I know that the "how to fail" gimmick is often used jokingly, but it is often more useful to look at what failed rather than what succeeded to avoid a Survivor bias
@@viniciusschadeck4992 Survivor bias is when people focus only on what succeeds and overlook what does not, sometimes resulting of a false conclusion of how the results came to be.
@@viniciusschadeck4992its useful to consider the chance of survivors bias. A lot of successful games having thing X doesnt mean that thing X is necessarily the key to success, it could be that a whole lot of really bad games have it too, and its not necessarily good or bad or potentially it could even be actively bad, but the good things the successful ones outweighed it
This one is especially interesting to me because of how well-regarded FTL: Faster Than Light is yet it does over half of the things listed in this video (I myself have 1500 hours in it), especially if you compare the base game to it's largest overhaul mod "Multiverse" which mostly attempts to patch up most of the "failures" of the base game (including some that aren't even mentioned in this video, such as losses that are out of the player's control or having the game end in the same place always), yet there are some people who are like "no we don't like this we want more similar to vanilla FTL."
FTL also succeeds at a lot of these though. Items, upgrades and crew members do combine in different ways that allow for different approaches and some enemies interact directly with these different attack approaches in meaningful ways. A run where you focus on boarding enemy ships, might give you trouble when you cross paths with an automated ship, unless you're boarding with lanius crew, in which case it's basically a free win. Some ships are really weak against drones, while others might have a lot of ways to fend off your drones. The main issue is that there are a few weapons that are clearly better than the rest. (Looking at you, Flak Cannon Mk I)
@@bobthebox2993 The #1 thing FTL fails at (from this video) is the "impossibly high skill ceiling" part and generally needing to pour dozens of attempts in to memorize a lot of the events. I didn't say that FTL failed at everything here though, it's obviously a good game but it just surprises me that it succeeded despite these very glaring problems. Some people are just masochists though.
I remember one time I made a small roguelike game and the way items worked was similiar to risk of rain, you could get multiple of the same item and your items could activate other items and at some point in the game you would get so absurdly op that your items activated constantly until your target was dead
This is a comment either adding or subtracting for your self esteem, but it’s actually a comment-like, so it’s a random outcome every time you read it no matter what you do to effect it.
Sorry, , but just stick to it man, it happens to the best of us, sometimes we make a Noita, and sometimes you can turn out a Spelunky, but just keep trying and I'm sure you can make something no one will care about eventually.
Don't forget to throw in the possibility of RNG completely destroying any hope of continuing a run! Gotta keep the good players on their toes, by negating all skill with a luck wall!
@@ethanhayes7038 I was more thinking of traditional rogue likes, like Dungeon Crawl Stone Soup or Tales of Maj'Eyal, where sometimes you can just get a bad hand, and have a couple very strong enemies spawn too early and destroy you. (mostly an issue with DCSS, because TOME4 has counterplay to bad RNG, and potential revives, but it's still present)
Reaching the end of the game the player comes across a wall that rolls a 1D100000 dice. which requires the dice to land on the highest number. Furthermore to reach the end of the game takes a very very long time. Each run taking 1000 areas to teach the end of the game to even have a single roll to try and get past the wall. And if the roll fails it's right back to the start of the game again, with zero upgrades and not keeping any items from privious runs. ... That could actually be a good game if done right.
Great way to fail: Introduce RNG elements arbitrarily but also inconsistently, in parts of the game that rely on core game mechanics, and put little or no thought in how the RNGs are generated. Random.Range is more than good enough for these functions. That way, the game will feel hard no matter your skill level!
i think its time to make a frustrating roguelike with items that dont tie into eachother, and requires extreme reflexes in an arena with no random generation
I would love to see how to fail at making a boomer shooter or doom clone. This is honestly a great way to learn though at what mistakes you shouldn’t make!
“Just add a dash mechanic and call it good” God this is so incredibly, painfully accurate. Seems like every dang game nowadays has a dash and it feels like something the devs thought was a “must have” rather than it actually complementing gameplay and movement mechanics
I like dashes tho, they feel nice and that’s probably why. But it always feels cool when someone implements something different (especially when they do it well)
A lot of roguelikes and games in general nowadays just don't know how to make enemies fair or proper player scaling. They just throw bullshit at you and then when you die the response is "lolmao you're just mad you died skill issue"
And roguelites are like, lol, keep grinding and you'll beat it eventually! And mobile games are like, lol, pay $1.99 for a boost and you'll beat it! And AAA games are like, lol, turn the game to casual story mode!
This is the funniest series of videos on gamedev. Actually If we take pointers from it to what not to do then we could actually make good games. It is very educational if looked from the correct prespective
I love videos about hot to fail. It's really funny, but also informative :D I really want to see more videos like this and would be good to see about tower defense games :)
That might be good, I'll have to learn more about what actually makes a TD game good first though. But eventually I'm sure I'll make an episode on that.
Here me out tho. If you go to steam and click the "roguelike" tab under "categories" then you'll see games like Elden Ring, Armored Core 6, Lords of the Fallen, Remnant 2, Jedi Survivor, and other games that are in no way going to be even confused with roguelikes because according to steam is just a fucking joke and as long as your game gets enough attention then it's a rogue like... Also according to steam Soulslike falls under the Roguelike category lol
[Insert Self-Esteem Boosting Comment] I really enjoy this series. Always funny. Actually had some time to pop in and watch the premiere. I need to get back into game dev so I can start failing too! 😄
This is true! A fighting game for sure takes some incredible amounts of balancing, and I suppose any Moba would too, especially if different options are provided for playable characters or units. Great point! :D
It's gotten to the point for me that seeing the roguelike tag on steam is a red flag. So many of them have great mechanics and would be excellent as other genres like metroidvanias or action platformers, but are held back by a sea of messiness because of procedural generation, RNG, weird items and design choices, insane difficulty for no reason, etc.
