guild-1bit-aesthetic

Top-Down Pixel Aesthetic (Guild / OutsideRPG)

Visual rendering direction for the outsiderpg-platform-vision world. The product concept (“real world = game world, every indie shop is a guild hall”) needs a rendering. The rendering is top-down retro pixel art — nostalgic, welcoming, legible at small sizes.

The ship target is concrete: a top-down view of the actual Dungeon Books interior. Longer horizon: a top-down view of Jersey City where real indie shops (starting with the NJ pilot three) appear as locations you can enter.

Filename kept as guild-1bit-aesthetic.md for stable links. Strict 1-bit was the starting point; the direction has since broadened to include limited-palette pixel where color carries meaning.

The north star: “a wizard shopping on eBay”

Captured 2026-04-14 from Panat. The aesthetic isn’t a discrete “game view” sitting next to the dashboard. It’s an ambient fantasy-pixel layer over the whole app. Web development in a fantasy world. Magic and technology intersecting at every surface — the shop, the product grid, the cart, the cursor, the scroll behavior.

Examples of the intent:

• Pixel sprites drifting through page headers as you scroll
• Cursor trails that emit sparks
• Loading states as arcane sigils spinning
• Hover feedback with particle bursts
• Transitions between pages that feel like moving between rooms
• Empty states that show a sleeping dragon on a pile of inventory

The shop-interior top-down map (Phase 1 below) is one strong instance of this treatment. But the treatment is app-wide, not scoped to a single surface. Every interactive surface in Guild should feel like the same fantasy world rendered through modern UX.

Reference points for “delight as identity, not decoration”

• Duolingo’s owl: character pervades the product. Every screen feels like the owl is there, even when it’s not on-screen.
• Animal Crossing UI: menus, cursors, transitions, SFX all carry the world’s character. Not decoration — continuity.
• Stardew Valley shop menus: literal inventory UI rendered in-world register.

The risk: delight-as-friction

Every animated sprite is attention the user spends not-shopping. Failure mode: performative flourish that makes the site feel slow, distracting, or twee. Working constraint: every animation must serve feedback, continuity, or identity. No decoration-only sprites. If it doesn’t do one of those three jobs, cut it.

Performance ceiling

Ambient sprites + cursor effects + pixel shop scene + shaders running at once can melt low-end Android. Discipline through:

• Dev-only perf overlays (FPS, draw calls, GPU memory) on from day one
• Budget per surface (e.g., background ambient layer = X sprites max)
• Fall back to reduced-motion mode for users who opted in (respect prefers-reduced-motion)

Tech implication

This framing strengthens the game-engine-choice lean toward PixiJS + @pixi/react: the library is designed to mix Pixi-rendered elements with React DOM at any level of the component tree. Ambient pixel layers alongside product grids is the case it’s built for. Phaser’s scene-owning lifecycle doesn’t fit “pixel sprites drifting through a product listing header” as naturally.

Product anchor

This aesthetic note is subordinate to outsiderpg-platform-vision. The visual choices here are in service of rendering that product. Three phases:

Phase 1 — Shop interior (near-term)

Top-down view of Dungeon Books’ interior. Counter, shelves, event table, kiosk location, bathroom, Carrie at the counter, whoever’s at the kiosk right now, members currently present. Functions as: an at-a-glance “who’s in the shop” surface for members, a “what the shop feels like” marketing surface for new members, a canvas for NFC check-ins to visibly drop a sprite into the scene.

Concrete deliverable: one scene, composed from off-the-shelf assets, rendered on the kiosk + portal. This is tractable — not a game, a view.

Phase 2 — Jersey City streetscape (medium-term)

Top-down view of a JC area containing Dungeon Books and the other participating shops. Each real shop = one building on the map. Tapping/clicking a shop = viewing its interior (Phase 1 scene for that shop). Foot-traffic visualization uses cross-merchant check-in data from cross-merchant-identity.

