Roguelikes, Persistency, and Progression 

Hey everyone! You might be thinking “wait, haven’t I already seen this video?” and the answer is - sort of! I wasn’t happy with the original version of this video, and found myself agreeing with some of the more negative comments that I received. Ultimately, while GMTK will always and inevitably involve my personal opinion, the original version of this video leaned too heavily into that and made it seem like my preferences were “correct”, and everyone else’s was “incorrect”. That’s not true, and so I’ve tried to make the video more balanced and evenhanded - by looking at the advantages and disadvantages of both including and not including persistent upgrades. I also felt like the original video had comments that seemed to be in favour of things I’ve vehemently fought *against* on Game Maker’s Toolkit! That being inaccessibility / gatekeeping and restrictive genre definitions. So I’ve reworded those. Anyway, I’m sure some people will be just as negative about me changing the video - but I think this was the right move for me and the channel. GMTK is a tricky beast because it balances opinion and education and, sometimes, I’m just going to get that completely wrong. In the interest of transparency, can find the original video here - ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-eXEQ8VBqXBY.html Cheers. Mark

I can't believe Mark's video died and got sent back to the beginning At least it kept some of his progress

My favorite Roguelike is Flappy Bird.

About that "weirdo backwards difficulty curve": That is only true if you look at the game as a whole over multiple runs. In each single run, the difficulty increases the longer you play. What's important, is that you as a player are constantly challenged. If your character is stronger on the 20th run you will get further in the game, but you will also have to play through harder levels which only now you are equipped to beat. To me that's just rewarding game design.

I like Hades' approach to the difficulty curve, first letting you power up, then unlocking extra difficulty (including restricting the previous power ups) after you've beaten the game.

Fact is I don't play those games for the feeling of "achieving something by my skills" but rather for the pleasure of playing and discovering things.

Rewatching this, there's two points I'd like to bring up. Firstly, I think there's a third pillar of the idea of a 'roguelike/lite' that was missed (which is missing for the variety of meme 'X is my favourite roguelike' comments) being 'an RPG system supported by either random or semi-random elements' (most frequently stat changing item drops). Most of the games mentioned as jokes involve procedurally generated challenges, but the way the player interacts with those challenges frequently remains static. With 'Arcadelikes', as I will dub games with procedural generation and perma death but no RPG systems, the player's ability to interact with the world changes far less drastically than it would in a true Roguelike. Arcadelikes usually use randomness to test the player's execution within the given system (to prevent rote memorisation and/or brute forcing one's way through a tricky section by just dying on it until you happen to the exact set of inputs). Roguelites, on the other hand, usually use randomness as a way to generate non-repetitive content in the place of the generic 'grinding' you might see in a normal RPG. Roguelikes, on the other hand, frequently are about testing the player's ability to make decisions, achieved by having a large number of variables. Thus, not only does the environment need to change, but so does the specifics of how the player interacts with it. I think it's also important to point out that, even in a standard Roguelike, the difficulty of the game isn't a flat line but rather a wide bar that contains any number of different lines. With an RNG influenced upgrade system, different runs will require a different level of mastery to be completeable. I'm sure anyone who's played a Roguelike for any length of time has run into 'the god run' where they get a windfall early on and ride that windfall far further than other runs would take them. As a player's skill increases, the number of RNG layouts they can potentially complete increases, but if you try to measure 'how long until a player beats the game once', that's going to depend on how long until the game spits out a seed that crosses their current skill level.

Most good roguelites make up that weird difficulty curve with increasing enemy strength and complexity. Compare the enemies from the hell area in Rogue Legacy to the ones in the first few rooms and its night and day. The attack patterns and more complex and deal more damage, as well as having more health, thus demanding more skill of the player to beat them.

I respect Mark Brown. He ended up not liking a piece of content he put out, so he fixed it until he did.

Another positive with Dead Cells is after you beat the game once the game lets you increase the difficulty for more content.

I prefer Rogue-lites because I'm bad at video-games, plain and simple.

