Two scopes

Strictness (1–5)

A fine dial on top of Early / Mid / Late. Same export — different mood: “I am still building” vs “I only want RTA-ready pieces.”

• 1Very forgiving — many runes get a role; weak subs and near-miss lines often pass. Good while farming and powering up.
• 2Relaxed — still friendly; more Keep than Sell on mid-tier runes.
• 3Balanced — default for your chosen stage. Second-chance check can mark a rune (Relaxed) in Target.
• 4Strict — needs more real stat lines and stronger rolls; fewer “almost good” Keep. No second-chance pass.
• 5Very strict — only strong, clean runes keep roles; borderline pieces move to Sell faster. For trimmed inventories.

Charts → table

Click a verdict slice, number tile, or role row — the Rune Table opens with the same filter. Enter or Space on a chart row works too (keyboard). Five chart tabs answer “where are my runes?” without leaving the Dashboard.

Copy summary (next to filters) pastes a text snapshot for Discord or notes — preset, depth score, filters, verdict counts, slot totals.

Three parts of the score

• Speed depth 35% Runes with total SPD sub ≥ 18 (roll + grind). Full bar ≈ 250 such runes.
• Power depth 35% 6★ +15 count. Full bar ≈ 600 runes.
• Elite quality 30% Average efficiency of your best 50 runes (uncapped internally). Strong boxes land near 110% average here.

How the 0–100 score is built

Three independent bars add up to 100 points max. Your counts are shown on the dashboard cards; underneath you see how many points each part contributed.

• ◇Speed (up to 35 pts) — each Rare+ rune whose total SPD sub (roll + grind) is at least 18 counts. Example: 125 such runes ≈ half the speed bar; 250+ fills it.
• ◇Power (up to 35 pts) — every 6★ at exactly +15. 600+ maxes this bar — endgame inventory depth.
• ◇Elite (up to 30 pts) — average efficiency of your best 50 runes (uncapped, so it can go above 100%). Starts contributing above 80% average; tops out around 110% average.

Suggested stage

Apply suggestion only copies the preset — per-rune verdicts change when you pick Early / Mid / Late yourself.

Columns that matter

Toolbar & summary chips

• ◇Chips above the grid — quick counts: total shown, Ancient, equipped, +15, Keep (for the current filter).
• ◇More Filters — Ancient only, hide Target column, reset filters.
• ◇Active filter chips — click the × on a chip to drop one filter without clearing everything.
• ◇Target column — hidden by default; turn off “hide Target” to see Keep quality lines and Grind/Gem hints in-grid.

Forge Score vs Eff%

Eff% is the classic SWOP-style efficiency (every sub line scored the same way vs Legend max rolls). Forge Score is our separate 0–100 rating: stat tiers (main list ≠ sub list), roll fullness, synergy multipliers, duplicate penalty — not Eff% and not verdict.

Eff% — formula and Legend baselines

Computed in the app from your SWEX rune (not from Sell/Keep). Main always counts as a full 1.0 “slot” in the score sum; innate and each sub add fractional points from roll size only (no stat-type preference except flat penalty).

score = 1.0 + innateTerm + Σ subTerm
subTerm = subValue ÷ (maxPerRoll × 5) — subValue is base roll only (grind not included in Eff%).
innateTerm = 0.5 × (innateValue ÷ (maxPerRoll × 5))
Flat innate or flat sub (HP / ATK / DEF): multiply that line’s term by 0.5.
Eff% = min(100, round(score ÷ 2.8 × 100, 1 decimal))

Legend 6★ max per single roll (denominator uses ×5 for five upgrades on subs/innate):

Depth charts can use uncapped Eff% (>100% possible in raw math); the table column caps at 100%.

Forge Score — formula

raw = (mainPts + subPts + innatePts) × mainSubSyn × subSubSyn × dupSub
Score = round( min(100, raw ÷ 295 × 100) )
Sub values use roll + grind (subRuneValue). Roll quality = value ÷ max (capped at 1.12). Calibration constant 295 maps typical runes (~55–75 raw) onto 0–100.

