diff --git a/ROADMAP.md b/ROADMAP.md
index 144e224b0..36f1ec980 100644
--- a/ROADMAP.md
+++ b/ROADMAP.md
@@ -105,7 +105,7 @@ Ranked toward beta; detail in [docs/roadmap.md](docs/roadmap.md), gaps in [docs/
 1. Camera colour of our own: measured calibration for the popular models (50–90 h)
 2. CR3 for every Canon body (25–45 h)
 3. Render-fidelity suite against Lightroom, then tuning (30–60 h)
-4. Full-resolution zoom (8–15 h)
+4. ~~Full-resolution zoom~~ (done 2026-10-12)
 5. AI masks: Subject / Sky / People (40–70 h + model decision)
 6. Compressed ORF and HE NEF (28–55 h)
 7. Lens profile database of our own (40–80 h)
@@ -192,6 +192,7 @@ Ranked toward beta; detail in [docs/roadmap.md](docs/roadmap.md), gaps in [docs/
 | 2026-10-12 | minor | Mamiya MEF decoded and imported: ZD verified (241); headline numbers unchanged |
 | 2026-10-12 | minor | Epson ERF decoded: 3 more verified models (240); headline numbers unchanged |
 | 2026-10-12 | minor | Georgian (`ka`) interface language added (every catalog message) |
+| 2026-10-14 | minor | Full-resolution zoom (gap 4) closed: soft proofing gets zoom windows; headline numbers unchanged |
 | 2026-10-12 | minor | UI/UX remaining 40–70 h → 34–58 h (accessibility audit landed); headline numbers unchanged |
 | 2026-10-14 | minor | Camera-coverage sweep (#1327): 30 models verified on 40 CC0 files (coverage.py 515 → 545 verified); sensor verification does not establish colour parity (#1277); headline numbers unchanged |
 | 2026-10-12 | minor | 22 more CR3 bodies verified, EOS R6 Mark III white balance: RAW decoding 77% → 83%, RAW ~56% → ~60%; mainstream 49% → 51%, essentials 64% → 65%; full unchanged at ~61% |
diff --git a/crates/engine/src/media.rs b/crates/engine/src/media.rs
index 6d1bfe0de..22d76528d 100644
--- a/crates/engine/src/media.rs
+++ b/crates/engine/src/media.rs
@@ -2189,6 +2189,40 @@ mod tests {
         assert!(mean(&a) > mean(&b) + 10.0, "two stops brighter after: {} vs {}", mean(&a), mean(&b));
     }
 
+    // LR-VIEW-ZOOM: what the loupe shows at 1:1 is what an export at full size writes. A window of
+    // the zoomed frame (the photo's own pixels) is pixel-for-pixel the export, at the centre and
+    // at a corner, with tone, colour and detail edits on a camera-sized photo
+    #[test]
+    fn one_to_one_window_pixels_are_export_pixels() {
+        use lightcraft_pipeline::{OutputDepth, OutputSpace, PixelWindow};
+        let mut s = crate::Session::with_demo();
+        let id = s.active().unwrap();
+        let p = s.catalog.photo(id).unwrap().clone();
+        let (w, h) = (p.width as usize, p.height as usize);
+        assert!(w.max(h) > 2560, "larger than the preview cap: {w}×{h}");
+        let mut d = (*s.develop_of(id).unwrap()).clone();
+        d.light.exposure = 0.7;
+        d.light.contrast = 25.0;
+        d.color.saturation = 20.0;
+        s.set_develop(id, d, "Edits").unwrap();
+        let export = s.render_export(id, w, h, OutputSpace::Srgb, OutputDepth::U8).unwrap().image;
+        assert_eq!((export.width, export.height), (w, h), "a full-size export");
+        for win in [PixelWindow { x: w / 2 - 256, y: h / 2 - 256, w: 512, h: 384 }, PixelWindow { x: 0, y: h - 300, w: 400, h: 300 }] {
+            let tile = s.region_job(id, w, h, win, true).unwrap().run().rendered.unwrap().image;
+            assert_eq!((tile.width, tile.height), (win.w, win.h));
+            let mut worst = 0u8;
+            for y in 0..win.h {
+                for x in 0..win.w {
+                    let (a, b) = (tile.data[y * win.w + x], export.data[(win.y + y) * w + win.x + x]);
+                    for k in 0..3 {
+                        worst = worst.max(a[k].abs_diff(b[k]));
+                    }
+                }
+            }
+            assert!(worst <= 1, "window {win:?}: the 1:1 view differs from the export by up to {worst} levels");
+        }
+    }
+
     /// The thumbnail of a photo whose preview-level source is decoded is cut from it: the file (here
     /// one nothing can decode) is not read again.