My favourite one is roguelike deckbuilder because they’re always clearly inspired by Slay the Spire except they end up just making it again but literally almost always worse (except for visuals, but almost anyone could make better visuals than spire lmao)
I just love how your videos are slowly pushing me towards the BEST way of failing - towards never trying =) You make it sound so hard and hopeless, that giving up seems more and more of an option rather than an excuse! /j
Don't give up, game dev is hard, but seeing the ways others have failed can actually help you avoid the same pitfalls, that's all this series is about really. :)
Ironically enough I have been contemplating a roguelite game for a while now but I am no game dev Just, writer and voice actor with too many ideas and no skill... But have a friend who's good at art... But that's about as far as we got cause none of us can code ;3;
If your aiming to make an annoying game that just isn't fun to play, make the game extremely long. What do I mean by extremely long? I don't know, 100 floors? If you really want too, maybe even longer. But dont spend too much time making every floor different. Thats just making every floor slightly unique. This way, you'd be playing for hours on a single run and could suddenly die to permadeath. Also, add instant death holes and spikes.
@@EdKolis Most of roguelikes have less (like, 30 or 50 floors), but Lost Labyrinth has 150 floors. You can also win by collecting 9 pieces of staff, but the game spawns approximately 1 piece per 10 floors, which makes 90 floors, which is close. Touhou Genso Wanderer Reloaded has dungeon or dungeons with nearly 100 floors, but I don't know precisely, didn't play it. Also some roguelikes make you walk all the way back from the bottom, which makes them twice longer.
You can make a fast character that fights really fast and interesting ennemies. Nope, that's bad, make a slower character who has to frame perfect dodge / counter the slow boss attacks. And the HP bar is huge. And there is 3 phases. Yeah this is perfection, the player has to hit the 0,001 frame, that's skill and fun at all right ? There is good music and design after all. And they can speedrun it, right ?
Direct control of a single unit, with a grid like style of movement where combat takes place in the same instance (And is not some other space like Final Fantasy), in which time only moves forwards as you take actions. Perma death, Proc Gen, and so on are not necessary (granted this is rare). If the above definition is true, it's almost certainly a roguelike. Meaning, you can't call 99% of indie games Roguelikes because of those two qualities alone.
Only as I was making the first episode did I realize the similarities, hopefully I got my own spin on it though, while of course taking inspiration! :D
Lol great video. I'm actually working on a Roguelike and this is helpful, makes me realize some things you mention in the video that I wasn't aware I was doing like doing certain things because that's how other roguelikes do it, basically trying to fit my game into the genre instead of making it the way I want as you said.
Don't let a genre define what a game should be. Some of my favorite games are the ones that blur genres together to make something weird and interesting.
Playing quite a bit of those I notice another trend that makes for great “worst” rugelite: shty meta progression that adds 0 depth aside from flat stat changes combined with locking any progress behind area/bosses. 15 min run and you died to first boss despite killing 1000 enemies? Yikes, not even one meta currency to progress
"After all, the only skill the player should have to improve at is reflexes." Bro... you just described why I hate deepwoken. These are words I've been wanting to speak but have never been able to.
So basically "Hotline Miami"? Reflex/precision based combat, super hard, static levels, weapon veriety is mostly just cosmetic, few enemy types, and camera tilting back and forth with shifting colors to get that nausia feeling. Great game though 👍
I haven't been failing anymore, because it's been a while since I stopped trying game design. Have you ever made videos about interactive fiction and visual novels?
That first point is so good... and I think that's something I started as such a victim of... Trying to pick a genre first... now I understand, games are just games, genres are just catergories we put them in to define them, but you can take aspects of many genres in one game and still have it work.
the part i love about this series is that by simply thinking about what he says in reverse it becomes a tutorial on how to make a GOOD GAME. thats right. this is actually a tutorial on how to SUCCEED in making games!
Vampire survivors is called a RougeLike by the devs. But it doesnt have much weapon sinergy, everything just destroys in a different way. And there is no random generation, most of the maps just loop. Heck the enemies arent even random. And the upgrades you can get, are mostly just small stat boosts. They failed at being a RougeLike, but succeeded in being a Bullet Heaven.
yeah, and they've pretty much succeeded at creating a new genre being called "survivors like" spawning a bunch of copy cat's and even legit games based of that game play loop.
@@Artindi I personaly dont like "Survivor Like" as that name feels a little to much like survival (a very different genre) for my liking, also it sounds little too much like the game is inspired by Surviverio, which is a cheep copy of Vampire Survivors.
@@Swagpion I relate to that, I don't really like it either, but they didn't ask me when the world decided it was a thing. :/ Side note: it's "SurvivorS like". with an S at the end, but yeah, still sounds too close to a plan old survival if one isn't paying attention.
Make sure that the gameplay at the beginning of the roguelike is exactly the same as the gameplay at the end of the roguelike, but with bigger numbers. Nothing brings the player a better feeling of progression than doing 10000 damage to a million health enemy, when they were doing 10 damage to a 100 health enemy earlier in the game. As we all know, this is the best and only way to make sure the player knows they are growing while fighting stronger enemies.
Roguelike or roguelite, is the game like rogue or similar to rogue? I'm pretty sure nobody even knows what rogue is let alone play it. So why is the genre titled with rogue?
It is used basically as a marketing buzzword. Most of the audience do not know what it means, and it makes the game appear more interesting. If they called it "run-based game" or something, it would not be so interesting. Of course, there are also millions of people who have played Rogue and know what roguelike means, and will complain, but they are still a minority.
Oh wow great video! I don't know if I like more the minimalist art style of the video or the music that absolutely collides with the feelings you have while playing roguelik/tes 😂 Thanks for the non-white background I think that It could destroy me, nice touch.
I have another nice advice. Roguelikes are games about randomness, which means that winning or losing must depend on your luck. If you are lucky, the game will give you great items and lots of resources, and if not, the game will give you bad items and not much resources, so you will lose. Not every roguelike run should feel fair.
No, i don't think so. Roguelikes can feel being about luck, but it's actually about the Synergies and specialization. Certain Classes are more favored than others and therefore should require a higher Stat increase from Perks and Gears to be somewhat balanced. I feel like it should be better to encourage experimenting with perks, Gear and/or Classes rather than just copying other Games. Also gathering Knowledge as a Player about mechanics and all is also a big part of a Roguelike. Meta Progression can also be a thing to take into account.