I read the thumbnail as Roguelikes and Depression and was like yea same

But we all know one thing's for sure: Minesweeper is the best rogue-like

What I would say about Rogue Legacy is that, the game's difficulty doesn't decrease over time, rather that the game's goals change over time. When you first walk into the castle the game is difficult, but the aim is to survive one or two rooms.because that should get you all the money you need in order to purchase an upgrade. Then the price of upgrades increase. So now when you enter the castle you need to survive ten rooms to get enough coins - you're stronger, but the goal is also greater and more difficult. Your character is stronger, but if you don't play well enough to progress through those 10 or so rooms, then you don't get anywhere in the game. In a roguelike you have this same idea, except the goal isn't X amount money for upgrades, the goal is 'reach the final floor'. It's a fixed goal instead of an iterative one that changes to suit you. Sure, in Rogue Legacy you still need to beat the entire game in one go and when you do that you'll be much stronger. But the game is designed in a way that requires your character to be stronger from the first room to really stand much of a chance. I much prefer this video to the original you released on the topic, but I still think it unfairly categorises roguelite games as "pointless because you'll win eventually"

Whats great about dead cells however, is that in BC 4 and 5, the difficulty is so much higher than earlier boss cells, that you need to master parrying and timing of all enemies. It feels like it starts out as a rogue lite and eventually morphs into a roguelike

When I eventually beat a roguelite with my persistent upgrades, it never feels like it was only because of the upgrades that I won. The moment that you win is, as your graph here shows, the moment where skill and difficulty are matched. If my final run was too easy, then I would feel like it was just the upgrades, but that almost never happens; it's still thanks to my own skill that I managed to push through and win a little, or a LOT, sooner than my upgrades might otherwise allow. However, this still relies on the designers getting it right. The meaningfulness of the upgrades, the rate you get them, and the amount of the game that's reasonably playable without the upgrades, it all has to be carefully balanced to make that union point of skill and difficulty a satisfying one... but it can be done.

Huh. Well, damn, this actually shows a healthy amount of intellectual honesty to me, Mark. This is why I'm always here for these vids. Even if they are just re-uploads.

The Binding of Isaac does both. After meeting certain criteria you can unlock different characters that benefit a slightly easier playstyle, said characters will eventually get a permanent starting item/ability that they didn't start with, and you can donate money to the shop to make it so shops have a permanently bigger selection as you play. However, the game also moves the "difficulty goalposts" so as you get better and your normal runs get stronger, the game also gets harder with more difficult floors and new bosses to defeat and new floors and new bosses to reach after the "end" of the game. Playing fast allows you to reach a difficult boss rush mode for additional upgrades but these and other diversions are completely optional, mostly there to allow the more skilled players to be challenged.

I like how FTL has a lot of different cruiser designs and layouts, each of which provide entirely new ways to play.

"Well, time to go get killed again." I do so love this.

Awesome video Mark :) ! Though I really liked the old version as well, I must admit this one feels more "balanced", with great points on the positive and negative aspects of roguelikes and roguelites ! Thanks !

Thank you for taking your time to resubmit this video. I was one of those people that found it to be a bit too hard on roguelites, which I personally prefer heavily over roguelikes due to the persistent elements, and frankly, I was shocked by how much I was disagreeing with you... for the first time ever! This video I think does a much better job outlining the pros and cons of both. My main complaint with the first video was that the "backwards difficulty curve", as you put it, is actually more of a balancing act, rather than an inherent flaw of the subgenre. In fact, I would say that in regular roguelikes with no persistent upgrades, the difficulty for each run is actually higher, as you need to better tackle more content, and the later content is also more difficult than the earlier content, so your improvement rate over time must match that difficulty increase and in fact surpass it to actually beat the game. This is why those games are so taxing to the player, not because the difficulty is constant. On the other hand, in roguelites with persistent upgrades, the difficulty is actually closer to constant, as the content deeper into the game gets more difficult, but so does your power. It can be decreasing if the rate and power of those upgrades is much higher than the difficulty ramp-up of the content, but that is different on a per game basis depending on what the developer wants out of it.

Coming back to this with Hades fully out now, it does things amazingly well for a Roguelite. The upgrades aren't all that great. And mostly needs good play to take advantage of your upgrade. And the God Mode is wonderful for players who aren't great at the game, but can be turned off for the more hardcore players.

Dead Cells' difficulty curve is really odd, because it's at its hardest points both in the very beginning and very end (i.e. 5 bc's)

The Binding of Isaac's unlocks definitely change the game's difficulty, both making it easier and harder. A 100% save file has a dozen or so characters with different abilities and stats, plus hundreds of items in the world, as well as almost double the length of a run along with alternative floors (when you first start there's 6 levels, after a while it goes out to 10+). The sheer amount of items, that all interact with each other, means when you first start the game you get decent combos, but as you go the possibility of getting god like combinations both increases and decreases due to the bloated item pool. However, player skills also has to increase to proceed as the new floors are much harder, and some characters like The Lost (1 hit kill) are amazingly hard to play as.