     #[test]
diff --git a/crates/ui-egui/src/panels/detail.rs b/crates/ui-egui/src/panels/detail.rs
index 68b5346e3..44427565e 100644
--- a/crates/ui-egui/src/panels/detail.rs
+++ b/crates/ui-egui/src/panels/detail.rs
@@ -402,12 +402,16 @@ fn request_window(app: &mut LightcraftApp, c: &WindowCtx, v: &WindowView) -> Opt
     let job = if c.interacting { job.draft() } else { job };
     let overlay = if v.before { lightcraft_pipeline::Overlay::None } else { c.overlay };
     let job = if v.before { job } else { job.with_hidden_grading(app.ui.grading_hidden_for(c.id)) };
-    let job = job.with_overlay(overlay);
+    // the After window is proofed like the whole-frame render under it; the Before side is
+    // never proofed (its whole-frame render isn't either)
+    let proof = app.ui.soft_proof.then_some(app.ui.proof);
+    let job = job.with_overlay(overlay).with_proof(if v.before { None } else { proof });
     // keyed as the renderer keys the texture (a display profile changes the key)
     let job = app.renderer.job_for(v.slot, job);
-    // The overlay is part of the tile's identity too, so changing mask view/colour never leaves a
-    // stale region tile on screen while its replacement renders.
-    let look = job.settings.hash64() ^ overlay.key().rotate_left(17);
+    // The overlay and the proof are part of the tile's identity too, so changing mask view/colour
+    // or the proof never leaves a stale region tile on screen while its replacement renders.
+    let proof_key = if v.before { 0 } else { proof.map_or(0, |p| p.key()) };
+    let look = job.settings.hash64() ^ overlay.key().rotate_left(17) ^ proof_key.rotate_left(41);
     let view = crate::region::RegionView { photo: c.id, before: v.before, key: job.key, full: (fw, fh), window: win, settings: look };
     // windows of other photos, looks and zooms of this side can't be drawn any more
     app.region_tiles.retain(|(before, _), t| *before != v.before || t.is_current(c.id, (fw, fh), look, c.interacting));
@@ -533,15 +537,16 @@ pub fn show(app: &mut LightcraftApp, ui: &mut egui::Ui) {
     let drawn_long = target_rect.width().max(target_rect.height()) * ppp;
     // the whole frame at about canvas size; the zoomed part of it is a window render (below)
     let look = d.hash64();
-    // Diagnostic views that need whole-frame context, soft proofing, or reads too far for one
-    // window still render the whole frame at the size it is drawn. A mask overlay is window-safe.
+    // Diagnostic views that need whole-frame context, or reads too far for one window, still
+    // render the whole frame at the size it is drawn. Mask overlays and soft proofing (both act
+    // pixel by pixel) are window-safe.
     let sizes = crate::region::ViewSizes {
         drawn_long,
         canvas_long: canvas.width().max(canvas.height()) * ppp,
         native_long,
         texture_side,
         draft,
-        windows: !app.ui.soft_proof && overlay_supports_window(overlay),
+        windows: overlay_supports_window(overlay),
     };
     let mut plan = crate::region::plan(&app.ui.settings, sizes);
     if plan.window_edge.is_some_and(|edge| app.window_refused == Some((id, look, edge))) {
@@ -623,10 +628,9 @@ pub fn show(app: &mut LightcraftApp, ui: &mut egui::Ui) {
         app.renderer.request(Slot::Before, job, 90);
     }
     // zoomed past what the whole-frame render holds: also render the part on screen, at no more
-    // than 100 % (the GPU magnifies beyond that), for each side that is shown. Mask overlays use
-    // the same window; diagnostic overlays that need whole-frame context do not.
-    let windowable_view =
-        hover_key.is_none() && !app.ui.soft_proof && overlay_supports_window(overlay) && crate::region::same_aspect(display_aspect, aspect);
+    // than 100 % (the GPU magnifies beyond that), for each side that is shown. Mask overlays and
+    // soft proofing use the same window; diagnostic overlays that need whole-frame context do not.