This is a real design project, not just asset curation. Open questions:

• Map scale: abstracted (Pokemon town, ~6–12 buildings) vs. geographic (actual JC blocks)? Probably abstracted. A recognizable JC-flavored layout that isn’t a literal map.
• Is the player an avatar that moves, or is this a viewing surface only? Probably viewing surface initially — no character movement, no fog of war, no game loop. Just a living diorama.
• How do new shops enter the map as they join the network?

Phase 3 — OutsideRPG world (long-term)

Multi-city maps, cross-map travel, expansion as more shops join across geographies. Scope of outsiderpg-platform-vision. Not this note’s problem yet.

Visual lineage (reframed)

The reference points are not Nethack / Brogue / Caves of Qud. Strict 1-bit roguelike rendering of a real modern city would read as hostile or alienating. The correct references are:

• Pokemon Gold/Silver/Crystal — towns where shops are buildings you enter. Shop interior renders as a separate scene with shopkeeper and counter. Welcoming, legible, iconic.
• Earthbound / Mother 3 — modern-day towns in charming 16-bit pixel. Department stores, drug stores, diners rendered with personality.
• Stardew Valley — pixel town with shops identified by owner (Pierre’s, Clint’s, Gus at the saloon). Exactly the “real people behind the counter” energy we want.
• Ultima IV–VI top-down — the Dicegoblin blog’s reference. Fantasy-flavored but the composition grammar translates.

1-bit remains in scope for specific surfaces — print (receipts, thermal labels, zines, signage) and anywhere a stark graphic treatment reads better than color. But 1-bit is not the spine of the top-down world.

v1 spine commitment — Mini Medieval / Fruitpunch24

Locked 2026-04-15 after visual evaluation and Carrie’s input.

v1 asset spine: Mini Medieval + Mini Medieval UI by VEXED. 8×8 grid, Polyphrog’s Fruitpunch24 palette, CC-BY 4.0, $8.25 total.

Why this over the other candidates

• Demonic Dungeon was flagged by Carrie as too dark / hell-like. Even though “Dungeon Books” is the shop’s name, the in-shop vibe is cozy indie bookstore, not heavy-metal dungeon crawl. Palette vibe matters more than literal name match.
• Urizen is visually flat. CC0 + huge content, but 1-bit / one-color-per-tile is utilitarian, not expressive. Usable as filler library; wrong for hero art.
• mrmotarius packs carry too much personal signature. Using MoreGoblins or Lambda Rogue means the Guild brand reads as “mrmotarius’s visual voice.” VEXED is deliberately referential (SNES JRPG / classic CRPG lineage); Mini Medieval reads as “tradition” not “one artist’s hand.”
• 8×8 hits the nostalgia layer harder than 16×16. SNES-era 16×16 is polished. NES / Game Boy / RuneScape Classic 8×8 is the MUD-adjacent pre-SNES register, closer to the mud-lineage narrative. Matches the warm-nostalgic feeling we want.
• Growing universe in one palette. Mini Medieval main (v2.3.1, Feb 2026) + MM UI (v1.1, Nov 2025) + planned expansions (dungeons, caves, VFX, additional enemy races). Long runway without leaving Fruitpunch24 coherence.
• Polyphrog palette provenance. Fruitpunch24 is a widely-used indie-pixel palette with community tooling. Recognizable lineage, not idiosyncratic.

The “8×8 leaves room for imagination” principle

The 8×8 constraint is not a compromise — it’s a design move. At that resolution there aren’t enough pixels to over-define a character’s face, expression, or identity. The sprite is a silhouette and a few gesture pixels. The member’s imagination fills in everything else. The character is the member, not the artist’s rendering of the member.

Consequence: don’t fight the grid. Don’t try to add detail 8×8 can’t hold. Let silhouette + color + motion do the work. This is why the sprite works for every member, not just members who happen to match a specific illustrated hero.

Upscale strategy

• Character screen / hero-size display: 8× (64×64) or 12× (96×96 — chunky, very RuneScape-portrait). Pick after testing in-browser.
• Nav avatar / small presence indicator: 4× (32×32) or 6× (48×48).
• Tile rendering for future scene surfaces: 4× (32×32 per tile).

All at nearest-neighbor interpolation. No exceptions.

