion variations isn't that large of a tree anyway (in term of number of nodes)
Most of the decisions you make don't actually end up creating a unique path. Using a tree is a bad metaphor; a better one would be a stream. You're floating down the stream, and there's an obstacle you have to decide to go left or right around. But after the decision you pass the obstacle and see that the two forks merge back into the same stream.
Well that's because the real end of the ME3 is the end of the Citadel DLC. Anytime I wanna play ME, I start with the first, play through all 3 until I hit Citadel and call it a day
The reason Mass Effect's decision trees are so good, is because Garrus made sure to calibrate them extensively
Why is shepherd about to punch lady Shiva?
I just wanna know where Erin got that dress
decision trees and stories are great, too bad so much development time is spent on "romance" instead of actually interesting choices more linked to the main story and actual gameplay.
I thought that the not so secret model is behavior tree
Tbh, I adore Tali so much, that as fem Shep I don't romance Garrus, so he and Tali can get together. She (or rather both) totally deserve that.
4.57 : the screen clearly shows 978,447,237,120.
Isn't it 978 BILLION. And yet the voiceover says 978 TRILLION.
Is it me who counts wrong or is it them.
It's them. Sloppy writing/narration.
What about AI ...
in the near future id like to see a series based on me my rule power wealth might and my past life in my past life of me AKA MRM AKA PREZ MRM THE GREAT I
Best decision I made playing the trilogy was to romance Jack. I've talked to a lot of people about the game and I've never found another person who chose her during their first run.
It's a shame the end suuuuuuuucks
All those branches gone to waste... because why would anybody NOT pick Garrus? 😆
I might be an idiot saying this but wouldn't it be better to have a group of check flags in a bitmask or a FSM instead of Trees?
Yeah, hard to top Mass Effect for this subject.
Well, except that they screwed the ending. Something like Undertale would be a better overall example.
I prefer Male Shepard and Tali love.
That was my choice too. XD Tali is just too sweet, even moreso when she develops into a complete and total badass in ME2.
@@AegixDrakan Ain’t that the truth. That’s why I made a fanfiction about her and Shepard. Called Shepard and Tali Parallel universe on Archive of our own. And why whenever I play through again, I always go with Tali every time.
Mass Effect romances are a terrible example of how to make your decision trees feel organic.
remember when they got memed to high hell after trying to be smart on video games
i wouldn’t consider this a reliable source
But, like, how do you make one?