🖥️ PC2020Supergiant GamesHARDHades rebuild deck
HARD
Feel here is arithmetic — cancel windows, a 0.12s input buffer, four frames of hitstop — and an agent will hand you plausible numbers that produce mush you cannot diagnose by playing.
You will get the loop working. Then the real work starts.
Ask for the frame data as a table and actually read it, because feel here is arithmetic and an agent will hand you plausible numbers that produce mush. The specific failure is the cancel: leave it out and every swing commits you for its full recovery, and players will call the game unresponsive without being able to tell you why. Everything stacked above the combat — boons, rooms, shop, meta-currency — is a weekend of JSON and it works. The twenty thousand voiced lines underneath it, and the system choosing which one you have earned, are what took the studio three years.
- Difficulty
- HARD · 3 of 5
- First playable
- A project
- The original
- 55 team-years
- Released
- 2020 · PC
- Genre
- Action Roguelike
Build one biome of an isometric action roguelike: one weapon, four room layouts, one boss. Canvas 2D, or three.js orthographic. Not the underworld. Requirements: - The dash is the universal cancel and it is the load-bearing detail. It interrupts any attack once its active frames have passed, carries invulnerability for 0.15s from frame two, and any input during an animation lock is buffered 0.12s and fires the instant the lock ends. Without the buffer and the cancel the combat is sticky, and no particle system fixes that. - Attacks defined as data with explicit windup, active and recovery frame counts, so cancel windows are inspectable rather than guessed. - Enemy telegraphs are fixed-duration windups with a ground decal growing to exactly the area about to be hit. Windup length is the difficulty dial; never make an attack unreadable to make it hard. - Rooms are clear-locked: exits open when the last enemy dies. Four layouts, different cover and spawn geometry. - Boons as data-driven modifiers hooked onto verbs the player already has: attack, dash, special. Never a new button. Rarity scales one number, not the effect itself. - A boss with three phases, each adding a telegraph rather than health. - Feel: four frames of hitstop on a connect, arcing damage numbers, and death returning you to a hub with a persistent upgrade. Definition of done: I dash out of the recovery frames of a heavy swing, take nothing from a telegraph I was standing in, and a boon from room one visibly changes my dash all run.
- More than twenty thousand fully voiced lines, roughly 300,000 words, with a priority system deciding which one you have earned right now
- Boon combinatorics across eight gods and six weapons, tuned over nearly two years of early access so that almost any build can clear
- The art direction and Darren Korb's score, which are what make the fortieth attempt feel different from the fourth
- Encounter design across four biomes, where the rooms are composed rather than spawned
Reward-first doors — you pick the next room by the reward shown above the exit, not by where it leads.
It converts a linear crawl into a chain of small informed commitments, which is most of what people mean when they say a game is well paced. Drop it into a level select, a mission board, a dungeon corridor or a skill tree and it improves all four, at the cost of one label.
ScopeAn evening— build just this, not the game
Build a reward-first door chooser as a single self-contained HTML file, DOM or Canvas 2D, no build step. Requirements: - Generate a twelve-room run as a directed acyclic graph. Each room offers two or three exits. - Every exit is labelled with the reward behind it before you commit: a stat boon, currency, a heal, a shop, or a high-risk room paying better. You are choosing the reward, not the room. - Draw rewards from a weighted pool with pity rules, which are the interesting part: never the same reward type on three consecutive doors, always at least one heal offered within any five rooms, and down-weight a boon type already held at maximum. - Clicking a door resolves the room instantly, no combat, and applies the reward. - Keep a run log of every door taken and refused, with the accumulating build beside it. Definition of done: I finish a run, the log shows the doors I turned down, and no five-room stretch in it went without a heal on offer.
Not a clone — Hades re-themed and cut down to something you can finish. The angle is the important reel: it is what turns a project into a weekend. 5,880 ways to land.
Fathomless Flatland
Hades, but 2D instead of 3D, set in cosmic horror. Rendered in PS1 low-poly with texture warping.
