Into the Breach Design Postmortem
Вставка
- Опубліковано 14 тра 2024
- In this 2019 GDC session, Subset Games co-founder Matthew Davis details the Into the Breach design process from early drafts to the final balancing decisions. Davis dives into years of cut content and iteration to show how Subset Games approached the difficult design challenges of making Into the Breach.
Register for GDC: ubm.io/2yWXW38
Join the GDC mailing list: www.gdconf.com/subscribe
Follow GDC on Twitter: / official_gdc
GDC talks cover a range of developmental topics including game design, programming, audio, visual arts, business management, production, online games, and much more. We post a fresh GDC video every day. Subscribe to the channel to stay on top of regular updates, and check out GDC Vault for thousands of more in-depth talks from our archives.
Say's he isn't a good public speaker, knocks it out of the park with a clear, engaging presentation. You can see him smiling while talking about something he's passionate about.
A good exemple that it's more important to have someone who knows and love what he's talking about that, than a generic placeholder speaker who's only good at speaking.
@@randomrandom450 I hate the high rising terminal that is pervading speech these days.
Thing about speeches is it has to be organic than sounding scripted. He really did like a speech of a pro.
Came here to say this. He was quite nervous to start but blew it away before long.
I think the real brilliant part (or at least what I appreciated most) was how the game is about saving people and protecting VIPs than it is chasing down and killing every enemy. I’ve sacrificed pilots and Time Pods before to try to save as many buildings as possible, and it’s a great kind of bad feeling, that feeling that you’re being heroic.
It is ! , it adds more strategy in terms of not being "kill everything" but rather resist and try to save the most important things
My most memorable run was with Abe Ishmalu as pilot and the laser squad. It was the second last round of the last battle and i had to reset the turn because i messed up. I realized that in order to win i had to sacrifice Abe by using him under a falling rock.
Ikr. It feels so shitty when you don't have any other option but to let a building get destroyed.
On the final island level there are those power generator thingies which are basically a substitute for buildings, and when they get destroyed you don't feel as bad, because people didn't die.
@@bratprica6383 Which makes sense within a gameplay context, because you don't have to defend them as stringently when it's the last mission of the run.
this game is so darn **smart**.
Agreed 1000%
I've never seen anyone take a drink from a milk carton during a talk before. My life has been improved by the experience.
It was water. One of those "Boxed Water Is Better" things.
@@BackfallGenius Awww. I was enjoying the expanded horizon.
That post-swallow wince though, maybe it's vodka............... or esophageal ulcers. 🙃
@@CodeOptimism what, just because a man has a certain hairline and talks about chess that means he's Russian and therefore has a drinking problem?
I mean, I believe it.
@@CodeOptimism or it's chilled?
-> worried his talk will tank
-> gives one of the best and most informative talks about design process
Matthew Davis has the kind of quiet confidence about game design you only see in people that are scary good at it
06:28 - "we're not looking to make these big 40-hour games."
>Looks at my 100 hours on FTL and 85 hours on ITB.
"Oh."
I think they meant that it takes 40 hours to finish one campaign-type of game.
I have 200+ hours on FTL, and 40h on into the breach. I stopped playing once I got all the achievements, because the game was "done" for me. While in FTL I still enjoy launching a game and try something with a random ship.
It's the difference between 40+ hours of content and 40+ hours of replay-ability.
Haha honestly. I haven't even unlocked the third island and I'm at least 50 hours in.
Right? Though something like Into the Breach is sorta like a mix of Solitary and Chess with progression elements, it's a delightful brain exercise.
Into the Breach is an incredible game, so tightly designed it really feels like a classic. It's like a more accessible version of Chess puzzles.
What an incredible talk. It's rare to see developers being able to explain their design process with this degree of conceptual precision and clarity. And it really shows in their games!
PLEASE invite this man back. I've watched tons of GDC talks, but this one easily makes my top 5. Such a genuinely engaging speaker with some really provoking philosophies on design in general.
True!
Could you share your top, please? :)
What other talks make up your top 5?
It's so refreshing to see an artist that isnt totally full of themselves. I love how they obviously value being efficient and concise. "Perfection is not attained when there is nothing left to add, but when there is nothing left to remove" -Some Dude (paraphrased)
I've never anyone with such an analytically precise understanding of their own game before like this.
I mean they are just two dudes. And this is the SOLE programmer to boot.
Subset Games have produced 2 of the most incredible games to ever exist. To think that they both came from the blood, sweat and tears of these two people is amazing.
Hey, we are building a new game, would you mind checking it out? it's in our website
Matthew is brilliant on every axis. what a wonderful talk.
didn't expect to see you here :D
make video game vids please! would love to see more content from you besides MTG!
