Reported: pegging the TRAX fader changes nothing audible, the ladder never reaches red, and the
ladder does not scale with the fader.
Cause (two defects in the same setters):
- UserGain wrote the OUTPUT DEVICE's volume. NAudio's WaveOut ignores that, so the fader was inert.
- LocalGain (the desktop-mute follow) recomputed the sample-chain volume WITHOUT the user gain, so
every duck/mute silently discarded whatever the creator had set.
Fix: one `EffectiveVolume(user, local)` = MusicVolume × user × local, applied to the sample chain by
both setters. The meter sits AFTER that stage, so the ladder now scales with the fader and can reach
the peak band on a loud track — exactly what pegging should look like. Max user gain is +6 dB, so
"pegged" means the file at unity, not a magic boost.
Guard: MusicLevelTests pins the arithmetic (default half, pegged double, mute multiplies to zero),
that the user gain survives the mute follow, and that nothing may drive the output device's volume
again — the no-op that made this invisible.
Reported: the TRAX bar moves, but its sampling looks far slower than the mic meter.
Cause: the TRAX number was a raw instantaneous peak per metering block — a stutter sampled by the UI
timer, with no smoothing. The mic looks fluid because it is smoothed at audio-callback rate.
Fix: the player now smooths its measured level with the SAME curve the mic uses
(MeterScale.NextLevel: fast attack, gentle decay), and the dock's meter clock runs at ~30 Hz instead
of 20. Two ladders that move alike, because they are the same ballistics.
Guard: MeterScaleTests pins the attack/decay behaviour alongside peak-hold.
Creator: "why is there no meter for when trax is playing?"
Routing answer: TRAX rides the desktop loopback, so it already moved the GAME ladder — there was no
separate TRAX signal in the mix. But that ladder cannot answer the question a deaf creator actually
has: "is the music producing sound at all?" The play glyph cannot either — an open file with a dead
decoder still reports Playing.
So the player now measures its own output: \`MusicPlayer\` wraps the volume stage in an NAudio
\`MeteringSampleProvider\` and exposes \`Level\`; \`ITraxPlayer.Level\` keeps the fake honest. The VM maps
it through the same display scale (\`TraxMeterFraction\`), zeroed when not playing, and the TRAX bay
carries a 24-segment ladder above its fader.
This reverses my earlier "no ladder, nothing honest to show" — the honest thing arrived with a real
measurement. TraxBayTests flipped accordingly (it now asserts the ladder AND its source).
Unit 9, the one that matters most for this creator.
MASTER STRIP (top, nearest the preview):
- one ladder fed by the MEASURED mix — `AudioMixer` now measures the buffer the encoder receives
(`MasterLevel`, `MasterImbalanceDb`), so the master cannot show a source instead of the output
- `-12.4 dB` readout, and ONE L/R flag that appears above a 6 dB gap instead of a second ladder
- bitrate / FPS / dropped / duration from the existing health poll
MONO OUTPUT:
- `AudioFold.ToMono` averages each stereo pair into both channels, applied after the limiter and
the sync delay, so the stream AND the recording get what was folded
- `MonoOutput` is persisted (`Audio.MonoOutput`) and restored at startup: an accessibility switch
that resets each session is an accessibility failure
- one click in the strip, remembered forever
Guard: AudioFoldTests (hard-left reaches the right at half amplitude, hard-right — the inaudible
case — reaches the left, centred material keeps its level, mono buffers untouched) and
MasterStripTests (the strip binds the measured mix + the switch, the store round-trips it, and the
fold is the tested helper rather than new arithmetic). 33 audio tests green.
Unit 7. The bay shows the track name (left-click toggles playback, right-click opens the picker —
the label stays the control), a play/pause button, the duck badge, and a dB level.
No ladder here, deliberately: nothing measures the music bed, and a decorative ladder would break
the measured-not-painted rule. A test asserts the bay has no SegmentMeter.
The level is real: `ITraxPlayer.UserGain` is a new creator gain multiplied onto the fixed design
volume, separate from `LocalGain` (which follows the desktop mute). The VM's `TraxVolumeDb` is the
same dB scale as the other faders, so unity means the bed's design level.
Unit 6. Mirrors the mic bay for the desktop/game bed, plus the badge that matters to a creator who
cannot hear the duck: it shows while the ducker is actually attenuating, and it prints the gain the
mixer applied (−12.0 DUCK by design).
- `AudioMixer.DuckGain` exposes `AutoDucker.CurrentGain` instead of leaving it internal.
- VM: `GameVolumeDb` / `GameVolumeDbLabel` (fader face), `GameMeterFraction` (measured level),
`IsGameDucking` + `GameDuckLabel` refreshed on the level tick — the same moment the ducker runs,
so the badge cannot lag behind the mix.
- The bay: header chip + source, ladder, mute dot (uses the meter band, so mute reads at a glance),
dB readout, fader.
Guard: GameBayTests — the dock binds all four, `AutoDucker.DuckGain` is −12.04 dB and the label is
produced through the same dB scale as the faders, and the mixer property exists so the badge cannot
become a hardcoded number.
`AudioLevelMeter` measured one number for everything. It now also keeps smoothed per-channel
levels with the same attack/decay, plus `ImbalanceDb`. Mono samples report both sides equal (a mono
source is centred), so the two never disagree about whether audio exists.
The dock shows ONE ladder per source — two ladders per bay was visual noise for a rare failure —
but this creator is profoundly deaf on the right, so the one case that matters is "the sound is
panned away from me". That is now measurable and will drive a flag, not a second ladder.
