diff --git a/src/chain/chainProfile.js b/src/chain/chainProfile.js
index 9eadf99..428ce77 100644
--- a/src/chain/chainProfile.js
+++ b/src/chain/chainProfile.js
@@ -17,7 +17,7 @@ const DEFAULT_DECIMALS = 12;
 
 // family: 'monero' (subaddress + transfer_split), 'utxo' (BTC/LTC-style), 'mimblewimble' (Grin).
 const PROFILES = Object.freeze({
-    WOW:  Object.freeze({ id: 'WOW',  symbol: 'WOW',  decimals: 11, meanBlockTimeMs: 120000, family: 'monero',       uriScheme: 'wownero'  }),
+    WOW:  Object.freeze({ id: 'WOW',  symbol: 'WOW',  decimals: 11, meanBlockTimeMs: 300000, family: 'monero',       uriScheme: 'wownero'  }),
     XMR:  Object.freeze({ id: 'XMR',  symbol: 'XMR',  decimals: 12, meanBlockTimeMs: 120000, family: 'monero',       uriScheme: 'monero'   }),
     BTC:  Object.freeze({ id: 'BTC',  symbol: 'BTC',  decimals: 8,  meanBlockTimeMs: 600000, family: 'utxo',         uriScheme: 'bitcoin'  }),
     LTC:  Object.freeze({ id: 'LTC',  symbol: 'LTC',  decimals: 8,  meanBlockTimeMs: 150000, family: 'utxo',         uriScheme: 'litecoin' }),
diff --git a/src/game/difficultyConfig.js b/src/game/difficultyConfig.js
index 6af1af7..67d7c87 100644
--- a/src/game/difficultyConfig.js
+++ b/src/game/difficultyConfig.js
@@ -13,13 +13,16 @@
  * 5. Treasure placement (farther from exit = riskier)
  * 6. Number of monsters (future enhancement)
  * 
- * Block times:
- * - Wownero (WOW): ~2 minutes (120 seconds)
- * - Monero (XMR): ~2 minutes (120 seconds)
- * - Bitcoin (for reference): ~10 minutes
- * 
- * Note: Previously docs said WOW was 5 min but it's actually 2 min like XMR.
- * We keep the config flexible for future adjustments.
+ * Block times (measured against live daemons; see chainProfile.js meanBlockTimeMs):
+ * - Grin (GRIN): ~1 minute
+ * - Monero (XMR): ~2 minutes
+ * - Litecoin (LTC): ~2.5 minutes
+ * - Wownero (WOW): ~5 minutes  (measured 5.01 min/block over 1000 blocks — an earlier "correction"
+ *   to 2 min was WRONG; WOW is a SLOW chain, closer to BTC than XMR for calibration purposes)
+ * - Bitcoin (BTC): ~10 minutes
+ *
+ * Slow chains (WOW/BTC) can't get a timer-driven house edge from a single dungeon without it
+ * becoming an unplayable slog → they want MULTI-LEVEL depth. See NETWORK_TUNING below.
  */
 
 const DIFFICULTY_PRESETS = {
@@ -47,7 +50,7 @@ const DIFFICULTY_PRESETS = {
         targetHouseWinRate: 0.3           // 30% house wins (player-friendly)
     },
 
-    // Normal - balanced for Wownero (2 min blocks)
+    // Normal - balanced baseline (~2 min blocks, i.e. XMR; WOW is actually ~5 min — see header)
     normal: {
         dungeon: {
             width: 45,
@@ -144,32 +147,32 @@ function getDifficultyPreset(cryptoType = 'WOW', overridePreset = null) {
     return { ...DIFFICULTY_PRESETS.casino, presetName: 'casino' };
 }
 
