diff --git a/SKILL.md b/SKILL.md
index fa4b776d..2c930ea9 100644
--- a/SKILL.md
+++ b/SKILL.md
@@ -181,13 +181,13 @@ Subtitles have three dimensions worth reasoning about: **chunking** (1/2/3/sente
 
 **Worked styles** — pick, adapt, or invent:
 
-**`bold-overlay`** — short-form tech launch, fast-paced social. 2-word chunks, UPPERCASE, break on punctuation, Helvetica 18 Bold, white-on-outline, `MarginV=35`. `render.py` ships with this as `SUB_FORCE_STYLE`.
+**`bold-overlay`** — short-form tech launch, fast-paced social. 2-word chunks, UPPERCASE, break on punctuation, Helvetica 18 Bold, white-on-outline, `MarginV=90`. `render.py` ships with this as `SUB_FORCE_STYLE`.
 
 ```
 FontName=Helvetica,FontSize=18,Bold=1,
 PrimaryColour=&H00FFFFFF,OutlineColour=&H00000000,BackColour=&H00000000,
 BorderStyle=1,Outline=2,Shadow=0,
-Alignment=2,MarginV=35
+Alignment=2,MarginV=90
 ```
 
 **`natural-sentence`** (if you invent this mode) — narrative, documentary, education. 4–7 word chunks, sentence case, break on natural pauses, `MarginV=60–80`, larger font for readability, slightly wider max-width. No shipped force_style — design one if you need it.
@@ -263,7 +263,7 @@ One sub-agent = one file (unique filenames, parallel agents don't overwrite each
 
 ## Output spec
 
-Match the source unless the user asked for something specific. Common targets: `1920×1080@24` cinematic, `1920×1080@30` screen content, `1080×1920@30` vertical social, `3840×2160@24` 4K cinema, `1080×1080@30` square. `render.py` defaults the scale to 1080p from any source; pass `--filter` or edit the extract command for other targets. Worth asking the user which delivery format matters.
+Match the source unless the user asked for something specific. Common targets: `1920×1080@24` cinematic, `1920×1080@30` screen content, `1080×1920@30` vertical social, `3840×2160@24` 4K cinema, `1080×1080@30` square. `render.py` currently renders at 24 fps and scales each source's long edge to 1920 pixels (1280 in draft mode). Edit the extraction settings when another delivery target is required; `--filter` belongs to `grade.py`, not `render.py`. Worth asking the user which delivery format matters.
 
 ## EDL format
 
diff --git a/helpers/render.py b/helpers/render.py
index 0d02cffa..b56d7cd8 100644
--- a/helpers/render.py
+++ b/helpers/render.py
@@ -10,7 +10,7 @@
 
 Optionally builds a master SRT from the per-source transcripts + EDL
 output-timeline offsets, applies the proven force_style (2-word
-UPPERCASE chunks, Helvetica 18 Bold, MarginV=35).
+UPPERCASE chunks, Helvetica 18 Bold, MarginV=90).
 
 Usage:
     python helpers/render.py <edl.json> -o final.mp4