You could also add ways the player can gain power if they find themselves in a doomed run. Optional levels to farm for power that require some skill to get into, or a "gambling room" that rerolls all of your items. The run shouldn't feel fair, but if it feels hopeless too early the player might just restart, and IMO the best gameplay is wrestled from the maws of impending doom
TLDR: Roguelikes are RPGs, or Solitaire is a roguelike. Fight me. I've actually made a traditional roguelike game a couple times before, as a personal hobby and kind of meditation, though it definitely wasn't my first attempt at making a game. I, too, STRONGLY suggest you NOT make a roguelike game as your very first. They have a huge list of features that require a fair bit of knowledge, insight, and tools to implement. Instead, do a little warmup by making the simplest kind of game you can imagine. Even something like Pong or Asteroids will teach you an enormous amount about what you did not know you did not know, and that will help you when you decide you're ready to make a roguelike. And when you do, let your inspiration and experience guide you. Don't constrain yourself to a specific style of game. If that's really what you want to make, it's what your game will become. Traditional roguelikes are top-down, turn-based RPGs with procedural generation, randomized elements, and tile graphics, where you control a single main character and all progress is lost upon failure. They typically track a high number of statistics for all objects in the game from player to NPCs to environment. Combined with a crazy number of possible features, equipment and items, the world becomes deeply compatible and interactive and allows for an immense amount of strategy and variation. Most of them also have a victory condition, upon which a "run" is considered a "win." The most famous old roguelikes tirelessly spent decades accounting for every possible unique combination of their hundreds or thousands of game elements. Modern roguelikes mechanically incorporate procedural generation, randomized elements, and usually significant or total loss of progress upon the end of a run. More specifically, they use a game loop in which even the earliest elements of gameplay are revisited upon reaching a point of success or failure. Increasingly commonly, what have become known as the subgenre of "roguelites" (which are just a type of roguelike now much as a square is a more specific type of rectangle) add meta-progression in which some amount of progress is gained and/or retained between iterations of the main game loop (e.g. "each run".) The requirement for turn-based gameplay, or top-down/tile-based graphics have been cast to the wayside and modern roguelikes are increasingly action-based and now extend to all genres from platformers and ARPGs to shooters, card, and strategy games. They also often include controlling an entire team of characters, and even multiplayer co-op, and it seems to be every day that some new element is tested in the genre and risks spinning off into something new. Hence, "roguelite" can be treated the way that "arena shooter" or "battle royale" are both subgenres of shooters (whether or not they're FPS.) It's like back when people were really prickly about whether or not you specified a game was an FPS or a third-person shooter. This can only be pushed so far though before the genre loses all consistency, so I think the defining factor of what is and isn't a roguelike is now the inclusion of randomized elements, regular loss of progress, and reuse of most or all content in the main game loop, but it has to be specified that you embody a role in the game world. (I.E. it's an RPG. Because otherwise, solitaire is a roguelike. You see why classification is difficult?) It's no longer mutually exclusive with other genre of games like it used to be, just as RPG and FPS are no longer sufficient to set expectations for what kind of experience a game will provide. It's no longer referencing the Rogue package of features as a whole in the same way that Soulslike now basically only means having a stamina bar for attacks and dodging, a dodge roll with i-frames, and save points to unlock that refill your resources.
Yes, I want one of my first games to be a Roguelite, NO, its not because I think they're easy and take no effort lmao I've just already thought of a lot of different ideas for the world and the game I want to make, and I'm taking influence from a lot of other things, namely, The Binding Of Isaac with the overall gameplay and combat style of Breath Of The Wild. And an art style I can only describe as Dark Fantasy Slime Rancher lmao, for those smooth and cartoonish textures, but the overall atmosphere of a Souls game.
"Make a game as fun as possible first" these are the games i want to make. But then I fail at theorycrafting because one of them is a tower defense. A turn based tower defense. Why turn based? I like micromanaging towers but my speed in real life is too slow. So to take off the pressure, it's turn based. (The stop motion kind) But the first game I want to make... idk what it is. The current itteration of it (in my head and on paper) is a space invaders kind of game cause the enemies are in the sky ahead of you. The difference? You move across a grid and shoot them down in all 4 cardinal directions. You also move in those directions and move one gridspace at a time. But the gimmick that stayed near the start of this idea?... Is you switch between air and ground like some sort of mech. I think it'd be fun and it seems original enough. Nothing is like it.
Defining rogue like. I go pew pew pew, ahhh I dead, hmmm I take this now, pew pew ! Ahhh I dead........ WOAH LOOK! Pew pew! I WIN! Oh, that's level one? fuck.
Oh for sure other genres need plenty of balancing as well, and I did only say "probably" But I appreciate the feedback and would love to hear what other genres you think require a more delicate balancing and playtesting. :)
@@Artindi I would argue rogue-likes actually require some of the least balancing out of any kind of game out there, the idea being that "bad" runs through RNG are expected, so imbalance can ultimately be dismissed as an inevitability that you just need to persist through. At least, thats what a lot of them seem to rely on, with maybe a few systems designed to negate some of the "rogue" nature to be more forgiving, but I've played quite the handful since everyone seems to be just shoehorning the model right on in there like its nothing. Things like MMOs, which require measuring global player-to-player scaling vs enemy scaling at the same time, while also keeping an eye on the inevitable powercreep that occurs through persistent updates can be more difficult. Fighting games, while not more complex, can be astronomically more difficult to get "right" since the crowd that plays them varies so greatly and has such high expectations and can be very picky. The strategy games that resemble excel spreadsheets might be the most daunting of all. I'm confident I could make an ok roguelike. I am not confident I could make something like those in a competent manner, unless it was really, really barebones. Nowadays, roguelike just seems to be an attractive gimmick for people who need games to generate their own content without having to micromanage or manually pad things out later. I'm not saying its easy, especially if it's expansive, but I don't see a massive balancing act, just a pile of tedious work with the occasional emergent property you may not have accounted for initially.
