Context windows: correct the cost figure in both tooltips

overlap/length (~28%) is the per-window redundancy, not the whole-clip cost.
The real figure is stride-based: at 145/40 each window advances 105 new frames,
so a windowed scene does ~1.45x the per-frame work regardless of length. That
is offset by attention, which is quadratic in one pass and near-flat when
windowed: ~1.05x at 200 frames, 0.68x at 305, 0.42x at 500. Net: roughly
break-even around 200 frames, a real win past ~300.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

movie_editor/frontend/engine_settings.js

@@ -227,7 +227,7 @@
       deps: [{ name: "segmented_detailing" }, { name: "detail_mode", value: "repair" }],
       hint: "How much noise the crop gets re-noised to before its 3-step refine (0.85 is the official LTX 2.3 recipe's own value). Higher = more freedom to genuinely reconstruct the region (fix bad anatomy), risking drift from the surrounding frame; lower = closer to a plain upscale — looks 'detailed' as interpolation but doesn't actually repair it. If the result looks upscaled but not corrected, raise this. Only used in 'repair' mode." },
     { name: "context_windows", label: "Context windows (long scenes)", kind: "bool", default: false,
-      hint: "EXPERIMENTAL: denoise a scene that's LONGER than the model's comfortable window as overlapping windows instead of one giant pass — ComfyUI core's own mechanism, audio-aware on LTX (it maps each video window to its audio window and re-slices anchors, guides and JoyAI memory per window). Engages only on scenes longer than the window length below; shorter scenes are untouched and pay nothing. Cost isn't a flat multiple — the total work is about the same, plus the overlap being computed twice (~28% at the defaults). Needs ComfyUI v0.29.0 or newer; on older builds it's skipped with a note in the scene report. UNVALIDATED LIVE." },
+      hint: "EXPERIMENTAL: denoise a scene that's LONGER than the model's comfortable window as overlapping windows instead of one giant pass — ComfyUI core's own mechanism, audio-aware on LTX (it maps each video window to its audio window and re-slices anchors, guides and JoyAI memory per window). Engages only on scenes longer than the window length below; shorter scenes are untouched and pay nothing. Cost at the defaults (145/40) is about 1.45× the per-frame work, since each window re-does its 40-frame overlap — offset by attention getting cheaper the longer the scene is (quadratic in one pass, near-flat when windowed). Roughly break-even around 200 frames, a net win past ~300. Needs ComfyUI v0.29.0 or newer; on older builds it's skipped with a note in the scene report. UNVALIDATED LIVE." },
     { name: "context_window_length", label: "Window length (frames)", kind: "int", default: 145, min: 9, max: 2049, step: 8, dependsOn: "context_windows",
       hint: "Window size in real frames. A scene at or below this length skips windowing entirely, so this doubles as the engage threshold. Keep it at or under the length the model already generates well in one pass — the point is to stay inside that range while the scene as a whole goes past it." },
     { name: "context_window_overlap", label: "Window overlap (frames)", kind: "int", default: 40, min: 0, max: 512, step: 8, dependsOn: "context_windows",

samplers.py

@@ -2484,7 +2484,7 @@ class FunPackLTXAVSceneChainSampler:
                 }),
                 "context_windows": ("BOOLEAN", {
                     "default": False,
-                    "tooltip": "EXPERIMENTAL: denoise a scene LONGER than the model's native window as overlapping context windows instead of one giant pass (ComfyUI core's own comfy.context_windows, LTX2 branch — nothing ported). Each step splits the scene into windows of context_window_length frames, denoises each, and fuses the overlaps back together. Core's LTXAV path is genuinely audio-aware: it unpacks the packed AV latent, maps each video window onto its proportional AUDIO window, and re-slices denoise masks, keyframe_idxs and guide_attention_entries per window — so i2v anchors, mid-scene guides and JoyAI memory keep working inside a window. Cost is NOT a flat multiple: total tokens stay roughly the same and windowed attention is cheaper than full-length attention, but the overlap is recomputed (overlap/length extra work, ~28% at the defaults). Engages ONLY when the scene is longer than context_window_length — shorter scenes are untouched and pay nothing. Off by default. UNVALIDATED LIVE.",
+                    "tooltip": "EXPERIMENTAL: denoise a scene LONGER than the model's native window as overlapping context windows instead of one giant pass (ComfyUI core's own comfy.context_windows, LTX2 branch — nothing ported). Each step splits the scene into windows of context_window_length frames, denoises each, and fuses the overlaps back together. Core's LTXAV path is genuinely audio-aware: it unpacks the packed AV latent, maps each video window onto its proportional AUDIO window, and re-slices denoise masks, keyframe_idxs and guide_attention_entries per window — so i2v anchors, mid-scene guides and JoyAI memory keep working inside a window. Cost at the defaults (145/40): about 1.45x the per-frame work, because each window re-does its 40-frame overlap, offset against attention getting CHEAPER the longer the scene is (attention is quadratic in one pass, near-flat when windowed) - roughly break-even around 200 frames, a net win past ~300. Engages ONLY when the scene is longer than context_window_length — shorter scenes are untouched and pay nothing. Off by default. UNVALIDATED LIVE.",
                 }),
                 "context_window_length": ("INT", {
                     "default": 145, "min": 9, "max": 2049, "step": 8,