Table of Contents
- Scope & Purpose of this Novel
- Genre Name & Canon
- Defining Mechanics
- Disqualifying Mechanics
- Borderline Mechanics
- History
- Historical Influences
- Experience
- Hordes
- Close but no cigar
- Crossing the streams
- Mechanics Deep Dive: XP Gems and Map Attractors
- Mechanics Deep Dive: The Dash Button & Telegraphs
- Mechanics Shallow Dive: Invisible Bullets
- Mechanics Opportunity: Enemy Movement Patterns
- Left As An Exercise For The Reader: Scaling

Vampire Survivors didn’t invent the Horde Survivor genre, but it sure caused some problems in my life. I have played entirely too many games in this genre and spent too many hours, particularly when somewhat inebriated, in their sheer brain-numbing bliss. In an effort to turn idle comfort into productivity, as we all must under capitalism, I’ve decided to put a few of my thoughts on the genre to writing. I hope this will be useful to someone.
Scope & Purpose of this Novel
Within these excessively numerous pages I want to set out most of my thinking about the emerging Horde Survivor genre. From how I trace its lineage to how I delineate it from similar sibling-genres like the top-down roguelite action game to just an overly long encyclopedia of the entries in this genre I’ve played and what they all taught me.
This is my white whale. It’s my unhealthy obsession. I doubt many people care as much for this dumb-ass genre as I do. If there’s one post on this-here blog where you should probably go “good for you” and just skip to something else, this is it.
However, if you want to find out how deep the survivor hole goes, follow me.
Genre Name & Canon

I’ve chosen “Horde Survivor” as the genre name. There’s a few other options, primarily “Bullet Heaven”, but that feels wrong to use. It specifically arose from Vampire Survivor’s marketing materials (see above) and it feels weird to parrot that. But of course you can call it whatever. Since Google got rid of showing you how many results there are for a given search query in their ongoing effort to make the Internet more shit I can’t tell you which is the more popular term. (Note: if you do know of a search engine or a Google trick that can bring this stat back, I’d love to hear it! Comment here or drop me a note on Bluesky / LinkedIn. You used to be able to go to search tools on google to get that stat for a bit, but they’ve closed that loophole too.)
Obviously Vampire Survivors is the first game to talk about. The genre would not have blown up as it did without that game, and in many ways I consider it the most essential expression of the genre, but that is of course a circular argument. (Horde Survivors are defined by being like Vampire Survivors, therefore Vampire Survivors is the quintessential Horde Survivor) We’ll go into individual games and what cool stuff they tried later, but for now here’s a short list of games I’ve played that I consider part of the Horde Survivor genre:
- Vampire Survivors
- Brotato
- Deep Rock Galactic: Survivors
- Elemental Survivors
- Swarm Grinder
- Astro Looter
- Survivors of the Dawn
- Nordic Ashes
- Pesticide Not Required
- Cozy Space Survivors
- Soulstone Survivors
- Death Must Die
- 20 Minutes Till Dawn
- Halls of Torment
- Boneraiser Minions
- Extremely Powerful Capybaras
- Survivor.io
- NIMRODS: GunCraft Survivors
- League of Legends: Swarm
(it’s taken me the better part of half a year to write this post so this list is now badly out of date; anyway, you get the idea)
I’ve also played some of these games a lot! In 2024, Horde Survivor games became my most played genre. I’ve put nearly 40h into the demo of NIMRODS: GunCraft Survivor (a game that will fuck up my life when it comes out) alone. (Edit: It came out. It didn’t fuck up my life too bad. Turns out the “demo” was about 90% of the game) I’ve got 60h in Halls of Torment, 226h in Soulstone Survivors, 256h in Deep Rock Galactic: Survivor, and 150h in Vampire Survivors (plus roughly the same again on mobile). I’m saying this not to brag or for you to feel sorry for me, but to let you know how much of a brain parasite this genre has become for me. I’m desperate to distill into words how these games achieve that.
Right, to round up our canon section, here’s a few games where I’m on the edge whether or not they fall into the genre:
- Crab Champions
- Risk of Rain (1 & 2)
- Tesla vs Lovecraft
- Crimsonland
We’ll go over these outliers later and analyze what aspects of Horde Survivors (why am I capitalizing that? I’ll stop now) they lack and if that’s enough to push them out of the genre.
And finally here’s a few superficially similar games that I will argue definitely fall outside the genre (even if they very clearly inspired the genre)
- Nuclear Throne
- The Binding of Isaac
- Hades
Defining Mechanics
- Top-down perspective
- Auto-aim
- Mostly focused on movement
- Drastic increase in enemy numbers and strength
- Drastic increase in player strength
- Strong emphasis on more stuff on screen
- Experience pickups
- Levelup upgrades combine with each other in tons of ways
- Short runs (20-30 mins)
- Heavy emphasis on metaprogression
- “Breaking the game” as a desired outcome
- Individual run on a single map
- Weapon evolutions
- Incredible wealth of content: maps, characters, modes, modifiers etc
- Primary skill expression lies in build crafting:
- Meta knowledge
- Holding goals in mind
- RNG manipulation
- “Power now” vs “scaling” tradeoffs
- Runs are usually not close; either win huge or die quickly
Each and every one of these mechanics is negotiable (although you will notice that they all apply to Vampire Survivors, unsurprisingly) but if a game lacks too many of them it’s probably not in the genre. Of course genre is always a spectrum and also a multi-layered thing; a game can be in multiple genres, reasonable people can disagree on what those genres are, and everything is made up anyway. It’s just video games.
However:
Disqualifying Mechanics
Gotta get some spicy takes in here. These mechanics IMO move a game firmly out of the genre.
- Focus on aim
- Specific skill usage
- Elaborate boss fights
- Individual small rooms
Borderline Mechanics
These exist in a fair number of games in my canon but they make me go “hmm do we really need these in Horde Survivors?”
- Telegraphed enemy attacks you have to dodge
- Enemy projectiles that can’t be destroyed
- Dashes
- Other active skills that skill check timing and position
I think I’ll say more about those as we go into individual games that feature them. Soulstone Survivors, you’ve been put on notice.
History
Okay, so let’s start by tracing back the genre to its many roots. I think there are three influences on the genre, two of which will seem glaringly obvious and the third will seem like a complete shitpost, but I’m ready to defend it as 50% shitpost 50% sincere.
Bullet Hell Shmups