Here's another way to fail: give the player 2 upgrade options after a stage, both of which are antisynergistic to the player's chosen class. This way, he has to pass and get a measly 10 or 20 gold as compensation. Surely feels great.
Im not a Coder so i wouldn't be able to make a game I would love to play. but It always irked me how POORLY copy and pasted a lot of these roguelikes are without leaning into any fun innovation, style, or mechanic,. Idk how many Vampire survivor clones there are at this point.
Make sure your weapon and item descriptions are just flavor text so the player has to open a wiki every time they find something new to make basic informed decisions. Make sure there is no reliable way to heal damage. After all, taking damage in a highly random game is ALWAYS the player's fault. Make sure the weapons the player finds vary dramatically in power, and balance enemy health against the strongest weapons the player will almost never have. Make sure NPCs that might help or even provide tradeoffs are randomly absent from 90% of runs. If there's a way to skip the early game, make sure unlocking it is more difficult than beating the game normally, or make it a secret the player will never find.
The “empty room with no random generation” is becoming weirdly common these days for some reason Other than that, great video. I’m starting work on a roguelike myself and this inspired me to rethink some design decisions
In this video you create an idea of a good roguelike, by saying waht a good roguelike is and then saying "do the opposite if you want to fail", meaning that you could just do that "do the opposite" thing at the end, rather than pausing all the time to do the thing. Also, I, the viewer, who wants to fail, is also positioned as an idiot, who does not understand anything AND someone who does and chooses to do the opposite. I think. Therefore I fail.
this video is just Inscryption in a nutshell, a game that only ever received one real update because it launched so bad that it was full on illegal, and thats without how scummy the game was and also, weirdly, Hades, but Hades was actually pretty cool, i wish the devs would make a Transistor sequel tho
Well you are basically describing spelunky 2 and that was quite successful so i guess the takeaway is that its best to just make a game bug free And aestetically pleasing and have an established franchise and famous TH-camrs who play your game
I really don't like the original Rogue or this genre of likes and lites at all. I know there's a niche market for it but I feel that it overstayed its welcome when there were way too many coming out on Steam around a decade ago.
My rouglike game for graduation work is the most imposible. Not because one of the every possible enemy might want to turn his collider off and become mr. Invincible or smth....
Excuse my ignorance but even though i got the sarcasm, i stll feel like roguelike are the easiest & laziest to make compared to the other genres. It feels like a chore & the market for it is oversaturated. It's a just a new spin of nes/snes game difficulty but more annoying because you get better by unlocking stuff then losing the run Example: Ninja Gaiden or Catlevania 1 vs any roguelike of today. In roguelikes, you beat it by unlocking new skills/weapon/gear by dying alot whereas the nes games didn't have that. It was just you & your reflexes
I know that the "how to fail" gimmick is often used jokingly, but it is often more useful to look at what failed rather than what succeeded to avoid a Survivor bias
Looking at what does well is super great, but looking at what does poorly is often just as helpful. :)
i like survivors, you mean, like vampire survivors? that shit is GREAT. Or you mean, other time of survivors and i don't get it?
@@viniciusschadeck4992 Survivor bias is when people focus only on what succeeds and overlook what does not, sometimes resulting of a false conclusion of how the results came to be.
@@Artindi i totally are thought about other stuff LOL
@@viniciusschadeck4992its useful to consider the chance of survivors bias.
A lot of successful games having thing X doesnt mean that thing X is necessarily the key to success, it could be that a whole lot of really bad games have it too, and its not necessarily good or bad
or potentially it could even be actively bad, but the good things the successful ones outweighed it
This one is especially interesting to me because of how well-regarded FTL: Faster Than Light is yet it does over half of the things listed in this video (I myself have 1500 hours in it), especially if you compare the base game to it's largest overhaul mod "Multiverse" which mostly attempts to patch up most of the "failures" of the base game (including some that aren't even mentioned in this video, such as losses that are out of the player's control or having the game end in the same place always), yet there are some people who are like "no we don't like this we want more similar to vanilla FTL."
FTL also succeeds at a lot of these though. Items, upgrades and crew members do combine in different ways that allow for different approaches and some enemies interact directly with these different attack approaches in meaningful ways.
A run where you focus on boarding enemy ships, might give you trouble when you cross paths with an automated ship, unless you're boarding with lanius crew, in which case it's basically a free win. Some ships are really weak against drones, while others might have a lot of ways to fend off your drones.
The main issue is that there are a few weapons that are clearly better than the rest. (Looking at you, Flak Cannon Mk I)
@@bobthebox2993 The #1 thing FTL fails at (from this video) is the "impossibly high skill ceiling" part and generally needing to pour dozens of attempts in to memorize a lot of the events. I didn't say that FTL failed at everything here though, it's obviously a good game but it just surprises me that it succeeded despite these very glaring problems. Some people are just masochists though.
@@bobthebox2993
Burst Laser Mk II.
I remember one time I made a small roguelike game and the way items worked was similiar to risk of rain, you could get multiple of the same item and your items could activate other items and at some point in the game you would get so absurdly op that your items activated constantly until your target was dead
oooh that sounds fun is it on the internet
This man just roasted the fuck out of the entire concept of Geometry Dash
Ah, yes, my favorite rougelike
did you just describe the development cycle of risk of rain 2
Mobile indie games all follow your guide. Thanks
my guides are quite popular for that platform for sure! ;)
This is a comment either adding or subtracting for your self esteem, but it’s actually a comment-like, so it’s a random outcome every time you read it no matter what you do to effect it.
Have you done "How To Fail At Making A Simulation Game" yet?
Not yet, but eventually I will. :)
"MEME COMMENT"
boost!
Instructions unclear, accidentally developed Hades
Sorry, , but just stick to it man, it happens to the best of us, sometimes we make a Noita, and sometimes you can turn out a Spelunky, but just keep trying and I'm sure you can make something no one will care about eventually.
hell of a game.
@@vizthexis… is that a pun?
@@nobleninja94 yes.
Finishing Hades is a great wake up call. "Oh wow, the credits keep going and going and going"
Sea of stars as well.
Don't forget to throw in the possibility of RNG completely destroying any hope of continuing a run! Gotta keep the good players on their toes, by negating all skill with a luck wall!
