Video Game Sources // Field manual for people who play82 entries · 2026
Index / Genres & Recommendations / VGS-GEN-0157
Genres & Recommendations

What Actually Defines a Roguelike? A Real Answer

VGS-GEN-0157 · REV.A · Written 2026-08-02 · Last checked 2026-08-02 · 8 min

The short answer

A roguelike is a game built around runs that end in permanent death, a layout that’s regenerated or reshuffled each time, and permadeath that actually matters — you lose your progress and start over. The term traces back to a single 1980s computer game called Rogue, and the definition has been arguing with itself ever since.

§01What is the original definition of a roguelike?

The strict definition comes from a small group of genre fans who, in 2008, wrote up what’s often called the “Berlin Interpretation” — a set of traits based directly on Rogue, the dungeon-crawling computer game from the early 1980s that gave the genre its name.

Those traits included things like:

  • Randomly generated dungeon levels, different every playthrough
  • Permadeath — when your character dies, the save is gone, not just paused
  • Turn-based movement rather than real-time action
  • Grid-based movement, historically tied to a single character per tile
  • Complexity that comes from combining simple systems (items, monsters, traps) in surprising ways

Under this narrow reading, almost nothing released commercially in the last two decades actually qualifies. That’s not a flaw in the definition — it was written by and for a community of people who play traditional, ASCII-based dungeon crawlers and wanted a word that meant something specific. The problem is that the wider gaming public adopted the word “roguelike” long before most players had ever heard of the Berlin Interpretation.

§02So why do people call Hades or Slay the Spire roguelikes?

Because the industry and most players use a looser, functional definition: a game with procedurally generated or reshuffled runs, permanent death that sends you back to the start, and progress that resets each attempt but often carries forward as meta-upgrades. That’s different from the strict original definition, and both uses are common.

This looser sense is sometimes called “roguelite,” a word coined specifically to separate action games with roguelike structure from the turn-based dungeon crawlers the original term described. In practice, most people — including a lot of press coverage and storefronts — use “roguelike” and “roguelite” more or less interchangeably, which is exactly why the argument never settles.

Under the loose definition, a huge range of games qualify:

  • Twin-stick shooters with randomized rooms and gear
  • Card-battling games where your deck resets between runs
  • Platformers with procedurally arranged levels and one-hit permadeath
  • Turn-based tactics games where a wipe means starting the campaign over

None of these use a grid of ASCII characters or turn-based movement in the original sense, but they all share the run-based structure that made Rogue distinctive: you don’t save and reload your way past failure, you start again with a new layout and (usually) a little more knowledge or a permanent unlock to show for it.

§03What are the actual core ingredients, stripped of jargon?

Three things show up in almost every game people call a roguelike, regardless of which camp they belong to: death that costs you the run, a world that’s different each time, and progress that resets but isn’t entirely wasted.

  1. Permadeath. Dying ends the attempt. You don’t reload a manual save from ten minutes ago; you go back to the beginning, or to whatever earlier point the game defines as “start.”
  2. Procedural or randomized structure. Levels, item drops, enemy placement, or even story beats are generated or shuffled by rules rather than hand-placed identically every time. No two runs look exactly alike.
  3. Runs, not campaigns. The game is built to be played in relatively short, self-contained attempts, and the loop of “try, fail, learn, try again” is the actual game, not an obstacle between story missions.

A lot of the genre debate is really about how you feel about the save system. Traditional roguelikes and their action-oriented descendants both reject the idea of saving mid-attempt to dodge consequences. That puts the roguelike family in direct contrast with the way most other genres evolved — our piece on how the save game evolved covers how thoroughly the industry moved toward frequent, forgiving saves, which makes permadeath-based design feel almost deliberately old-fashioned by comparison.

Hand-drawn dungeon maps and dice representing the procedural, run-based structure that defines a roguelike

§04Why does the genre argument get so heated?

Because it’s really a proxy argument about difficulty, respect for a game’s history, and whether a marketing label should be allowed to stretch until it means almost nothing. None of those questions have a clean answer, which is why the debate resurfaces every time a new game gets tagged “roguelike” on a storefront.

Purists worry that if “roguelike” can describe both a text-based dungeon crawler from the 1980s and a fast, colorful action game with voiced characters and a soundtrack, the word has stopped doing useful work. If everything with random level layouts and permadeath is a roguelike, the term can’t tell a shopper anything specific about pacing, tone, or genre lineage.

Meanwhile, players who came to the idea through modern action games often don’t care about the etymology at all — they care that “roguelike” reliably signals a certain kind of replay loop: short attempts, escalating risk, and the promise that failure still teaches you something useful for next time. That’s a genuinely useful shorthand even if it annoys genre historians.

