<![CDATA[Pengembangan game horor konvensional yang statis sering kali gagal mempertahankan tingkat ketegangan pemain. Penelitian ini mengimplementasikan prototipe serious game horor 2D adaptif "SUKMA" menggunakan metode Game Development Life Cycle (GDLC). Sistem mengintegrasikan ESP32, Pulse Sensor Amped, broker MQTT Eclipse Mosquitto, dan Godot Engine (.NET/C#) dengan algoritma Finite State Machine (FSM). Logika FSM membagi kondisi biologis pemain menjadi tiga state: Calm (<85 BPM; 5 NPC hantu aktif), Tense (85–100 BPM), dan Panic (>100 BPM) yang memicu penyesuaian berupa pengurangan hantu, penurunan agresivitas, dan efek visual red vignette. Pengujian Black Box (11 skenario) mencapai tingkat keberhasilan 100%. Latensi transmisi data MQTT berlangsung secara real-time dengan rata-rata 42,5 ms ( ms). Sementara itu, User Acceptance Test (UAT) terhadap 20 responden menghasilkan nilai penerimaan 84,80% (Sangat Baik). Hasil ini membuktikan efektivitas integrasi sensor detak jantung, MQTT, dan FSM dalam menciptakan pengalaman bermain horor yang adaptif dan personal. Conventional horror game development is often static and fails to maintain consistent player tension. This study implements "SUKMA", an adaptive 2D horror serious game prototype developed using the Game Development Life Cycle (GDLC) method. The system integrates an ESP32 microcontroller, Pulse Sensor Amped, Eclipse Mosquitto MQTT broker, and Godot Engine (.NET/C#) powered by Finite State Machine (FSM) logic. The FSM categorizes player physiological conditions into three states: Calm (<85 BPM; 5 active ghost NPCs), Tense (85–100 BPM), and Panic (>100 BPM), which triggers negative adaptations including reduced ghost count, decreased aggressiveness, and a red-vignette visual effect. Black Box testing across 11 scenarios achieved a 100% success rate. Data transmission latency via MQTT averaged 42.5 ms ( ms), well below the 100 ms real-time threshold. Furthermore, User Acceptance Testing (UAT) with 20 student respondents yielded an 84.80% approval score (Very Good category). The results demonstrate that combining heart-rate biofeedback, MQTT protocol, and FSM effectively creates a real-time, adaptive, and personalized horror gaming experience.]]>