Ah, the Binding of Isaac approach.
@@ethanhayes7038 I was more thinking of traditional rogue likes, like Dungeon Crawl Stone Soup or Tales of Maj'Eyal, where sometimes you can just get a bad hand, and have a couple very strong enemies spawn too early and destroy you.
(mostly an issue with DCSS, because TOME4 has counterplay to bad RNG, and potential revives, but it's still present)
Ah yes darkest dungeon approach (though the RNG can also work in your favor when your dudes survives multiple death door check back to back)
Reaching the end of the game the player comes across a wall that rolls a 1D100000 dice. which requires the dice to land on the highest number. Furthermore to reach the end of the game takes a very very long time. Each run taking 1000 areas to teach the end of the game to even have a single roll to try and get past the wall. And if the roll fails it's right back to the start of the game again, with zero upgrades and not keeping any items from privious runs.
... That could actually be a good game if done right.
Great way to fail: Introduce RNG elements arbitrarily but also inconsistently, in parts of the game that rely on core game mechanics, and put little or no thought in how the RNGs are generated. Random.Range is more than good enough for these functions. That way, the game will feel hard no matter your skill level!
i think its time to make a frustrating roguelike with items that dont tie into eachother, and requires extreme reflexes in an arena with no random generation
Exactly! That's what I'm talking about!
Actually if you do it right it could be fun, come to think of it. :)
Maybe if it wasn't defined as a roguelike, I think maybe it could be good
Ah yes, an arcade game! XD
That basically already exist, it's called Hotline Miami
@@FloatingOerit's more strategy than reflexes
I would love to see how to fail at making a boomer shooter or doom clone. This is honestly a great way to learn though at what mistakes you shouldn’t make!
There are so many categories I want make a how to fail video for. One day I'll have the time. :)
Oh that one's easy. Prodeus
Thank you for reminding me to reexamine every single one of my game design decisions in excruciating detail
Oh, happy to help. :)
I like how you both roasted and validated the roguelike community here.
One of my greatest achievements. ;)
guys with the berlin interpretation tattooed on their backs sobbing tears of joy as their decade-long passion project gets its third download
Meanwhile a roguelite first person shooter called Rogue 2 gets half a billion dollars in sales its opening week
“Just add a dash mechanic and call it good” God this is so incredibly, painfully accurate. Seems like every dang game nowadays has a dash and it feels like something the devs thought was a “must have” rather than it actually complementing gameplay and movement mechanics
I like dashes tho, they feel nice and that’s probably why. But it always feels cool when someone implements something different (especially when they do it well)
A lot of roguelikes and games in general nowadays just don't know how to make enemies fair or proper player scaling. They just throw bullshit at you and then when you die the response is "lolmao you're just mad you died skill issue"
And roguelites are like, lol, keep grinding and you'll beat it eventually!
And mobile games are like, lol, pay $1.99 for a boost and you'll beat it!
And AAA games are like, lol, turn the game to casual story mode!
The Binding of Isaac: Repentance
This is the funniest series of videos on gamedev. Actually If we take pointers from it to what not to do then we could actually make good games. It is very educational if looked from the correct prespective
yeah, hopefully pointing out the dumb things we do sometimes will help us recognize them before we do them. :)
I love videos about hot to fail. It's really funny, but also informative :D I really want to see more videos like this and would be good to see about tower defense games :)
That might be good, I'll have to learn more about what actually makes a TD game good first though. But eventually I'm sure I'll make an episode on that.
@@Artindi I'm very appreciate that
Agreed, they are a pain in the ass to write, but they are easily worth the pain.
Instructions clear, made a bad game
Haha! That's excellent. Particularly laughed at wasting time arguing about definitions.
guilty.... :/ Glad you enjoyed!
just call your game Roguelike, doesn't necessarily need to be a Roguelike
technically counts i guess
Bit less confusing I suppose. :)
That's what everyone does these days anyway.
I like the creative way you're using to show how a decent game should feel like
Here me out tho. If you go to steam and click the "roguelike" tab under "categories" then you'll see games like Elden Ring, Armored Core 6, Lords of the Fallen, Remnant 2, Jedi Survivor, and other games that are in no way going to be even confused with roguelikes because according to steam is just a fucking joke and as long as your game gets enough attention then it's a rogue like... Also according to steam Soulslike falls under the Roguelike category lol
yeah, when you break it down far enough, almost any game is a roguelike. lol :)
[Insert Self-Esteem Boosting Comment]
I really enjoy this series. Always funny.
Actually had some time to pop in and watch the premiere.
I need to get back into game dev so I can start failing too! 😄
Glad you could make it! Even if you don't do a lot, game dev is always a great hobby, even if you fail. :)
0:20 fighting games, which require years to make and balance even at lower levels of polish:...
This is true! A fighting game for sure takes some incredible amounts of balancing, and I suppose any Moba would too, especially if different options are provided for playable characters or units. Great point! :D
It's gotten to the point for me that seeing the roguelike tag on steam is a red flag. So many of them have great mechanics and would be excellent as other genres like metroidvanias or action platformers, but are held back by a sea of messiness because of procedural generation, RNG, weird items and design choices, insane difficulty for no reason, etc.
My favourite one is roguelike deckbuilder because they’re always clearly inspired by Slay the Spire except they end up just making it again but literally almost always worse (except for visuals, but almost anyone could make better visuals than spire lmao)
I just love how your videos are slowly pushing me towards the BEST way of failing - towards never trying =) You make it sound so hard and hopeless, that giving up seems more and more of an option rather than an excuse! /j
Don't give up, game dev is hard, but seeing the ways others have failed can actually help you avoid the same pitfalls, that's all this series is about really. :)
A rougelite that I think followed all of these steps perfectly is: Splatoon's Side Order DLC
Hello, I'm Brazilian, and I liked your video style, and I'll follow more content like this, thank you very much for making me laugh!
Happy you enjoyed it! :D
Ironically enough I have been contemplating a roguelite game for a while now but I am no game dev
Just, writer and voice actor with too many ideas and no skill...