+    let windowable_view = hover_key.is_none() && overlay_supports_window(overlay) && crate::region::same_aspect(display_aspect, aspect);
     let window_ctx = WindowCtx {
         id,
         overlay,
diff --git a/crates/ui-egui/src/region.rs b/crates/ui-egui/src/region.rs
index b242d30be..953008e0d 100644
--- a/crates/ui-egui/src/region.rs
+++ b/crates/ui-egui/src/region.rs
@@ -77,8 +77,8 @@ pub struct ViewSizes {
     /// `None`: no drag.
     pub draft: Option<f32>,
     /// Whether a window can be drawn over the whole-frame render (not with a diagnostic overlay
-    /// that needs whole-frame context, soft proofing, or a read too far for one window). Mask
-    /// overlays are window-safe.
+    /// that needs whole-frame context, or a read too far for one window). Mask
+    /// overlays and soft proofing are window-safe.
     pub windows: bool,
 }
 
@@ -225,7 +225,7 @@ pub fn plan(settings: &crate::state::AppSettings, v: ViewSizes) -> LoupePlan {
     let drawn_long = if v.drawn_long.is_finite() { v.drawn_long.max(8.0) } else { canvas_long };
     let scale = v.draft.map_or(1.0, |s| if s.is_finite() { s.clamp(0.1, 1.0) } else { 1.0 });
     if !v.windows {
-        // the whole frame at the size it is drawn (soft proof or a whole-frame diagnostic has no
+        // the whole frame at the size it is drawn (a whole-frame diagnostic has no
         // window); while a slider drags, a canvas-sized draft as with windows: a drag at 1:1 must
         // not render the photo's every pixel each frame
         let long = if v.draft.is_some() { drawn_long.min(canvas_long) } else { drawn_long };
@@ -422,7 +422,7 @@ mod tests {
         assert_eq!(drag.window_edge, still.window_edge);
     }
 
-    // Given no window (a whole-frame diagnostic or soft proofing), the view at rest renders the
+    // Given no window (a whole-frame diagnostic), the view at rest renders the
     // whole frame at the size it is drawn, but a slider drag drafts at the canvas size like the
     // windowed path
     #[test]
@@ -626,7 +626,7 @@ mod tests {
         }
     }
 
-    // Given a mode whose picture can't be a window (soft proofing, a whole-frame diagnostic, or a
+    // Given a mode whose picture can't be a window (a whole-frame diagnostic, or a
     // spot read too far for one render), the whole-frame render is as big as the view
     #[test]
     fn without_windows_the_whole_frame_is_rendered_at_the_drawn_size() {
diff --git a/crates/ui-egui/src/tests_preview_limit.rs b/crates/ui-egui/src/tests_preview_limit.rs
index dfddb9c4f..070921256 100644
--- a/crates/ui-egui/src/tests_preview_limit.rs
+++ b/crates/ui-egui/src/tests_preview_limit.rs
@@ -261,6 +261,29 @@ mod in_the_loupe {
         assert!(w >= 1300 && hh >= 700, "it covers the canvas: {w}×{hh}");
     }
 
+    // Given soft proofing on at 1:1 (LR-VIEW-ZOOM), the view is a canvas-sized frame plus a
+    // native window carrying the proof, as without it, instead of a whole-frame render at the
+    // photo's full size; switching the proof off swaps the window back to the plain look
+    #[test]
+    fn soft_proofing_at_one_to_one_is_a_window() {
+        let (mut h, native) = detail();
+        h.request("engine.execute", json!({"command": "view.zoom100"}), T);
+        h.settle(SETTLE);
+        let plain = h.app.region_view.expect("a window render without the proof");
+        let r = h.request("engine.execute", json!({"command": "view.softProof", "params": {"on": true, "destWarning": true}}), T);
+        assert_eq!(r["ok"], true, "{r}");
+        h.settle(SETTLE);
+        assert!(h.app.ui.soft_proof);
+        let main = rendered_long_edge(&h);
+        assert!(main <= 2560 && main < native, "native {native}: the whole-frame render is {main}");