Roguelite action games

Idle games

Historical Influences
The most direct influence in my opinion are top-down shooters, particularly ones with a roguelite flavoring to them. So Nuclear Throne, Hades, The Binding of Isaac come to mind. From them, we get our perspective and some of the core dodge baddies gameplay. We also get the combinatorics of upgrades which is central to the genre, inherited from the roguelike genre in general.

The big differences here are in which kinds of skills are being checked and what the pace and structure of a run is. Other differences include run length, number of enemies on screen, and the nature of your attacks and how they grow in power throughout a run.
The most obvious place to start is skill checks. By this I mean “what type of thinking and acting is required to do well in the game”. We can break skill checks down as broadly or narrowly as we like, but for our purposes, I would suggest that the following are not usually Horde Survivor skillchecks, while they are very commonly found in top-down roguelites:
- Aiming precision and prediction
- Active cooldown management (when to press which button)
- Offensive resource management (mana, skill charges, other resources you can burn for a tactical advantage)
- Dodging of enemy missiles or AOE zones (this one deserves a big fat asterisk because of games like Soulstone Survivors and Swarm)
- Memorization of enemy attack patterns (small asterisk for Nordic Ashes and Swarm)
At an even simpler level we can say that, unlike in top-down roguelites, in Horde Survivors you rarely use inputs other than your movement buttons and you rarely dodge things other than enemies. In its most pure form, a Horde Survivor treats even enemy bullets as enemies; Vampire Survivors does this. This means there are no special rules; you can just kill bullets and be safe.
Other influences come from two very different places: Bullet Hell shmups and idle games. Let’s deal with Bullet Hells first.
The term Bullet Heaven makes it clear that there is some kind of inversion going on here, relative to the Bullet Hell genre. That inversion is the inherent meaning of projectiles on screen: bullet hells use them to create super-fast mazes and obstacle courses for you to navigate while treating player attacks primarily as an aesthetic; that is to say, more attention is paid to dodging enemy projectiles compared to keeping your own attacks on target. I say this as someone with a very superficial knowledge of bullet hell games, so please correct me if I’m wrong!
Anyway, bullets bad == bullet hell, bullets good == bullet heaven.