Main stat tiers (0–100)

Each main: mainPts = tier × rollQuality. Roll quality uses slot max mains on 2/4/6 (e.g. CDmg slot 4 max 80, SPD slot 2 max 42); slots 1/3/5 use Legend sub max ×5 for that stat type.

Sub stat tiers (0–100)

Each sub: subPts = tier × rollQuality × slotMul. On slots 2 / 4 / 6, flat HP / ATK / DEF subs: slotMul = 0.88 (else 1). Unknown stat: tier 45 (main fallback 50).

Innate (prefix): same sub tiers × roll quality × 0.38 (INNATE_WEIGHT_VS_SUB). Counts in synergy averages; not summed as a full sub line.

Synergy multipliers

Main ↔ sub — average over all subs (innate excluded from sub list): listed pair → multiplier; unlisted pair → 0.94 (anti-synergy). Pairs are unordered (ATK% + CDmg = CDmg + ATK%).

Sub ↔ sub — average over all sub pairs; unlisted → 1.00.

Duplicate subs: same stat twice (or more) → multiply by 0.88 for each extra line (0.88^(count−1) per stat name).

Hover Score for live breakdown (main / subs / innate pts and × multipliers). Chip color: ≥88 high · ≥72 mid · below low. Sort by Score for “stat quality”; sort by Eff% for classic efficiency.

Reading a row

• ◇Colored chips — grade, role, verdict, Eff%, Score. Set / main / subs are plain text.
• ◇Equipped — search can find the monster name in the row; useful for “what is this rune on?”
• ◇Ancient — A-with-dot before grade; Ancient only in More Filters.
• ◇HP% 13 [3] — total on rune; bracket = grind bonus only. Gem = circular-arrows icon.
• ◇Header filters — verdict / role / grade / set / slot / main under column titles; click column headers to sort (default Eff% down).

Huge exports: first 500 rows for speed — Load all … for the full filtered list. Search highlights matches in the row.

Share & export

Export CSV — spreadsheet of the filtered list. Share (toolbar) — read-only link for monsters/runes (see App Settings for modes).

Filters update the URL to #runetable?… — bookmark or send the link to reopen the same view in this browser.

The path each rune takes

Two answers per rune

Build roles

Each role is a PvP/PvE build template — allowed mains on 2 / 4 / 6 and which substats count.

Seventh outcome: valuable stats, no exact build — keep for future monsters or trade fodder carefully.

What “fits a role” means

A rune must pass several checks for that role at your chosen stage. Think of it as a checklist — not one magic number.

• ◇Main stat on 2 / 4 / 6 — e.g. slot 4 must be ATK%, HP%, or another allowed main for that build.
• ◇No forbidden substats — stats marked “exclude” for that role (wrong stats block the match).
• ◇Must-have stat — e.g. Bomber often needs ATK% somewhere (sub or main).
• ◇Enough useful lines — minimum count of good substats for that slot group (1/3/5 vs 2/4/6).
• ◇High-roll anchor — on Mid/Late, at least one included sub should be a real “high” line for Hero or Legend (depends on preset).
• ◇Overall stat strength — substats should be reasonably close to strong lines for that stage, not just present.

God Roll, Duo Roll & Universal

Fit Score

0–100 — how closely substats match the chosen role. It does not change Verdict; use it to compare two Keep runes for the same build. Shown in Target on Keep as Fit 67 etc.

Verdicts by rune level

1. Upgrade
   Below +9 — too early to judge; level it first.
2. Finish
   +9 to +11 — has a role (or God/Duo) or can reach one with a grind; worth finishing.
3. Keep
   +12 and up — fits a build; keep in inventory.
4. Gem
   +12+ — role OK but has a bad flat sub (HP / ATK / DEF) worth replacing with a gem.
5. Grind
   +12+ — no role yet, but one grind on SPD / HP% / DEF% / ATK% could unlock a build.
6. Reapp
   Legend +12+ — no role; matches reapp rules (set, main, efficiency cap).
7. Sell
   +12+ — no fitting build and no grind/reapp path; fodder or sell.

Target column

Early / Mid / Late

Stage changes every rune’s Role and Verdict — not the account progression score at the top.