What we retained from earlier drafts

• Urizen stays in the library as CC0 fallback — use when Mini Medieval doesn’t have a needed asset and we can’t wait for a Polyphrog-palette alternative.
• Chroma Noir UI parked as alternative 1-bit UI register if Mini Medieval UI feels too busy on specific surfaces.
• Demonic Dungeon series skipped — palette too dark for the shop’s brand despite the literal name match.
• mrmotarius, IKR, Lobit, Radiant Dungeon skipped for v1 but kept documented for future reference in case a specific gap appears.

Attribution

Single CREDITS line on portal footer satisfies CC-BY 4.0 for Mini Medieval + MM UI both. If we later add other VEXED packs, same line covers them. No per-surface attribution required.

Tiers of treatment by surface

Surface	Treatment	Why
Shop interior / city map	Limited-palette colored pixel (16-color Mystic-16 or Cartoon-Candy, 16×16 grid)	Welcoming, warm, Pokemon/Earthbound lineage. Real places rendered with affection, not brutalism.
Portal UI (buttons, frames, icons)	1-bit UI kit (Chroma Noir UI, 8×8) OR matching colored UI (Demonic Dungeon UI, 16×16)	Pick one. Locking UI is more visible than locking scenes.
Member avatars / class icons	Limited-palette matching the map palette	Class identity legible, member sprites visible on the shop-interior scene
Print collateral (receipts, thermal labels, zines, signage)	Strict 1-bit (Urizen, Kenney mono, MRMOTEXT)	Print fidelity, cheap reproduction
NFC cards	Hybrid — 1-bit card back (survives thermal printing if needed), colored front art (premium feel)	Physical artifact, manufactured; different constraints
Activity feed flavor	Matches map palette	Consistent with the scene sprites are drawn from

This is the “option 2” from the previous revision — 1-bit spine for print, colored top-down for screen world — with the split made explicit by surface.

Reconciliation with existing direction

guild-visual-direction frames the aesthetic as cassette futurism + 80s sword & sorcery (module art, cross-hatching, 1–2 color print, CRT/phosphor). That direction is still valid for merchandise and marketing collateral — posters, NFC cards, manual-style documentation, the storefront aesthetic around member tiers. It is not the right treatment for the OutsideRPG world map. Clean ownership:

• Product world (kiosk, portal, map, interiors): top-down pixel, colored
• Print / manual / merchandise: cassette futurism + module art
• UI chrome on the product: pick one pixel UI kit, commit

---

Asset packs — quick reference table

All evaluated packs. Details captured 2026-04-14/15. Verify license on pack page before use — terms can change.

