Ludo Atlas · Genre Handbooks · Fashion Sim¶

Genre Handbooks · Volume 4. Positioning: the genre that makes "styling" its core verb. The theme sets the question, the wardrobe supplies the material, and the scoring grades the paper — wound together into a loop of "read the question, pick pieces, swap one piece and retry"; long-term drive comes from collecting, story and display. This page covers outfit scoring, wardrobe content and art costs, narrative integration, social display and monetization compliance. Companions: Game Design Handbook (core loops and scoring balance) · Art & Audio Handbook (clothing assets and outsourcing management) · Live-Ops & Growth (seasons, monetization and long-term cadence) · Legal, Patents & Competition (probability disclosure and content compliance). This page carries no external links; benchmarks are widely known works only, and the figures are common magnitudes — calibrate them against measurements in your own project.

1. Positioning and Core Loop¶

Fashion sim (in the Chinese market it is commonly called the "dress-up game") is one of the pillar genres of the female-oriented market: it started from the dress-up play of browser and early mobile titles, and the Nikki series (Nikki UP2U: World Traveller, Love Nikki and Shining Nikki, from 2013 onward) turned "collect clothes, style to clear stages" into a mobile top seller; on the handheld side there is the clothing-store and stock-selection play of the Style Savvy line, and the dress-up systems inside life sims are cousins from the same root. If you can dress yourself, you can play — there is almost no barrier to entry.

In one line: fashion sim is the genre that makes styling the first-order pleasure. Players use clothing as vocabulary and outfits as sentences, answering one question after another — "what do I wear to this occasion?"; the scoring system grades the paper, and the wardrobe supplies the vocabulary. Success rests on two things: scoring must be explainable (§3.1), and individual pieces must keep arriving (§3.2, §3.5); if either link drags, players shift from "I want to build one more outfit" to "I'll go look up a guide".

The core loop, written as a verb chain:

read the brief (occasion and style) → browse the wardrobe → select and style → submit for scoring → change one piece per the feedback and retry → claim rewards and new pieces

All three scales of the loop need payoff points: one styling session is measured in minutes, scored the moment you submit; one chapter is measured in hours, settled by a run of levels plus story progression; one season and one compendium are measured in weeks and months, closed out by collecting, contests and the season settlement.

Drawing boundaries against adjacent genres:

Adjacent genre	Boundary
Life sim	There, dressing up is part of self-expression, with no scoring or levels; in fashion sim every outfit must pass a grading system
Dating sim	There, the main loop is advancing a relationship and clothing is only an item that moves affinity; fashion sim makes the clothing itself the main course
Management sim	Running a clothing store is about stock, pricing and foot traffic; fashion sim is about "does this outfit answer the question" — the business side is at most a side dish
Collect-and-raise games	There, collecting is compendium completion itself; in fashion sim every piece must be wearable, scorable and displayable — collecting is a means, not the end

A self-check question: swap the clothing for furniture, pets or weapons — does the gameplay still hold? If the answer is "it still holds", what you are making is probably a collection-management game; if the answer is "it doesn't", what you are making is fashion sim.

2. Player Experience Goals and Benchmark Titles¶

Experience goals (in priority order):

Expression delivered: satisfy "looks the way I imagined" first, "scores high" second; players will take a second look at their own finished outfit — that moment is the core appeal of the whole genre.
Judgment you can explain: a score must trace back to specific reasons (style mismatch, missing pieces, outfit completeness); change one piece along the feedback and the score moves — only then is scoring not a lottery.
Wardrobe growth: the collecting itch comes from "one more piece to go" and from "the new arrival unlocks a new solution"; the supply cadence on the free path decides long-term retention.
A stage for display: outfits must be seen; photo albums, contests and peer review give expression an epilogue — without display, collecting motivation halves.
The pull of story: occasions and motives come from the narrative (an interview, a banquet, a stage) — that is what makes scoring criteria meaningful and keeps levels from being pure score farming.

