head_trim_frames default 14 (production-proven; shot-start sync fix)

INSTRUCTIONS.md

@@ -156,6 +156,7 @@ a GGUF is selected as the DiT (existing limitation, stated in the tooltip).
 | `low_vram` (encoder) | ON if Gemma encoding OOMs; slower but ~3 GB. |
 | resolution | 544x960 is a fast preview. For dialogue, raise height to 864+. |
 | `hires_factor` | 1.0 = off (default). The background AutoFinish upscale usually covers you. |
+| `head_trim_frames` | Leave at 14 (default). Trims the unstable first ~0.5s of each shot, where identity morph and lip-sync error concentrate. |
 
 ## 9. After updating this patch — one-time step
 

nodes.py

@@ -1140,11 +1140,12 @@ class JoyEcho_Generate:
                                "audio static level).",
                 }),
                 "head_trim_frames": ("INT", {
-                    "default": 0, "min": 0, "max": 24,
+                    "default": 14, "min": 0, "max": 24,
                     "tooltip": "Trim this many frames (plus matching audio) from the START of every "
                                "shot. The model's first frames morph out of the reference/memory "
-                               "content - a split-second flash of the reference image. 0 = auto: "
-                               "trims 8 when a reference_image is wired, none otherwise.",
+                               "content, and shot-start lip sync is at its worst there. 14 is the "
+                               "production-proven value (THE WITNESS ran it). 0 = auto: trims 8 "
+                               "when a reference_image is wired, none otherwise.",
                 }),
                 "decode_tiling": (["auto", "on", "off"], {
                     "default": "auto",