diff --git a/port/src/input.c b/port/src/input.c index 84ade086..e400e5c5 100644 --- a/port/src/input.c +++ b/port/src/input.c @@ -293,8 +293,9 @@ static int mouseCaptureMode = 1; /* WI-1 default: Quake-style click-to-lock + static int captureArmed = 0; /* user has clicked to lock (capture mode) */ static int windowFocused = 1; static int mouseAimSpeed = 16; /* aim-mode sensitivity, percent (B3: 50 -> 25 M-29 -> 16; still overshot at 25) */ -static int gepdSens = 25; /* D194 Input.GepdSens: GEPD SENSITIVITY setting, range 1..80 - (20 -> 25 after M-123 user playtest: "a bit slow/unsensitive") */ +static int gepdSens = 38; /* D194 Input.GepdSens: GEPD SENSITIVITY setting, range 1..80 + (20 -> 25 "a bit slow" -> 30 user-calibrated match point + -> 38: user asked defaults ~20-35% faster than that) */ static int aimBand = 20; /* aim mode: usable stick range above the 60 gate */ /* D194/D238: default 100 -> 40 (M-123 user calibration). The old gain * (MOUSE_TURN_GAIN=6 stick/px) saturated the game's quadratic natural-turn @@ -303,7 +304,9 @@ static int aimBand = 20; /* aim mode: usable stick range above the 60 * leave that mode". At 40% (2.4 stick/px) the mid-speed view rate matches * the GEPD reticle's angular pace at GepdSens=30; flicks still reach full * turn speed, only later. */ -static int mouseTurnSpeed = 40; /* hipfire yaw sensitivity, percent */ +static int mouseTurnSpeed = 50; /* hipfire yaw sensitivity, percent + (40 = M-123 match point at GepdSens=30; 50 = the user's + "~20-35% faster" default request, same +25% as gepdSens) */ static int menuPointerSpeed = 100; /* front-end cursor speed, percent */ static int mouseInvertY = 0; /* 1 = mouse-down looks up */ static int mouseYScale = 100; /* extra vertical (pitch) sensitivity, % */