Semantics that mattered: a silent side is compared at the −60 dBFS floor, so a fully one-sided
signal reads ≈54 dB rather than "nothing to compare". My first version returned 0 for exactly the
case the feature exists for; the test caught it.
Guard: StereoMeterTests — hard left, hard right (the inaudible case), balanced stereo, mono
centred, and silence-is-not-an-imbalance. 31 audio tests green.
The state colours the view-models hand to bindings — top bar, REC/ON-AIR dots, health banner,
meter bands, toast tints — were hex literals in C#. They now resolve from Themes/Tokens.xaml
through `Helpers/ThemeBrush`, so the values live in one place. Without an Application (unit tests)
the helper returns a neutral grey: visibly wrong rather than silently plausible.
Four tokens the code needed and the document lacked, added to DESIGN.md (so the doc-is-the-contract
guard demanded their brushes): info-container #3a3b52, success-container #1d5c38,
warning-container #b8860b, test-container #9c6f1c (the test-stream gold bar).
Also fixed by this: the top bar still painted the RETIRED brand red #e94560 when live — the
standard's #ff5451 now, and `OnAirSignTests` (which pinned the old value) says so. BarOn/BarOff
became properties rather than static fields, because a static field would resolve tokens before
the Application exists.
Verified: DesignTokensTests, GameMeterHonestyTests, NotificationAreaIntegrationTests,
OnAirSignTests, SessionTeardownTests — 38 tests green.
The broadcast picture is part of the app, so it stops carrying its own colours. Alert cards, the
ticker and the branding flash now paint from OutputPalette, which Mirrors DESIGN.md for the brand
tokens and owns the alert box's six-event accent language (gold/violet/pink — content, not
chrome; the standard has no purple and should not grow one, but they live in ONE file instead of
57 scattered literals).
It is a separate type because the renderers run on the frame-pump thread and must not touch
application resources. The duplication is made safe by OutputPaletteTests: it parses
Themes/Tokens.xaml and fails if any mirrored token drifts, and fails if a renderer reintroduces a
colour literal or "Segoe UI".
Ticker and flash glyphs now use the bundled Inter instead of Segoe UI. Neon red is the standard
live red — one red in the product, including the credit.
Fixed on the way: the palette's font fields were static pack-URI initialisers, which throw
TypeInitializationException in any host without a WPF Application (the test process caught it);
they are lazy now with a name fallback.
In the Socials drawer, under the show/hide switch: text colour, an optional background plate
with its own colour and opacity, and a text drop shadow. The bar had been hard-wired
white-on-transparent since it shipped.
- `SocialsConfig` gains five additive properties (BarTextColor, BarBackgroundEnabled,
BarBackgroundColor, BarBackgroundOpacity, BarTextShadow).
- Persistence: five new Socials columns, added by the per-column idempotent migration pattern
(a half-migrated DB heals instead of failing), written in Save and read back in Load with
defaults for pre-existing rows.
- `SocialBarRenderer.Render(entries, SocialBarStyle?)` — the look is a separate value type so the
renderer stays a pure "entries + look" function and the preview can never read a half-applied
config mid-edit. A malformed hex falls back to white/black instead of throwing inside a render
tick.
- The plate hugs the entries (rounded 20px, padding) so it reads as a caption bar, not a band
across the frame. Off by default, because transparent is what creators have today.
- The colour pickers reuse `BorderColorPalette` — one catalogue of the app's colours, per the
TASK 60/63 rule.
Guard: SocialBarAppearanceTests paints the real strip and inspects pixels — default is white with
no plate, a red text colour reaches the glyphs, a 55% blue plate yields half-transparent blue
pixels, and the shadow adds dark pixels that were not there before.
- Source.IsBrandLogo persisted (additive migration); layer healed on load, removal blocked
- Image stored as an asset; settings keep the asset id and file name
- App Settings: Brand Logo file field with Browse
- Default placement top-left; per-scene eye uses the existing element visibility
Test: BrandLogoTests; layout/background tests pass (34). Not pushed.
The transition started with no from-frame, so BlendFrame returned the new scene at once.
Start now requests a capture and the frame pump supplies its last output on the next tick.
Test: TransitionFadeTests. Docs: ai.md, TASKS.md.
Not pushed.
Unit 2 of slimming the mandatory read. Verbatim move again - the four capture
blocks leave ai.md unchanged in content, and one stub carries the law.
- docs/ai/capture.md: 'Screen backdrop capture', 'Webcam capture', 'Webcam
background effects' and 'Vertical (9:16) tier camera placement', byte-
identical, grouped under backdrop / webcam / vertical headings, with a
'Related in ai.md' footer.
- ai.md stub keeps: the locked-Background model and NormalizeBackgrounds;
ActiveBackgroundImage is fed by static art only (2026-08-23 regression);
window:/picker: keys are session-scoped and a live pin beats auto-detect;
the settled WinRT interop paths (TryGetDataUnsafe, from-DXGIDevice); the
VideoFrame seam, own-catch subtype ladder and no fallback camera; one app-
level webcam identity with live-lock attainability; CameraManager as the one
effects processing point with per-session mask smoothing; geometry-only
vertical tier and the model-level rect lock.
- docs/ai/index.md gains the capture.md row; Services/index.md (2) and
TASKS/task-38-window-backdrop.md repointed.
ai.md 1,975 -> 1,558 lines. No source touched; no push.
ai.md is the mandatory read every session; at 2,702 lines / ~64k tokens it
costs the whole subsystem history even when the task never touches it. This
is a verbatim move, not a compression - nothing is deleted and each stub is
written from the moved text.
