The Quest for El Dorado

Deep Dive for an iOS Roguelite Translation

Project context: Translate the core loop of The Quest for El Dorado (Reiner Knizia, Ravensburger, 2017) into a single-player iOS roguelite. The "Dominion -> Slay the Spire" pattern, but for a racing deckbuilder. No direct digital port exists; the genre slot — deckbuilder racer on iOS — is essentially empty.

1. The mechanics, precisely

A turn from the explorer's seat. You start with an 8-card pauper deck — four basic Travelers, two Sailors, two Explorers — and shuffle it into a personal draw pile. Each turn you draw up to a hand of 4 cards and choose, in any order, to either play cards onto the hex map to move your pawn or spend cards as currency at the central market. There is one resource: coins printed on cards. Movement cards each show a terrain type — green machetes for jungle, blue paddles for water, yellow coins (which double as currency) for villages — and you pay a hex by playing one or more cards whose symbols match its cost. A jungle space marked "2 machetes" eats a 2-machete card whole, or a 3-machete card with one pip wasted, or two 1-machete cards combined. Crucially, surplus pips can chain into the next hex of the same terrain type — a 3-machete card can carry you through three consecutive 1-machete jungle hexes in one motion, but cannot spill into a water hex. After you've spent every card you wanted to spend, you discard your entire hand (used or not), draw 4 new cards, and pass. (UltraBoardGames rules, BGG entry)

The market and deck construction. The center of the table is an Explorer-style row of six face-up card piles, each at a fixed coin cost from 1 to 4 — Sailor (1), Traveler (1), Explorer (2), Scout (2), Captain (3), Cartographer (4), plus rarer 4-cost specialists. To buy, you discard cards from hand whose total coin pips meet the sticker price; the bought card goes to your discard, not your hand, so it cycles in next reshuffle. Unlike Dominion you do not get a free buy per turn — your whole turn can be one purchase, or a 4-hex push, or a 2-hex push followed by a small purchase with the leftover. Every purchase is a tempo trade: cards spent as coins are not moving you, and a fat new card might not see your hand for two reshuffles. (Tabletop Together review, Punchboard review)

Why typed-cards convert deckbuilding into a route puzzle. Dominion's terminal/non-terminal split is purely mathematical — cards combo because of "+actions" and "+cards" tokens. El Dorado's combo space is spatial. The board in front of you might offer two routes to the next checkpoint: a five-hex jungle corridor (machetes) or a three-hex water-then-village dogleg (paddles plus coins). Your current hand of 4 and your deck composition over the next three reshuffles must agree with one of those routes. You cannot just "play the highest card." A 4-coin Cartographer in your hand is a brick on a jungle hex. The whole strategic loop becomes: read the terrain ahead, prune your starter deck of off-color cards (there are remove-card market spaces explicitly for this), and bias your buys toward the route you committed to. That commit-to-a-color is what gives the deckbuild stakes. (Thoughtful Gamer breakdown, Meeple Mountain review)

Blockades, caves, special hexes, and the El Dorado dash. Between map tiles sit blockade tokens — barrier hexes that cost a fixed number of coins paid all at once (no chaining, no movement reward) and once paid are removed, opening the way for trailing players too. This creates a real dilemma: pay 4 coins to bulldoze the gate yourself, or sit a turn behind a rival and let them eat the toll. Cave tokens scattered on certain hexes give a one-time bonus when you stop on them (an extra coin, a card-removal, a free move) but each token can only be claimed once per game — the leader scoops them, the chasers find empty stones. The map is built from 10 double-sided modular tiles plus 8 barrier tiles, advertised as 100,000+ combinations, and the final tile always contains the El Dorado hex. The end-game trigger is binary: first pawn to enter El Dorado wins immediately. There is no scoring, no second place, no tiebreak — the race ends the instant a pawn touches gold. (Board Game Quest review, BGE encyclopedia entry)

2. What makes Quest for El Dorado work as a tabletop game

Knizia's "tighten everything" design philosophy. Knizia's stated design ethos is "Simple games, but then the people bring themselves into it. And you see out of the simplicity, a second level of depth." El Dorado is the deckbuilder reduced to its load-bearing studs: one currency, three terrain types, four cards per hand, one binary win condition. There is no VP track, no end-of-game scoring, no attack cards, no trash-this-then-gain-that wildcards. The entire genre's ornamental scaffolding has been sanded off, leaving a deckbuilder a 10-year-old can learn in ten minutes that still rewards a 1.94-weight strategist for fifty plays. (Game Developer profile, Tabletop Trove on Knizia)