-// Per-network difficulty tuning, SOLVED by the balance sim (src/sim/calibrate.js) so cryptoType
-// finally shapes difficulty instead of being a dead parameter. Two independent levers:
-//   sizeScale ∝ √(blockTime/ref) — PACING: keeps a run's length a consistent fraction of the block
-//     interval (GRIN sprint … BTC epic), applied to EVERY preset; the timer then contributes a
-//     stable baseline. Reference network = WOW/XMR (2-min blocks) = 1.0.
-//   monsterSpeed (movesPerPlayerMove) — the house EDGE lever; calibrated for the CASINO preset
-//     (70% target) only, so it's applied ONLY to casino. On slow chains (BTC) the timer barely bites
-//     a ~30s run, so the monster carries the edge (and caps ~64% even maxed — a real property of
-//     block-timed play, not a bug).
-// CAVEAT: the sim bots don't actively evade the monster, so monsterSpeed is a STARTING POINT that
-// under-provisions vs skilled humans — validate/retune with live telemetry. The sizeScale half is
-// bot-robust. Operator env overrides (DUNGEON_*, MONSTER_*) still win. Kill-switch: NETWORK_TUNING_DISABLED=true.
+// Per-network SIZE tuning so cryptoType shapes difficulty (it used to be a dead parameter).
+//   sizeScale ∝ √(blockTime/ref) — PACING: a run's length as a consistent fraction of the block
+//     interval, applied to EVERY preset. Reference = XMR (2-min blocks) = 1.0. Clamped [0.7, 1.6]
+//     so single maps stay playable.
+//
+// DELIBERATELY NO monster-speed lever. The sim proved a supernaturally fast monster (a) feels like
+// cheating and (b) doesn't even WORK on slow chains — WOW (5-min blocks) caps ~64% house-win even
+// at a 2.2× monster on a 111×55 map, because a single dungeon clears in ~1 min << the block, so the
+// timer (the fair, on-theme "race the block" edge) barely bites. The real edge lever for slow chains
+// is MULTI-LEVEL DEPTH (cumulative run length across levels), which is the next build. Until then,
+// slow chains (WOW/BTC) run honest-but-under-target on the timer; the monster stays at its fair
+// preset speed. Operator env overrides (DUNGEON_*, MONSTER_*) still win. Kill: NETWORK_TUNING_DISABLED=true.
 const NETWORK_TUNING = {
-    WOW:  { sizeScale: 1.00, monsterSpeed: 1.08 },
-    XMR:  { sizeScale: 1.00, monsterSpeed: 1.08 },
-    LTC:  { sizeScale: 1.12, monsterSpeed: 1.11 },
-    BTC:  { sizeScale: 1.60, monsterSpeed: 2.20 },
-    GRIN: { sizeScale: 0.71, monsterSpeed: 0.94 }
+    GRIN: { sizeScale: 0.71 }, // ~1 min blocks
+    XMR:  { sizeScale: 1.00 }, // ~2 min (reference)
+    LTC:  { sizeScale: 1.12 }, // ~2.5 min
+    WOW:  { sizeScale: 1.60 }, // ~5 min (measured) — wants multi-level, not one giant map
+    BTC:  { sizeScale: 1.60 }  // ~10 min — wants multi-level
 };
 
-// Fold the per-network tuning into a resolved preset (size for all presets, monster for casino).
+// Fold the per-network SIZE scale into a resolved preset (pacing; monster left at the fair preset speed).
 function applyNetworkTuning(preset, cryptoType) {
     if (process.env.NETWORK_TUNING_DISABLED === 'true') return preset;
     const t = NETWORK_TUNING[String(cryptoType || '').trim().toUpperCase()];
     if (!t) return preset;
-    const out = {
+    return {
         ...preset,
         dungeon: {
             ...preset.dungeon,
@@ -177,10 +180,6 @@ function applyNetworkTuning(preset, cryptoType) {
             height: Math.max(12, Math.round(preset.dungeon.height * t.sizeScale))
         }
     };
-    if (preset.presetName === 'casino' && t.monsterSpeed) {
-        out.monster = { ...preset.monster, movesPerPlayerMove: t.monsterSpeed };
-    }
-    return out;
 }
 
 /**
diff --git a/src/sim/calibrate.js b/src/sim/calibrate.js
index 378a793..85c9d62 100644
--- a/src/sim/calibrate.js
+++ b/src/sim/calibrate.js
@@ -24,7 +24,7 @@ const { BOTS } = require('./simBots');
 const { meanBlockTimeMsFor } = require('../chain/chainProfile');
 const { DIFFICULTY_PRESETS } = require('../game/difficultyConfig');
 