There is another influence that is a little more subtle. In bullet hells you have masses of bullets moving in mostly predictable manners and you have to find a way through them given very limited space. In bullet heavens, you have tons of enemies moving in mostly predictable (but much simpler) ways, and even though you generally have infinite space in a level (exceptions such as Brotato and DRG:S notwithstanding) you have effectively limited space due to the way enemies spawn (usually in circles around you). Some games recognize this and use the hit box of bullet hell games; Elemental Survivors stands out as a game that really doubles down on the tiny pixel size hitbox of bullet hells, and it plays great for that.

So that leaves us with the historical influence of the idle game. Games like Cookie Clicker, AdVenture Capitalist, or Idle Champions of the Forgotten Realms are very tidy and simple math engines. They’re the purest form of “number go up” brain stimulation. You have some sort of trickling income (sometimes based on actual clicks you perform) and various ways of investing said income. You can make a given source of income produce more baseline, you can open a new source of income, or you can get various modifiers that cross-multiply income streams.

For instance, in cookie clicker, the first income stream you unlock is the “cursor”. There are simple upgrades for the cursor, such as making it twice as effective, and there are multiplicative upgrades, like making cursors better for every non-cursor object you own.
What this does is constantly shift around the value of investing in one income stream over another. It’s trivial to do the math and see what investment gives you the best c/s/c (cookies per second per cookie, of course), but once the multipliers come in, things get more interesting. This is especially true if you don’t look up what multipliers will come your way in the future. Maybe buying more grandmas doesn’t seem cost effective past a certain point, but eventually you get an upgrade where every grandma you have multiplies the output of, say, farms, which may now be your most effective income stream, and so suddenly you re-assess the value of grandmas under capitalism.
It’s a weird game.
Horde Survivors do a similar thing: you tend to have weapons and upgrades, and the upgrades generally multiply an aspect of a weapon. For instance you might get projectiles +1; this would be great on a weapon that only shoots one projectile (representing a +100% damage increase), but not so great on a weapon that already shoots 10 projectiles (representing only a +10% damage increase). Maybe projectile range is initially useless on a short range weapon because it stops on first target hit, but eventually perhaps you’ll unlock target piercing where said projectile can go through multiple targets, and range becomes a relevant stat.
Also if we zoom out for a moment, the feeling of putting together an engine that works pretty much by itself and then sitting back and watching it go while from time to time adding to said engine and seeing how much of an improvement you made: this feels like it’s the same thing for idle games and pure bullet heaven games. There are for sure execution decisions to make even in Vampire Survivors: where do you go, when do you turn around, can you squeeze past these enemies, etc. But in the grand scheme of things they don’t matter nearly as much as which upgrades did you get, how did you combine them, what did you prioritize, etc. Knowing the timestamps when particularly challenging waves spawn in and whether or not you have the DPS and survivability to deal with them is more important to your success than being really good at movement and reaction time.

At the risk of getting ahead of ourselves, this is one reason why I call horde survivor design that emphasizes reaction time and precision cowardly design. It shies away from what makes this genre so different.
Experience
Probably the most defining mechanic in horde survivors is that experience isn’t earned immediately when an enemy is killed, but rather created as an object in the world, most commonly some kind of gem or glowing rock. The player needs to approach the xp obj within a certain range to pick it up. This is one of the easiest rules of thumb to see if a top down roguelike falls within the horde survivor category: does it have a pickup range stat you can invest into?

At a visceral level, this is mostly about the dopamine hit that comes from seeing your character hoover up tons of shiny objects, their arrival usually marked by ding! sounds that slowly ascend in pitch as you fill up your XP bar. The payoff of course is the level up, which tends to come with a lot of audio-visual ceremony as well. But there’s more to these little xp gems. They complicate movement decisions in interesting ways, both tactically and strategically. They’re neat.
But where did they come from?