Benchmark titles (play them yourself before breaking them down; watch especially how scoring feedback explains itself and what players do when stuck):

Title	What to learn
Shining Nikki	The engineering and staging spec of 3D dress-up; outfits as narrative units; the long-term structure of contests and seasons
Love Nikki	The scoring system and chapter progression of 2D dress-up levels; light text plus high-density clothing updates
Style Savvy	Turning "read the customer's needs" into core gameplay: the small loop of selecting, matching and presenting
The Sims 4	Proof that dress-up works as self-expression: it holds without level scoring, and display and sharing can drive on their own
Animal Crossing: New Horizons	A lightweight model for clothing design and sharing: custom patterns and the gifting loop

3. Design Essentials¶

3.1 Outfit Scoring Mechanics¶

The scoring system has to do two jobs at once: judge right from wrong, and teach the method. Its structure starts from three accounts:

total score = base score (rarity) + fit score (tags times weights) + combo bonus (sets and series)

The base score lays a floor by rarity and returns the value of "owning";
The fit score is the main account: each piece carries 2–4 style tags and values (6–10 master styles across the whole catalog), a level supplies a set of weights, and the two are multiplied and added;
The combo bonus rewards "dressing completely": full sets, matching color families and matching series score lightly, encouraging a coordinated outfit rather than stacking one piece.

Design discipline: the fit score must take 60–80% of the total, so that "the right low-rarity piece" reliably beats "an expensive random pile"; fail that, and scoring degrades into a power comparison, with free players frustrated the whole way (§3.5).

The four-piece kit for a level's question: the occasion (write the scene in narrative language, as in "a class reunion — comfortable yet presentable"), the weights (multipliers for primary and secondary styles, as in simple ×3, fresh ×1.5, gorgeous ×0.5), the thresholds (clear / excellent / perfect, three tiers), and the feedback (an explanation beyond the score, as in "missing a clean top").

Three acceptance criteria:

Feedback speaks plainly: "An interview calls for dignity; this looks too casual" beats "elegance value insufficient"; number-minded players can optimize against the thresholds, and intuition players can guess the right direction from reading the question.
No one-size-fits-all: scan the wardrobe with a script across all levels; the top-scoring outfit in any two levels should overlap by no more than two or three pieces; when one outfit beats everything, adjust the weights or rotate the themes.
Thresholds calibrated against the free-to-obtain wardrobe: work the clear line backwards from the best free-path solution, leaving one or two tenths of headroom; paying buys faster and more complete, never the "only solution" (§3.5).

3.2 Wardrobe Content and Art¶

The wardrobe starts with slot taxonomy before output volume: head (hairstyle, makeup, headwear, earrings), torso (tops, bottoms, dresses, outerwear), limbs (socks, shoes, handwear, handhelds), atmosphere (backgrounds, effects, poses); start with 8–10 core slots, then add by production capacity — finer slots mean more combinations and higher cost.

The books on the 2D and 3D routes:

Route	Per-piece content	Cost magnitude	The price
2D variations	Each piece drawn as a layer over a uniform pose, sharing a base body	Half a day to two days per piece	Low freedom of motion and camera; change the pose and the whole set is redrawn
3D models	Modeling, pattern-making, materials, rig fitting	Several days to two weeks per piece	The highest cost; clipping across layered cloth is long-term maintenance debt
Hybrid	The main body follows one route; staging and effects get separate layered treatment	Between the two	Freeze the spec as early as possible, or you rework on both ends

Production discipline (for pipelines and outsourcing management see the Art & Audio Handbook):

New clothes must mix with the old: run a mix-and-match spot check as each new piece enters the pipeline (sampling the combination matrix against catalog pieces); a one-off that only works within its own set was made for nothing.
Deliver in series: each version or chapter produces one themed series as a batch, with a unified style and a matching copy set, so outsourced acceptance has a reference pack and samples.
Don't miss the by-products: icon, compendium card, thumbnail, acquisition copy and try-on data — every piece comes with this whole set; miss one and you rework.
Establish the clothing chapter of the Art Bible first: era, silhouette, palette, materials and subject boundaries written clearly, or the outsourced pieces will assemble like a sampler platter.

3.3 Narrative Integration¶

The fashion subject comes with scripts built in: a designer's rise, the road to top model, reviving a family brand, climbing a competition ladder — all suit a chapter structure of "one occasion per scene"; the flow is fixed as:

story intro → 1 to 3 styling levels → story reaction (relationship progress and rewards) → new pieces and new hooks

Level motivation comes from the story: "you have an interview tomorrow" lets players understand the scoring criteria better than "level five"; the occasion description directly shapes how players read the question.
Scoring criteria and story demands must agree: if the story says "plain and understated", scoring cannot hand high marks to the gorgeous — this is the most frequent spot where narrative and gameplay undercut each other (§7).
Clothing can be a relationship prop: events where you save a friend in a pinch or pick clothes for a partner wire the wardrobe into social and romance lines (build the romance line to the standards of the matching genre handbook).
Budget text and staging separately: dialogue, branches, portrait variants and voice are all additive costs; estimate them at visual-novel magnitudes before writing (estimation methods in the Production Handbook).
Long-term serialization: the story serializes by version, with chapter themes aligned to outfit themes; one batch of clothes tells one story, and the story teases the next batch.