This isn’t a new pattern. Genre labels shift and blur constantly in gaming — the same kind of argument happens around whether a “Metroidvania” needs backtracking on a connected map, or whether a “soulslike” needs anything more than difficulty and a specific stamina system. Language in games tends to follow what sells and what press coverage repeats, not what a small founding community intended.

§05Does a roguelike need to be turn-based?

No, not anymore. Turn-based movement was part of the original Berlin Interpretation because Rogue itself was turn-based, but most games commonly called roguelikes today are real-time action games. The turn-based requirement only applies if you’re using the strict, historical definition.

This is probably the single biggest split between the old and new meanings of the word. A true traditionalist would say a real-time action game simply cannot be a roguelike no matter what else it does, because turn-based tactical decision-making was central to what made Rogue interesting — you had time to think about every step. A more relaxed reader points out that plenty of games nobody disputes as “roguelike-adjacent” are real-time, and that gatekeeping the term this tightly makes it useless for describing anything released in the last fifteen years.

There’s no umpire here. Both readings are internally consistent; they’re just answering different questions. One is asking “does this game descend directly from Rogue’s mechanical DNA?” The other is asking “does this game create the same feeling of run-based risk and restart that made Rogue’s descendants popular?”

§06Where did the genre actually come from, and why does it matter now?

It grew out of small, mostly free computer games made by hobbyists and university students in the 1980s, long before “indie” was a marketing category. That scrappy, low-budget origin is a big part of why roguelike design resurfaced so strongly once independent development became commercially viable again.

Procedural generation and permadeath are both cheap in the specific sense that matters to a small team: they multiply the amount of content a limited team can produce without hand-building every level, and they turn a modest amount of hand-built content into a large number of distinct playthroughs. That’s a huge part of why the genre boomed alongside the broader indie scene — for more on how small teams reshaped what counted as a legitimate, commercially serious release, see our piece on how indie games became a serious part of the industry.

It also connects to a broader shift in how difficulty and failure are treated across gaming history. Older games were often punishing because arcade cabinets needed players to spend more coins, a pattern explored in how arcades shaped the games we play today. Roguelikes inherited some of that unforgiving spirit deliberately rather than for coin-op economics, which is part of why the genre gets discussed alongside broader questions about why classic games were so punishingly difficult in the first place.

§07Is “roguelike” even a useful label for buying decisions?

Mostly yes, if you treat it as a signal about structure rather than a strict genre guarantee. It tells you to expect runs, permadeath, and randomized layouts — not whether the game is turn-based or real-time, story-driven or abstract, slow or fast.

When deciding whether a game tagged “roguelike” is actually worth buying, it helps to ask a few practical questions instead of relying on the label alone:

  • Does progress carry over between runs at all, or does every failure erase everything?
  • How long is a typical run — minutes, or a couple of hours?
  • Is the randomization surface-level (item placement) or does it change the actual challenge (enemy types, level shape, available builds)?
  • Is failure treated as content — new information, new options — or purely as a setback?

Those answers matter more for whether you’ll enjoy a specific game than whether it technically fits the Berlin Interpretation. If you’re weighing whether a particular roguelike-tagged release is worth its asking price, the same general reasoning in our guide on deciding if a game is worth full price applies — replay value from procedural variety is a real selling point, but only if the underlying moment-to-moment play holds up across dozens of attempts.

It’s also worth noting that this genre confusion isn’t unique to roguelikes. Plenty of long-running series get relabeled, rebooted, or reframed as tastes shift, which is part of a broader pattern covered in why so many long-running series reboot themselves. Genre words are marketing tools as much as technical descriptions, and they drift for the same commercial reasons franchises do.

§08So what’s the honest, short answer?

There isn’t one single agreed definition, and treating either camp as simply wrong misses the point. Strictly, a roguelike is a turn-based dungeon crawler descended from Rogue. Loosely — and far more commonly today — it’s any game built around procedurally varied, permadeath-driven runs. Both uses are legitimate; they’re just answering different questions about lineage versus feel.

Readers digging into genre history more broadly can find related context in our Game History section, and anyone specifically shopping for roguelike-style games might find the general reasoning in our Buying Advice archive useful for figuring out which of these run-based games are worth the time investment before committing.

Leave a Comment

VIDEO GAME SOURCESParts list · 82 entries
© 2026 Video Game Sources · Drawn not to scaleIndependent — we buy the hardware ourselves
About  ·  Contact  ·  Privacy Policy
© 2026 Video Game Sources