"Wouldn't write a movie about it" Have you seen battleship? I say go for it
these guys are the gods of game-making, not sure if they realized it yet
They are probably too humble to ever realize it.
I wouldn't say THE but their divinity os unquestionable
57:00 - "wow" indeed! The enemy spawning and movement make for seemingly hand crafted puzzles. Props for an astute reaction stemming from a great question!
I would have bet money that the AI was purposely directed. I thought it placed enemies in ways to take advantage of the load out I was currently running, or taunt me with set ups for weapons and upgrades that I didn't take. Wow indeed.
It's the minecraft music design philosophy.
@@OnyDeus it kind of is directed, I think, with the enemies choosing randomly from thier move and attack options, never stepping on a pod, giving you room to always find a solution to the puzzle and stuff
15:40 Everyone who has played the final game knows exactly how much threat telegraphed attacks can still provide (spoilers: it's a lot)! It's so awesome to hear how they worked through all that uncertainty and uncharted territory to make an amazing, unique game!
I personally love the choice of 2/3/4 islands, sorry that people got confused, but I love the flexibility.
Choosing 2 islands for a quick run, or really pushing some of my lease fav mechs down to 4 to see what they can do.
Idk, I enjoy the choice and appreciate it a lot, ty :)
3 and 4 are most fun for me, but I completed 2 with one squad (blitzkrieg)
Silverkingx2 yeah i agree entirely, i think it was a great design decision
This! I fully agree on the ability to choose and it works so well in all scenarios. (This topic came up around 43:00)
From the game design point of view it was a clear fail as it led to no reward. That's where they had an opportunity to extend the story/progress. If you would have managed to finish the boss with two islands you'd be teleported to some very final island with a real final boss to get rid of the threat once and for all.
@@yrussq but how would that work with replayability? and some squads work best at 2 islands done, some at 3 or 4, meaning each squad has an easier time with their own island preference. Each island you do also makes the final one slightly harder (with more vek, and more elites)
I agree that it is strange story wise for there to be an option of 2/3/4 islands, but gameplay wise the reward is the rewards from the islands if you play them well, the risk is if you dont get good things from the next island
My 2018 GOTY. I just think it's brilliant. And unlike most games, I think (though impossible to know) it's a timeless game that will hold up for decades if not longer. The elegance & design simplicity is all in favor of pure gameplay
Absolutely. Will make my nephew play that shit.
Holy crap, dude. You say you're not good at talking in front of an audience then proceed to give one of the most comprehensive presentations I've ever seen. I mean I could probably create Into The Breach in Unity just based on this presentation.
This is a brilliant talk, some of the scenarios this game throws at the player are incredibly sublime and clever. You get the sense there is a lot of magic underneath enemy attacks, some amazing algorithm that somehow knows how to get the player guessing on 'best of a worst situation scenarios' while not revealing the mechanics that make it work, its like the algorithms are conspiring intelligently to ensure the player is thinking and having fun, its amazing how fair and balanced it all feels. So much respect for the thinking and design that went into this title.
Subset Games are gods of game design. So much work and talent put into making the most mechanically elegant and fiendishly repayable games ever. One of the best talks ever made at GDC and the rapturous applause at the end shows how much respect these guys have in the industry.
"Everything follows the game design" Amen
Now I understand why I sucked so bad at this game, my first attempt I did extremely poorly and it didn't clicked to me and I got discouraged by the restarting the timeline mechanic.. but the game became super interesting when I realized that I could trick the enemies into killing each other, but for some reason I never went back it.. I will definitely give it another try because this reminded me how cool that moment was..
The big mistake made was to have progression on the mechs rather than progression on the buildings or defenses. The expectation of players was already on preserving units but the game itself only reinforced that problematic mindset. Consequently players who misunderstood found it hard while the ones who realised that units were disposable found it easier.
One of the best game design talks I've seen. Thank you very much for doing this talk, Matthew Davis.
I always thought the energy grid maintained temporal cohesion for the mechs. Since it plays with the time travel mechanic. They needed a Buble of temporal stability in that region. That also pairs well with the last missions.
I thought that it maintained some kind of defense around the island that prevents Vek from entering because when grid fails, a bunch of enemies spawns.
Im fascinated by this game. I rewatch this video often to absorb the developer’s perspective
Loved this game! It satisfied me more than XCOM 2 did
51:50 Damn I was sure that the percentage was secretely higher
I just started a game where my 2 first damages were resisted and I was 100% the game was cheating to go easy on me. Damn it's so good to know it's completely random. :)
Yeah it always seemed to me that the game would likely give me a resistance if I played perfectly in the battle and could still not prevent the attack. One of my favourite things about this game is that it's unfair
To be fair, answering that the chance is totally random is still the right thing to do if it's actually manipulated. Manufacturing close calls is worthless if everyone knows you're doing it.