It’s hard to trace this mechanic through previous games. My best guess right now is that this is traced back to the Shmup genre. The closest antecedent in a very similar genre is probably Nuclear Throne (2015). There is one interesting difference between Nuclear Throne and the horde survivor genre when it comes to XP: XP in Nuclear Throne quickly times out. If you don’t go and pick it up right away, you miss out on the xp altogether. This is a mechanic that’s closer to DOOM (2016) in that it strongly incentivizes forward motion and makes taking safe shots from a distance not a viable strategy. We should ask ourselves why horde survivors have not replicated this aspect of the XP pickup object mechanic.
Hordes
Having a lot of enemies is not exactly new territory for videogames. Note that there is a palpable difference between more enemies and enemies with more HP. There’s something viscerally different felt by the player. HP is an abstraction. Hordes of enemies do not need to be interpreted. They’re massive and immediately felt as an escalation.

There have always been games trying to put as many enemies on the screen as the hardware could support. Even Asteroids can put quite a few of the eponymous space objects on the screen based on how you prioritize shooting stuff. But over the years as we improved our understanding of the gameplay in these genres, we tended to reduce down to smaller numbers of enemies, trying to make decisions around each enemy and making space for dodge or aim gameplay. The switch back to quantity over quality came around the turn of the millennium with a little known Finnish game called Crimsonland.

Crimsonland’s main selling factor is right there in its title: you will murder so many enemies that you’ll turn the green grass of the level crimson. The audacity of that notion appealed to a 20 year old me who’d previously downloaded mods for games from DooM to Duke Nukem 3D to make them bloodier. But in the course of putting lots of enemies on the screen and honing a difficulty curve that saw you dodging through gaps in an ever increasing horde, 9 Ton discovered an incredible formula that they’d go on to hone with games like Tesla vs Lovecraft (they’ve always been great at names).

Close but no cigar
So why don’t I consider Crimsonland and Tesla vs Lovecraft part of the canon? Three reasons:
- XP is gained immediately on kill
- The level up upgrades are insufficiently transformative
- Focus on aiming precision and movement prediction (“leading”)
There’s more, but it’s mostly nitpicks like most power spikes coming in the form of time limited powerups so you never feel permanently overpowered and the fact you only use one weapon at a time. But Crimsonland got so close.
In a way, not even Nuclear Throne got this close. And that’s in spite of the fact NT has experience gems and represents the first game with this camera angle that I could find that did both XP gems and rogue-like upgrades on level up.
Of course when I say “it got so close”, don’t misunderstand me. I’m not saying that Nuclear Throne almost was a great game. It absolutely is a great game and it’s doing its own thing. It’s a beautiful ballet of speed, precision, violence, risk, and mutations. Yes the design space of the upgrades is pretty flat (there’s not an awful lot that is mutually multiplicative, for instance) and yes the “find better weapons during your run” generally ends with everyone using the same weapon (auto crossbow FTW), but I don’t think Vlambeer were trying to create a thousand hour game.
Crossing the streams
The game that made the genre blow up was Vampire Survivors, of course, but I want to direct your attention to a special crossing of the streams. Vampire Survivors was itself a, shall we say, homage to the 2021 Android game Magic Survival. I think we inadvertently got a lot of the really good stuff from the limitations of the mobile format. Obviously twin stick move and shoot *can* be done on a phone interface, but it’s not the easiest thing to control and if you want your game to have a more laid-back vibe you probably want to drop shooting. Things like the experience gems and the variable reward chests also feel *mobile* to me: they’re laser focused on delivering regular but variable dopamine hits. They elevate ceremony to ludicrous levels (see also: the chest opening animations in VS)

But I’m also not sure VS would have blown up like it did if Luca Galante had just made a mobile clone of a mobile game. I think that taking this simplified loop and having it recontextualized in a PC context added a lot. Something that was merely an accommodation for the simpler input paradigm of phones became an almost brazen design choice on PC: the thought that you’d control a character’s movement but not aim their shots in a mouse and keyboard world was scandalous! Being on PC I think also helped the game find content creators willing to stream it and players willing and happy to binge it for many many hours. Here in the west at least phone games are still mostly seen as little time machines that let you skip 5 boring minutes in the dentist’s waiting room or half an hour of your bus commute. A PC game is a much more serious thing that demands to be engaged with fully.