This is a much improved video from the original It takes a real sense of humility to admit when we're wrong about something So I just want to say thank you for going through all this effort Thank you And good stuff as usual

“Unchanging difficulty” *ascension mode*

I value my time (and enjoyment, bc I do end up feeling like it was for nothing when I completely start over) over some nebulous sense of skill at a game.... so yeah, sign me up for Hades and Dead Cells haha

I remember when the term "rogue-lite" started showing up, it was used to describe games such as the Binding of Isaac, to distinguish them from turn-based "games that are like Rogue". Crazy how now BoI is being called a 'pure' roguelike, with rogue-lite used for games that stray even further from the formula. I've no problem with terminology changing over time, but now that the slope is confirmed to be slippery, there's an obvious problem. "Games that are like Rogue" (turn-based, tile-based, etc.) will never go away, simply because they're so easy to program for how deep their systems can be, which makes them a popular genre for amateur developers who want to try creating a game completely from scratch. And there's a real community of players too. The question is what to call them? They're Roguelikes, obviously, but now that that term has moved beyond a genre and mainly just describes level generation systems more than anything else, discussion in that niche community can maybe get muddied.

"A low skill player may never finish the game." = me 1000 hours in Binding of Isaac.

So happy to have found your channel Mark! I don't develop games, but I'm learning so much information I subconsciously knew from years of gaming, but probably wouldn't be able to explain. You're a great teacher!

Slay the Spire is my favourite Roguelike for various reasons, one of which is how perfectly it avoids both of the main issues that Roguelikes can have (both of which were pointed out in this video): 1. The issue of lower-skill players never being able to finish the game is avoided by having 20 different difficulty levels (or 'Ascension Levels' as the game calls them), allowing a very wide spectrum of players of different skill levels to be able to pick up the game and play through it comfortably. Not to mention the fact that the game also offers two different endings for each run; a more challenging ending that involves collecting 3 keys to go to Act 4 to defeat the Corrupt Heart, and a significantly easier ending that simply involves defeating the Act 3 boss; further increasing the spectrum of skill-levels that can enjoy this game even more. 2. The issue of lost runs feeling like a 'waste of time' is avoided by having the game not be very long (especially if you're going for the easier ending), as well as unlocking new cards and relics to potentially find on future runs. Most importantly though, apart from its roguelike element, Slay the Spire is a very deep turn-based strategy game. That level of depth lends itself greatly to new playthroughs because you'll always be learning new things and coming up with new ideas. So even in a lost run, you'll still gain some new things to think about in future as long as you didn't lose on purpose. It wasn't a waste of time.

I like both, but I really like it when the game gives me the option to grind past a difficulty bottleneck just in case that's how I'm feeling that day.

I love this remake of the last video, I feel like it's a solid improvement. Still split between roguelikes and roguelites. If I had to pick one, I'd probably say I prefer roguelikes because it means I'm improving at the game.

I like my every game to have a sense of meaning and achievement. I'll always go for the lites

I for one love the feeling of becoming more powerful in a game. Being the kind of player who's not complete garbage, but not very good unless I start no-lifing, it allows me to relax, have satisfaction and not sweat my ass off.

I feel something is missing from thoses definitions of "roguelikes" and "roguelites" One of the key points of earlier roguelikes was the consistency of game rules between player and game entities (ie: monsters, npc, objects) In rogue, if you drink a teleportation potion, you get transported to a random position. if you throw that potion on a rat, it get transported to. Likewise, in Spelunky, spikes will kill you as well as the monsters, but they are harmless if you walk through them. (I belive I've picked up on this idea from an interview of Derek Yu) This types of rules place a strong emphasis on learning the game (it work on top of the randomization). While a new player have to learn how to not get killed by traps, an advanced player can learn how to use said traps. This is why I'm highly anticipating Noita later this year. I'd be interested to know if you have thought on this @GMTK

Definitely more balanced arguments, and the tone fits way better now because of it. Really well done, and well argued. The only point I'd add is, as a (generally) bigger fan of roguelites, I don't necessarily agree that they're all inverse linear difficulty curves. Your argument at the end kind of contradicts the added counterpoints: yes roguelites can buff the player to an extent, but not creating safeguards for pure OP grinding is more the fault of the design than the (sub)genre. At some point you likely need to prove your skills as a player to get more unlocks, so the question is more about how to make those unlocks rewarding while also maintaining a larger difficulty curve. Which is a difficult balance to be sure (I think it works better as thresholds/set goals than with currency/shops), but to me, blaming roguelites as a whole is like throwing away checkpoints in a platformer because it's not the 'core' experience.