This talk makes me think of the Cursed Problems talk, and how they basically had to persevere until they "uncursed" themselves and made a game that perfectly fit the hook
I love how naturally and easily he explains that something as simple-looking as ITB has so many complex thoughts and concepts behind it and why it doesn't have this "run of the mill" feeling to it which most turn-based games have these days.
48:47 "I love the feeling of getting rid of bad design now."
Well, I can't say I understand enjoying it, but it's far more preferable to cut out the cancer than spend more time suffering by trying to piece together a hopeless mess. I experienced this when writing some code and getting to the degbugging stage.
@@nomoturtle1788 Just imagine being able to get rid of a regret that keeps you up at night. Feels pretty good.
I really appreciate this window into your creative process - this is one of my favorite games, your hard work is much appreciated! I'm reminded of something that I read in an article about Mo Willems, a children's author whose work is lauded for the depth it achieves in spite of its apparent simplicity - his editor said something to the effect of "You have no idea how hard it is to produce something that simple!".
I like the comparison to chess. In fact, the chess-like nature of ITB is exactly what has persuaded me to buy and play this game. I don´t regret anything. It is awesome. Thank you very much, Subset Games. I enjoy Into the Breach more than XCOM and your choices were right.
The mistake they made was to not realise early on the similarities to Chess which is itself about telegraphic strategic gameplay. While you don't know which piece will move you do know how they'll move which is the fundamental line drawn between the two games.
By adding unit progression, more varied behaviours, tech trees, damage types and layer upon layer of complexity the clearly broadcasted strategy became obfuscated by "noise" that made the game harder to predict. Other games like Slay The Spire are showing that it's more satisfying to be presented with a clear problem and being offered clear solutions to choose from rather than being subject to RNG or something so complex that you're just muddling through.
This telegraphic strategy is the perfect antidote to TRPGs which are bogged down in the problems inherited from old PnP d20 RPGs where a carefully planned strategy is undone by rolling a 18 against a 19 skill check. It feels cheap to lose like that and worse still is that its not even that satisfying to roll a 20.
I was so convinced that the AI is always making sure that there is a way for the player to not lose or something, and for some reason I am glad that there is not and that the whole experience is genuine
Clicked on this talk while reading a thread on game design, didn't expect to lose 1 hour of my day to this!
Very engaging and interesting talk
I first played it in 2020. Without a doubt one the tightest designed games I've played in recent memory! I hope other future turn-based tactics games take note!
I freaking loved this. I loved listening to the dev be able to express their design choices and how each of those choices they made in their processes affected all of the other aspects of the final product. Also, I am trying really hard for myself to learn that work you put into systems or designs that don't make it to the final release are not "wasted" work as it is work that pointed you to the correct solution in the end
Its always a pleasure to hear a competent professional talk about his work. I'm sure not many developers can talk like that about their games.
I always thought the maps were procedurally generated....damn.
I was just playing it few days ago and realized maps are not procedural generated.
I randomly picked this game up during thanksgiving thinking it was like ff tactics. I’ve been completely enthralled by this game. Seems simple but so complex. It’s pacific rim meets chess. Each mech unit plays so different and I find myself falling In love with certain teams and the *oh shit* moment you have when you find out a synergy of the team! The game keeps me coming back again and again just to unpack it all.
Legend in the making. Looking forward for great things from Subset Games!
Awesome, great inspiration from this team, game and GDC.
I've watched this talk before. Got FTL, watched that postmortem, came back to watch this one. I love these talks.
These guys and guys like Bay 12 Games are simply doing amazing work.
I bought both FTL and Into the Breach. I bought Into the Breach twice. One on PC and one on Switch.
I really love this guys voice. it has a very unique quality and sounds so nice
Thank you for this GDC and Matthew Davis.
This talk was amazing. I feel like I'm a kindred spirit to the Subset guys, and hearing a success story from one of them is awesome.
wow. this talk is exceptional and ITB is a gorgeous puzzle game from aesthetics to design itself. what a gem.
21:00
I thought the power grid would work for some kind of ground stabiliser stopping the vek from emerging. Explaining why vek appear everywhere when the grid collapses.
One of the best talks in GDC, ever!