I think this is also what made their content strategy so successful. The post-release content strategy of Vampire Survivors can probably be best summarized as “more”. VS embodies the saying that “quantity has a quality of its own”. Sure, the base content probably wouldn’t stand up to a lot of repeat play, especially once a player progressed through the metasystems and unlocked sufficient permanent power to trivialize most runs. But not to worry: there’s always MORE around the corner. Generously sized and ludicrously underpriced (usually around $2), each DLC for Vampire Survivor reactivated its user base, bumped it up on people’s radars undoubtedly leading to more sales, and built massive amounts of player trust in the developer.
But it’s somewhat unserious, right?
And I think that’s why it worked so well. It was news in and of itself that a game this cheap looking and, well, actually cheap would be the indie darling of the year. Something that looks (and often is) quite brainless, fawned over by so many streamers usually playing brainier fare? It’s bound to make you sit up and go “what the fuck?”
A third stream, perhaps, is that Luca Galante’s penultimate job before becoming very very successful with Vampire Survivors was at a social online casino studio. I don’t know enough about the dark art of casino game design, but I do know that those Sith motherfuckers sure know how to sink the greatest possible number of hooks into your brain and get you well and truly addicted. I think this shines through in a lot of VS’s design.
Mechanics Deep Dive: XP Gems and Map Attractors
The first mechanic I’d like to do a deep dive on is the humble XP gem. It would be easy to dismiss it as “just ceremony for ceremony’s sake” and “somehow feeling like mobile design”. Whether it was the design intention or not, the XP gem is the beating heart at the core of this genre.
Tactical Movement
Let’s start simple. XP gems drop near murdered enemies, like in real life. This has a few implications. First, merely keeping your distance and sniping things from range is a non-starter. Eventually you’ll have to walk up to the scene of the crime and pick up your piece of candy.
Where there was one enemy, there tend to be more enemies in the future. This means that often a player is faced with the choice of “pick up this giant horde of XP gems now at the risk of taking damage or play it safe but risk falling behind the level curve”. Or better yet, cheat the system and somehow deftly move in just close enough to trigger the XP gem attraction sequence and turn around before taking any damage. This is the closest a pure a horde survivor comes to precision skill expression: skirting an enemy, all but touching a baddie.
Relatable.
Map Attractors
If the map was a big empty stage, there would be very little reason to move any meaningful distance across it; aside from maybe being swamped by enemies and needing to get some distance. It seems likely that a pattern of farming enemies in one location, usually by circling around a group of them, and then picking up XP would emerge. There are situations in many HS games where that is indeed the best thing to do, at least for a little while. But we want players to make interesting decisions, right? And a lot of the mechanics that usually generate decision points, such as cooldowns and resources, are generally excluded from the genre.
You could say HS players make meaningful decisions in the level up screen and leave it at that, but I think if we can wring just a little bit of interest around decisions you make in between level ups our game is probably more interesting for it.
Enter: Map Attractors.
By this I mean any mechanic that entices the player to move to a certain location on the map, usually after scouting the map; games with fixed knowable attractors exist, and it’s what VS does, but more modern HS design seems to go toward random placement that requires scouting.
In VS these are the preplaced powerups on all (most?) maps, usually at very long distance from the player. This has a fun secondary function of rewarding investment into movement speed beyond the level where you’re swift enough to dodge enemies: if you don’t buy movement speed, you probably won’t have the time to visit all powerups in the 20 minutes of a normal VS run.

In DRG:S, the attractors are meta-resources (until you’re maxed out and sitting on 2k+ of each, like I am), run resources (gold and nitra), health crystal or healing zones, and the Huli Hoarder. But DRG:S also does a great job of balancing power growth vs challenge so that for me at least it always remains in the optimal flow zone of the friction spectrum, and so having to move lest you get overwhelmed by enemies or creatively using the tunneling system to separate yourself from the swarm also serve as gameplay loops that function like reverse attractors.
In Elemental Survivors this is shops and villages, in Rogue: Genesia it’s the altars, in Halls of Torment it’s preplaced scrolls and random chests, in Pesticide Not Required it’s the fishing and mining mini games, and so on. It’s hard to think of an HS game that doesn’t have Map Attractors. Even Brotato, whose levels are featureless boxes scarcely bigger than your screen, spawns trees to keep you moving.
XP gems vs Map Attractors
So this is usually the core tension in an HS game, particularly the early stages: do you remain in one place so you can farm levels optimally or do you go and chase after map attractors right away? Of course, it’s not one or the other; players constantly go back and forth between the two modes. Maybe try to farm up levels until you feel powerful enough or fast enough to dash across the map and pick up an attractor, then go back to farming in place, etc.
Managing how much time you spend in one mode vs the other is one of the more interesting cerebral challenges in the genre. Run timers add some pressure to this choice: you often don’t have the time to go everywhere, do everything. Games like DRG:S or Survivors of the Dawn that break the run into many sub levels add a lot more pressure as either time or kill count will end the level, whether you’ve gone where you wanted to go or not.
Mechanics Deep Dive: The Dash Button & Telegraphs

