Phase 0: Planning — From Idea to Blueprint¶

Every project starts somewhere. Dilder started with a question: how hard is it to actually build a Tamagotchi?

The Idea¶

The original Tamagotchi was a 1996 keychain toy with a tiny LCD, three buttons, and a virtual pet that needed feeding, attention, and care. The concept is simple. The implementation, for a modern DIY version with a proper e-ink display, 3D-printed case, and programmable firmware, is substantially more involved.

The goals for Dilder:

A functional virtual pet device you can hold in your hand
A documented build process — hardware, software, enclosure, everything
A public project with a community that can build their own
An honest look at AI-assisted development — all prompts logged

What Phase 0 Covered¶

Phase 0 was purely planning. No hardware ordered, no code written. Just research and documentation:

Hardware research — identifying the right display (Waveshare 2.13" e-ink), compute (starting with Pico W, Pi Zero WH later), and input options (tactile buttons)
Enclosure concepts — rough proportions, button layout, display-to-case ratio
Project structure — phases, documentation format, GitHub repo
Website and community planning — this site, Discord, Patreon

Key Decisions Made¶

E-Ink Over LCD¶

E-ink was chosen over a small OLED or TFT LCD for a few reasons:

Standby current — less than 0.01µA in standby. Perfect for a device that needs to run on a small battery.
Readability — high contrast in any lighting condition, no backlight needed
Aesthetic — the Tamagotchi style aesthetic benefits from a high-contrast, "paper" look
Tradeoff — 2-second full refresh, 0.3-second partial refresh. Animation is possible but limited.

Pico W First, Pi Zero Later¶

Development starts on the Raspberry Pi Pico W — it's what we have on hand, it's cheap (~€6), and it's ideal for prototyping the display and input system with MicroPython. Instant boot, simple USB workflow, no OS to configure.

The Pi Zero WH is planned for Phase 5 when we need Linux, a filesystem, networking features, or more compute for a personality/AI system. The display and button wiring is nearly identical between the two boards.

Tactile Buttons Over Capacitive Touch¶

Five discrete 6×6mm tactile buttons (up, down, left, right, center/select) beat capacitive touch for a first prototype because:

Breadboard-friendly — no soldering required for the test bench
Tactile feedback matters for a game device
Zero additional components needed (Pico W's internal pull-ups handle everything)

Prototype Concept¶

The enclosure concept settled on an "iPod Nano" style layout — landscape rectangle, display on the left covering ~75% of the face, compact d-pad cluster on the right.

v2 dimensions:

Metric	Value
Case outer	88 × 34 × 19mm
Display window	57 × 27mm
Active pixel area	48.55 × 23.71mm (250×122 px)

See the hardware docs for full enclosure specs

What's Next¶

Phase 1 is hardware assembly. The Pico W and e-ink display are on the bench. The goal: flash MicroPython, wire up the display and buttons on a breadboard, and get something — anything — to show on the screen.

When the display shows its first pixel, Phase 1 is done.