In regards to the difficulty curve for both genres, it doesn't feel like the game’s getting easier, it's more that you're becoming better, and gaining access to more difficult levels, until you beat the game for the first time; at which point I can imagine many roguelites/likes would get boring. The Binding of Isaac actually changes the difficulty of the same levels as you progress, I'm not exactly sure how or when, but I think after your first full run, and after each successful run for that matter, the game increases the difficulty of all levels so as to keep it challenging. The game also withholds like more than half of its content to be unlocked after your first successful run. I think that concept should be in all roguelike/lite games, otherwise it's pointless to continue playing once you 'beat the game' which you were mostly purely hooked on due to the challenge. Static difficulty -> progress ending with successful run -> increase of difficulty -> endless repetition.

Personally, Hades is the first rogue anything I've ever played. And I love it. I tend to be pretty bad at certain game types and I'm not the kind of person to keep playing a game I'm sucking at. For me, games being able to be beaten but sheer force of will and time isn't necessarily a drawback. There are people that just can't get good at certain formats for whatever reason. I have barely played any soulslike or fps games because I'm just so bad at them. Even if I'm deeply fascinated by the world and story (which is often my favorite part of games). I'm the person that plays games on easy mode because I don't want to get mad or frustrated or test my skill. I want to live in the world, and games that allow me to do that while also allowing another player to speedrun it at max difficulty that I would crap myself even dreaming of are probably some of my favorites. In short, I don't think ambiguity in player skill is always a bad thing. Especially if there's something that says, at the beginning, "Hey, if you're here for the story, or just don't wanna sweat about this, turn this on. You're not a worse player or a bad gamer for it." I think there's a lot of stigma with just playing a game to enjoy it and not be challenged, and it applies to many genres of game. And I think that whole concept is stupid. Games are meant to be enjoyed in whatever capacity you've come to games to find enjoyment. Be it a challenge, an escape, a brain bender, or a companion filled jaunt. Genre and the fulfilling goal of a game are both important, and one shouldn't be belittled or sacrificed in the name of the other.

Today on Toolkit; will Mark Brown stop mentionning Spelunky? Sources say 'no.'

I've spent the last few months wondering why Gungeon keeps my interest, but Dead Cells doesn't. Both have really great combat systems, so... what gives? I think you finally put it into words; in Gungeon, I don't feel like I'm grinding to unlock the tools I need to win. But that's exactly what it feels like to play Dead Cells.

Every time I watch one of your videos I realize I’ve already watched it. I love you.

Happy to have stumbled on your channel. Always have interesting videos. 👍

Hey Mark! Tbh I'm kind of glad to see the reupload and like it, as the last one did feel a bit matter-of-fact about roguelikes being better than roguelites. This one seems to acknowledge personal bias much more, which I appreciate. And overall, as an aspiring game designer I'm always glad to see videos like this examining the pros and cons of different yet similar genres. So thanks! I can certainly understand the preference for roguelikes, as the necessity of skill to emerge victorious is certainly a desirable feature for some, and the backwards difficulty curve of roguelites can be a bit odd. Although I'm personally a bigger fan of roguelites. I think for me that's because I'm such a big fan of RPGs, where you're ramping up the power of your character over time, and it always feels so rewarding for me when I finally reach the level of power I've spent all that time grinding for. So if I'm playing a roguelike, and spend all this time gathering power, and especially if I get a lucky run and get some sort of really strong random power up, then seeing myself die and get booted back to level 0 is so disheartening for me.

Honestly, I was more on your side of the argument when I first watched this video, but now that you have remade it, I've realized the depth and complexity afforded to the player with a rogue-lite system may actually be much better. I for one love pure and simple games, if you don't need an experience bar, don't use it. However, if the game is still possible to complete if the player decides to completely ignore that mechanic, than it is actually offering a lot of autonomy to the player on how difficult you want your game. Rather than a player having to choose Easy medium or hard, they can almost use their experience bar as a difficulty slider that can be tuned to their own liking.

