Why Early Access?

“Jingjie: Probability Cloud - Nagumo's Dilemma is a systems-heavy carrier command game about uncertain scouting, report confidence, deck timing, and strike commitment. If we use Early Access, the goal is to let players help us test whether that command loop is readable and rewarding over repeated play.

We would also use Early Access to tune the adaptive opponent. The game can build a bounded tactical profile from play and after-action evidence, then use it to adjust enemy pressure in later battles. Player feedback is important because this kind of self-learning AI must feel understandable, fair, and earned rather than random or cheating.

Early Access is not intended as crowdfunding, and players should judge the game by the playable build available at the time they buy it, not by future promises.”

Approximately how long will this game be in Early Access?

“If we enter Early Access, our current target would be roughly 9 to 12 months. That estimate may change if player feedback shows that onboarding, scenario structure, or core command readability needs more work than expected.

We will update this section if the schedule changes in a meaningful way.”

How is the full version planned to differ from the Early Access version?

“The planned full version is expected to expand beyond the initial teaching-focused build with more scenarios, clearer progression, stronger debrief and after-action tools, broader carrier-operation situations, refined enemy counterplay, and additional polish for settings, localization, performance, and accessibility.

We also plan to keep developing the game's adaptive / self-learning AI layer. The direction is not an all-knowing opponent, but a fair strategic opponent that learns from visible tactical patterns: over-trusting reports, predictable search corridors, late validation, unsafe fuel margins, or repeated strike timing. Future work may add richer player tactical profiles, more varied next-battle pressure dimensions, clearer debrief explanations, and settings that let players reduce or disable adaptation.

Those plans may change as we learn from players. Our priority is to make the intelligence chain understandable: players should be able to explain why a strike hit, missed, or arrived too late.”

What is the current state of the Early Access version?

“The current build contains a playable carrier-command slice centered on blue-water search, scout planning, contact reports, confidence, report age, error area, action tradeoffs, deck pressure, CAP / strike readiness, and outcome debriefing.

It also contains a first playable version of the adaptive opponent stack. Inside a scenario, the enemy can react to exposure evidence without reading hidden player UI. Across after-action analysis, the game can turn replay features into a bounded adaptive policy for future pressure, such as decoy pressure, course changes after contact, tempo compression, search-gap exploitation, or fuel-margin traps.

It includes a packaged Windows build, basic settings such as language, volume, font size, fullscreen, and adaptation controls. It does not yet represent a finished campaign. Wider content, save / checkpoint behavior, balance, onboarding clarity, long-form progression, and the full potential of the self-learning opponent are still under review.”

Will the game be priced differently during and after Early Access?

“If we launch in Early Access, the price should reflect the content and polish of the current build. We currently expect the Early Access price to be lower than the eventual 1.0 price, with any increase tied to substantial added content and polish.

Any price change would be communicated before it happens. Regional pricing may differ by territory through Steam.”

How are you planning on involving the Community in your development process?

“We plan to use Steam forums, playtest feedback, bug reports, update notes, and focused surveys to understand where players become confused or where the command loop becomes compelling.

The most important feedback areas are first-run comprehension, contact and confidence terminology, action-drawer readability, scenario pacing, performance, accessibility, debrief clarity, and whether the adaptive AI feels fair. We especially want to know when the opponent's learning feels like meaningful counterplay, and when it feels opaque, too punishing, or insufficiently explained.

We will not treat development as a simple feature vote, but community feedback will help us decide what needs explanation, tuning, or redesign before 1.0. If we ever use external playtest data to improve the adaptive model beyond local game telemetry and internal simulation, we would communicate that separately and clearly.”