- docs/ai/frame-pipeline.md: 'Frame cadence', 'Per-element composite timing +
scene dump', 'What the first instrumented take said', 'Render budget',
'Scene compositor' and 'Live frame pipeline', byte-identical, plus a short
'Related in ai.md' footer and two stale 'below' pointers corrected.
- ai.md keeps two stubs carrying the invariants that must not be missed
(overlay on every render path, VideoFrame.Epoch, StripPool threads,
Release-only render numbers; arrival-stamped rawvideo, non-blocking submit,
stop order, burned _outputIndex judge) and links the new file.
- docs/ai/index.md + docs/index.md route the split; schema.md inventory lists
the new tier.
- 4 index-row references and 2 TASKS.md anchors repointed; ai.md's Services
row too.
Measured: ai.md 63.6k -> 47.1k tokens, 2,702 -> 1,975 lines. Progressive
disclosure per schema.md convention 5. No source touched; no push.
The vertical tier is a centred band crop of the SAME 1920x1080 master (band x
656.25, w 607.5, h 1080), so the 16:9 camera placement (x~1408) falls outside the
cropped window and the creator is simply not in the vertical stream. Each scene
now keeps its own vertical position, size and shape; switching back to 16:9
restores the desktop layout exactly (the creator's locked table, 2026-10-17).
Geometry only - NO render-path change. The preview and the compositor both read
the element's own X/Y/W/H + ClipShape, so the tier switch swaps the LIVE geometry:
- Models/VerticalCameraLayout.cs (new): the pure math - the band, its two 607x540
areas, the six spots (top/bottom x left/centre/right), the 32px inset and the
270x152 default tile (480x270 in the 1080x1920 output) in the bottom-right spot,
the corner the 16:9 default webcam uses.
- Models/WebcamSceneConfig.cs: the vertical placement lives in seven additive
columns (VerticalX/Y/Width/Height/ClipShape/RectWidth/RectHeight; NULL = "this
scene has no vertical placement yet"). The 16:9 layout stays in the old columns
and is parked in the transient Landscape* fields while the vertical tier is
live; Begin/EndVerticalLayout + CommitLiveLayoutToVertical do the swap. The
X/Y/W/H overrides clamp every write into the band - outside it the camera is
invisible in the vertical output.
- Models/SceneElement.cs: X/Y/Width/Height became virtual so the derived element
constrains every writer (drag, resize handle, typed fields) in ONE choke point
instead of three writers each re-implementing the rule.
- ViewModels/MainViewModel.Webcam.cs + .Streaming.cs: IsVerticalTier/VerticalBand,
ResolveVerticalCameraLayout(s) (the tier swap), MoveWebcamInVerticalTier /
SnapWebcamInVerticalTier (the six-spot snap, by the tile's centre),
EnsureVerticalCameraLayout; ApplyStreamQuality is the tier-change hook. The
Live scene gets its small tile automatically on the first vertical switch (the
vertical stream would otherwise have no camera at all); the other scenes keep
their 16:9 placement until the creator drags them into the band.
- Controls/PreviewPane.xaml.cs: the drag/resize route through the VM snap under
the vertical tier instead of free-floating over the master.
Controls/LeftPanel.xaml: the webcam properties gained a Geometry section
(X/Y/W/H) for the manual entry the creator asked for, bound to the LIVE
placement, gated on the VM's IsWebcamSelected (NOT SelectedElement.IsWebcam -
the TASK 51 test finds that panel by binding text and would match twice).
- The store: seven guarded ADD COLUMNs, the INSERT/SELECT extended, and the
X/Y/W/H/ClipShape columns stay the 16:9 layout - while the vertical tier is
live the save writes the parked values there and the live ones into Vertical*.
Test: ytLive.Tests/VerticalWebcamLayoutTests.cs - the ONE Good Dog test. The
vertical tier brings the Live camera INTO the band as a small tile; dragging it to
the top-right lands it top-right inside the band; back to 16:9 the desktop layout
is restored exactly (1408/778/480x270); both placements survive a save + reload.
Full suite 429/429, 0 warnings. LayerReorderPersistenceTests.RealMouseDrag fails
ALONE on a clean tree in this environment - pre-existing flake, proven by stashing
this work, not a regression.
Derivative work: OBS has no built-in per-aspect-ratio layout storage - the
maintainers' answer is "create a second scene and add the scene you want to send
as source" (obsproject.com/forum/threads/161307) - and Streamyard's dual-aspect
guide says the same ("no built-in way to store separate layouts per aspect ratio",
so you rely on manual workarounds and plugins):
https://streamyard.com/blog/dual-aspect-ratio-streaming-tool
Our per-scene, per-tier stored placement, resolved on the tier change, is the
step past both.
Memory in the same commit: ai.md (new section), TASKS.md + TASKS/task-57-*.md
(Unit 1 shipped, Unit 2 = the Chat scene still open), HANDOFF.md, MyMistakes.md
(the per-tier geometry-swap recipe, the binding-text panel trap, the store's
webcam-identity rule) and the Models/Services/ViewModels/Controls index rows.
The cadence work ended with a question the log could not answer: the compositor
manages 6.8 composites/s at ~141ms while the 10:03 take did 29-34ms on the "same
scene shape" (full-render split=0 elements=6 dynamic=5). Nothing in the stats line
said WHICH of the six elements cost what, and nothing recorded the scene itself, so
the two takes could only be compared by eye. This closes that gap and nothing else:
no fix is attempted until a real take names the culprit.
- Services/Compositor/CompositeTimings.cs (new): per-window collector.
Add(stage, ticks) accumulates; Describe(renders, minMs = 0.1) prints the averaged
breakdown slowest-first, dropping sub-0.1ms noise.