+        let proofed = h.app.region_view.expect("a window render with the proof");
+        assert_eq!(proofed.full.0.max(proofed.full.1), native, "cut from the photo's own pixels");
+        assert_ne!(proofed.settings, plain.settings, "the proof is part of the window's look");
+        h.request("engine.execute", json!({"command": "view.softProof", "params": {"on": false}}), T);
+        h.settle(SETTLE);
+        assert_eq!(h.app.region_view.map(|r| r.settings), Some(plain.settings), "back to the plain window");
+    }
+
     // Given the user capped the preview at 1600 px, then 1:1 is rendered no larger than that
     #[test]
     fn an_explicit_limit_still_caps_one_to_one() {
@@ -646,10 +669,16 @@ mod in_the_loupe {
     #[test]
     #[ignore = "a measurement, not a check"]
     fn profile_slider_drag() {
-        for (size, percent) in
-            [([1400.0, 900.0], 0.0), ([1400.0, 900.0], 100.0), ([1400.0, 900.0], 400.0), ([2800.0, 1700.0], 100.0), ([2800.0, 1700.0], 400.0)]
-        {
+        for (size, percent, proof) in [
+            ([1400.0, 900.0], 0.0, false),
+            ([1400.0, 900.0], 100.0, false),
+            ([1400.0, 900.0], 400.0, false),
+            ([2800.0, 1700.0], 100.0, false),
+            ([2800.0, 1700.0], 400.0, false),
+            ([1400.0, 900.0], 100.0, true),
+        ] {
             let (mut h, native) = detail_in(size);
+            h.app.ui.soft_proof = proof;
             h.app.ui.zoom = if percent == 0.0 { crate::state::Zoom::Fit } else { crate::state::Zoom::Percent(percent) };
             h.settle(SETTLE);
             h.app.session.begin_interaction("Exposure").unwrap();
@@ -673,9 +702,10 @@ mod in_the_loupe {
             ms.sort_by(|a, b| a.total_cmp(b));
             let (w, hh) = region_tile(&h).unwrap_or((0, 0));
             eprintln!(
-                "PROFILE {}x{} canvas, {percent:>4}% (0 = fit) of {native} px: tick→shown median {:.1} ms, p90 {:.1} ms, max {:.1} ms; whole-frame render {} px, window tile {w}×{hh}",
+                "PROFILE {}x{} canvas, {percent:>4}% (0 = fit){} of {native} px: tick→shown median {:.1} ms, p90 {:.1} ms, max {:.1} ms; whole-frame render {} px, window tile {w}×{hh}",
                 size[0],
                 size[1],
+                if proof { ", soft proof" } else { "" },
                 ms[ms.len() / 2],
                 ms[ms.len() * 9 / 10],
                 ms[ms.len() - 1],
@@ -819,18 +849,6 @@ mod in_the_loupe {
         assert_eq!(h.app.region_tiles.len(), 2, "{:?}", h.app.region_tiles.keys().collect::<Vec<_>>());
     }
 
-    // Given soft proofing at 1:1 (a mode that draws one image, so no window), the whole-frame
-    // render is as big as the view, not a canvas-sized upscale
-    #[test]
-    fn soft_proofing_at_one_to_one_renders_the_whole_frame_at_full_size() {
-        let (mut h, native) = detail();
-        h.app.ui.soft_proof = true;
-        h.request("engine.execute", json!({"command": "view.zoom100"}), T);
-        h.settle(SETTLE);
-        assert_eq!(h.app.region_view, None);
-        assert_eq!(rendered_long_edge(&h), native);
-    }
-
     // Given a window the photo's spots make too big for one render (refused), the loupe falls back
     // to the whole frame at the drawn size, and tries windows again when the look changes
     #[test]
diff --git a/docs/gaps.md b/docs/gaps.md
index 1fa9d705d..3bd61c294 100644
--- a/docs/gaps.md
+++ b/docs/gaps.md
@@ -18,7 +18,7 @@ Hours are Opus 5.5 agent wall-clock hours (see [target-app-parity.md](target-app
 | 1 B | **Camera colour is estimated, not measured** (LR-PROF-CAMERACOLOR). Raws start from a per-file fit to the camera JPEG, 5 bundled profiles or 52 spectral matrices; others use a neutral matrix (a black-and-white camera JPEG gives its tone curve only, #577; dark JPEG tones also guide the toe, #1307) | ΔE vs camera JPEG 2.9–10 after fitting (raw-parity.md); 0 chart-calibrated models vs 1,500 Adobe Standard profiles | every raw shooter sees colour that is close but not Lightroom's; some get neutral-matrix colour | 50–90 | [raw-parity](raw-parity.md#colour) |