Love the way you describe how the difference between learning from failure and succeeding, especially in video games

Thank you for making this video - I was wondering if I needed to go start my own case study on the differences between different rogueli*e games' progression system, but this gives a fine baseline for me to work with to help make my game!

You should've looked at Risk of Rain, it has a great system where you can do easy mode, and learn the ropes, but not unlock new characters, items, or achievements. Or do medium or hard mode where you can have your stats recorded and unlock new characters and items.

Well, about roguelites... there's nothing really stopping you from starting the game from scratch and seeing how quickly you can beat it again, while trying to rely on persistent upgrades as little as possible. Take Hades for instance, you can actually strip yourself off of all the upgrades you've accumulated. Not to mention that the game also allows you to upgrade the game's difficulty itself later on. On a side note, I think Enter the Gungeon and Isaac actually can be considered roguelites. Sure, the guns you unlock in the gungeon are not given to you at the start but there *_are_* a lot more powerful ones you can unlock, some that are even run-winning. Same thing for Isaac, better quality items > less harsh RNG, not to mention characters can be 'upgraded' to have a starting item (example: the D6 for Isaac - giving you more versatility in future runs) So the real question is... if the game lets you add more powerful stuff into the RNG pool of your runs, does that also make the game a roguelite one?

This is excellent. I would love to see a video on replayability and the pros and cons of New Game+ in games of other genres

i believe hades removes that skill or time feeling, as all upgrades only really improve your chances, or give you more health, so if you get hit, it's still your fault.

The guns unlocked in gungeon do change runs difficulties, as many of those you unlock as you progress are better, dealing more dps and having better effects

3:57 funny enough this "weird" difficult curve is quite common in RPGs if you play Dragon age, Witcher, Pillars, Fallout or any RPG the game is harder when you begin and when you get to some specific ability of your character the game becomes easier as time passes. Specifically if you play on the hardest difficulties that is really easy to notice.

Roguelites aren't very different than RPGs. You level up your character and situations become easier and easier as your character and equipment improves.

Not sure if it was implemented the same way when you made this video (I remember when it used to just be a lernie drop), but the heat system in Hades is a really great way they've counterbalanced the persistent upgrades from the mirror of night. As someone who's played since day one, the recently introduced Extreme Measures rank 4 difficulty modifier has proven to be way more skill testing than anything else they've put in. Broken tactics and cheesy builds aren't enough for it and I've still only cleared the final encounter once and died to it quite a few times. I hope Supergiant keeps making games forever.

I'm not a huge fan of rogue-likes in the first place, but I'm personally a fan of Mystery Dungeon style games where rather than needing to do the entire game in one go, you do a series of smaller procedurally generated levels with story bits in between. That way, you're still going through the challenge starting from scratch, but you're definitely making progress along the way. My personal favorite is Dokapon: Monster Hunter for the GBA.

Thanks for the reupload! I enjoyed it just as much both times, but there is one thing that bothers me a bit. In your section about gungeon, you said guns are no stronger than the ones you can find at the beginning. I'd like to believe this is true, but after a couple hundred hours in the game, I've found it to not be so. While you can beat the game with the guns you can find at the beginning, guns bought in the breach such as the black hole gun, yari launcher, or even the clone item, found after finishing a certain side quest, can make the game extremely easy if found. Anyway, I just wanted to put my opinion out there. Thanks for the amazing content!

I honestly love both, they are 2 of my favourite game genres! my attention span can be wary at times especially for games but i find both rogue types easy to pick up again and I just keep going back to them!

One of my favorites is one way heroics, which has both the option for rougelike and roguelite play in the form of maniac mode.

I love your channel, I would really like to see you do a video about atmosphere in horror games and how it's done.

Crypt of the Necrodancer did it best in my eyes. Of course, there's the standard Rougelike mode where you go through Area 1, 2, 3, and 4 in order without dying. However, the regular and probably more often taken path (and the one you showed in the video) is the segmented Area run, where you start at Area 1 and just get to the bottom of *that* to unlock Area 2. Because the game's areas are way more drastically different than other rougelikes, because you have to learn new enemy patterns, music, it's encouraged to play this way so you can learn the Area and really get good at it before moving onto the next. The segmented Area runs also have persistent upgrades, ranging from health to items, which can make the game significantly easier. However, these upgrades don't persist though the All Areas mode, as every upgrade is unlocked from the start in that mode, except for health upgrades. This system makes the difficulty curve natural, letting players start from whatever Area they want. It also encourages them to try All Areas after beating the game, because they now have the knowledge they got through the segmented run, so they won't have to die on an enemy not knowing what the hell they do, an issue that I tend to see a lot in modern rougelikes. Another issue they fix is making All Zones marathon safe, because if shit hits the fan they can always go into a segmented area and finish the run there. (Thats what they did in AGDQ, for example.)