Compared to Dominion: the same engine-build, but movement is the points. Dominion's deckbuild outputs a number — VP — that is invisible during play and only resolved at game end. The thrill is abstract; the feedback delayed. El Dorado's deckbuild outputs physical position on a shared map every single turn. Every card you buy will, two reshuffles later, either move your pawn three hexes or sit dead in a hand on the wrong terrain. That converts Dominion's spreadsheet-y engine-tuning into a visceral, glanceable tug-of-war: you are behind, you can see you are behind, you can see the cave token the leader just claimed, and your next purchase is overtly an answer to that. The race chassis gives the deckbuilder permanent dramatic stakes that Dominion has to manufacture with attack cards and curses. (Meeple Mountain review, BGG review thread)

What the race gives that economic deckbuilders can't. Three things. First, visible progress — you don't need to read the state, you just look at where the pawns are. Second, instant feedback on every purchase — a Captain hits the table, gets shuffled, and within two turns you either moved further with it or didn't. The judgment is cheap and frequent. Third, finish-line drama: the last 90 seconds of an El Dorado game, with two pawns within four hexes of the gold square and one player counting machetes in their discard pile, is structurally impossible in Dominion because Dominion has no spatial home stretch. The race is the dopamine engine the genre was missing. (Punchboard review, Board Gamers Anonymous review)

Expansions that thicken the loop. Heroes & Hexes (2018) adds a tavern row of one-shot Hero cards (drafted alongside the main market), familiar cards that piggyback on plays, and a fourth terrain — demon hexes that demand specific Hero plays — plus curse tokens that punish shortcut-takers with random debuffs. The Golden Temples (2019) is a standalone successor with new map content and a temple-stop mechanic for mid-race objectives. The expansions confirm something useful for a digital design: the base loop is under-leveraged. There is room to layer modifiers, tokens, and one-shot cards without breaking the chassis. A roguelite that adds a "Hex Pool" or "Relic Pool" is well within the design grammar of the IP. (Heroes & Hexes review at Board Game Quest, Shut Up & Sit Down review, What's Eric Playing on Heroes & Hexes)

3. Existing digital attempts

Official digital El Dorado: none. As of April 2026, Ravensburger has shipped no app version on iOS, Android, or Steam. Their digital portfolio is heavy on jigsaw puzzles and family titles (Labyrinth has a port; Disney Villainous has one); El Dorado is conspicuously absent. The Ravensburger product pages list only physical SKUs for the base game and both expansions. (Ravensburger US product page, Ravensburger UK product page)

Community implementations are thin. El Dorado is not on Board Game Arena as of the most recent BGG threads requesting it (2023-2024), despite a multi-year stream of user requests. There is no active Tabletopia room, no widely-shared Tabletop Simulator workshop mod that the BGG community pins. The closest you get to playing it online is hand-rolled solo variants distributed as PDF on BGG — Solo/Co-op Variant, 7 Caves Solo Variant, and a more recent Solo/Co-op/Competitive file with a rising-flood timer mechanic. These confirm an appetite for solo play that Ravensburger has not addressed. (BGG: Board Game Arena thread, BGG: Digital version request, BGG: Solo/Co-op variant file, BGG: 7 Caves Solo variant)

Adjacent genres on PC and mobile. Deck RX: The Deckbuilding Racing Game (Meteorbyte Studios, Steam) is the only marketed product that combines the words "deckbuilder" and "racing" — same comp as in the Heat doc — but it is vehicular combat coded as racing, six themed worlds, eight characters, 50+ car upgrades, PC-first. It does not touch El Dorado's terrain-typed-card-vs-hex puzzle at all. Dominion has an official iOS/Android client (Temple Gates Games) and proves that paid-premium pure deckbuilders work on phones — the engine-build feels great on touch, the bottleneck is content depth and AI quality. Slay the Spire mobile is the financial proof point: $9.99 premium, no energy mechanics, six-figure download counts. Dawncaster is the mobile-native roguelite deckbuilder reference for UI density on a phone screen. None of these are races. (Deck RX on Steam, Deck RX studio page, Dawncaster on Steam)

The empty lane. A turn-based, deckbuilder-driven, terrain-typed racing roguelite on iOS does not exist. Knizia's loop is the cleanest existing tabletop blueprint for that lane, the IP is unclaimed digitally, and the appetite is documented in every BGG thread asking when an app is coming.