 | 2 B | **CR3 beyond four verified bodies** (LR-IMP-FORMATS): `crx.rs` refuses vertical-tile C-RAW, partial-subband and other plane coding, multi-tile 0x200 and adaptive low-frequency gains, and those files fall back to the preview; CR3 sRAW / mRAW and HEIF CR3 too. Verified: M50 / R100 / R8 / R5 Mark II. EOS R7 C-RAW (odd-height QP map) decodes since a45c2b66 and R6 Mark II C-RAW is reported working on 0.6.0, but neither is verified yet (no corpus test or recorded check) | measured: Canon is the largest bucket (33% of photographers) and its share whose camera fully works is the lowest of the big four, 73%; 18 of 22 popular Canon bodies write CR3, 4 verified (raw-parity.md); R7 rule checked only on paired RAW / C-RAW captures ([cr3.md](cr3.md)) | +6.2 points of photographers whose camera works, the biggest single raw gain | 25–45 | [raw-parity](raw-parity.md#remaining-work-ranked-by-photographers-unlocked-per-hour) |
 | 3 B | **Render fidelity unmeasured** (LR-BEHAV-RENDER-FIDELITY): tone, highlights, texture / clarity / dehaze, NR and sharpening are tuned by eye | no side-by-side suite; a retune needs a new process version (only V1 exists) | edits imported from Lightroom look different; the look "isn't Lightroom's" | 30–60 | [raw-parity](raw-parity.md), [process-versions](process-versions.md) |
-| 4 B | **Original-pixel detail at 100%** (LR-VIEW-ZOOM, P0): Detail, Before/After, Compare and Reference are sharp at 1:1–8:1 (zoom windows, #323). Mask overlays and soft proofing still render the whole frame at the drawn size instead of a window: sharp up to the GPU texture side, but a full-size render per edit at 1:1 | checklist row 🟡 | editing with a mask overlay or soft proof on at 1:1 is slow on large files, and blurry past the texture side | 4–8 | [ui-parity](ui-parity.md) |
+| 4 | ~~Original-pixel detail at 100%~~ (LR-VIEW-ZOOM): closed 2026-10-12, see Closed recently | checklist row ✅ | — | 0 | [ui-parity](ui-parity.md) |
 | 5 B | **AI masks: Subject, Sky, People, Background, Landscape** are heuristics; Object / Describe run SAM 3 but its download has no mirror configured | `crates/segment`; ai-masks.md; MASK rows 67% | masking is how most Lightroom users edit today | 40–70 + model decision | [target-app-parity](target-app-parity.md#ai) |
 | 6 B | **Compressed ORF** (every Olympus / OM body since ~2008) | measured: 66 of 78 Olympus / OM models preview-only; `orf.rs` refuses; OM bucket 3% of photographers, 3% of them fully work | +2.6 points; best gain per hour of any raw gap (~0.2) | 8–15 | [raw-parity](raw-parity.md#compressed-orf-a-clean-room-starting-point) (the two Olympus patents that describe the method) |
 | 7 | **Nikon High Efficiency NEF** (HE / HE★ on Z 8, Z 9, Z 6III, Z f, Z50II, Z5II when set to HE) | named and preview only (#193); needs a pure-Rust JPEG XS decoder | +1.1 points (only shots taken in HE; lossless works on the same bodies); low gain per hour (~0.04), so not a beta blocker | 20–40 | [raw-parity](raw-parity.md) |
@@ -60,6 +60,8 @@ Beta-blocking gaps (B) total **~230–430 h**; with the stability and platform w
 
 ## Closed recently (kept for calibration)
 
+- 2026-10-12: original-pixel detail at 100% (gap 4, LR-VIEW-ZOOM, #323): soft proofing gets zoom windows like
+  Detail, Before/After, Compare, Reference and mask overlays; 1:1 tested equal to the full-size export.
 - 2026-10-11: Sony ILCA-99M2 profile.
 - 2026-10-11: Nikon edge bars on bodies without CropArea (#976, #914): 36 model / size / depth groups measured.