- SceneCompositor.Render / CompositeLayers / BakeStaticBase / BlitBakedElement /
BlitCachedLayer + SceneGraph.GetBakedBase: trailing optional CompositeTimings?
Every stage is timed (backdrop, base, flash, social, ticker, zoom, scale, overlay,
bake backdrop, bake overlays) and EVERY element gets its own entry keyed by
z-index + name ("#2 Webcam"), with the nested "raster" sub-stage recording the
paste-cache rasterize miss - the split between "re-samples every tick" and "slow
paste" without a second session. Bake-pass elements are prefixed "bake #..." so a
one-off rebake cannot masquerade as a per-composite cost. Callers that pass
nothing (thumbnails, existing tests) are unaffected and pay nothing.
- FramePump: owns the collector, threads it through RenderScene/RenderFull/
GetBakedBase/CompositeLayers, and logs "FramePump parts (N renders/5s): ..." once
per 5s window beside the stats line (averaged per render, so it compares directly
with "avg render"), then resets it. Also dumps the scene - "FramePump scene dump:"
header plus one line per element (z-index, name, kind, source type, visibility,
rect, opacity, clip) - at pump start and on every scene change.
- One Good Dog integration test (ytLive.Tests/CompositeTimingTests.cs): a real
composite of a full-canvas element plus a 64x64 one through the real collector
must rank the full-canvas element first, carry the raster sub-stage and the fixed
stages, and the dump must name both elements with z-index/type/rect and echo the
split point.
Derivative work (AGENTS.md): OBS's answer to "which source is slow" is the Source
Profiler - average execution time of each source's tick and render functions
(https://docs.obsproject.com/reference-libobs-util-source-profiler, util/profiler.h).
Same shape here: one timer per element per composite, reported as an average, plus
the stage split this case needs.
Docs in the same commit: ai.md (new subsection "Per-element composite timing + scene
dump", the closed gap corrected in "Frame cadence"), Services/index.md (new file row
+ FramePump/SceneCompositor rows), TASKS.md (unit marked shipped, next = creator
take), HANDOFF.md (rewritten: unpushed commit, next step, verify.sh whitelist hole).
Also recorded this session (from the creator): the launch command is
`dotnet run -c Release` - Release is the DEFAULT, Debug only when explicitly named,
and perf takes must read Release. ai.md -> Run updated.
Full suite 428 tests, 0 warnings; the only failures are the known audio env flakes
(both pass in isolation). Scope check passed.
- Model: Google's official MediaPipe Selfie Segmenter (landscape, Apache-2.0 per model card),
converted to ONNX by tools/selfie-seg/convert_tflite_to_onnx.py (custom op
Convolution2DTransposeBias -> ConvTranspose) and verified against MediaPipe's own runtime
(verify_against_mediapipe.py: max |diff| 5e-5, 100% binary agreement). Embedded resource, 430 KB.
- Services/BackgroundEffects: IPersonSegmenter seam, OnnxPersonSegmenter (onnxruntime 1.31.0 CPU,
never throws), BackgroundEffectProcessor (None passthrough, Remove = alpha mask). Not wired yet.
- Tests: Remove integration test (fake segmenter) + bundled-model contract test vs MediaPipe
reference mask. Full suite 407/408 — the one failure is the documented interactive-desktop flake
RealMouseDrag_OnTheLayerList_PersistsTheReorder (passed in the run before; code path untouched).
- Docs: ai.md reversal + section + licensing guard, THIRD-PARTY-NOTICES 11-12, TASK 51, indexes,
MyMistakes recipe, HANDOFF.
References: obs-backgroundremoval (approach only — ONNX Runtime on CPU, 2 threads; GPL-3.0, no code
used) https://github.com/royshil/obs-backgroundremoval ; MediaPipe image segmenter
https://ai.google.dev/edge/mediapipe/solutions/vision/image_segmenter
Creator corrected the "loupe" zoom on 2026-10-08: the click anchors a fixed
capture region (box/2 centred on the click) and the box is a live 2x view of
that region — dragging the box never changes its content; it magnifies the
fully-composited frame (every layer, incl. webcam) and draws LAST (n+1).
Rectangle shape is 4:3 ("more of a 640x480 aspect").
- MainViewModel.Zoom.cs rewritten: ActivateZoom anchors CaptureX/CaptureY
(independently clamped); UpdateZoomMove relocates the box only; the per-frame
BuildZoomFrame/ZoomFrameBuilder path is gone.
- SceneCompositor: skips the active ZoomSceneElement in every layer loop and
BlitZooms it last via the new pure ZoomMagnifier (2x bilinear, temp snapshot
via ArrayPool, round clip).
- Preview: ZoomBox hoisted above CanvasGrid as a sibling; its brush samples the
whole composed preview at the fixed capture (2x).
- Tests: ZoomMagnifierTests (box-over-blue shows green capture at 2x incl. an
extra red "layer" — the loupe/backdrop-only regressions both fail it) +
round-clip; ZoomBoxInteractionTests rewritten (anchored capture, move-doesn't-
change-capture, 4:3/9:16/square). 404 pass, 2 known environment flakes.
Right-click in the live preview now procs a box that is actually visible and
populated with the zoomed region, and each Zoom View Shape sculpts the box per
spec. OBS-style derivative: magnifier box shows the composited region and dims
the surroundings (same pattern as OBS "zoom" / preview region highlights).
Root causes of the invisible box (two WPF traps, recorded in MyMistakes.md):
- Empty <VisualBrush Stretch="None"/> used as an OpacityMask paints nothing,
so the mask alpha is 0 and the WHOLE box (border included) went transparent.