You should have mentioned that a few of the games you mentioned include mechanics that allow players to increase the difficulty, even after significant improvements to either their skill or characters. Slay the Spire with ascension mode, Dead Cells eith god cells and Hades with the pact of punishment.

I choose Roguelites. I've never managed to beat an actual roguelike before I got tired of the game.

It's the same video, but's still different! :D

Honestly I really could care less if I’m “good at a game” but rather that it is fun to play and a great experience.

Can't believe I'm just finding this channel, it's brilliant

Thanks for making this Video. Now I can link it anytime someones doesn't know that rogue-like and lites are two different variations of the same genre.

Thank you for the semi-reupload!

This is first time I see someone splits these two types of games as rouglikes and rouglites. As far as I've seen from other sources "rougelite" was usually used as a more fair synonym for what is generally today called rouglike, in order to make difference between games that are actually like rouge (and should be called rougelike), and what is popularly today called "rougelike", that has just several elements from rouge.

One game that does something interesting with progression is Shiren the Wanderer (1995, Super Famicom). The game was an early attempt at the home console implementation of the genre in its strictest sense (procedurally generated levels, permanent death, turn-based RPG combat on a grid, unidentified items whose effect must be determined by cautious experimentation, the whole shebang) yet was far ahead of its time in its _metaprogression_ (roguelike jargon for progression that persists between runs). It has sidequests that take multiple runs to complete, which in the dungeon unlock new items to find and NPC allies to recruit and in the game's towns open new stores for you to shop in. But its most interesting form of metaprogression is probably the way it lets you save resources between runs. In some of the game's towns, there are storehouses where you can stash your items, whose state persists between games. In one town you can keep a sword in the storehouse and, every time you pass through, take it to the local blacksmith to be upgraded. But since you can't keep an item in the warehouse _and_ use it in the dungeon, actually using that sword you've been improving for so long is a huge risk! Every run is a gamble on the combined value of all the resources you choose to take with you that those resources will be sufficient for you to win the game. To fully benefit from the progression, you have to risk _losing_ it. Completing the sidequests does make the game easier, but even once you've done them all -- which doesn't take that that long -- the game is still bitch hard. The storehouses represent the biggest change from the classical roguelike rules, yet they form a balanced risk/reward system that, if you're not careful, will crush your dreams much faster than it took to build them up. Ultimately the game is more accessible than the classical roguelikes but not, I think, actually that much easier.

well deadcells implemented a cool feature where the game favors people that are good at the game like killing 60 enemies without getting hit and the only upgrade that really effects the game is the potion

I disagree, Rougelites do have a traditional difficulty curve. The accumulation of power over time doesn't invert the curve, but lowers the y intercept. The the whole of the game gets easier, but the later levels are still much harder than the start.

I'm quite fond of Roguelites. Neon Chrome is one of my favorites, with it being persistent in it's u locks and stats, but each successfully completed playthrough unlocks another difficulty tier. It gets painfully difficult over time

I know this is an older video but after watching the most recent video about Deathloop, I had to come back to this. Huge fan of the Roguelike/lite genre but Dead Cells stands out to me in particular because of the Boss Cell system or Difficulty system they have. Even after getting literally all of the permanent upgrades the game is still quite challenging plus there is a secret area only available for those on the hardest difficulty. I would like to see more roguelites take inspiration from Dead Cells because it is the only roguelite, specifically for me, that I can keep coming back to. Makes it go from lite to like.

Very good video and extremely informative. As a Gamedev looking into making my own roguelike, this was a good entrance point, so thank you. One thing I would have liked to see added, is that a lot of roguelites also have climbing difficulty through "ascensions", "hard mode" etc. For example Slay the Spire gets you more unlocks for your character each level up, but each time you beat the game you also unlock one higher difficulty, ramping up to 20. Beating the game on Ascension 20 is still extremely hard, and you need the experience and the unlocks. So both is possible

Playing through hades right now, first time playing this kind of thing

I was so confused watchig that videos... "Oh, a new videeo on rogue likes... Wait, didn't i watch a video like this before? Eh, must be from another RU-vidr!" Then at 5:39 i really got that feeling of Dejà Vu.. I went to the commentr section and oh! It's a reupload lmao. Great work dudo. You made me laugh too ahah