But have a friend who's good at art... But that's about as far as we got cause none of us can code ;3;
If your aiming to make an annoying game that just isn't fun to play, make the game extremely long. What do I mean by extremely long? I don't know, 100 floors? If you really want too, maybe even longer. But dont spend too much time making every floor different. Thats just making every floor slightly unique. This way, you'd be playing for hours on a single run and could suddenly die to permadeath. Also, add instant death holes and spikes.
Yeah, if each level doesn't feel like progress and like your in a new environment, it gets stale real quick.
Yeah, whose bright idea was it to make a roguelike with 100 floors anyway...
@@EdKolis Most of roguelikes have less (like, 30 or 50 floors), but Lost Labyrinth has 150 floors. You can also win by collecting 9 pieces of staff, but the game spawns approximately 1 piece per 10 floors, which makes 90 floors, which is close.
Touhou Genso Wanderer Reloaded has dungeon or dungeons with nearly 100 floors, but I don't know precisely, didn't play it.
Also some roguelikes make you walk all the way back from the bottom, which makes them twice longer.
@@tehteh9893 I was thinking of Angband, which originally had 50 floors but it was expanded to 100. 150 is just insane!
Devil May Cry games have 100 floors in bloody palace.
Would play again.
You can make a fast character that fights really fast and interesting ennemies.
Nope, that's bad, make a slower character who has to frame perfect dodge / counter the slow boss attacks. And the HP bar is huge. And there is 3 phases. Yeah this is perfection, the player has to hit the 0,001 frame, that's skill and fun at all right ?
There is good music and design after all.
And they can speedrun it, right ?
Roguelike have dashes and need Reflexes? What are you talking about? TRUE roguelike is rogue himself, angband, dcss, Pixel Dungeon and so on.
Direct control of a single unit, with a grid like style of movement where combat takes place in the same instance (And is not some other space like Final Fantasy), in which time only moves forwards as you take actions.
Perma death, Proc Gen, and so on are not necessary (granted this is rare). If the above definition is true, it's almost certainly a roguelike. Meaning, you can't call 99% of indie games Roguelikes because of those two qualities alone.
Permadeath is utterly core to being a roguelike. It's one of Rogue's most defining features along with procgen.
Thought this channel had way more subscribers than it does, great work! It's like terrible writing advice but game dev!
Only as I was making the first episode did I realize the similarities, hopefully I got my own spin on it though, while of course taking inspiration! :D
Lol great video. I'm actually working on a Roguelike and this is helpful, makes me realize some things you mention in the video that I wasn't aware I was doing like doing certain things because that's how other roguelikes do it, basically trying to fit my game into the genre instead of making it the way I want as you said.
Same here, except i am at the planning phase and i was leading the game towards a bad direction.
Don't let a genre define what a game should be. Some of my favorite games are the ones that blur genres together to make something weird and interesting.
Im about to release my demo, let me know if I failed.
Playing quite a bit of those I notice another trend that makes for great “worst” rugelite: shty meta progression that adds 0 depth aside from flat stat changes combined with locking any progress behind area/bosses.
15 min run and you died to first boss despite killing 1000 enemies? Yikes, not even one meta currency to progress
"After all, the only skill the player should have to improve at is reflexes."
Bro... you just described why I hate deepwoken. These are words I've been wanting to speak but have never been able to.
Instructions unclear, mistakenly made a successful Roguelike
You had one job... one job! ;)
So basically "Hotline Miami"? Reflex/precision based combat, super hard, static levels, weapon veriety is mostly just cosmetic, few enemy types, and camera tilting back and forth with shifting colors to get that nausia feeling. Great game though 👍
I haven't been failing anymore, because it's been a while since I stopped trying game design. Have you ever made videos about interactive fiction and visual novels?
Yes, there is a "How to fail at interactive stories" video, but most people don't like it. It's pretty harsh and blunt. Good luck! :D
That first point is so good... and I think that's something I started as such a victim of... Trying to pick a genre first... now I understand, games are just games, genres are just catergories we put them in to define them, but you can take aspects of many genres in one game and still have it work.
Me looking at the Vampire Survivors genre...
I wonder if Cult of the Lamb has done any of these yet
Playing roguelike games is a subtle sign of masochistic tendencies.
It would be fun to see how to fail at triple A developement, as an executive
the part i love about this series is that by simply thinking about what he says in reverse it becomes a tutorial on how to make a GOOD GAME. thats right. this is actually a tutorial on how to SUCCEED in making games!
psa: your indie game doesn't need to be played infinitely, it just needs to be fun with the first playthrough, ideally for a few hours.
I feel attacked /s (I should make more small games)
Vampire survivors is called a RougeLike by the devs. But it doesnt have much weapon sinergy, everything just destroys in a different way. And there is no random generation, most of the maps just loop. Heck the enemies arent even random. And the upgrades you can get, are mostly just small stat boosts.
They failed at being a RougeLike, but succeeded in being a Bullet Heaven.
yeah, and they've pretty much succeeded at creating a new genre being called "survivors like" spawning a bunch of copy cat's and even legit games based of that game play loop.
@@Artindi I personaly dont like "Survivor Like" as that name feels a little to much like survival (a very different genre) for my liking, also it sounds little too much like the game is inspired by Surviverio, which is a cheep copy of Vampire Survivors.
@@Swagpion I relate to that, I don't really like it either, but they didn't ask me when the world decided it was a thing. :/
Side note: it's "SurvivorS like". with an S at the end, but yeah, still sounds too close to a plan old survival if one isn't paying attention.
i cant break up the long empty hallways without making more long empty hallways
The "try to fail a little less next time" strategy is exactly how I've been playing rogue fable 3 but i don't think that's intended ;-;
You mean that's not how you're supposed to play roguelikes?
The very first one helped me the most. Thank you. But I did actually want a rogue like because it’s the hardest. The combinations my dude…
Make sure that the gameplay at the beginning of the roguelike is exactly the same as the gameplay at the end of the roguelike, but with bigger numbers. Nothing brings the player a better feeling of progression than doing 10000 damage to a million health enemy, when they were doing 10 damage to a 100 health enemy earlier in the game. As we all know, this is the best and only way to make sure the player knows they are growing while fighting stronger enemies.