Example (illustrative)

Legend Swift slot 2, main SPD, subs: HP% 18, DEF% 12, flat ATK, CR% 6 — same rune, different presets:

• EarlyKeepBruiser — anchor rules softer
• MidGemRole OK — replace flat ATK
• LateSellLines too weak for late-game bar

Strictness & Rune Rules

Strictness (1–5) on the Dashboard makes checks tougher: more useful substats required, higher rolls to match a role, stricter God/Duo bars. 1–2 — more runes pass; 4–5 — only strong pieces; relaxed retry is off at 4–5.

Expert edits live in Rune Rules — use Save & Recalculate after changing thresholds there.

Simple vs full Rules

How the pieces connect

Result → Role, Verdict, Target in the Rune Table. Plain overview → How scoring works in this Guide.

Engine tab

1 — Constants (8 stats)

One row per stat (SPD, HP%, ATK%, DEF%, CRate, CDmg, ACC, RES). Totals use roll + grind on the rune, same as in the table.

• ◇Base — Mid-game Hero anchor for that stat (e.g. SPD 16).
• ◇God +% — God Roll line = Base × (1 + %). One huge sub can save a rune as Universal.
• ◇Early −% / Late +% — scales the High Roll grid by account stage.
• ◇Legend −% — Legend runes use a slightly lower HR line than Hero at the same stage.
• ◇Duo −% — Duo line = High Roll × (1 − %); pairs like SPD+HP% must both clear their lines.

Show threshold previews — read-only God / High Roll / Duo tables (stage × grade). Change numbers above → previews update live; Save & Recalculate applies to all runes.

2 — Evaluation Policy

Global strictness on top of your Dashboard stage. Expert fields:

Roles tab

Six default builds (Fast CC, Classic DPS, Bomber, Tank, Bruiser, Slow DPS). Pick one in the left list; the big sheet is that role’s formula.

• ◇Enable formula — off = role ignored in matching.
• ◇Accepted mains — slot 2 / 4 / 6: comma list (SPD, HP%, …) or None.
• ◇Sub-stats grid — per stat × Early/Mid/Late: Include (counts toward min lines), blank, or Exclude (blocks the role).
• ◇Must have — one required stat on the rune for that stage (e.g. Bomber → ATK%).
• ◇Slot requirements — e.g. slot 2 must be SPD for Fast CC on Mid.
• ◇Min stats — how many Include lines needed on 1/3/5 vs slot 2 / 4 / 6 groups (must-have does not add to this count).
• ◇Anchor (High Roll) — checkboxes: at least one sub must reach Hero or Legend HR line for that stage.

Edits apply on change — table refreshes without the footer button. You can + Add role for custom builds (keep at least one role). Reset Constants only restores the 8-stat grid, not these formulas.

Verdict rules tab

Fine-tunes Gem, Grind, Reapp on top of the main verdict tree (Upgrade / Finish / Keep / Sell).

Save & reset

• ◇Roles — auto-save on every change.
• ◇Constants, Policy, Verdict rules — use Save & Recalculate (footer).
• ◇Reset to Defaults — full factory reset of rules (careful).

Keyboard

• /Runes → Table — focus search from any tab (not while typing elsewhere). Open Table to see the field.
• ↑ ↓Runes → Table — scroll the grid when that tab is open and focus is not in a filter box.
• PgUp PgDnRunes → Table — page through the grid faster (same conditions).
• Enter SpaceDashboard charts — activate a clickable verdict / role / set / slot row (same as click) → opens filtered table.
• Enter SpaceMonsters roster — open the focused monster row.

Mouse & links

• ◇Drag & drop a SWEX .json onto the page — loads into active Data 1 (same as the upload button).
• ◇Dashboard → Table — click chart bars, verdict tiles, or role lines to filter the table in one step.
• ◇Monster rune slot — click a slot in unit detail → Rune Table focused on that rune.
• ◇#runetable?… in the address bar — share your current filters; Copy summary on Dashboard for a text dump.
• ◇Filter chips on the table — remove one active filter with × instead of resetting all.

App

• ◇App Settings — language, share link modes, four database slots.
• ◇Theme toggle (header) — light / dark.
• ◇After an update, verdicts look off → hard refresh, then once Save & Recalculate in Rune Rules.