I can't believe you didn't mention one of Dead Cells greatest mechanics at all, Boss Cells, which makes ending the game for the first time seem like a tutorial. After you 'beat" the game the first time, you are given a Boss Cell from the end boss, then you put it at the start of the area where you respawn and the difficulty increases and you have to beat the final boss again to get the next Boss Cell. With each new Boss Cell things get more exciting, new enemies appear, they deal more damage, new weapons and routes become available, you have fewer (or none) flask recharges to heal and you can also upgrade the forge further. This is how the game avoids that 'reverse' difficulty curve you talk about, and still gives new content which each new difficulty. 5 Boss Cells active is extremely hard, even with everything upgraded, very few people can even get past 2 Boss Cells,

I like how in The Binding Of Isaac, every time you come across a new upgrade it can be viewed in an archive. You can see what you've found and how many things there are left to find, so in a way you're making progress as a collector.

I like how Risk of Rain 1 and 2 do it, by letting you unlock new characters, items, and even bonus modes. It lets you feel like you're making progress without forcing you to beat the game or making it demonstrably easier.

*Reuploads video* “Well, time to go get killed again”

That "weird backward difficulty curve" is just a function of any RPG style level up system. I've been playing one of the souls games, and I was banging my head against a wall fighting one of the bosses, so I leveled up a little until I could beat them. The fact that it was my time investment and not my skill which beat the boss doesn't bother me.

Super interesting video Mark!!

I'm incredibly bad at video games, yet i cherish rogue-likes for some strange reason. Something about finally FINALLY getting to the heart of The Binding Of Isaac (which i've done a miserable 10~ times winning 4 after 300 hours lmao) and beating it made me so happy i couldn't get enough.

I feel like you still undersold the luck aspect here. You say rogue lites have an ambiguity to whether or not you won based on skill (even though most can be won upgradeless) but don't rogue likes have the same issue in whether or not you won on skill or just got lucky enough with drops to win?

My first real experience with a game of this genre is Enter the Gungeon, at first it was a bit frustrating to not really have a sense of progression other than making it to the next floor or room but i discovered that I loved the feeling of "just one more run" in the hopes that i get better gear, loot, weapons etc. I have over 500 runs in the game and about 175 hours clocked and I love it! Easily one of my favorite games

Game makers toolkit is so educational about a topic that isnt discussed in such a manor very often and these videos are relaxing as fuck. I truly love these videos.

Bro the progression in Roguelikes for me personally is the literal progression you make in the game after you play for a while. After you die 50 more times in Enter the Gungeon, you finally beat the boss and see what the next level has in store. Because the difficulty is unchanging, it provides a progress tracker for your skill

I cant believe that he did this video without even mentioning Isaac :,v

This negative slope of difficulty curve in roguelites reminds me a similar response curve in a feedback loop of an adaptive system. You actually said something along these lines (with how low level you can finish that game). My point is, we can look at it as a game dynamically adjusting to player's skill. I'm okay with that. Love your videos, BTW. :)

Excellent video! Thanks again for the quality analysis and content. I generally like my games a little easier so I personally prefer roguelites. I find that I need that knowledge that I'm always making progress in order to not get discouraged and stop playing a game; I would feel like my time was wasted otherwise. I've recently been playing children of morta, a top down hack and slash roguelite rpg, which I feel like balances the progression and difficulty well. As you play through the game your characters earn levels (netting you skill points to spend on new abilities) and you can spend money earnt during runs through dungeons on persistent upgrades. As the game progresses, the dungeons get more and more challenging. The game is definitely harder at the start when a character has no levels, but difficulty so far has kept pace well; as soon as my character clears one area, the next one is usually too hard to beat on the first go, and requires me to get better to learn how to use new abilities better or more efficiently, in order to surpass it. Lastly, the only way for characters to heal is to pick up health potions which only have a chance of dropping randomly. Otherwise, there is no healing. Combat is fair and balanced, but mistakes are hard to take back. Getting enough money in one run for a large upgrade requires you to survive through most of the run, and thus make less mistakes.

“that creates in a weirdo backwards difficulty curve that brings in an unpleasant ambiguity over whether your success is due to your increasing skill, or the game’s falling difficulty” If I’m being honest, I’m pretty sure the only people who care about that are the ones who are already good at really hard games. XD

Rogue Legacy also has ascensions, where you clear all the bosses and restart the dungeon with the same character at a higher difficulty. My new favorite Slay the Spire is *freaking hard*, mostly because the card draws are randomized and the bosses hit like freight trains.