Display comes in four layers: for yourself (photo album, photo mode, one-tap output with frames and watermarks); peer review (themed contests, pairwise voting — anonymous, vote-limited, tier-matched); competition (leaderboards and seasons, with tiered rewards but no shaming of the losers); playing together (borrowing clothes from friends, co-op events and mini-games).

Keep rewards light: a guaranteed participation prize, ranks split into tiers only — low-scoring outfits are treated kindly, so the urge to express does not turn into social pressure.
Fairness comes from mechanics: anonymous entries, anti-vote-farming (vote limits plus server-side tallying), tier-matched pairing — newcomers never share a stage with veterans who own full wardrobes.
Scope UGC first: custom dyeing, patterns and stickers lead; free drawing comes later; design and naming bring in moderation, copyright and minor-protection issues.
Sharing is growth: share images carry watermarks and event tags — the cheapest organic traffic in this genre; build composition templates into photo mode; don't expect players to crop their own.

3.5 Monetization Structure and Compliance Boundaries¶

Monetization has four layers (for operating cadence and spend see Live-Ops & Growth); only the structure is given here — calibrate the numbers against your own retention and choke points:

Direct purchase: individual pieces, full sets and bundles at clearly listed prices — the genre's word-of-mouth bedrock;
Monthly cards and battle passes: daily resources plus wardrobe perks, carrying steady revenue and daily return visits;
Limited draws: seasonal themed sets obtained by probability draw, structurally identical to the card pools of gacha RPGs: probability disclosure, pity and draw records follow the same standards (Legal, Patents & Competition; Live-Ops & Growth), and this page does not repeat them;
Stamina and tickets: level consumption and recovery cadence regulate how fast content is consumed — set the parameters before setting the outputs.

One red line: main-line clear thresholds must be calibrated against the free-to-obtain wardrobe (§3.1); paying buys "more choices and a faster pace", never the "only solution"; if a paid-exclusive piece is the only solution, the scoring system becomes a toll booth (§7).

Compliance boundaries — walk through them at greenlight (each expanded in Legal, Patents & Competition and the Multi-platform Launch Playbook): probability-draw disclosure, pity and records; minor spending limits, real-name verification and playtime controls; marketing copy may not use absolute terms, and claims about looks or originality need substantiation; real brand marks, celebrity likenesses and fashion photography can all be protected subject matter — art reference is not usable material; UGC content must pass review, with minor-protection measures built alongside.

Ledger discipline: multiply per-piece pricing by content volume first; confirm it "sells back its cost" before opening a production line; the standards for this ledger and for operating metrics follow Live-Ops & Growth.

4. Technical Essentials¶

The engineering difficulty concentrates in three places: dress-up rendering, testable data and scoring, and wardrobe search plus saves; for general architecture see the Programming Handbook.

Dress-up rendering

2D route: anchor points and layer order need a written spec (each slot's texture origin goes into the asset spec; display order rules like "outerwear over the hem" go into a table); misalignment between combined pieces is 2D dress-up's most common accident; recolor via palettes and shaders, not redraws. Previews must open in a snap: preprocess thumbnails and atlases — never render the full outfit live at runtime.
3D route: skinning and cloth simulation, multi-body fitting, and a cloth toggle on mobile; build clipping detection into a tool (bounding-box spot checks plus a visual-inspection matrix of critical combinations) — fixing one piece and breaking three is the norm, and the detection pipeline is worth more than after-the-fact patches.

Data-driven design and testable scoring

One row per piece: slot, tag vector, rarity, source, icon, compendium copy, live status; one row per level: weights, thresholds, copy, rewards; code only reads tables, and formulas live apart from tables. A sold asset cannot be deleted; retiring it only affects its acquisition entry point.
The batch scoring script is the core tool: all levels × all wardrobe search, producing each level's top solution and distribution, rerun on every new release to keep new pieces from breaking old levels (§3.1); unit tests cover empty slots, duplicate pieces, extreme weights and borderline thresholds.

Wardrobe UX and saves

Search experience at the thousand-piece scale is retention: filter by slot and tag, sort by color and recency, favorites, outfit presets, and a one-tap "generate a draft from the current level's weights" button.
Save fields: ownership, favorites, presets, level progress and season data all serialized in full, with a version number and migration functions; for live-service titles, add server-side records of purchases and draws, server-side contest tallying, and share-image rendering specs.

5. Content Volume and Workload Reference¶

The following are common magnitudes for comparable projects, for scoping estimates; they are not a commitment.