- ZoomBox lived inside CanvasGrid while its VisualBrush sampled CanvasGrid — a
circular reference WPF renders as nothing. Hoisted ZoomBox to a sibling of
CanvasGrid inside a new PreviewCanvas wrapper (same 1920x1080 space).
Shape semantics (MainViewModel.ActivateZoom):
- Circle: square, click point = center.
- VerticalRectangle (phone): tall 9:16, click point = top-left corner.
- HorizontalRectangle (tablet): wide 16:9, click point = top-left corner.
- Boxes stay fully inside 1920x1080 via Math.Clamp.
UI: ZoomRectOutline/ZoomCircleOutline shapes replace the square Border;
circle clips the box to an EllipseGeometry; brush Visual released when inactive.
Zoom View Shape dropdown now uses the standard YtComboBox style.
Test (Good Dog rule, one integration test): ZoomBoxInteractionTests verifies
aspect/anchoring/clamp per shape and the matching outline+clip in the real
MainWindow. Full suite 404 passed, 0 warnings.
Memory: HANDOFF.md zoom state updated; WPF VisualBrush traps appended to
MyMistakes.md.
A composite splits into four horizontal row strips across four DEDICATED
background threads; the calling thread takes a strip too (so four workers
never mean five-way oversubscription), and anything under 64 rows runs
inline. BlitContentRaw, BlitBakedElement and BlitOverlay now fan out.
Output is byte-identical — strips already partitioned by row.
The demotion deliberately does NOT come from Parallel.For. The pool is
shared, so Thread.CurrentThread.Priority = BelowNormal inside it leaks
into every unrelated work item that thread later runs — a full test run
after trying it broke AudioPipelineTests.Mix_HonorsProviderGains_AndGameMute_
KillsTheLoopback. Established guidance agrees: use your own producer/consumer
queue with your own Threads and set their priority yourself
(https://stackoverflow.com/questions/524483/net-backgroundworker-objects-thread-priority,
https://stackoverflow.com/questions/1578599/c-sharp-lower-thread-priority-in-thread-pool).
Helpers/StripPool.cs is that queue. Four threads is the measured sweet
spot: fewer leaves render over budget, more steals cycles from the game
and the encoder and made render worse.
The scaled blit's column LUT moves from a shared static buffer to
ArrayPool<float>.Shared with a finally — a single static buffer is a data
race the moment strips run in parallel.
Gate: SceneCompositorTests.StripParallel_RepeatedRenders_AreByteIdentical —
8 renders of a scaled round webcam + mirrored image + master-sized flash over
a short backdrop must be byte-for-byte identical, and must differ from a
backdrop-only render so it cannot pass vacuously. It caught a loop-variable
capture (new Thread(() => Worker(i)) saw i == Workers) that crashed the test
host on the first run.
Handle the case where an opaque source exactly covers the destination
canvas with 1:1 dimensions (even if offsets differ in calculation). This
avoids per-pixel alpha branching and mirrors the existing memcpy fast
path. Byte-identical for opaque sources.
Add timing for social bar, ticker, flash delegates and scratch acquire,
and extend the 5s stats line to report them (ms per render). No behavior
change, only measurement.
The creator is DHH and lipreads their own recordings, so a webcam that
judders is an accessibility defect, not a polish item. Take
ty-20261005-1001-821-0001.mp4 holds 25.6 distinct webcam images/s against
the camera's 30fps with 281 gaps of 50-100ms - roughly one hitch in five -
in a file that muxes flawlessly. Nothing in the app could see that: rawvideo
carries no per-frame timestamps (ffmpeg stamps by ARRIVAL at the declared
60fps), and the stats line divided every average by the number of frames
SUBMITTED, which deadline padding always fills to the full target. So the
pump reported "300/300 frames, avg render 12.4ms" while producing 60
distinct composites.
Helpers/FrameCadence.cs is one pure counter fed from three points:
MediaCaptureFrameSource logs the device's own SystemRelativeTime stamps
against our arrival times (TryAcquireLatestFrame keeps only the LATEST
frame, so falling behind is silent by design), and FramePump logs the
distinct webcam Epoch values that reached a composite. Read together they
name the culprit - device/USB, our reader, or our compositor - instead of
guessing. Render/resolve/wait now divide by the real render count, and
"frames per 5s from R renders" is the duplicate-on-lag ratio stated
honestly.
An out-of-order timestamp counts as a frame but produces no gap and does not
become the next interval's reference, so one bad stamp cannot invent a
multi-second hitch twice.
Also corrects a stale ai.md line (the "webcam renders at ~18fps" figure
came from a different session and counted every newly rendered layer) and
re-records the background-removal ruling as permanently out. Records the
agreed-but-unbuilt Units 1-2, the no-fallback-camera ruling, and the
SetFormatAsync dead end found by probing the 19041 projection.
Suite 402/402, 0 warnings. The one red, LayerReorderPersistenceTests'
RealMouseDrag, fails identically on a stashed clean tree - it needs an
interactive desktop session this shell lacks.
root cause (proven, not inferred): contentDetails.boundStreamId on
liveBroadcasts.insert does NOT bind the broadcast. Our diagnostic reads
lifeCycle=created privacy=private autoStart=true bound=NONE
boundMatches=NO-BINDING | streamStatus=active
created is Google's "incomplete settings" state, so enableAutoStart has no
bound stream to fire from; close-out then 403s invalidTransition because
created -> complete is not a legal edge. A reusable stream accepts bytes for
a broadcast that isn't watching it, so a perfectly healthy 99MB RTMP session
is exactly what an unbound broadcast looks like -- every observable pointed
away from the binding.