 - 2026-10-10: Nikon lossy-after-split NEF (#633), ORF 12.8 bpp (#651), HDR gain map / PQ AVIF / float
@@ -80,6 +82,7 @@ Beta-blocking gaps (B) total **~230–430 h**; with the stability and platform w
 | 2026-10-14 | minor | Gap 21: de, es, fr, ja and uk catalogs filled (97–99%); 55–90 h → 40–70 h |
 | 2026-10-14 | minor | Gap 10: 17 uncompressed Olympus and Samsung models verified on one CC0 file each (254 models) |
 | 2026-10-14 | minor | Gap 29 added: the HDR merge, tested on real 67 MP brackets, keeps moving clipped highlights and merges in batches now; AI Denoise on merged DNGs, JPEG response curves and per-frame memory remain |
+| 2026-10-14 | minor | Gap 4 closed: soft proofing at 1:1 renders a window (mask overlays already did); 1:1 pixels tested equal to export; only diagnostic overlays (Visualize Spots, Point Color range, HDR range) stay whole-frame |
 | 2026-10-13 | patch | Gaps 10, 22 and 23 each had two rows (two branches edited them; the merges kept both): the older copies are removed, and `cargo xtask parity` (in `ci`) now fails when a gap number is listed twice |
 | 2026-10-13 | minor | Closed recently: Sony ILCA-99M2 profile |
 | 2026-10-14 | minor | Gap 10: 21 Sony and Leica models verified on one CC0 file each (258 models) |
diff --git a/docs/parity-checklist.md b/docs/parity-checklist.md
index 043aebb50..0db3c6262 100644
--- a/docs/parity-checklist.md
+++ b/docs/parity-checklist.md
@@ -812,6 +812,7 @@ Latin stays upright. Advanced Japanese composition remains open. Regression cove
 | 2026-10-14 | minor | LR-IMP-CAMERA-COVERAGE: Nikon D600, D610 and D5000 NEFs drop the filler and noise columns on the right (a band), and the D600 and D610 measure their white level inside the image (#1627); status unchanged (🟡) |
 | 2026-10-15 | patch | LR-VIEW-BEFOREAFTER: side by side and stacked panes share click-zoom animation for whole-frame images, window tiles and input; render requests keep the final target (#1821); status unchanged |
 | 2026-10-14 | patch | LR-MASK-BRUSH: ⇧-click with the brush draws a straight stroke from where the last stroke ended, tapering when the brush changed in between (`cmd:mask.brushStroke` `connect`); status unchanged |
+| 2026-10-14 | minor | LR-VIEW-ZOOM 🟡 → ✅: soft proofing gets zoom windows (mask overlays already had them; row text corrected), 1:1 pixels tested equal to the full-size export, slider-drag latency at 1:1 measured on Linux |
 | 2026-10-13 | patch | LR-VIEW-COMPARE, LRC-LIB-COMPARE: ⌘/Ctrl- and Shift-clicks on the filmstrip in Compare change the selection as in the grid; the two photos shown stay selected and the active photo is unchanged (#187); statuses unchanged |
 | 2026-10-13 | minor | KEYC-SELECTONLYACTIVE: Edit → Select Only Active Photo (⌘⇧D) lands as a new row; status 🟡 (headless-tested, native macOS menu path unverified) |
 | 2026-10-13 | patch | LRC-LIB-CATALOG-IMPORT: a catalog whose tables are all still in its `-wal` file (main file schema format 0) opens instead of failing as an unsupported schema format; status unchanged |
diff --git a/docs/target-app-parity.md b/docs/target-app-parity.md
index 07ae89fd7..557f37a6a 100644
--- a/docs/target-app-parity.md
+++ b/docs/target-app-parity.md
@@ -157,7 +157,7 @@ essentials replaces its guessed 45% raw look with 84% × 0.75 (+1 point).
