Production plan
Six months, $30,000, one finished micro-game. No discovery phase, because the prototype already answered those questions.
What is already built
Everything in this list is in the public playable build, and none of it is being funded twice:
- Multiplayer sync across devices and tabs (Realtime Database + BroadcastChannel fallback), syncing IDs and coordinates only
- 3D plate assembly with twelve primitive ingredients, drag, stack and rotate
- The hairdryer heat model and six cook stages with emissive transitions
- The CRT KDS: tickets, rush timers, bump-bar keys
- Date-seeded procedural specials, reviews and cosmic calorie scoring
- Web Audio synthesis: K-OWL 88.5 FM and eight sound effects, zero downloaded audio
- Zero-auth onboarding with client-side avatar compression
- Kiosk mode, PWA install, Capacitor Android/iOS builds
Schedule
| Month | Milestone | Done when |
|---|---|---|
| 1 | The HUD refactor. KDS becomes a translucent overlay on the live 3D bench; ticket claiming loads that order onto the plate. | The cook never changes screens, and every ticket action works from the keyboard. |
| 2 | The scatter, and the fauna. Physics on abandonment; Ceiling Clinger, Vent Tick and Bin Barnacle in the kitchen. | Switching tickets mid-plate is funny three times in a row in an unprompted playtest. |
| 3 | Art pass. The establishing exterior, the bench, ingredient silhouettes, the sign, the receipt. Contract artist engaged. | It looks intentional on a phone at 360 px wide. |
| 4 | Audio and feel. Contract audio designer on the scatter, the chute, the chime family and the K-OWL generative bed. Juice pass on every interaction. | Playable and legible with sound off, and noticeably better with it on. |
| 5 | Accessibility and the cold start. External accessibility audit and remediation. Single-player mode where the seeded generator plays the diner. | The audit's findings are closed, and a solo player has a complete session. |
| 6 | Release. Web and itch.io launch, Android/iOS wrappers, and the architecture write-up published. | Public, free, $0/month, and a stranger has clipped it. |
Structured playtests run at the end of months 2, 4 and 5 — twelve sessions total, deliberately including players on slow Android devices and locked-down machines.
Budget
Grant requested: $30,000. Total project cost $50,250, the balance being the applicant's prototype work already delivered.
| Line | Basis | Amount |
|---|---|---|
| Contract 3D artist — environment, ingredients, fauna | 18 days @ $600 | $10,800 |
| Contract audio designer — synthesis, generative bed, SFX | 8 days @ $600 | $4,800 |
| Lead design & engineering (applicant) — HUD refactor, scatter, single-player, release | 20 days @ $450 | $9,000 |
| External accessibility audit and remediation | 3 days @ $700 | $2,100 |
| Playtest programme | 12 sessions, participant payment + facilitation | $1,800 |
| Contingency | 5% | $1,500 |
| Total grant | $30,000 |
| Not funded by the grant | ||
|---|---|---|
| Prototype development already delivered | 45 days @ $450, in-kind | $20,250 |
| Hosting and infrastructure, in perpetuity | Free-tier static hosting and database | $0 |
| Total project cost | $50,250 |
Contract rates are set at or above standard Australian games day rates for the relevant discipline; nobody on this project works for a credit. Rates are quoted excluding GST.
Risk register
| Risk | Likelihood | Consequence | Mitigation |
|---|---|---|---|
| Two-player-only games have nobody to play with | High | Fatal to a session | Always-open public room; single-player mode in month 5; the link-in-a-group-chat loop is the primary acquisition path |
| The joke wears out by order three | Medium | Fatal to the product | Playtests at 3, 6 and 10 orders specifically measure this; the procedural board and the fauna exist to supply novelty per order |
| WebGL performance on old mobile | Medium | Excludes a target audience | Primitive geometry, no shadows on low tier, device-matrix testing from month 2 rather than at the end |
| Free-tier limits exceeded by a viral spike | Low | Site goes down at the worst moment | Only IDs and coordinates are synced, no media; hard client-side write throttles; static hosting absorbs reads |
| Contractor availability in a 6-month window | Medium | Slips the art or audio month | Both contractors briefed and scheduled before month 1; art and audio months are independent and can swap |
| Scope creep toward a service game | Medium | Never ships | No roadmap in the design. Features that imply retention are out of scope by definition |
After release
The game stays up, free, at zero cost, indefinitely — that is the point of the architecture. The one ongoing commitment is the published write-up of the zero-account, zero-cost multiplayer pattern, so the next Australian micro-studio can skip the part we had to work out.