Absolutely gorgeous and useful talk! Not the biggest fan of the game, but from the game design point of view it was clear right away what a tremendous job they've accomplished to distill the gameplay to that thinnest point of balance. It's nice to know how much effort these guys have put to get the game feel right - that's how the games should be made! Bravo!
i really enjoy learning stuff about these games. indie games are the crux are true creativity in the modern gaming world. AAA is tired and done.
Coming from 2 years in the future, your statement is even more true now than it was when you made it. I am constantly amazed by how jaw dropping and innovative indie games are becoming as AAA continue to get worse.
Wonderful presentation, also shout-out to the FTL assets in the prototype version of the game!
Thank you for the great talk, really interesting insights into your game design process.
I'm looking forward to playing your game when it comes out on Linux. Keep up the good work!
I love Into the Breach the way it is, but I got to say that i would love to play the game with that worldwide aspect, management and 20+ hours gameplay
Great talk, thank you so much for sharing.
Oooh I need to play more of this game
god daaaaaammnnnnnn, that was highly entertaining and informative!!!! so cool. so cool.
I especially loved the questions at the end there, interesting and varied little tidbits of info that enhanced the knowledge the talk gave. :)
Damn, what a great and incredibly interesting talk!
One of my favourite games.
def in my top 5 gdc talks!
Good talk.It was interesting to see what the game could have been. Also as a bonus I learned that kinkajou and occiput are actual words. (from the crossword puzzle :)
Brilliant game and speaker! I personally love the variety of squads and achievements, and wonder how the balance is so good
Played it, loved it
The corruption and panic mechanic and others could be a nice DLC or extra mode
have watched this a couple of times because i find it really interesting, love the game - also finally thought of an example when he's talking about the 4 or 5 turn battles being really fun! Card games from the final fantasy games have that with Triple Triad etc., little puzzle mini games on a small board with varied rules going on (FF8 in particular had some wild shit going on with rules spreading etc)
what i love about subset games that they do better than any other developer, is the presentation of information, each game is so incredibly complex but due to the way the information is presented it comes across as so simplistic to pick and play at the start.
I'm kind of a big fan of complex systems, and wanting to explore them is something that drives me in coding games, but just like their experience suggest, a lot of complex mechanics doesn't always translate into interesting gameplay. It's still worth exploring them in prototypes though because sometimes "so complex it's practically random" can lead to something more interesting than actually random.
System doesn't necessary need to be complex to have complexity. Get Rock, paper, scissors; tic-tac-toe; coin-flip and dices, combine them together with decent rules and dependencies keeping each stage separated but influenced by others and you'll get a complexity some strategic/tactical games would envy of.
Is it complex? No. Any of these games are well known to anyone. Would it have complexity? Hell yeah it would.
@@yrussq That sounds reasonable, but perhaps because I'm weak with that kind of games (chess, go etc.) I'm drawn to one or two complexities on top, which exhibit more chaotic evolutions.
@@FreeScience If we are talking about game design in my opinion total randomness means that you've failed in your design of a complex system. Complex on the other hand usually means it's hard to forecast the consequences but if you have invested time to study how the system works you can predict and manipulate it to your own use. Random has nothing to master - pure chance.
What you are talking about is kinda mixture like a max 95% hit chance. Or that 15% building resist in ITB. Or event cards in Monopoly. I like that too :)
Thank you. This was awesome.
great talk. thank you for uploading :)
Loving the Hoplite credit, I NEVER thought that game would grip me as much as it did when I accidentally found it
great talk! i love ItB so much so this was so damn interesting to listen to
This makes me like this game even more. Both these games are just fantastic, its really clear that Subset cares deeply about their design process and games and it really shows.
Great talk thank you so much for sharing this.
7:38 hoplite is one of the few games that stay on my phone and even when I upgrade my phone I transfer it cause it's that good!
Hoplite is the best phone game ever IMO.
Check out Antiyoy, it's also a good turn-based strategy for Android.
Amazing talk, really incredible!
Good to see them referencing Hoplite as part of their inspiration. Fantastic game - if you are watching this and haven't tried it, do make an effort to get hold of it, it's a really pure tactical game and very fun and challenging in a really satisfying way.
Fully agreed, I love that he gave Hoplite a glowing shoutout. It’s really a perfect phone game. So easy to pick up and play but packs a surprising amount of complexity. Can definitely see the influence in Into The Breach.
Absolutely wonderful.
I have to say that Into the breach feels very much the way you described it on this video. I enjoyed more FTL, and I'm pretty sure if you didn't "lose" that much time you could have make a really good game. It lacked on content for me (essentially objectives you need to do for a mission, very repetitive), but I still enjoyed my time on it.
Very interesting. FTL is my fav game of all time and ItB is not too far behind. Hope to see more from Subset Games in the future!
