announcement
Summer 2026 — Umbrella Update
Summer update connecting literate programming, InterDead’s independent C++ Game Core, and the formation of IRON CREED.
This summer at Zhovten Games was mostly about turning several parallel experiments into one development pipeline.
What began as a practical problem inside InterDead led to a methodological study, then to a clearer game architecture, and eventually helped formalize a separate engineering practice around the studio.
These developments turned out to be more closely connected than we expected.
1. Literate Programming
We started with a concrete problem in InterDead.
The scenario needed to remain an independent linguistic source rather than becoming inseparable from a particular engine.
That question led us back to Donald Knuth’s concept of Literate Programming and eventually to a broader methodological review of reproducible source generation, executable documents, and LLM-assisted development.
The research follows WEB and compares several contemporary routes around the same underlying problem: how explanation, source code, execution, and verification can remain connected.
The companion implementations cover CWEB, noweb-like C++, Org Babel, Quarto, Jupyter, and R Markdown.
We also examined a newer boundary case: LLM-assisted development.
A prompt alone is not a literate source, but we found that AI-assisted work can preserve the useful properties of literate programming when it follows a stricter chain:
explanation → specification → code → tests → artifact
The important part for us was not the historical exercise itself.
The research returned directly to the game project that had started it.
Research: Literate Programming: Donald Knuth, WEB, and Contemporary Workflows
Research and companion implementations: Zhovten Games / literate-programming
IRON CREED adaptation: Why Documentation Matters
2. From the Scenario to the Game Core
The research helped us formalize a clearer architectural model for InterDead.
The scenario remains a separate linguistic source.
The implementation moves into an independent C++ Game Core that contains the underlying game logic without tying it to a specific presentation environment.
The current structure is:
Scenario → C++ Game Core → Unreal projection → Web / WASM projection
The projections are not intended to maintain separate copies of the game logic, but to provide different interfaces to the same underlying core.
This is now the stage we are actively building.
The GDD is close to completion.
The scenario has been written.
The next major task is turning that structure into the executable core of the game.
3. IRON CREED — Something Else Escaped
While we were building pipelines, writing research, fixing infrastructure, documenting methods, and trying to keep all of this under control, something else happened.
One of our recurring characters escaped.
IRON CREED became the applied IT and engineering practice of Zhovten Games.
What had existed partly as a character and partly as an internal way of thinking about engineering gradually became a separate public contour for development, infrastructure, automation, security, technical research, documentation, and verification.
This gives us a place for engineering work that deserves to exist beyond a single project.
It also gives us somewhere to put the side research that keeps refusing to remain side research.
IRON CREED does not replace the game studio; instead, its methods, tools, and research return to Zhovten Games and directly influence how we build our projects.
About IRON CREED: IRON CREED — Applied IT and Engineering Practice
Where This Leaves the Project
The intermediate result of the summer can be summarized as one connected chain:
game and narrative design → methodological research → reproducible development pipeline → independent C++ Game Core → multiple projections → engineering methods that return to the studio
For us, this summer was largely about making sure that when we build the visible game, there is a coherent system underneath it.
The next stage is implementation.