 
 | Discount | Mainstream | Essentials | Evidence |
 |---|---:|---:|---|
-| Interaction fidelity / UI clarity | ×0.95 | ×0.95 | brush live preview gaps and feathering (#517, #518); crop artefacts and limits (#731, #742, #770, #771); Compare white-balance selector (#743–#745); low-res slider previews (#330); dialogs taller than the window (#781); mouse-wheel scroll (#714); 100% zoom capped (gap 4) |
+| Interaction fidelity / UI clarity | ×0.95 | ×0.95 | brush live preview gaps and feathering (#517, #518); crop artefacts and limits (#731, #742, #770, #771); Compare white-balance selector (#743–#745); low-res slider previews (#330); dialogs taller than the window (#781); mouse-wheel scroll (#714) |
 | Stability on real machines | ×0.92 | ×0.92 | launch and quit crashes or silent startup failures (#250 macOS 12 Intel, #315 Windows Intel GPU, #431 Intel Mac, #260, #462, #783 Linux / Nouveau, #790 Windows installer); tests run in CI on FreeBSD only (gap 9) |
 | File exchange with Lightroom users (mainstream) / opening files people send (essentials) | ×0.92 | ×0.95 | Adobe XMP sidecars change colours (#290), XMP CameraProfile and Upright ignored (#204, #205), catalog import fails (#599), same Light values render differently (#523, #209), no write-back to `.lrcat`; for casual users: CR3 beyond verified bodies, compressed ORF, HE NEF open as previews (#473, #485, #556), HEIC trouble (#693) |
 | **Product** | **×0.804** | **×0.830** | |
@@ -235,13 +235,13 @@ open as previews, and colour on every non-DNG raw is estimated rather than measu
 
 **Alpha gate:** all six core workflows (import, cull, develop, save and reopen, batch looks, export)
 work end to end on macOS; see the table in [roadmap.md](roadmap.md#alpha-gate). The partial rows (CR3
-variants, compressed ORF, HE NEF as previews; capped 100% zoom) degrade but don't block a workflow.
+variants, compressed ORF, HE NEF as previews) degrade but don't block a workflow.
 
-**To beta:** ~14 points of readiness, **~340–600 h**: the nine `B` gaps in [gaps.md](gaps.md)
+**To beta:** ~14 points of readiness, **~340–600 h**: the `B` gaps in [gaps.md](gaps.md)
 (~230–430 h), plus burning down the ~92 open core-path user issues (40–80 h) and stability work
 around them. Beta needs: CR3, ORF and
 HE NEF decoding from sensor data; measured colour for the popular models and a fidelity suite with
-scores; full-resolution zoom; Subject / Sky / People masks; a lens profile database for the
+scores; Subject / Sky / People masks; a lens profile database for the
 popular lenses; tests in CI on macOS, Windows and Linux.
 
 ## How hours are calibrated
@@ -303,6 +303,7 @@ P2 items 12–25 h; look tuning, performance, packaging, hardening 25–45 h.
 | 2026-10-14 | minor | Localization ~52% → ~55% (mean of the 12 key languages): de, es, fr, ja and uk catalogs filled |
 | 2026-10-12 | minor | UI/UX remaining 34–58 h → 32–54 h (canvas tools by keyboard); headline numbers unchanged |
 | 2026-10-14 | minor | Merge and HDR: merge 100% → 88% (LR-MERGE-HDR ✅ → 🟡 after testing on real 67 MP brackets; gap 29); area percentages unchanged |
+| 2026-10-14 | minor | Gap 4 (full-resolution zoom) closed: removed from the interaction-fidelity notes, the alpha-gate partial rows and the beta needs; headline numbers and the ×0.95 interaction factor unchanged |
 | 2026-10-12 | minor | UI/UX remaining 40–70 h → 34–58 h (accessibility audit landed, ui-parity.md); percentages unchanged |
 | 2026-10-14 | minor | Camera-coverage sweep (#1327): 30 models verified on 40 CC0 files (`coverage_sweep_corpus.rs`, `coverage-sweep.tsv`; `docs/raw/coverage.py` 515 → 545 verified rows); the counts above are not re-measured here. Sensor verification does not establish colour parity (#1277); headline numbers unchanged |
 | 2026-10-12 | minor | 22 more CR3 bodies verified on one CC0 file each (raw-parity.md): use-weighted share whose camera decodes 85% → 92%, RAW decoding 77% → 83%; full 60.7% → 61.2% (~61%), mainstream depth 72.1% → 74.0% (~49% → ~51%), essentials depth 77.5% → 78.0% (~64% → ~65%) |