3sec in and I had to subscribe. Amazing voiceover. Good content. Thank you.

Roguelites are definitely my preference. I like the different types of progression, and reward for meeting those goals.

I personally prefer roguelites, but I'm not sure how much of that is just to spite the elitists who dislike them for being accessible to "casuals".

TL;DR: Give me (unlockable) items or options that reduce the game difficulty (along with a drawback like less credits or worse gun RNG in Enter the Gungeon, for example) that can be turned on or off at will so if you feel like you can go without you can-unlike permanent stat or ability upgrades. Some games already do the opposite: making the game harder for more rewards, and so I feel this is a perfectly fine addition. I do definitely prefer the model of a roguelike such as Enter the Gungeon, with progression from unlocking different guns and characters, over something like Spelunky. I think it keeps the game fresher for longer and you may find that one of the later characters, who will usually be relatively harder, such as the Robot only having armour and no health (or whichever character in Binding of Isaac who has soul hearts but no regular hearts), might be more suited for your playstyle. As for rougelites, I can't really comment, as the only one I've played is Rogue Legacy, and I didn't particularly like it (I thought it was cool, just not the game for me). I'm pretty average-not that bad, but certainly not great-at video games, so I do find it hard to really get into stuff like Spelunky and Binding of Isaac where all I have to rely on is my own skill. That being said, I also just kinda prefer the slow accumulation of power, like in JRPGs, my favourite genre. I don't want to go into each area overpowered, but there's also something comforting knowing that even if I can't beat that big monster now, I can definitely beat that big monster in the future, and it gives me something to work towards. And if I'm feeling particularly good at the combat system, I can come back while underlevelled and try my luck. I don't feel the same pull with "If you keep playing you'll (maybe) get good enough to be able to complete the level", and I do agree that it can suck spending 30 minutes on a run and not really getting anything tangible for it. Which isn't to say I particularly dislike difficult things. I've just started Celeste and died 150 times on the first two levels but I'm having a great time, and I want to eventually get all 175 red strawberries and complete at least all the B-side levels. It's very different as like Super Meat Boy, the reset time is very short, and you can try again immediately. With roguelikes, while I know shortcuts exist, that also cuts out any items you could've gotten from earlier levels and you'll usually be underpowered. And while I've beaten the High Dragun on Floor 5 in Gungeon, I've only gotten the elevator to Floor 2, so I can't practice Floors 4 and 5, easily my weakest ones. Not to mention the hidden Floor 6, to which there is no elevator for obvious reasons. If I want to get good at something, I'll just do it over and over again, like practising the same lick on guitar until you can play it perfectly. I don't want to have to play the entire song just to get to the solo and mess it up yet again like I know I bloody will. Perhaps roguelikes just aren't the kind of game for me. My favourite games of all time are Danganronpa 2 and Rune Factory 4, Danganronpa 2 being a murder mystery visual novel and Rune Factory 4 being a Harvest Moon-esque game with more of a focus on combat, and having upgradable skills for everything, I believe up to and including walking, eating, and sleeping. I can't really say whether I prefer roguelikes or -lites, but what I'd want in either of them is a way to sort of modulate the difficulty (so I guess I want roguelites). I'm not saying I want a difficulty level or slider, but I do want the potential to make the game easier. For example, in Gungeon, there could be an item you equip in the hub that makes less enemies spawn or makes all the enemies a lower level, with the downside of reducing (perhaps heavily) how many credits you earn or perhaps also making all the guns that spawn a lower level (I think less enemies would still be a positive, even with weaker guns). Or perhaps an item that makes the game speed 75% to make it easier to dodge bullets, with an equal downside. A "crutch" is the term I'm thinking off. I don't want to get to the point where I've gotten all the upgrades I need and the game isn't a challenge anymore, but I also don't want to be stuck banging my head against the proverbial wall for hours with nothing to show for it.

Returnal I keep coming back to this video

One interesting approach to this is the Roguelike Gameboy game Cave Noire which is divided into 10 levels that unlock as you beat each in turn. The 1st level is stupidly easy but the 10th is super hard and long. Each individual run is distinct, starting from the beginning but you feel progression as you unlock levels. The game is also divided into 4 separate dungeons with different winning goals. 10 levels each.