Project form	Clothing volume	Timeline magnitude	Notes
Mechanics prototype	30–60 placeholder pieces	2–4 weeks	Color blocks instead of art; validates scoring and the loop only
Small complete title	200–400 pieces, with levels matched to chapters	6–12 months	Art takes the lion's share; the story is produced in parallel
Long-running mobile product	600+ pieces at launch, 60–120 per version	1 year and up, with ongoing investment	Output capacity is locked to the operating cadence
High-spec 3D product	Thousands of pieces in total	Multiple years for a team	Clothing and staging are the cost body

Per-piece costs and magnitude conclusions (rough measures; calibrate against your own): a 2D variation takes half a day to two days (drawing, layer separation, try-on, icon); a 3D piece takes several days to two weeks (modeling, materials, rig fitting, mix-and-match clipping tests); one styling level (weights, copy, thresholds, scan verification) takes half a day to two days; one story chapter (5–10 levels plus dialogue and staging) takes weeks. Clothing assets are usually the largest slice of the whole project's art budget (a common measure is over half; for pipelines see the Art & Audio Handbook) — work out capacity first (pieces per person per month), then set the content plan and update promises; capacity failing to match consumption is this genre's main accident, because players can burn through a month's output in a week; the cutting priority, first to last: reuse clothing (recolors of the same design, edits of the same model) > reduce staging specs > reduce story branches > shrink the compendium — clothing is the main course, so the knife goes there last.

6. How to Start the First Prototype¶

Goal: spend 2–4 weeks answering one question — can the loop of "pick clothes, score points" make someone voluntarily build another outfit; use color-block placeholders throughout, and draw no art.

Week 1: the clothing skeleton. 8–10 slots, 30–60 placeholder pieces (color block plus tags plus values), and one acquisition entry point (level rewards or a shop) so new clothes can reach the player.
Week 2: scoring and levels. One scoring formula (base score plus fit score), 5–8 levels (each with weights and three-tier thresholds), and explainable feedback copy; write a scan script alongside to check that the top solution is not the same outfit across levels.
Week 3: close the loop and add a story shell. Acquire → style → score → reward → acquire again; add 3–5 screens of dialogue to the first chapter to make the "occasion" clear, and watch whether players read the question.
Week 4: playtest and tune. Find 5 people who have never played; record only, prompt never: do they read the question, can they explain the score, do they swap one piece and retry after failing; note the spots where "stuck means quit".

Success criteria (all observable):

Testers can explain the score in their own words ("this scene calls for elegance, and I dressed too casually") rather than "no idea why".
With art and audio stripped out, testers will replay levels for the next piece of clothing, 30+ minutes in one sitting.
Any two levels have different top configurations, and the scan script can verify this within 10 minutes.
At least 3 of the 5 testers adjust their outfit on their own and retry after failing, and can say why.

Not during the prototype: 3D, cloth, social systems, gacha, real currency, localization; verify "is styling fun" first, then do the launch engineering.

7. Common Pitfalls¶

Scoring is unexplainable: players cannot tell where the score gap comes from, and gameplay degrades into guide-reading. How to avoid: feedback specific to tags and pieces, in the language of the occasion (§3.1).
A single optimal solution: one universal outfit beats every level, and collecting and styling lose their meaning at once. How to avoid: weight rotation plus solution-space scanning (§3.1).
Rarity crushes styling: high-rarity pieces dominate on raw numbers, scoring becomes a power comparison, and free players are frustrated. How to avoid: let the fit score take the lion's share; calibrate thresholds against the free wardrobe (§3.1, §3.5).
Wardrobe search out of control: past a thousand pieces, clothes can no longer be found. How to avoid: filters, favorites, one-tap drafts (§4).
Clothing output collapse: updates cannot keep up with consumption, and promised sets go missing. How to avoid: compute capacity before making promises; amortize costs with reuse and recolors (§5).
Clothes with no stage: no levels, story or display to use them in — the clothes have nowhere to go. How to avoid: give each batch of clothing a batch of uses; every release is an event (§3.3, §3.4).
Clipping and fitting debt: 3D clothing clips as soon as layers pile up. How to avoid: put detection tooling and combination-matrix spot checks into the pipeline (§4).
A paywall in the way: the main line's only solution is purchasable. How to avoid: "the free path can clear the game" is the red line; paying sells choices and pace (§3.5).
Draw standards and social pressure: undisclosed pity rates and public punishment for low scores both push well-disposed players away. How to avoid: the same disclosure rules as card pools; tiered rewards, anonymous peer review, and a prize for taking part (§3.4).