diff --git a/docs/ui-parity.md b/docs/ui-parity.md
index e0bf171ba..e5e0c6e5f 100644
--- a/docs/ui-parity.md
+++ b/docs/ui-parity.md
@@ -1,6 +1,6 @@
 # UI parity
 
-> **Last reviewed:** 2026-10-10 · **Last updated:** 2026-10-12 · **Change:** minor (canvas tools by keyboard; focus-order test; ~50 unnamed hand-drawn widgets named) · **Target:** Adobe Lightroom Classic 15.5 / Lightroom 9.6
+> **Last reviewed:** 2026-10-10 · **Last updated:** 2026-10-14 · **Change:** minor (zoom row: soft proofing sharp at 1:1; canvas tools by keyboard; focus-order test; ~50 unnamed hand-drawn widgets named) · **Target:** Adobe Lightroom Classic 15.5 / Lightroom 9.6
 
 How LightCraft's interaction compares with Lightroom's: views, panels, tools and handles, modifier
 keys, shortcuts, zoom, context menus and feel. Rows for individual menu items and shortcuts live in
@@ -27,7 +27,7 @@ assessment and the notes that don't fit a row. Visual reference for agents: `pla
 | Crop, Upright, transform handles | handles, aspect lock, angle, overlays (O / ⇧O), guided Upright | ✅ all of them; overlay cycle on ⇧O | 85% | crop opacity of cut-out area (15.5) |
 | Masking tools | brush (A / B, flow, density, auto mask, pressure), linear / radial, range masks, AI masks, pins | brush, linear, radial, colour / luminance range, add / subtract / intersect / invert, masks panel, pins | 60% | brush live preview (#517 / #518), no pen pressure, no depth range, AI masks, Feather / Edge controls (15.5) |
 | Remove / heal | content-aware remove, heal, clone, visualize spots | heal, clone, auto source, visualize spots, red / pet eye | 65% | content-aware (patch synthesis) remove, generative remove, detect distractions |
-| Zoom and navigation | Fit / Fill / 1:1 / 2:1 … 11:1, Navigator, spacebar pan | Fit to 800%, pinch zoom, pan, Navigator-like | 60% | **original-pixel detail at 100% with a mask overlay or soft proofing** (no zoom window there; Detail, Before/After, Compare and Reference have them; P0 gap LR-VIEW-ZOOM) |
+| Zoom and navigation | Fit / Fill / 1:1 / 2:1 … 11:1, Navigator, spacebar pan | Fit to 800%, pinch zoom, pan, Navigator-like | 60% | diagnostic overlays (Visualize Spots, Point Color range, HDR range) render the whole frame instead of a zoom window; everything else, mask overlays and soft proofing included, is sharp at 1:1–8:1 |
 | Library culling and metadata | ratings, flags, labels with keys, painter, keywording, metadata panel, filter bar | ratings / flags / labels and advance, keyword list and sets, keywording box, metadata fields, smart albums, filter bar | 85% | painter tool, quick collection, some Classic keys |
 | Shortcuts | Lightroom desktop and Classic keymaps | Lightroom-desktop keymap with Classic additions, editable | 85% | deliberate differences below; no Classic keymap layer |
 | Context menus | everywhere | 26 sites (grid, filmstrip, albums, folders, keywords, masks, presets…) | 70% | unaudited against Lightroom's menus item by item |
@@ -136,6 +136,7 @@ a real screen reader on any platform yet.
 | 2026-10-13 | minor | Library culling: backtick cycles the flag; ⌘↑ / ⌘↓ raise / lower flag status |
 | 2026-10-13 | minor | KEYC-COMPARE: `=` / `-` grid thumbnail size keys noted in the shortcuts section |
 | 2026-10-13 | minor | Shortcuts: Home / End and Shift+Left / Right in the two library grids, Cmd+O (Open Library…) and Shift+H (Visualize HDR) get bindings and keyboard tests; percentages and the dated counts above are unchanged |
+| 2026-10-14 | minor | Zoom row: soft proofing gets zoom windows; LR-VIEW-ZOOM gap closed |
 | 2026-10-12 | minor | Accessibility: an AccessKit tree audit test over every view, Develop tool and 13 dialogs; unnamed fields, slider ranges and the Auto Stack slider fixed; 30% → 55%; UI remaining 40–70 h → 34–58 h |
 | 2026-10-11 | minor | Zoom row: Compare and Reference get zoom windows; what remains of LR-VIEW-ZOOM is overlays and soft proofing |
 | 2026-10-10 | major | Created: measured command and shortcut counts, assessment by area, estimates; moved the shortcut conflicts section here from the checklist |