-const BASE_BLOCK_MS = 120000; // WOW/XMR — the reference the size scale is relative to
+const BASE_BLOCK_MS = 120000; // XMR (2-min) — the reference the size scale is relative to
 
 function parseArgs(argv) {
     const a = { target: 0.70, preset: 'casino', bot: 'explorer-greedy', runs: 160, cadence: 320,
diff --git a/test/networkTuning_ultra.test.js b/test/networkTuning_ultra.test.js
index 3cdc7d9..b47d6cf 100644
--- a/test/networkTuning_ultra.test.js
+++ b/test/networkTuning_ultra.test.js
@@ -14,27 +14,31 @@ beforeEach(() => { KEYS.forEach(k => { SAVE[k] = process.env[k]; delete process.
 afterEach(() => { KEYS.forEach(k => { if (SAVE[k] === undefined) delete process.env[k]; else process.env[k] = SAVE[k]; }); });
 
 describe('per-network difficulty tuning', () => {
-  test('WOW is the reference (sizeScale 1.0) — casino dims unchanged', () => {
-    const t = applyNetworkTuning(casino(), 'WOW');
-    expect(t.dungeon.width).toBe(DIFFICULTY_PRESETS.casino.dungeon.width);
-    expect(t.dungeon.height).toBe(DIFFICULTY_PRESETS.casino.dungeon.height);
+  test('XMR is the reference (sizeScale 1.0) — casino dims unchanged; WOW (5-min) scales up', () => {
+    const xmr = applyNetworkTuning(casino(), 'XMR');
+    expect(xmr.dungeon.width).toBe(DIFFICULTY_PRESETS.casino.dungeon.width);
+    expect(xmr.dungeon.height).toBe(DIFFICULTY_PRESETS.casino.dungeon.height);
+    // WOW is a slow chain now (5 min) → bigger map than XMR.
+    expect(applyNetworkTuning(casino(), 'WOW').dungeon.width).toBeGreaterThan(xmr.dungeon.width);
   });
 
-  test('dungeon size scales with block time: GRIN < WOW < BTC', () => {
+  test('dungeon size scales with block time: GRIN < XMR < WOW', () => {
     const g = applyNetworkTuning(casino(), 'GRIN').dungeon.width;
+    const x = applyNetworkTuning(casino(), 'XMR').dungeon.width;
     const w = applyNetworkTuning(casino(), 'WOW').dungeon.width;
-    const b = applyNetworkTuning(casino(), 'BTC').dungeon.width;
-    expect(g).toBeLessThan(w);
-    expect(w).toBeLessThan(b);
+    expect(g).toBeLessThan(x);
+    expect(x).toBeLessThan(w);
   });
 
-  test('monster speed is tuned for casino but NOT for other presets', () => {
-    expect(applyNetworkTuning(casino(), 'BTC').monster.movesPerPlayerMove).toBe(NETWORK_TUNING.BTC.monsterSpeed);
-    // normal keeps its own monster speed (tuning was calibrated for casino only)
+  test('tuning scales SIZE only — the monster stays at its fair preset speed', () => {
+    // No cheating-fast monster: every preset keeps its own movesPerPlayerMove.
+    expect(applyNetworkTuning(casino(), 'BTC').monster.movesPerPlayerMove)
+      .toBe(DIFFICULTY_PRESETS.casino.monster.movesPerPlayerMove);
     expect(applyNetworkTuning(normal(), 'BTC').monster.movesPerPlayerMove)
       .toBe(DIFFICULTY_PRESETS.normal.monster.movesPerPlayerMove);
-    // ...but normal STILL gets the pacing size scale
+    // ...but both get the pacing size scale.
     expect(applyNetworkTuning(normal(), 'BTC').dungeon.width).toBeGreaterThan(DIFFICULTY_PRESETS.normal.dungeon.width);
+    expect(NETWORK_TUNING.BTC.monsterSpeed).toBeUndefined();
   });
 
   test('unknown network is left untouched', () => {