Pack	Creator	Grid	Palette	License	Price	Link
Urizen 1-bit tileset	vurmux	12×12	1-bit	CC0	Name-your-own	https://vurmux.itch.io/urizen-onebit-tileset
Kenney 1-bit pack	Kenney	16×16 (verify)	1-bit + color variant	CC0	Name-your-own	https://kenney-assets.itch.io/1-bit-pack
Demonic Dungeon	VEXED (v3x3d)	16×16	Mystic-16	CC-BY 4.0	$5(50$10)	https://v3x3d.itch.io/demonic-dungeon
Demonic Dungeon — UI	VEXED	16×16	Mystic-16	CC-BY 4.0	$3.25(50$6.50)	https://v3x3d.itch.io/demonic-dungeon-user-interface
Demonic Dungeon — Building Interior	VEXED	16×16	Mystic-16	CC-BY 4.0	$3.25(50$6.50)	https://v3x3d.itch.io/demonic-dungeon-building-interior
Radiant Dungeon	VEXED	16×16	Cartoon-Candy	CC-BY 4.0	$3.25(50$5)	https://v3x3d.itch.io/radiant-dungeon
Mini Medieval	VEXED	8×8	Fruitpunch24	CC-BY 4.0	$5(50$10)	https://v3x3d.itch.io/mini-medieval
Mini Medieval — UI	VEXED	8×8	Fruitpunch24	CC-BY 4.0	$3.25(50$6.50)	https://v3x3d.itch.io/mini-medieval-user-interface
Chroma Noir — UI	VEXED	8×8	1-bit	CC-BY 4.0	$3.25(50$6.50)	https://v3x3d.itch.io/chroma-noir-user-interface
MoreGoblins	mrmotarius	12×12	Goblin22 (22 colors)	mrmotarius¹	$5+	https://mrmotarius.itch.io/moregoblins
Lambda Rogue (WIP)	mrmotarius	12×12	1-bit	mrmotarius¹	Name-your-own	https://mrmotarius.itch.io/lambda-rogue
MRMOTEXT	mrmotarius	8×8	1-bit	mrmotarius¹	$5+	https://mrmotarius.itch.io/mrmotext
RAS: Overworld	IKnowKingRabbit	16×16	12-color	IKR²	$8.75+	https://iknowkingrabbit.itch.io/ras-overworld
MAS: Roguelike	IKnowKingRabbit	8×8	12-color	IKR²	$3.75+	https://iknowkingrabbit.itch.io/mas-roguelike
Lobit old-school tileset core	helicity	unspecified	unspecified	Unclarified³	Free	https://helicity.itch.io/lobit-old-school-tileset-core
v3x3d (VEXED) catalog	VEXED	—	—	—	—	https://v3x3d.itch.io/
IKnowKingRabbit catalog	IKR	—	—	—	—	https://iknowkingrabbit.itch.io/
mrmotarius catalog	mrmotarius	—	—	—	—	https://mrmotarius.itch.io/
Dicegoblin blog — source article	Lars	—	—	—	Free	https://dicegoblin.blog/going-retro-using-pixel-art-on-your-virtual-table-top/
Modern Interiors	LimeZu	16×16 (verify)	colored	paid (verify)	paid	https://limezu.itch.io/moderninteriors
LimeZu catalog	LimeZu	—	—	—	—	https://limezu.itch.io/
Pita catalog (cozy fantasy)	Pita	—	—	—	—	https://pita.itch.io/
Finalbossblues catalog (Time Fantasy series, RPGMaker-native)	Finalbossblues	16×16 (verify)	colored	paid (verify)	paid	https://finalbossblues.itch.io/

License footnotes:

1. mrmotarius: commercial OK, modification OK, redistribution/resale prohibited, attribution appreciated not required. Explicit prohibition on NFT/blockchain and AI/ML use.
2. IKR: commercial OK with attribution required, modification allowed for personal use, redistribution prohibited, no generative AI used in creation.
3. Lobit unclarified: pack page doesn’t specify; comment thread asking about commercial use got no creator reply. Panat’s read: likely CC0-in-spirit (itch.io silent-default behavior). Safe for prototyping; get explicit written clarification before commercial production deployment.

Grid families for mixing:

• 8×8: Mini Medieval + MM UI, Chroma Noir UI, MRMOTEXT, MAS Roguelike
• 12×12: Urizen, MoreGoblins, Lambda Rogue
• 16×16: Kenney (verify on download), Demonic Dungeon + DD UI + DD Interior, Radiant Dungeon, RAS Overworld

Pixel art does not mix across grid sizes within a single scene. Typography/UI (8×8) can live on a different grid from scene tiles (12×12 or 16×16).

Per-pack notes and ranking for Phase 1 (shop interior)

Phase 1 is a specific rendering task: compose a top-down scene of the inside of Dungeon Books. Rank packs against that.

Tier 1 — buy/use now for Phase 1

• Demonic Dungeon — Building Interior (VEXED, $3.25 on sale, CC-BY 4.0, 16×16 px, Mystic-16) — https://v3x3d.itch.io/demonic-dungeon-building-interior

  • Best single pack for Phase 1. Tables, carpet, beds, dressers, shelves, chairs, countertops, fireplace. Inn owner, peasants, knight as NPCs — retool as Carrie, Nachi, customers. Animated pet = shop cat.
  • Gap: it’s “inn/tavern” flavored. Bookshelves-as-bookshelves don’t exist in it — shelves will read as medieval shelving. Acceptable stylization, not a dealbreaker.