4. The translation problem

What gets stripped. The big losses are social. Lead-change trash talk — the table-wide groan when someone leapfrogs from third to first by chaining a freshly-bought Cartographer through a four-hex water stretch — has no single-player analog. Physical hex laying — the ritual of picking which two-sided tile lands where for tonight's race — is half the game's variety pitch (100,000+ combinations) and on a phone collapses into "the procgen made another map." Drafting in front of a table — watching opponents buy the last Captain — turns from social signaling into a UI list. None of these are showstoppers, but they are the warm half of the game's appeal and they cool off in translation.

What gets added (the digital meta-layer). Five candidates worth weighing:

Procedural jungle generation with constraint solving. The 10-tile/100k-combo pitch is a procgen brief in disguise. Generate maps under hard constraints: minimum two viable color routes from start to finish, at least one blockade per map quadrant, exactly one cave per quadrant, one El Dorado tile pinned at a tunable distance.
Run-based map progression. A run = 5 jungles of escalating distance and hostility (jungle 1 is short and forgiving; jungle 5 is long, has demon hexes from Heroes & Hexes, and a named rival). Lose a jungle and you bleed an "expedition supply"; lose three and the run ends.
Draftable upgrade pool. Between jungles, draft 1 of 3 cards from a pool seeded by your current run's biome. Roguelite-standard pick-3.
Themed expedition equipment (relics). Persistent passives for a run: Brass Compass (first card played each turn gets +1 pip), Quinine Kit (curse hexes ignored), Sextant (peek at top 2 of deck before drawing). Two slots, drafted between jungles.
Daily challenges. Fixed map seed + fixed starter deck variant + fixed rivals; global leaderboard by finish-turn count. Tweet-the-screenshot social hook.

Recommendation: ship v1 with #1 + #3 + #4. Procgen is the spine — the game cannot exist without it. The card-draft is the run-shape. Equipment is the build identity that makes runs distinguishable. Daily challenges and full career mode are post-launch retention; cutting them keeps the MVP scope honest. (This mirrors the v1 cut chosen in the Heat deep dive, and for the same reason: a solo-dev premium roguelite needs three meta-mechanics, not five.)

What cannot translate cleanly. The physical scarcity of the central market — six visible piles depleting in real time, watched by every player — turns into a refresh row on a phone. Mitigation: an animated "rival just bought" ticker so the player feels the pool draining. The commitment cost of buying a card you won't see for two reshuffles loses sting in single-player because the AI's reshuffle timing is opaque. Mitigation: a small persistent "deck composition" inspector that shows hand-color distribution over the next two cycles — heretical to purists, necessary on a 6-inch screen.

5. Concrete iOS prototype spec

Target: iOS portrait, single-developer Unity or SwiftUI+SpriteKit, 4-6 month MVP, premium one-time purchase $6.99-$9.99.

60-second core loop. Tap a card in your hand of 4. The map highlights every hex you can legally reach with that card alone (color-matched, chained through same-color neighbors). Tap a destination hex; the pawn animates along the chained path, the card slides to your discard. Repeat with remaining hand. Or tap the market drawer at the bottom of the screen, drag coin-bearing cards onto a market pile to buy. Tap "End Turn"; the rivals' moves animate sequentially in 1-2 seconds total; your hand discards and redraws. A single jungle is ~10-16 such turns; ~3-5 minutes. A full run of 5 jungles is one 25-30 minute session — long for a phone, deliberately session-shaped like Slay the Spire.

Single-screen layout (phone portrait, 6.1" reference):

Top 55%: Hex map, scrollable/zoomable but defaulting to a framing that shows your pawn, the next checkpoint, and 2 rivals. El Dorado pulses at the far edge when within sight.
Middle 10%: Status strip — turn counter, deck composition pip-bar (green/blue/yellow ratio), expedition supplies remaining.
Bottom 35%: Hand of 4 cards (large enough for thumb-tap and drag), market drawer pull-tab on the right, end-turn button. Drag a card up onto the map to play; drag right onto the drawer to spend.

Card system.

Starter deck: 8 cards identical to the board game — 4 Travelers (1 machete OR 1 paddle OR 1 coin, each as their own subtype split 2/1/1), 2 Sailors (1 paddle), 2 Explorers (1 machete + 1 coin). Mirror the physical balance — there's a decade of playtesting behind these numbers.
Market grid: 6 visible piles in a horizontal-scroll drawer, costed 1/1/2/2/3/4. A pile depleting to zero remains as a ghost slot — refilled per board-game rules only on game-end conditions, not refreshed mid-run.
Visual language: green hexes/cards = jungle/machete, blue = water/paddle, yellow = village/coin. Bold flat-vector icons big enough to read on a 6" screen at thumb distance.

Run structure.