Roguelike or roguelite, is the game like rogue or similar to rogue?
I'm pretty sure nobody even knows what rogue is let alone play it.
So why is the genre titled with rogue?
It's just the first of that style.
It is used basically as a marketing buzzword. Most of the audience do not know what it means, and it makes the game appear more interesting. If they called it "run-based game" or something, it would not be so interesting.
Of course, there are also millions of people who have played Rogue and know what roguelike means, and will complain, but they are still a minority.
Don't make your enemies interesting or unique. Just give them a shitload of health and shitty AI. THAT'LL KEEP 'EM ENGAGED FOR HOURS!
this is how Neon Abyss was made
Pixel Dungeon is pretty good. Its also the only RL Ive played. Can anyone suggest others for the Android?
Oh wow great video! I don't know if I like more the minimalist art style of the video or the music that absolutely collides with the feelings you have while playing roguelik/tes 😂
Thanks for the non-white background I think that It could destroy me, nice touch.
I have another nice advice. Roguelikes are games about randomness, which means that winning or losing must depend on your luck. If you are lucky, the game will give you great items and lots of resources, and if not, the game will give you bad items and not much resources, so you will lose. Not every roguelike run should feel fair.
No, i don't think so.
Roguelikes can feel being about luck, but it's actually about the Synergies and specialization. Certain Classes are more favored than others and therefore should require a higher Stat increase from Perks and Gears to be somewhat balanced.
I feel like it should be better to encourage experimenting with perks, Gear and/or Classes rather than just copying other Games.
Also gathering Knowledge as a Player about mechanics and all is also a big part of a Roguelike.
Meta Progression can also be a thing to take into account.
You could also add ways the player can gain power if they find themselves in a doomed run. Optional levels to farm for power that require some skill to get into, or a "gambling room" that rerolls all of your items.
The run shouldn't feel fair, but if it feels hopeless too early the player might just restart, and IMO the best gameplay is wrestled from the maws of impending doom
TLDR: Roguelikes are RPGs, or Solitaire is a roguelike. Fight me.
I've actually made a traditional roguelike game a couple times before, as a personal hobby and kind of meditation, though it definitely wasn't my first attempt at making a game. I, too, STRONGLY suggest you NOT make a roguelike game as your very first. They have a huge list of features that require a fair bit of knowledge, insight, and tools to implement. Instead, do a little warmup by making the simplest kind of game you can imagine. Even something like Pong or Asteroids will teach you an enormous amount about what you did not know you did not know, and that will help you when you decide you're ready to make a roguelike. And when you do, let your inspiration and experience guide you. Don't constrain yourself to a specific style of game. If that's really what you want to make, it's what your game will become.
Traditional roguelikes are top-down, turn-based RPGs with procedural generation, randomized elements, and tile graphics, where you control a single main character and all progress is lost upon failure. They typically track a high number of statistics for all objects in the game from player to NPCs to environment. Combined with a crazy number of possible features, equipment and items, the world becomes deeply compatible and interactive and allows for an immense amount of strategy and variation. Most of them also have a victory condition, upon which a "run" is considered a "win." The most famous old roguelikes tirelessly spent decades accounting for every possible unique combination of their hundreds or thousands of game elements. Modern roguelikes mechanically incorporate procedural generation, randomized elements, and usually significant or total loss of progress upon the end of a run. More specifically, they use a game loop in which even the earliest elements of gameplay are revisited upon reaching a point of success or failure. Increasingly commonly, what have become known as the subgenre of "roguelites" (which are just a type of roguelike now much as a square is a more specific type of rectangle) add meta-progression in which some amount of progress is gained and/or retained between iterations of the main game loop (e.g. "each run".) The requirement for turn-based gameplay, or top-down/tile-based graphics have been cast to the wayside and modern roguelikes are increasingly action-based and now extend to all genres from platformers and ARPGs to shooters, card, and strategy games. They also often include controlling an entire team of characters, and even multiplayer co-op, and it seems to be every day that some new element is tested in the genre and risks spinning off into something new. Hence, "roguelite" can be treated the way that "arena shooter" or "battle royale" are both subgenres of shooters (whether or not they're FPS.) It's like back when people were really prickly about whether or not you specified a game was an FPS or a third-person shooter.
This can only be pushed so far though before the genre loses all consistency, so I think the defining factor of what is and isn't a roguelike is now the inclusion of randomized elements, regular loss of progress, and reuse of most or all content in the main game loop, but it has to be specified that you embody a role in the game world. (I.E. it's an RPG. Because otherwise, solitaire is a roguelike. You see why classification is difficult?) It's no longer mutually exclusive with other genre of games like it used to be, just as RPG and FPS are no longer sufficient to set expectations for what kind of experience a game will provide. It's no longer referencing the Rogue package of features as a whole in the same way that Soulslike now basically only means having a stamina bar for attacks and dodging, a dodge roll with i-frames, and save points to unlock that refill your resources.
Yes, I want one of my first games to be a Roguelite, NO, its not because I think they're easy and take no effort lmao
I've just already thought of a lot of different ideas for the world and the game I want to make, and I'm taking influence from a lot of other things, namely, The Binding Of Isaac with the overall gameplay and combat style of Breath Of The Wild. And an art style I can only describe as Dark Fantasy Slime Rancher lmao, for those smooth and cartoonish textures, but the overall atmosphere of a Souls game.
Mean comment!
Disgusted retort and mean reply!
"Make a game as fun as possible first" these are the games i want to make. But then I fail at theorycrafting because one of them is a tower defense. A turn based tower defense.
Why turn based? I like micromanaging towers but my speed in real life is too slow. So to take off the pressure, it's turn based. (The stop motion kind)
But the first game I want to make... idk what it is. The current itteration of it (in my head and on paper) is a space invaders kind of game cause the enemies are in the sky ahead of you. The difference? You move across a grid and shoot them down in all 4 cardinal directions. You also move in those directions and move one gridspace at a time. But the gimmick that stayed near the start of this idea?... Is you switch between air and ground like some sort of mech. I think it'd be fun and it seems original enough. Nothing is like it.