I can't put ITB down but sadly FTL doesn't grab me. I find the combat system confusing with all the charging and pausing and targeting.
But i'll try again and it wouldn't surprise me so much to be hooked for 100 hours just because a mechanic clicked.
@@keratubes Like Matthew said in the presentation (if I recall correctly?), FTL is significantly harder and more random. I would argue it is randomness made as good as it gets though.
The way everything fits the context and setting, every single mechanic, menus, story, death, the premise AND HOW IT IS USED to justify the entire design is genius. Gameplay is center. It goes beyond the "perfection" achieved in From games where death and healing items have lore justifications, ITB is not deep, but it is close to perfect in everything fitting in. My 2018 GOTY. I was addicted to it like I was addicted to Bloodborne, and I appreciated it on a similar level as well. Thank you, it was amazing to chance on this video.
This was awesome and I love the example showing advance wars, one of my favorite games of all time.
(One of the best puzzle games I've ever played)
I seriously think you guys are some of the most respectable and capable design teams around. I will without a doubt purchase whatever your next creation is. This video was a wonderful look into what makes you guys so good. As mentioned, "Design a 4 turn limit game." sounds hard. But following your process gets you these kinds of complexities naturally. I had so many issues with FTL, but recognized all of its brilliance at the same time. I hope you guys keep doing these retrospectives and developing your development process, because the things you've mined go well beyond game design truths.
I'm IN AWE. Such a great talk oh my god. They should have given him that extra half an hour to just talk about difficulty.
Great talk and great game!
I know I’m late to the party but this is just so engaging, I thought I was 5 minutes in we’ll I’m at minute 45
I kind of want to see what the game would be like if they stuck with the idea of winning key buildings that give you upgrades and rewards. It really seems like a neat idea (around 40 minutes into the video). It reminds me of the campaign for Armies of Gileinor where beating a mission gives you access to a new unit or ability.
We sort of have that right now with the missions to protect power plants, but this seems very interesting to expand on.
I can totally see them implementing different game modes, like “classic,” “save the facilities” (for the building rewards), and “corruption mode” (for the timer).
Amazing design philosophy producing an amazing game.
This game is a master piece amazing i love every things of it its just perfect i hope they add more stuff in it like they did on the last update
In terms of explaining mechanics with narrative devices, I don't think it's a bad idea to have the mechs powered by the energy grid around them. One of the most iconic mecha anime of all time, Evangelion, had the same conceit. The onboard batteries of the mechs could only keep them running for a few minutes once they were detached from the constant power supply of the grid, and their electrical umbilical cords were an important plot point. Of course, that show took a left turn, plotwise, that ITB wasn't going to, but I think it makes perfect sense that giant mobile robots shooting high powered lasers all over might need more energy than a small battery could give them, and more mobility than a large battery could give them.
ITB is one of the best pure games I've ever played. The most interesting thing about it is, unlike FTL, which requires a lot of behind-the-scenes calculations a computer is capable of, ITB could be played perfectly well as a board game with an 8x8 board, a few props, and two 8-sided dice.
paradox's recent imperator: rome could have used such a good game designer, able to kick ideas and focus on the ui, instead of excusing afterwards about 42 % on steam
Paradox's "game design by landfill" approach is a great example of a terrible philosophy that frequently produces strangely compelling games.
Wasn't sure if the concept was right for me. Got it today on discount and have to say it is executed brilliantly. It's a mech-bug-puzzler.
how can the ai be so simple? so often you need to use every move at your disposal to win.
Yeah, Hoplite! Amazing game!
wow the amount of trial and error was enormous and coding work for those is just astounding
45:45 I think this situation was only egregious when you had a map with tremor worms in it. Being unable to utilize knockback system mechanics made them really frustrating to deal with in large numbers.
Not sure if this matters, but "co-founder" is not spelled correctly in the description. Great talk, though, thanks for recording and uploading!
45:05 omg that is so true. When I play the game I always assumed that 4 islands means that I am supposed to play all of them. I always assumed that since its a tectics game there was a tactical element to it (go if you feel like one more island will crush you or sth), and that difficulty scaling was not perfect (usually in games even if difficulty scales it is almost never perfect and there is an optimal). I would never have assumed that it is just a personal choice based on what I feel like playing.
This is GOLD to me, ITB was my favourite game of 2018 (no kidding, no empase, just the truth) and I'm a game dev attempting making cool indie game !
Not just yours. It won PC Gamer's Best Game award. :)
@@mattolstein2999 Well, I don't read english press but between that and their interest in Mount & Blade, I like those people !
1. Qxh7+ Kxh7 2. Rh3#
Oh, yeah, the talk, great indeed !