Nine defined stages—from the 1.0 foundation through 3.1RR—form one progressive learning path.
Computer architecture becomes learnable when state stops being invisible.
Ark is an educational processor lab. It connects code, execution, machine state, guided practice, and verified AI help without hiding the architecture behind an abstraction.
From reading diagrams to running ideas.
Architecture courses ask students to connect source code, encodings, datapaths, memory, and control flow. Those connections are difficult to form when every piece lives in a different tool—or when the machine only appears after an error.
Ark keeps those layers in one place. A student can step a single instruction, watch its effects, test a hypothesis, and ask for an explanation grounded in the processor that is actually open.
A browser-native continuation, not a disconnected rewrite.
The web product is guided by the project’s own original instruction documentation.
The project’s own instruction architecture supplies the scalar, stack, control, and graphics semantics.
The browser profile turns that architecture into an inspectable lab with source, state, display, lessons, and AI verification.
What Ark promises students and instructors.
The processor is explicit
No silent fallback changes a student’s target. Saved files, AI runs, lessons, and history retain their processor identity.
Feedback comes from execution
Lesson outcomes and code-changing AI answers are grounded in bounded processor runs, not textual confidence.
The first experiment is open
Guests can use the complete simulator and practice lab. An account is needed only for AI and account-synced projects.
Boundaries stay visible
Unsupported instruction families and deployment-dependent features are named honestly instead of simulated by implication.