The red badge of cowardice, the white flag of surrender, this is a design crime on par with not putting a treasure behind a waterfall. Let me explain.
The Dash Button
Remember how we discussed above, in “Crossing the Streams”, how it was the audacity of its simplicity that made VS so successful? The extremely simple input scheme which may originally have been a concession to the mobile format became one of the genre’s unique selling points.
But simplicity gives designers anxiety. After all, if there is design space, shouldn’t we use it up? You know, like fossil fuels?
No! On both counts. Yes, you can absolutely add more skill expression to the genre by adding a dash or similar mechanic. You’re now both asking the player to manage a resource (the cooldown of said ability) and asking more of the player’s coordination. Both because dashing in the wrong place can be very dangerous indeed and because giving the player a dash gives the designer permission to create enemies that would not be avoidable in a world without a dash.
When you’re increasing complexity and skill expression in one part of your game, you often decrease it elsewhere, whether you mean to or not.
Enemy Attack Telegraphs

Look, I’m generally a huge fan of clear, unambiguous telegraphs. They’re essential in PvP games and they make learning and mastering PvE games much more accessible. Sure, you could ask players to react to animation and sound cues and memorize the timing and size of enemies’ big attacks, but you could also just project a red rectangle on the ground and have it gradually fill; when it’s full, the attack hits. There, area and timing perfectly telegraphed. Now your player can get better at dodging.

This is absolutely the right approach for top-down action roguelikes like Hades; but that’s not the genre we’re in. The more you ask players to step out of red rectangles, the fewer brain cycles your players will likely have for other things.

Also, remember the H in our genre stands for Horde. The temptation to put more and more of those special enemies that put boxes on the ground is huge. Here’s where we come to a gentle roast.
Soulstone Survivors: What Is You Doing?
A lesser discussed but very fun HS game is Soulstone Survivors. It has probably the most robust metagame / progression level of any HS game. Next to unlocking characters, you craft weapons for your characters, you unlock runes to customize them for a run, and then there’s this Skill Tree that would make Path of Exile blush:

It’s a very good game! I’ve gotten 227h of brainless fun out of it! That’s pretty good for $9.99!
But it’s also the most cowardly of HS games (League of Legends: Swarm notwithstanding). Every character (save one, which is OP in other ways) has a number of dashes; you can improve them with upgrades during your run; and they’re absolutely necessary for survival. That’s because enemies that telegraph heavy attacks you absolutely must dodge start appearing soon and eventually flood your game in ludicrous numbers. This means your game eventually looks like a chaotic mess of circles, lines, and rectangles all vying for your attention. Moment to moment, you can barely afford to think about anything but “which attack can I dodge by walking and when do I have to dash? Can I afford to tank this smaller hit?” In moderation this is a fun decision, but moderation is not a word Soulstone Survivors knows.
The various elements of an HS make this kind of skill expression not very fun in the long term. Enemies move and act predictably, you don’t have any meaningful decisions to make beyond “how do I get out of here”, and the thrill of action coming at you thick and fast wears thin pretty quickly.
Plus, with your own weapon effects stacked on top of the absolute mess of enemies, enemies telegraphs, hit effects, damage numbers, etc, the game becomes absolutely unreadable and silly, which is both the best and the worst thing a horde survivor can be. Look at this glorious screenshot:

Mechanics Shallow Dive: Invisible Bullets
We call this genre a bullet heaven because you, the player, spawn an unreasonable amount of bullets which tear through an enemy horde. Of course, in practice you often spawn things other than projectiles: you may create an area around yourself that damage enemies, you may create static damage areas on the floor, you may reach out in sickle-shaped whip attacks or project rotating flamethrower blasts, and so on.
Pushing this beyond its breaking point is the point. But sometimes the game will push you beyond the breaking point when you’re not done with the run yet. And maybe there’s still some enemies that you actually have to dodge, but you can’t see them anymore because of all the particle spam you put out into the world.
Enter: Special effects visibility.

