← systems
Laptop

Open laptop

A system architecture that isolates the cost-and-differentiation choice in one socket.

baseline: MNT Reform — CERN-OHL / GPL, full KiCad published
CPU i.MX8M Plus (socketed SOM)display 13.3″ 1920×1080 eDPbattery 8× LiFePO4 18650design open KiCad

MNT Reform is fully open — schematics, PCB, firmware, mechanicals. Its design concentrates the question: the processor lives on a swappable module, so everything around it is settled and low-cost to reuse, and the socket leaves one clean choice — reuse a stock module, or design a new one.

7 of 8 settled or cyclical — 1 on the frontier. Each part placed by cost and differentiation — the landscape, not a verdict. The true cost of building without knowing the market and the benefit, surfaced — so you choose by your goal.

Exploded

Decompose the product into its stack. Each layer is colour-coded: settled, cyclical, or frontier.

13.3″ eDP display Settled
Processor module (SOM) Frontier
Open motherboard v3.0 Settled
Keyboard + trackball (RP2040) Settled
LiFePO4 packs + BMS Settled

Bill of materials

Settled ×7 Cyclical ×0 Frontier ×1

Every part is a real, sourced component from the open baseline — no invented part numbers.

i.MX8M Plus SOM NXP Frontier · higher cost and risk — differentiation, SOTA, or science
Processor module — Quad Cortex-A53 on a socketed module (RK3588 / LS1028A / RISC-V alternates)

The socket isolates the cost-and-differentiation choice. Reuse a stock module to ship, or design an open RISC-V application SoC for the slot — energy-optimised, no vendor boot blob. openWafer can do either.

https://mntre.com/documentation/reform-handbook/hardware.html →
Reform motherboard v3.0 MNT Research Settled · mature — lowest cost and risk
Mainboard — open KiCad; carries system controller, bridges, hub, audio

Settled open design. Reuse.

https://mntre.com/documentation/reform-handbook/hardware.html →
LPC11U24 NXP Settled · mature — lowest cost and risk
System controller — Cortex-M0; power rails, charger, keyboard link

Mature MCU. Reuse.

https://mntre.com/documentation/reform-handbook/hardware.html →
LTC6803 + LTC4020 + INA260 Analog Devices / TI Settled · mature — lowest cost and risk
Battery management — 8-cell monitor + LiFePO4 buck/boost charger + V/I

Standard BMS chain. Reuse.

https://mntre.com/documentation/reform-handbook/hardware.html →
18650 LiFePO4 ×8 various (JGNE / Eremit) Settled · mature — lowest cost and risk
Battery cells — 3.2 V cells in two packs

Commodity cells. Reuse.

https://mntre.com/documentation/reform-handbook/hardware.html →
13.3″ eDP (Innolux / BOE / AUO) Innolux / BOE / AUO Settled · mature — lowest cost and risk
Display — 1920×1080 via SN65DSI86 DSI-to-eDP bridge

Standard panel + bridge. Reuse.

https://mntre.com/documentation/reform-handbook/hardware.html →
RP2040 Raspberry Pi Settled · mature — lowest cost and risk
Keyboard controller — Kailh Choc switches, open firmware

Open input controller. Reuse.

https://mntre.com/documentation/reform-handbook/hardware.html →
RP2040 + PAT9125EL Raspberry Pi / PixArt Settled · mature — lowest cost and risk
Trackball — 5-button optical trackball

Open trackball. Reuse.

https://mntre.com/documentation/reform-handbook/hardware.html →

Where openWafer can help

Reform concentrates the cost-and-differentiation question in one place: the processor lives on a socketed module, so the open motherboard, the LiFePO4 management, the keyboard, and the trackball are settled and low-cost to reuse. That leaves a clean choice at the socket — reuse a stock module to ship, or design a new one to differentiate. The frontier option there is an open RISC-V application SoC, energy-optimised, with no vendor boot blob, which openWafer can design and drop into the slot MNT standardised. The architecture makes either path an informed, isolated decision rather than a whole-system rebuild.

Open sources — decomposed and linked, not hosted