Build a browser game called "Fathomless Flatland". The pitch in one line: a action roguelike in the spirit of Hades, but 2D instead of 3D, set in cosmic horror. WHAT THIS IS Take the SHAPE of Hades — its core loop, and whatever makes that loop good — and rebuild it as your own game in a different world at a smaller scope. Use your own names, art and audio, which you want anyway: a reskin of somebody else's game is less interesting than a new one that works the same way. THE ANGLE — this is the important constraint, honour it above everything else 2D instead of 3D. Canvas 2D and sprites. The mechanics survive the dimension drop; the asset budget falls by an order of magnitude. This is what makes the project finishable. If a decision would push the scope back towards the original's, take the smaller option every time. THE THEME Cosmic horror. The threat should be mostly absent. Information the player uncovers should make things worse rather than better. The theme is not a coat of paint. Let it change what the mechanics mean — a reload, a health pack and a locked door should all be things that make sense in this world and nowhere else. LOOK PS1 low-poly with texture warping. Vertex-snap positions to a low-resolution grid in the vertex shader, use affine texture mapping so textures swim, dither to 15-bit colour, and fog the draw distance. REQUIREMENTS - Runs in a browser with no build step. Canvas 2D unless the angle genuinely requires 3D, in which case three.js. - Fixed-timestep update loop with interpolated rendering, so it behaves the same at 60Hz and 144Hz. - Every tuning constant in one CONFIG object at the top, exposed as live sliders in a debug panel. You will find the feel by dragging those, not by prompting. - All content — levels, entities, balance numbers — in JSON or plain data files, never inline in the logic. - Audio generated with the Web Audio API rather than asset files. - Respect prefers-reduced-motion: keep the fades, drop the shake and the parallax. BUILD ORDER 1. The core loop with placeholder rectangles, at the reduced scope the angle demands. No theme, no art, no audio. Make it fun as rectangles first. 2. The angle's consequences. Cutting to bots, or to turn-based, or to one level changes the design rather than just shrinking it — find out how and rewrite step 1. 3. The theme, as content and data. 4. The look, applied last as a rendering layer over a game that already works. Start by writing a one-page plan: the core loop, what the angle removes and what that frees you to do properly, and what you are deliberately leaving out. Wait for me to approve it before writing code.
- GDC Vault — Breathing Life into Greek Myth, Supergiant on how the dialogue actually works
- Vlambeer — The Art of Screenshake, where the hitstop and shake numbers come from
Half the time one of these already does what you wanted, and the other half it saves you a week of solving a problem someone documented in 1997.
One count per game. No account — it is tied to a salted hash of your IP, which is not stored in a form anyone can reverse. Every build adds 55 team-years to the counter on the homepage.
Then try these
Questions
How hard is Hades to rebuild with an AI agent?
Yes — every game is buildable at some scope. The question is how hard, and Hades is HARD (3 of 5). You will get the loop working. Then the real work starts. Feel here is arithmetic — cancel windows, a 0.12s input buffer, four frames of hitstop — and an agent will hand you plausible numbers that produce mush you cannot diagnose by playing. Reckon on a project to something you can actually play — not to something that matches the 55 team-years Supergiant Games spent.
What makes Hades hard to rebuild?
Feel here is arithmetic — cancel windows, a 0.12s input buffer, four frames of hitstop — and an agent will hand you plausible numbers that produce mush you cannot diagnose by playing. That is why it sits at HARD rather than a tier either side of it. If you want the idea without the project, the transferable part is reward-first doors — you pick the next room by the reward shown above the exit, not by where it leads. — an evening of work.
How long does it take to build a Hades clone?
Wrong question, and it is worth saying why. Nobody rebuilds Hades — the original took Supergiant Games on the order of 55 team-years, released in 2020 on PC. What you can do is reach something playable and genuinely yours in about a project, and then keep going for as long as it stays interesting. Any figure that claims otherwise is measuring a prototype and calling it a game.
What do I lose by building my own Hades?
More than twenty thousand fully voiced lines, roughly 300,000 words, with a priority system deciding which one you have earned right now. Boon combinatorics across eight gods and six weapons, tuned over nearly two years of early access so that almost any build can clear. The art direction and Darren Korb's score, which are what make the fortieth attempt feel different from the fourth. Those are the parts a prompt will not hand you.
Which AI coding agent should I use to build it?
Any of Claude Code, Codex or Cursor will handle this prompt. Paste it as your first message, let the agent scaffold the project, then iterate on feel — the second, third and tenth prompts are where a game actually gets good. Work in an empty folder or a fresh git branch so you can throw it away cheaply.
One email when new games go on the list. No other reason to email you, ever.