• Demonic Dungeon (main pack, $5 on sale, 16×16) — https://v3x3d.itch.io/demonic-dungeon

  • Base pack the Interior expansion depends on. Same palette, heroes, monsters, items. Provides sprite library for “people in the shop.”

• Urizen 1-bit tileset (CC0, 12×12, name-your-own) — https://vurmux.itch.io/urizen-onebit-tileset

  • Keep as the 1-bit spine for print collateral and fallback tile library. Modern theme includes urban tiles useful for Phase 2. Doesn’t participate in the 16×16 scene composition.

• UI kit — decide one:

  • Chroma Noir UI ($3.25, 1-bit, 8×8) — https://v3x3d.itch.io/chroma-noir-user-interface — clean and graphic; contrasts with the colored scenes (chrome stays graphic, world is colored)
  • Demonic Dungeon UI ($3.25, Mystic-16, 16×16) — https://v3x3d.itch.io/demonic-dungeon-user-interface — matches the scene palette and grid exactly
  • Recommended: Demonic Dungeon UI for Phase 1, because grid + palette match is the lowest-risk path when the ship target is a single scene. Chroma Noir UI is a valid choice if you want the “clean graphic chrome wrapping a warm colored world” tension.

Tier 2 — add for Phase 2 (JC streetscape)

• Mini Medieval ($5 on sale, CC-BY 4.0, 8×8, Fruitpunch24) — https://v3x3d.itch.io/mini-medieval

  • Structures: house, barn, tower, castle, ships, bridge. Wrong vibe for modern JC. Medieval. Skip unless we want JC stylized as a fantasy village, which is a real creative option (RPG-ifying the city) but a branding decision, not a default.

• Mini Medieval — User Interface ($3.25 on sale, CC-BY 4.0, 8×8, Fruitpunch24) — https://v3x3d.itch.io/mini-medieval-user-interface

  • Portraits, emojis, buttons, frames, icons, inputs, banners, cursors, scrolls, inventory elements. Matching UI for the Mini Medieval world if we go fantasy-village on JC.

• Radiant Dungeon ($3.25 on sale, CC-BY 4.0, 16×16, Cartoon-Candy) — https://v3x3d.itch.io/radiant-dungeon

  • Brighter palette than Demonic Dungeon. Alternative spine if the Mystic-16 dark palette feels too gloomy for a daytime JC streetscape.

• Kenney 1-bit pack (CC0) — https://kenney-assets.itch.io/1-bit-pack

  • Includes modern and feudal Japan themes in 1-bit. Useful as structural reference for a modern urban top-down, though 1-bit modern-city tiles are visually sparse.

Tier 3 — add only if specific gap appears

All three Mrmotarius packs share the same license: commercial OK, modification allowed, redistribution/resale prohibited, attribution appreciated not required, explicit prohibition on NFT/blockchain and AI/ML use.

• MoreGoblins (“Oh no, more goblins!“) ($5+, 12×12 px, Goblin22 22-color palette) — https://mrmotarius.itch.io/moregoblins
  • ~400 base tiles (extended doubles); x3 variant at 36×36; GoblinRL bonus at 8×16. Animated tiles, 6×6 pixel font. 12×12 matches Urizen.
• Lambda Rogue (WIP) (name-your-own, 12×12, 1-bit) — https://mrmotarius.itch.io/lambda-rogue
  • WIP tileset with 3× upscaled variant. Slots into the Urizen 12×12 family.
• MRMOTEXT ($5+, 8×8, 1-bit) — https://mrmotarius.itch.io/mrmotext
  • Bitmap text/tile building-block system. Base 256 tiles + extended (1000+ tiles + alphanumerics + punctuation) + x3 variant at 24×24. Typography/ornament role, not scenes.

