app+csdk+docs: 修正上一轮最近邻2×的退化（颗粒/边沿台阶）——照片改为从160×120单级6×放大（官方ImageViewer同款），实时缓冲与csdk回退为插值2×；实测冷区|dx| 0.76×官方（更平滑），119单测全绿

2026-10-06 17:57:17 +08:00
parent 4a3916a2b7
commit 595309735e
9 changed files with 124 additions and 19 deletions
@@ -61,12 +61,13 @@ typedef struct mag160c_render_cfg_t {
    /* Brightness offset for the Android LUT base; the official app passes 0. */
    int lut_brightness;
    /* 1 = replicate pixels in the 160x120 -> 320x240 stage instead of interpolating.
-       The official ANDROID path has no such stage at all (its SDK hands the app a
-       160x120 bitmap which ImageViewer draws with a non-filtering paint = nearest),
-       while the PC SDK's bilinear version fabricates an interleaved full/half
-       amplitude column pattern at the sensor Nyquist - visible as extra fine stripes
-       in the dark/cold regions.  Measured on device: fine-scale cold-region stripe
-       ratio vs the official 1.16-1.29x (bilinear) -> 1.00/0.93 (nearest). */
+       Measured on device, this is the WRONG trade: replication halves the effective
+       resolution (2-sensor-pixel granules, a coarse staircase along hot/cold edges,
+       reported by the user), while the interpolation only adds a faint full/half
+       amplitude column pattern at the sensor Nyquist.  A consumer that can draw
+       gray160 directly (like the Android app's photo path: one 6x bilinear step from
+       the sensor grid) needs no intermediate stage at all and gets both right.
+       Kept as a switch for experiments; leave 0. */
    int upscale_nearest;
    mag160c_render_ffc_cb_t ffc_cb;
    void *ffc_user;