5 jungles per run: J1 (5-6 hexes deep, 1 rival, no blockades) -> J5 (15+ hexes, 3 rivals, demon hexes, named boss).
3-4 AI rivals per jungle, drawn from a personality roster of ~8 (the Hoarder buys aggressively; the Sprinter never buys; the Cartographer-spammer; the Naturalist who specializes in one color).
Win a jungle = reach El Dorado first. Lose = a rival reaches first; you forfeit one expedition supply (start with 3).
Lose all 3 supplies = run ends. Reach jungle 5 with supplies remaining = championship final, named-rival fight.

Meta-progression.

Between jungles: pick 1 of 3 draftable cards (replaces or adds to deck) plus 1 of 2 equipment relics if a slot is open (max 2).
Between runs (persistent): unlock new starting decks (different 8-card profiles — "Trader" coin-heavy, "Naturalist" machete-heavy), new biomes (desert/river/mountain reskins with shifted terrain ratios), new rival personalities, new relics. Target 30-40 unlockables for a ~20 hour completion arc.

Monetization. Premium $6.99-$9.99, no energy gates, no IAP. Cosmetic pawn skins as a single $2.99 pack post-launch if needed. Avoid F2P monetization entirely — a roguelite run is a session, and gating sessions kills the loop. This is the same pricing logic as Slay the Spire mobile and Dawncaster.

Aesthetic direction. Lean into the illustrated South American jungle of the physical box without overdoing the pulp tone. Hex tiles in flat-vector with painterly illustration insets; pawn miniatures as 3D-rendered toy figurines on a 2.5D board (à la Mini Motorways' tactility). Sound: ambient jungle bed during planning, percussive thock on each hex traversed, brass-band sting on El Dorado entry.

6. Risks and unknowns

Top design risks.

Terrain-card balance breaks under randomized maps. Hand-designed tiles guarantee both routes are viable; procgen can produce a map with a 90% jungle path and a dead-end paddle alternative, collapsing the route-puzzle into "buy machetes, win." Mitigation: constraint-solver procgen — every map must validate ≥2 distinct color routes within 10% of each other on a baseline-deck simulation. Reject and reroll otherwise.
Solo pacing against scripted AI may feel like a puzzle, not a race. Without a human opponent's bluffing and visible regret, the AI's decisions read as predictable. Mitigation: personality archetypes with legible quirks (the Sprinter never buys — you can see their deck stays small), occasional surprise plays (a 5% "the Hoarder breaks character" event), and "dramatic" framing — a rival closing within 2 hexes triggers a UI alert and a sting, even if it was probabilistically expected.
Draft snowball. Deckbuilders draft pools love compounding — pull two coin-pip cards early and you can never not hit the 4-cost row. Mitigation: draft pools tier by current deck composition (offer counter-color cards more often as you skew); cap relic stacking per category.
Race feel collapses without other humans. The board game's lead-change tension is partly social — the groan, the heckle. Solo, it's a number on a screen. Mitigation: this is the unsolved core risk. Lean on tight camera framing, sound design, and rival personalities, but accept this is the project's biggest open question. (See the same risk in Heat §6.)
25-30 minute runs may be too long for a phone session. Slay the Spire runs are 45-90 min and players accept it; El Dorado's tighter loop can probably hit ~25 if jungle 1 and 2 are kept short.

What needs validation, in order.

Tabletop solo playthrough first. Play El Dorado solo with the 7 Caves or Solo/Co-op variant for 5-10 races. Confirm the loop survives single-player on a table at all. If the table version of solo bores you, the phone version definitely will.
Paper prototype the procgen. Hand-deal random tile combinations; test how often a baseline starter deck can finish in <16 turns. Tune constraint rules.
Greybox digital prototype — one jungle, one rival, no meta-layer. Measure: average run length on a phone, tap counts per turn, where the player's eye goes during AI animation.

Three questions for the prototype builder.

How much will you deviate from the 4-card hand? A 4-card hand is tight on a 6" screen but anchors the entire game's tempo. Increasing to 5 makes UX easier and runs shorter, but rebalances every market price. Pick a side before building.
Will rivals share the player's market or have their own? Sharing market preserves the "they grabbed the last Captain" tension but forces the AI's purchasing decisions to be transparent. Separate markets are easier to balance, lose the depletion drama. Pick before architecture is locked.
How sacred is the binary win condition? The board game ends instantly on first-pawn-to-El-Dorado, no second place. A roguelite typically rewards graduated outcomes (silver/bronze tiers, partial XP for losing). Heretical to fans, but losing a 25-minute run with literally zero progression reward is a steep retention cliff. Pick a stance.