IKnowKingRabbit packs (Aleksandr Makarov) — 12-color “darkest range” palette, commercial OK with attribution, redistribution prohibited:

• RAS: Overworld (Roguelike Asset Series) ($8.75+, 16×16) — https://iknowkingrabbit.itch.io/ras-overworld
  • 200+ tiles, 20+ animated, 45 weapon icons with battle animations, 45 item icons, 12 animated characters, 15 animated enemies. Designed to pair with MAS: Roguelike.
• MAS: Roguelike (Micro Asset Series) ($3.75+, 8×8) — https://iknowkingrabbit.itch.io/mas-roguelike
  • Same content spec as RAS Overworld at 8×8 scale.

Creator catalogs (for browsing remaining packs):

• v3x3d (VEXED) catalog — https://v3x3d.itch.io/
• IKnowKingRabbit catalog — https://iknowkingrabbit.itch.io/
• mrmotarius catalog — https://mrmotarius.itch.io/

Lobit — likely CC0 but unclarified

• Lobit old-school tileset core (helicity) (free, ZIP of PNG tiles) — https://helicity.itch.io/lobit-old-school-tileset-core
  • License status: pack page doesn’t specify; a commenter asked about commercial use, creator didn’t answer. Panat’s read (2026-04-15): default itch.io behavior plus silence from creator points to CC0-in-spirit — creator likely intended permissive use and just never formalized.
  • Practical guidance: usable for internal prototyping and experimentation now. Before any commercial deployment (kiosk in production, NFC card artwork, signage, merch), try once more to get written license clarification. If creator doesn’t respond after a direct ask, the “silent CC0” assumption is common on itch.io but not legally ironclad.
  • Tiles for dungeons/maps; character tokens are a separate pack. Intellivision-flavored pixel. Designed for Tiled editor. Inspired by the Dicegoblin blog post.
  • Initial release Feb 2023; character tokens Feb 2023.

Critical gap: modern urban top-down for Phase 2

None of the packs above do a modern real-city streetscape well. VEXED is fantasy; Urizen’s “Modern” is limited; Mrmotarius is 1-bit fantasy. For Phase 2 we will likely need to either:

1. Stylize JC as a fantasy-flavored village (use Mini Medieval / Demonic Dungeon and accept that Dungeon Books becomes a “bookshop in a fantasy town”). This is on-brand with Guild’s RPG framing and removes the modern-pack gap. Trade-off: loses literal-JC legibility.
2. Hunt for a modern top-down pixel pack in 16×16 with palette compatible with VEXED — possible but not yet scoped. Search terms: “modern top-down pixel,” “urban rpg tiles,” “contemporary top-down.”
3. Commission custom storefronts for the pilot shops — compose JC blocks from stock pavement/street tiles + custom shop facades. Higher cost, highest authenticity.

Phase 2 decision owed: which of these three paths? The fantasy-stylization path is the cheapest and most Guild-coherent; the commissioned-facade path produces the strongest brand artifact but requires an illustrator budget. Don’t pick yet — Phase 1 first.

Candidate packs for the modern/RPGMaker route (added 2026-05-10)

If path 2 (modern top-down pixel) becomes the choice — or if Phase 1 itself shifts toward RPGMaker / Ragnarok Online / Pokemon-interior framing rather than the current VEXED/Fruitpunch24 spine — these are the lineages to evaluate:

• LimeZu — Modern Interiors — https://limezu.itch.io/moderninteriors
  • Top-down modern interior tileset (cafe / shop / office furniture). Direct candidate for a literal Dungeon Books interior render. Includes character generator. Verify exact grid/license on the pack page before purchase.
• LimeZu catalog — https://limezu.itch.io/ — additional Modern* packs (Office, Exteriors, etc.) that share style; useful if Phase 2 streetscape goes modern-realistic.
• Pita catalog — https://pita.itch.io/ — cozy fantasy palette and character work; alternative if VEXED’s Mystic-16 reads too dark and Fruitpunch24 too cartoony.
• Finalbossblues — Time Fantasy series — https://finalbossblues.itch.io/ — RPGMaker MV/MZ-native sprite-sheet layouts (3×4 walk cycles, 8×8 character sets). Standardized format means Phaser parses it trivially and new class sprites slot into one code path. Strongest fit if the kiosk goes JRPG-flavored with animated NPCs walking around the map.