This is a setting Soulstone Survivors introduced. It turns down the opacity of any VFX created by a player action. This allows you to see where the dangerous boss is even through 12 layers of bladestorm and power blast explosions.

But it leads to an even more absurd and yet underwhelming extreme: eventually, the best thing to do is to make your own attacks invisible. This is best both for your frame rate and for general visibility of the game.
And it looks super dumb.
I mean, it’s not like you were reacting to your own attacks at this point anymore. Early in a game of VS, you might try to memorize the timing of the Whip attack and make sure you stand in just the right place when it triggers, but eventually you vomit so many damage effects around you at all times that it suffices to be roughly in the same zip code as the enemy.
And so you walk around peacefully, spamming animations but seeing no effects in the world, except that enemies mysteriously die around you.
This is the logical extreme of “more everything! Bigger effects! Fuck now I can’t see anymore! More transparency! Okay cool there’s room again! More effects!” I end most of my endless runs in SSS with my effects turned off.
But remember that the motivating fantasy here is bullet heaven. If you can’t see your bullets, are you even in heaven?
Mechanics Opportunity: Enemy Movement Patterns
So let’s wrap this up with what I think is one of the most underexplored mechanics in horde survivor games: enemy movement patterns.
Recall how the term bullet heaven derives from the older “bullet hell”. Games that often look like this:

The gameplay is mostly about finding the gaps or else squeezing between two bullets when they’ve separated sufficiently. (This is where that single pixel hitbox in the middle of the player sprite we mentioned before comes in handy).
Bullet Heaven games SORT of create this gameplay, but haphazardly and “organically”. As enemies spawn and block each other’s movement, you often get brief patterns that are reminiscent of the bullet patterns in bullet hell games, but to my knowledge no bullet heaven does this purposefully, where enemies are aware of each other’s position and try to for instance keep gaps open. This is interesting design space I think and someone should play around with it! Not me though, sounds like work. Unlike writing a 5700 word blog post. That is pure pleasure.
Left As An Exercise For The Reader: Scaling
This could be its own 5k word post, easily, but I’ll leave it as an exercise for you, dear reader. Scaling! HS games are about getting ludicrously strong and sure the “broken” state of trivially murdering hundreds of thousands of enemies is the power fantasy, but it shouldn’t be reached too easily or too often. Flow, remember, happens when the friction is high enough that we’re engaged but not so high we’re overwhelmed. So letting the player scale to mega monster stage each run is counterproductive.
How then do you make sure the player feels like they’re meaningfully gaining power while making sure the monsters remain at least somewhat challenging for let’s say at least 80% of a successful run? How do you make sure that scaling the monsters doesn’t invalidate the player’s power fantasy? (If you do 10x damage but the monsters have 10x HP, do you really feel stronger?) (This is an honest question; there are ways to make you feel stronger even in this mathematically stagnant position; they have to do with presentation and framing)
Also, the roguelike RNG nature of the game means that variance between really good and really bad runs can be extreme. Do you have monster scaling respond to player power? Do you scale agnostic of the player’s luck, simply by time? Do you use a proxy for player power, such as kills per minute, and scale your monsters based on this?
And finally and most importantly, how do you keep runaway cases from happening? Maybe the case where the player falls behind so much they can’t kill monsters anymore and thus don’t get xp and thus fall even further behind is actually good? Maybe that’s the failure mode you need to get the player engaged and angry at the enemies, ready to go again? But what’s the failure mode of this failure mode? Hint: it closely parallels the nasty failure mode of a League game where you’ve clearly lost against a superior team but must still play out another 10 minutes or so, held captive by Leaver Protection. How do we make sure we trigger a mercy kill?
And what about the positive feedback loop runaway case? This is analogous to a Civ game where you’re 25% of turns through the game and have 5x the science income of your next AI opponent. You have obviously won the game, but officially winning requires playing through a boring, low-to-no friction slop of foregone conclusions. How do we avoid this? Is there such a thing as an anti-mercy kill? Is that even DESIREABLE?
Please have your answers ready by next Monday.