CPUX Machine

CPUX means Common Path of Understanding and Execution.

It is the runtime form of Intention Space.

Where the earlier chapters define Pulse, Signal, Intention, Object, Design Node, and Field, this section explains how those concepts move together as a working machine.


The Core Movement

A CPUX begins when a Signal triggers a configured path.

That path is made of Intention Containers.

Each Intention Container brackets a Design Node with Objects:

O_holder -> DN -> O_reflector

As the CPUX runs, reflected Signals are absorbed into a Field. The Field becomes the represented situational state of that unit of work.


Why CPUX Is Not A Workflow Engine

CPUX may look like a workflow engine at first, but the intention is different.

A workflow engine usually asks:

What step comes next?

CPUX asks:

What situation has stabilised enough for the next Design Node to act?

That shift matters.

CPUX is not only sequencing execution. It is accumulating and testing represented situational reality.


The Main Runtime Parts

The CPUX machine has several cooperating parts:

  • CPUX definition: the configured path of Intention Containers
  • Intention Container: the smallest executable CPUX unit
  • O_holder: the perception-history accumulator
  • Design Node: the explicit computation unit
  • O_reflector: the outward reflection boundary
  • Field: the evolving situational state of the CPUX
  • FieldBoard: the per-user container of active CPUX Fields
  • Visitor: the executor that tests the Field and activates eligible containers
  • GridLookout: the native UI surface that can directly invoke receptor ICs

Together, these parts let computation proceed through represented perception and intention instead of hidden event chains.


Follow The Machine

The CPUX overview should be read alongside the chapters that explain its executable core:

These chapters are where CPUX moves from vocabulary into runtime behaviour.


Reading With The Green Light Sample

The book now uses a small Green Light movement example as a recurring sample anchor.

As you read the machine chapters, keep this path in mind:

GridLookout Cell
-> Signal: I_move_if_allowed
-> IC_move_if_allowed
-> DN_move_if_allowed
-> Signal: I_movement_result
-> O_reflector validation
-> direct result + Field absorption

The code snippets are intentionally small and contract-focused. They show what the reader can implement or inspect around the SDK boundary without exposing the private CPUX engine source.


Backend And Frontend Modes

The same CPUX idea can run in different environments.

In backend mode, the Visitor can run synchronously until no more containers are eligible.

In frontend mode, the Visitor may sleep at the golden pass and wake again when human action changes the Field.

Frontend mode should be understood primarily as a client-side runtime engine. It is not only a UI adapter or rendering helper. The CPUX engine can live on the client device, maintain active Fields, coordinate ICs, handle direct GridLookout action, and preserve the human perception loop at runtime.

The rendering platform may be ReactJS, Android, iOS, Flutter, desktop, terminal, or embedded UI. The frontend CPUX engine is the client-side coordination layer underneath that native surface.

This gives CPUX a single conceptual machine with different runtime policies:

backend:  run -> stabilise -> terminate
frontend: run -> stabilise -> sleep -> wake on human action

The Human Loop

Frontend CPUX includes a special requirement:

the originating UI Cell must receive the direct IC result without waiting for Field absorption or the Visitor's next pass.

This preserves immediate human perception.

The same reflected Signal still goes to the Field and subscribers. CPUX therefore supports both:

  • immediate human-facing feedback
  • ordered system-level stabilisation

This dual path is one of the most important ideas in the CPUX machine.