Google's migration guide states the rule outright: a liveBroadcast must be
bound to exactly one stream before it can start, and is not bound at creation.
That was in the docs read during the original design and was not applied.
Shipped:
- BindBroadcastAsync -> POST liveBroadcasts/bind?part=id&id&streamId (no body),
called from PrepareAndStartLiveAsync right after CreateBroadcast, before the pump.
Failure is fatal to the session: an unbound broadcast can never go live, so a
live-looking UI over one is a lie.
- boundStreamId removed from the insert body, and the streamId parameter removed
from CreateBroadcast. A parameter that looks meaningful but cannot affect the
outcome is what made the misdiagnosis possible.
- Replaced CreateBroadcast_With_StreamId_Binds_Reusable_Stream_At_Insert, which
asserted the disproven behaviour and therefore laundered a guess into a
guarantee, with tests for the omitted field and the real bind endpoint.
Error messages (creator ruling 2026-10-04): bind failures are not all "check the
sign-in". Six documented reasons exist and only one is auth; BindFailureMessage
gives each its own text, and the fallback passes the raw reason through verbatim
so an unmapped reason stays greppable. liveBroadcastBindingNotAllowed matters
most -- it means our ordering is wrong, and a generic re-auth message would hide
a bug in our own code behind a plausible excuse. See MyMistakes.md.
ai.md: also corrects the TASK 44 claim that liveChatId is only populated once the
broadcast is live. The diagnostic shows it present while lifecycle is created.
That claim was inferred from a sample's active-status guard, which is not a spec.
Suite 395/395, 0 warnings.
The run I asked for came back with three alarming lines:
lifeCycle=? privacy=? recording=? bound=NONE boundMatches=NO-BINDING
autoStart=unset autoStop=unset liveChatId=none | streamStatus=active
Every single broadcast field was empty, and the stream half was fine. That
combination is the tell: not a failed request, a wrong nesting level.
liveBroadcasts.list returns a LIST wrapper, so the broadcast is at items[0]. I
descended into items for the stream and read the response root for the
broadcast, so every lookup missed, and BoundMatches — handed a null it could
not distinguish from a real null — reported a confident NO-BINDING. The
diagnostic invented the bug it was written to find. Three lines of fabricated
alarm on the one run that was supposed to produce evidence.
Two fixes, and the second is the one that matters:
FirstItem() descends into items[0] for both halves, and returns default when
there is genuinely no item.
The line now prints boundMatches=unread when the broadcast could not be
read at all, so an absent parse degrades to "I don't know" instead of a
verdict. A diagnostic that guesses is worse than no diagnostic: this one
would have sent me off to rebuild a binding that may well be correct.
Test guards the regression: FirstItem_DescendsIntoTheListWrapper_AndReports-
AbsenceHonestly, including the empty-items and missing-items cases. Suite
390/390, 0 warnings.
Credit to the run for catching it. What it did establish, from the half that
worked: streamStatus goes ready -> active about 30s into the push. So the
ingest registers as active server-side, which combined with the 99MB muxed
yesterday means the stream half is verified healthy end to end. The broadcast
half needs one more run to say anything.
The logs answered the question and closed off a whole branch of investigation.
The long run: broadcast inserted 14:17:04.577, ffmpeg to
rtmp://a.rtmp.youtube.com/live2/5f6z-... at 14:17:05.686, and ffmpeg's own
muxing summary at 14:18:56.473 reads
video:101229KiB audio:1419KiB muxing overhead: 0.202949%
~99MB of real video plus audio over 112s, which is what 8000+128 kb/s is
supposed to produce. No socket error, no 4xx from ingest, no reconnect. And the
broadcast still never left `ready`; close-out 403'd invalidTransition again. The
8s run behaved the same way, so "YouTube needed longer" is disproven — 112s is
longer.
So the RTMP path is provably healthy. That is the finding worth keeping: it
kills the entire "the push is broken" theory, which is where this had been
heading for a week. The break is between ingest being accepted and the broadcast
being told to start.
Which puts the binding in frame, and the app was structurally blind to it.
Google documents contentDetails.boundStreamId and says enableAutoStart fires
"when you start streaming video on the bound live stream" — both valid on
insert, neither restricted against reusable streams. But the migration guide
also says a broadcast "is not bound to any streams at the time that it's
created", and the failure that produces is invisible from the RTMP side
precisely because a reusable stream will happily accept bytes for an unbound
broadcast. Ingest succeeding proves nothing.
Hence LogBroadcastDiagnosticsAsync, riding the existing health poll rather than
adding a timer. One greppable line: lifecycle status, privacy, recording,
boundStreamId *compared against the stream id we are actually pushing*
(BoundMatches: yes / MISMATCH / NO-BINDING / unknown), boundStreamLastUpdate-
TimeMs, autoStart/Stop, liveChatId presence, the stream's own streamStatus and
activeLiveChatId, and the CDN settings we declared.
Deliberately self-limiting: _broadcastSeenLive latches on live/liveStarting so
the diagnostic shouts only while the broadcast is stuck and goes quiet once it
is healthy. A permanent per-tick log line would be noise; this is the opposite.
One test, per the Good Dog rule: BroadcastBindingTests pins the matrix and
specifically pins `unknown` when there is no stream id to compare against, so a
missing expectation can never be misreported as a match — that is the failure
that would make this diagnostic lie.