Aesthetic note: these packs commit to a different lineage than the v1 Mini Medieval / Fruitpunch24 spine locked on 2026-04-15. Picking any of them is a v1-spine reconsideration, not an addition. Don’t mix grids/palettes within one scene. The Mini Medieval lock was reasoned (8×8 nostalgia register, MUD-adjacent, “leaves room for imagination”); JRPG/RPGMaker is a different argument (universally legible game grammar, animated NPCs, modern-shop-as-modern-shop). Both are defensible. The reframe trigger: if user testing shows the 8×8 Mini Medieval read is too abstract for Carrie’s customers, these are the packs to evaluate next.

---

Dicegoblin blog — source

“Going Retro: Using Pixel Art on Your Virtual Table Top” (Lars) https://dicegoblin.blog/going-retro-using-pixel-art-on-your-virtual-table-top/

Tool stack

Aseprite, Tiled, Kenney 1-Bit Pack, Lospec.com for palettes, Dwarf Fortress palettes as reference.

Technical specs worth keeping

• Create at 16×16, upscale to 64×64 (4×) with nearest-neighbor interpolation at every step
• Layered composition: background → floor → furniture → walls
• Wall graphics only on visible edges, not full blocks (overhead perspective)
• Randomized stamp brushes for natural terrain variation

Applied to Phase 1

• Tile grid: 16×16 (to match Demonic Dungeon Interior)
• Upscale: 4× for kiosk/portal = 64×64 per tile. Verify at actual kiosk DPI.
• Rendering: CSS image-rendering: pixelated globally; no exceptions on pixel surfaces
• Composition tool: Tiled. Export to whatever the portal/kiosk consumes (PNG, JSON tilemap, etc.)
• Map layer structure: floor (checkout carpet, entryway rug, back room concrete), furniture (shelves, counter, kiosk, event table), people (Carrie, staff, members), ambient (lighting, plants, signage)

---

Where this lives in the product

• Kiosk: full-screen shop-interior scene on idle. Member check-in drops a sprite. Leaderboard overlays on the scene, not separately.
• Portal (dungeon.club): shop-interior scene as the landing page. “Who’s in the shop right now” visible. Member avatar appears on the scene when they’re physically present.
• Discord activity feed: sprite-and-scene mini-renders when members check in, level up, unlock name level
• Print (receipts, labels, zines, posters): 1-bit Urizen + Kenney mono — distinct treatment, not the colored scene
• NFC cards: colored front (member sprite + class), 1-bit back (logo, URL). Hybrid treatment reinforces the split.

Open decisions

• Confirm treatment split: colored top-down for screen world, 1-bit for print. Working hypothesis.
• Pick UI kit: Demonic Dungeon UI (recommended for Phase 1 — grid + palette match) vs. Chroma Noir UI (graphic contrast with colored world)
• Phase 1 scene composition: list the fixed elements (counter, shelves, event table, kiosk location, bathroom, back room, door), then map Demonic Dungeon Interior tiles to each
• Readability test at actual kiosk DPI with nearest-neighbor rendering through the full pipeline
• Phase 2 stylization question (fantasy-village JC / hunt for modern pack / commission storefronts) — defer until Phase 1 ships
• Attribution: single CREDITS surface on portal listing VEXED packs (CC-BY 4.0 requires it)
• Confirm Kenney 1-bit pack’s actual tile size on download (for print pipeline)
• Email helicity re: Lobit license — park until clarified

Related

• outsiderpg-platform-vision — parent product direction this aesthetic serves
• cross-merchant-identity — cross-shop foot traffic that Phase 2 map would visualize
• guild-visual-direction — cassette-futurism / module-art direction (owns print, merchandise, marketing)
• guild-kiosk — kiosk surface that Phase 1 scene renders on
• guild-portal-ui-mockups — portal surface that Phase 1 scene renders on
• name-level — strongholds that could appear as pixel buildings on the Phase 2 streetscape
• rpg-loyalty-system-creative-direction — creative principles
• bbs-ansi-inspiration — adjacent retro-computing aesthetic (owns print surfaces alongside module art)