Defining rogue like. I go pew pew pew, ahhh I dead, hmmm I take this now, pew pew ! Ahhh I dead........ WOAH LOOK! Pew pew! I WIN! Oh, that's level one? fuck.
You need to play other genres if you think roguelikes need more balance than any other. ffs dude.
Oh for sure other genres need plenty of balancing as well, and I did only say "probably" But I appreciate the feedback and would love to hear what other genres you think require a more delicate balancing and playtesting. :)
@@Artindi I would argue rogue-likes actually require some of the least balancing out of any kind of game out there, the idea being that "bad" runs through RNG are expected, so imbalance can ultimately be dismissed as an inevitability that you just need to persist through. At least, thats what a lot of them seem to rely on, with maybe a few systems designed to negate some of the "rogue" nature to be more forgiving, but I've played quite the handful since everyone seems to be just shoehorning the model right on in there like its nothing.
Things like MMOs, which require measuring global player-to-player scaling vs enemy scaling at the same time, while also keeping an eye on the inevitable powercreep that occurs through persistent updates can be more difficult. Fighting games, while not more complex, can be astronomically more difficult to get "right" since the crowd that plays them varies so greatly and has such high expectations and can be very picky. The strategy games that resemble excel spreadsheets might be the most daunting of all. I'm confident I could make an ok roguelike. I am not confident I could make something like those in a competent manner, unless it was really, really barebones. Nowadays, roguelike just seems to be an attractive gimmick for people who need games to generate their own content without having to micromanage or manually pad things out later. I'm not saying its easy, especially if it's expansive, but I don't see a massive balancing act, just a pile of tedious work with the occasional emergent property you may not have accounted for initially.
Here's another way to fail: give the player 2 upgrade options after a stage, both of which are antisynergistic to the player's chosen class. This way, he has to pass and get a measly 10 or 20 gold as compensation. Surely feels great.
Do immersive sim next for an even better definition debate
Do how to fail at RTS
Im not a Coder so i wouldn't be able to make a game I would love to play. but It always irked me how POORLY copy and pasted a lot of these roguelikes are without leaning into any fun innovation, style, or mechanic,. Idk how many Vampire survivor clones there are at this point.
My rougelike is a deck-builder, so therefore i can ignore all your advice!
Make sure your weapon and item descriptions are just flavor text so the player has to open a wiki every time they find something new to make basic informed decisions.
Make sure there is no reliable way to heal damage. After all, taking damage in a highly random game is ALWAYS the player's fault.
Make sure the weapons the player finds vary dramatically in power, and balance enemy health against the strongest weapons the player will almost never have.
Make sure NPCs that might help or even provide tradeoffs are randomly absent from 90% of runs.
If there's a way to skip the early game, make sure unlocking it is more difficult than beating the game normally, or make it a secret the player will never find.
sounds like Noita :D
very nice! hope to see deckbuilding games to this series
step 1: make a rougelike
step 2: you're done, make a better genre FFS
The “empty room with no random generation” is becoming weirdly common these days for some reason
Other than that, great video. I’m starting work on a roguelike myself and this inspired me to rethink some design decisions
In this video you create an idea of a good roguelike, by saying waht a good roguelike is and then saying "do the opposite if you want to fail", meaning that you could just do that "do the opposite" thing at the end, rather than pausing all the time to do the thing.
Also, I, the viewer, who wants to fail, is also positioned as an idiot, who does not understand anything AND someone who does and chooses to do the opposite.
I think.
Therefore I fail.
Live. Die. Repeat. Without variation. Without change. There. Failed.
Now do a vid with failing at failing.
this video is just Inscryption in a nutshell, a game that only ever received one real update because it launched so bad that it was full on illegal, and thats without how scummy the game was
and also, weirdly, Hades, but Hades was actually pretty cool, i wish the devs would make a Transistor sequel tho
I think you mixed "repeated attempts" and "reading the wiki".. There are so many wiki-required rouge likes.
Watching these videos after already starting on a game makes them feel like personal attacks sometimes
"old man yells at clouds"
Not sure what ya mean....
I should not watch this videos while drinking, looks like I cant take sarcastic advices.
Gonna go and make a game to fail, bye!
Well you are basically describing spelunky 2 and that was quite successful so i guess the takeaway is that its best to just make a game bug free And aestetically pleasing and have an established franchise and famous TH-camrs who play your game
You perfectly explained why I can't stand playing roguelikes.
Fun fact: QWOP is a roguelike, but the learning curve is extremely steep
Instructions unclear: I accidentally made the binding of isaac
I really don't like the original Rogue or this genre of likes and lites at all. I know there's a niche market for it but I feel that it overstayed its welcome when there were way too many coming out on Steam around a decade ago.
My rouglike game for graduation work is the most imposible. Not because one of the every possible enemy might want to turn his collider off and become mr. Invincible or smth....
to be fair, one of the most well know roguelikes is risk of rain are those levels are technically random
I followed your instructions, but Spelunky is already a game.
from 2:20 to 2:42 was just slander to my game and you didnt even targetted at me
Excuse my ignorance but even though i got the sarcasm, i stll feel like roguelike are the easiest & laziest to make compared to the other genres. It feels like a chore & the market for it is oversaturated. It's a just a new spin of nes/snes game difficulty but more annoying because you get better by unlocking stuff then losing the run
Example: Ninja Gaiden or Catlevania 1 vs any roguelike of today. In roguelikes, you beat it by unlocking new skills/weapon/gear by dying alot whereas the nes games didn't have that. It was just you & your reflexes
Yeah, they are easy to make, but hard to make right. So lot's of indies try and most of them fail hard. Good points! :D
This is underrated, can you do sandbox next
That's a good idea. I for sure plan on making more of these. Just need the time to do it. :)
If you do the upposite then you will end up with Enter The Gungeon or Exit the Gungeon if you know about it..
How to succeed at making a rougelike: Don't make a rougelike, pick literally any other game genre
You sound like minute science channel, forgot the exact name but cool!