Suite 387/387, 0 warnings.
force-ssl is gone from the requested scope. Its grant ("see, edit, and
permanently delete your videos, ratings, comments and captions") is a strict
subset of auth/youtube's ("Manage your YouTube account"), and Google's own
liveBroadcasts.insert reference lists auth/youtube as an accepted ALTERNATIVE
to force-ssl, so every endpoint this app calls — channels, liveBroadcasts,
liveStreams, liveChat, playlistItems — is covered by the one we keep. Two
consent lines became one, and one fewer scope to justify. Removing a scope
cannot narrow capability when the scope being kept is the broader of the two,
and there is no forced re-consent: a refresh token keeps the scopes it was
originally granted, so existing sessions are untouched.
The playlistItems change is NOT being made, and the map entry recommending it
was wrong. It came out of the verification hunt on the theory that public data
should not ride on user credentials. Walking the code kills it: ChannelStats-
Service is built with one HttpClient, and channels?mine=true on the same
service genuinely needs the session. Splitting it means a second unauthenticated
client PLUS a Google API key, since every YouTube Data API request needs a
credential and the call 403s without one. So the fix would ship a new
extractable, quota-abusable credential inside the binary to replace a token the
app already holds, buying zero verification-surface reduction (the data is
public, no scope involved) and zero correctness. Strictly worse. Recorded as a
do-not-do so it is not "helpfully" reintroduced later.
Test: ExactlyOneScopeIsRequested_AndItIsTheBroadOne asserts on the const's
parsed VALUE rather than scanning the file — a file-wide DoesNotContain
fails on the doc comment that names force-ssl to explain its removal, which is
the same trap I hit and documented earlier today in the locator test. Suite
380/380, 0 warnings.
Store policy 10.2.2 forbids dynamic code inclusion, and this was the only
third-party binary the app ever downloaded and ran. It had already failed in
production: the pin aged out of BtbN's 14-day retention and the download 404'd
on the creator's first real recording attempt (2026-09-01). Bundling kills
that failure class permanently and deletes the startup network dependency.
- scripts/fetch-ffmpeg.sh populates a gitignored ./ffmpeg from a pinned
month-end BtbN LGPL-shared artifact, SHA-256 verified before extraction,
idempotent, FFMPEG_ZIP= override. 135 MB of DLLs in git would tax every
commit on this 9p tree; the URL + hash are the audit trail LGPL §6 wants.
- ytLive.csproj copies ffmpeg\** to output and publish (Always), and a
GuardBundledFfmpeg target fails the BUILD if the folder is missing, so a
fresh clone finds out before a take rather than during one.
- FfmpegLocator loses the downloader, the zip extraction and the
%APPDATA%\ytLlive\tools cache. Order is now bundle -> PATH, deliberately
inverted: a PATH ffmpeg may be a different major version or a GPL build,
and the bundle is what was pinned, license-vetted and tested. A missing
bundle throws loudly; the old "found something on disk" path could satisfy
a missing bundle with a stale copy, and silent drift is the worst failure
mode for a dependency certification inspects.
- ffplay.exe (18 MB) is not shipped; nothing calls it.
- THIRD-PARTY-NOTICES.txt: "downloads" -> "bundled" throughout, and the stale
build tag autobuild-2026-08-09-13-03 — the tag that 404'd — corrected to
autobuild-2026-08-31-13-27 with the exact artifact name and archive hash.
OpenH264 made precise: compiled INTO avcodec-63.dll, no separate binary.
Checked whether other third-party binaries needed bundling: no. e_sqlite3.dll
and WebView2Loader.dll already ship via NuGet into runtimes\win-x64\native\,
OpenH264 is inside avcodec-63.dll, the rest are managed. Two gaps remain that
bundling cannot fix, recorded in the map: the WebView2 Evergreen runtime is a
machine prerequisite (TASK 48 item 2), and Velopack's updater downloads and
executes, so its teardown is a compliance action.
Tests: 16 locator facts, suite 379/379, 0 warnings. The integration test runs
the bundled ffmpeg -version, which is the only assertion that proves the
sibling libav*.dll resolved — presence checks pass with the DLLs missing and a
shared build then fails to start. TheLocatorCarriesNoNetworkCapability keeps
10.2.2 from rotting back.
Analysis: TASKS/research-store-certification.md §4.
Reported 2026-09-27: click avatar -> Change Account -> cancel in the Google
chooser -> back to a signed-out app, then "cannot sign-in because the callback
url was not released".
Two independent defects:
1. ChangeAccountAsync ran `await SignOutYouTubeAsync(); await SignInAsync();`.
SignOutYouTubeAsync -> TokenStore.Clear() -> deletes ytLlive.auth, so the
working account's token was destroyed before the browser even opened. It is
now just `await SignInAsync()`: success overwrites the token via the
sessionChanged hook, cancel/failure keeps the current session. The explicit
sign-out was never needed.
2. Abandoning the browser gives the app no signal, so the OAuth attempt held
localhost:8765 for its full 5-minute timeout and a retry could not bind. A
new AuthenticateAsync now cancels and awaits the abandoned attempt first.
That takeover is serialised by a SemaphoreSlim: the first version cleared
the in-flight marker before awaiting, so a caller arriving in that window saw
a free port and raced to bind it -- reproducing the very failure it was
written to prevent. My own test caught that; reasoning had "proved" it fine.
Also adds an `openBrowser` ctor seam so tests never shell out to a real
browser, and corrects the ai.md line that documented "Logout / Change Account ->
SignOutYouTubeAsync" as intended behaviour.
Test: YouTubeAuthServiceTests.RetryingAfterAnAbandonedAttempt_ReleasesTheCallbackPort
Full suite 370/370, 0 warnings. Coverage gap noted in HANDOFF.md: the ordering
fix has no automated test because _youtubeAuth is hardcoded in MainViewModel
and ChangeAccountAsync is private.
