Readability and Designing for Attributable Failure: A Heuristic Framework and its application to Commercial Action-Roguelites: Case Study of Hades
Minenko, Egor (2025)
Minenko, Egor
2025
All rights reserved. This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-2025111428060
https://urn.fi/URN:NBN:fi:amk-2025111428060
Tiivistelmä
This thesis proposes a heuristic framework for readability and fairness in action-roguelites: practical rules that developers can apply to ensure legible hazards, fast fail-to-retry loops, and perceived fairness. Building on theory (failure attribution, flow, and learning-through-patterns), the study operationalizes readability with explicit thresholds for five items: telegraph window ( ≥ 500 ms), bullet-to-background contrast (≥ 3:1), HUD/VFX occlusion (≤ 10% of sampled frames), death-to-control latency (“friction seconds”, ≤ 5 s), and first-death attribution (player can state the cause in one sentence). A 0–2 scoring rubric and a compact protocol (clips, frame counts, sampled frames, timed restarts) support consistent evaluation.
We apply the checklist to a commercial action-roguelite called Hades, producing one-pagers (table + annotated frame + notes) and a game matrix. Across the sample, we observe recurrent failure modes: contrast clashes in dark biomes, telegraph windows collapsed by overlapping events, and celebratory VFX or damage numbers masking hazards; and effective patterns such as consistent telegraphs, palette discipline, UI/VFX layering that avoids occlusion, and near-instant restarts.
The contribution is threefold: (1) a printable checklist with definitions and thresholds; (2) a replicable protocol for evidence captures and scoring; and (3) comparative observations that translate into actionable guidance for developers seeking evidence-backed readability improvements.
We apply the checklist to a commercial action-roguelite called Hades, producing one-pagers (table + annotated frame + notes) and a game matrix. Across the sample, we observe recurrent failure modes: contrast clashes in dark biomes, telegraph windows collapsed by overlapping events, and celebratory VFX or damage numbers masking hazards; and effective patterns such as consistent telegraphs, palette discipline, UI/VFX layering that avoids occlusion, and near-instant restarts.
The contribution is threefold: (1) a printable checklist with definitions and thresholds; (2) a replicable protocol for evidence captures and scoring; and (3) comparative observations that translate into actionable guidance for developers seeking evidence-backed readability improvements.
