Real webcams hit the 23000-byte bound, but the S3 esp32-camera build
allocates width*height/5 = 15360 bytes for a QVGA JPEG frame buffer and
stops copying one 1 KiB DMA half-buffer short of it, so oversized frames
lost their tail - including the EOI marker cam_hal scans for (the NO-EOI
storm on staging; the small canvas-fake frames masked it). The bounded
encoder now takes a budget (14000 for the Sense, default 23000 for the
QEMU ESP32-CAM path unchanged), the downscale fallback re-runs its own
quality ladder, and a frame that cannot fit is skipped instead of sent
truncated.
The toggle's attention border reused the .velxio-ants class name, which
is ALSO the canvas selection outline whose ::after is position:absolute
- on the non-positioned button it anchored to the toolbar and drew a
marching-ant line across its full width. Renamed the button variant to
.velxio-btn-ants. The ants also showed while the simulation was stopped;
now they only run when the sketch is running and the camera is off (the
only moment the button is actually asking for a click).
El boton conservaba animation:'none' inline fuera del estado requesting, y el
estilo inline gana a la clase: .velxio-ants quedaba con animationName none —
medido con getComputedStyle en staging. Inline solo mientras solicita permiso;
en el resto, undefined para que la clase anime.
Cuatro tiras de gradiente repetido, una por borde, deslizando un periodo de
guion por ciclo — fondo y no border porque un borde CSS no puede animar su
dash-offset. currentColor hereda el color de estado (gris en reposo, rojo en
error). Se apaga en streaming/solicitando, y con prefers-reduced-motion.
Ojo del gato encerrado: el estilo inline usaba el shorthand background, que
pisa el background-image de la clase — pasa a backgroundColor.
Un sketch de ESP32-CAM existe para capturar, asi que esperar el click en el
toggle se lee como 'la camara esta rota': el guest arranca, esp_camera_init
completa contra el OV2640 modelado, y cam_hal agota el tiempo eternamente
esperando fotogramas que nadie envia — sin pedir siquiera el permiso.
Ahora la webcam se solicita al iniciar la ejecucion; el prompt del navegador ES
el consentimiento, y el toggle queda como apagador manual. Solo una vez por
ejecucion: parar el stream a mano no lo re-dispara.
User goal: ESP32-CAM live preview that works with any webcam,
regardless of resolution, brand, or scene complexity. The previous
fixed-quality 0.28 was fragile (intermittent decode errors on
moving/textured scenes) and capped visual quality unnecessarily.
Two-layer fix; either alone is insufficient:
LAYER A — Bounded JPEG encoder (frontend, this repo)
frontend/src/hooks/useWebcamFrames.ts:
encodeBoundedJpeg() walks a quality ladder [0.6, 0.5, ..., 0.1]
until the JPEG fits in MAX_FRAME_BYTES (23 000). If even q=0.1
overshoots — extreme HD/4K scenes — falls back to a 240×180
downscaled canvas at q=0.4. Guarantees every emitted frame fits
the deliverable budget regardless of webcam hardware.
The hook now exposes lastQualityUsed + lastDownscaled so UI can
surface when auto-tuning kicks in.
frontend/src/components/simulator/CameraToggle.tsx:
Tooltip shows "(auto-tuned to q=0.X)" or "(auto-downscaled, q=0.X)"
while streaming so users see what the encoder picked.
LAYER B — Multi-lap descriptor ring walker (qemu-lcgamboa, submodule)
Bumps the QEMU per-frame deliverable cap from 8 KiB to ~32 KiB by
letting the walker reset the descriptor ring up to 4 times per
VSYNC. Submodule pointer bumped to eb8b7a5d.
Combined, the demo now supports:
- Cheap 480p webcams: q=0.6, 5-10 KiB JPEGs, sharp
- Logitech mid-range: q=0.5-0.6, 8-15 KiB JPEGs, sharp
- HD 1080p webcams: q=0.4-0.6, 15-23 KiB JPEGs, sharp
- 4K complex scenes: downscaled, still readable
Documented as bug closure in:
test/test-esp32-cam/autosearch/15_universal_webcam_compat.md
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
First open-source end-to-end emulation of the AI-Thinker ESP32-CAM
in QEMU, paired with a browser webcam → firmware bridge so users can
develop camera sketches without hardware. Status: esp_camera_init()
returns ESP_OK; OV2640 chip-id verifies (PID/VER/MIDH/MIDL exactly
match the datasheet); GPIO 25 VSYNC NEGEDGE interrupt enabled by
the upstream driver. Final piece (cam_task accepting frames) is in
progress — descriptor walker fix landed in this commit.
Backend (Python/FastAPI):
- simulation.py: camera_attach/frame/detach WS handlers
- esp32_worker.py: ctypes binding to velxio_push_camera_frame +
feature-detection fallback for older DLLs
- esp32_lib_manager.py: forward camera commands to the worker stdin
- esp-idf-template/main/CMakeLists.txt: esp32-camera headers added
via add_prebuilt_library + REQUIRES driver (resolves i2c_master_*
symbols). LED_BUILTIN=2 fallback for sketches that hardcode it.
Frontend (React/TS):
- EditorToolbar.tsx: ESP32-CAM (and the rest of the ESP32 family)
added to isQemuBoard list — Run button now starts the QEMU bridge
for these boards instead of falling through to the AVR path
- useWebcamFrames.ts: getUserMedia → OffscreenCanvas →
toBlob('image/jpeg') → base64 → WS at ~10 fps
- CameraToggle.tsx: header button with status colors + frame counter
- SimulatorCanvas.tsx: render CameraToggle for esp32-cam boards
- Esp32Bridge.ts: sendCameraAttach/Frame/Detach + chunked btoa
- useSimulatorStore.ts: diagnostic log on compileBoardProgram
- components-metadata.json: regen including esp32-cam component
Submodule pointer:
- wokwi-libs/qemu-lcgamboa → ff8eee0 (camera devices commit on
davidmonterocrespo24/qemu-lcgamboa branch picsimlab-esp32)
Investigation + tests in test/test-esp32-cam/:
- 13 autosearch markdown docs (overview, SOTA, OV2640 spec, DVP/I2S
spec, build blueprint, blockers resolved, descriptor walker fix)
- 5 sketches (camera_init, sccb_probe, dma_smoke, frame_roundtrip,
webcam_demo) + 8 live + WS regression tests
- README with the user-facing flow
.gitignore:
- libqemu-*.dll.{pre-camera,new,bak} (rollback points, regenerated)
- wokwi-libs/esp32-camera/ (clone consumed by arduino-esp32 path,
not part of this repo)
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
David Montero Crespo