The reported cross-instance effect is NOT this cause and stays open -- no code
path can sign out a second instance (per-instance TokenStore, local delete, no
server-side revoke).
The dev multi-instance use case (one instance screen-capturing the other for
marketing/training videos) exposed a real bug. BuildStartPath in
MainViewModel.Recording.cs builds the start name with no dedup:
var name = RecordingFile.BuildStartName(DateTime.Now) + ".mp4";
var path = Path.Combine(folder, name);
Only FinalizeRecordingAsync calls UniquePath. So two instances that start
recording in the same minute computed the identical path
(ty-20260927-1454-0000.mp4) and both ffmpeg processes wrote one file -- a
corrupt take, not an error message.
Creator's fix (2026-09-27): add a millisecond count to the filename timestamp.
New format is ty-<yyyyMMdd>-<HHmm>-<fff>-<hh2mm2>.mp4. Both the start name and
the final name derive from _recordStartTime, so the ffff segment survives the
rename-on-stop and the final name cannot collide with another same-minute take
either. No UniquePath call needed, which keeps the start path a pure function of
the timestamp and the existing collision suffix behaviour intact.
Tests: 2 new (the same-minute guard and the rename-survival guard), 5 existing
assertions updated for the format. RecordingSaveDialogTests' stem fixtures
updated to match. 369/369 pass, 0 warnings.
Note: the first test run reported 367/367 because only ytLive.csproj had been
built -- the stale test DLL never picked up the new tests. The real count is 369.
Creator 2026-09-26: "the ticker should appear over the stream alerts video, not over
the entire preview window". It was a GLOBAL 1920x48 bar blitted last at a hardcoded
(0, 0) -- it covered the whole preview width, not the alert video.
The strip is now rendered at the alert box's own size and the box's origin travels on
the frame in a new VideoFrame.Placement ((int X, int Y)?). OriginX/OriginY default to
0 when Placement is null, so every full-canvas overlay (the branding flash) is
unaffected. Every ticker blit site now reads tickerFrame.OriginX/OriginY instead of a
literal 0, 0 -- SceneCompositor x2 and FramePump's static-bake Overlay path -- and
PreviewPane.xaml's AlertTickerElement binds the same rect (AlertTickerLeft/Top/
Width/Height), so the preview cannot drift from the recording the creator is judging.
Why the position rides on the frame rather than in a full-canvas frame: a 1920x1080
overlay is 8.3MB of large-object-heap garbage per tick, ~2.5GB churned over one 10s
alert at 30fps. A box-sized strip is ~370KB. The branding flash does render
full-canvas but RECYCLES one 8MB buffer and bumps Epoch, so it never allocates per
frame; the ticker allocates per call, so it has to stay small.
Also fixed: CopyStrip now clips ROWS to the target height. The pill rasterises at its
natural 48px, so an alert box shorter than that would have written past the end of the
target buffer once the strip became box-sized.
No alert box in the scene now means no ticker at all -- there is no global position
left for it, and this is the guard against the old bar quietly coming back.
Tests (3 new facts, 27 in the file):
ComposedOutput_PutsTheTickerInsideTheAlertBox_NotAtTheTopEdge -- renders through
the real SceneCompositor and asserts the pixels land at (620, 430) and NOT at the
top-left corner. The creator is judging compositing from local recordings, so the
OUTPUT side is the side that needed proving.
ASceneWithNoAlertBox_PublishesNoTicker
ABoxShorterThanTheStrip_ClipsRowsInsteadOfOverrunning
Updated AlertLayer_PublishesARealTickerFrameToThePreviewSink, which asserted the old
1920 width from a box with no geometry (defaulted to 1px); it now uses the product's
own default box (680x200 at 620,430, MainViewModel.Sources.cs:54-57) and pins the
frame size and origin.
Full suite 367/367.
Creator 2026-09-26: "the made with llamacasty flash should appear in all scenes,
not just live" and "should also appear in recordings". The presenter was
Start()/Stop()-ed from UpdateLiveVisuals()'s IsLive branch, so a recording made
WITHOUT ever going live carried no credit at all -- the exact case the creator hit
while judging compositing from local recordings.
It is now Start()ed once in the MainViewModel ctor and never stopped on live-state
churn. One start covers every scene, the preview, the stream and the recording,
because go-live and local recording are the SAME FramePump: both
Streaming.Operations.cs:68 and :199 call StartAsync with the same brandFlash:
delegate. Verified by reading both call sites, not assumed -- there is no second
encoder path that needed a "redirect". The licence gate needs no live branch at
all: IsPremium's setter already pushes BrandFlashPresenter.Enabled from anywhere.
Fixed a trap that app-lifetime exposed: Enabled = false stops the presenter's
DispatcherTimer to cut the advertisement mid-credit. When Start() was per-go-live
the next go-live restarted it; with a single app-lifetime Start() nothing would,
so a key entered mid-session would leave the credit dead until the process was
restarted. The setter now restarts the timer when re-enabling while _running.
Tests (12 facts in BrandFlashOutputTests, +2):
Credit_IsComposited_WithNoLiveSession_AndSoARecordingCarriesIt -- drives a real
never-live VM past the 5s first-flash delay and asks the frame the pump would
composite. Uses a new internal AdvanceBrandFlash seam; because Advance only
advances the cadence while the presenter is running, a credit coming out
proves Start() happened at construction.
ADowngradeMidSession_RestartsTheCadenceTimer -- asserts the premium/downgrade
edge decision directly (IsCadenceTimerEnabled) instead of sleeping through a
30-60s interval, which a synchronous test body cannot observe.
Full suite 364/364.
gramps