- several compile warning fixes
- multipin LED compile config
- release info (update page, JSON "info")
- WiFi scan fix if no networks found
- UI glitch when no presets are found fix
With multipin LED config it is now possible to assign GPIO to PWM RGB outputs.
Achieved by having length of DATA_PINS be divisble by lengt of PIXEL_COUNTS.
-changes save roughly 600bytes of flash
-made blurring faster by not writing the color and then reading it back but keeping it as a variable: on a C3, FX black hole goes from 55FPS to 71FPS
-added optional parameter to blur (smear) that can be used in combination with SEGMENT.clear(), blurring the frame without dimming the current frame (repeated calls without clearing will result in white). this is useful to blur without 'motion blurring' being added
-scale8 is inlined and repeated calls uses flash, plus it is slower than native 32bit, so I added 'color_scale' function which is native 32bit and scales 32bit colors (RGBW).
-added fully random palette function ('the old way', currently just used for initialization)
-changed randomness values to make it a little less random
-added 10% chance for pastel color palette
-now using swap() from std library for shuffeling
-changed function name
-moved update check from loadPalette() to handleRandomPalette() saving CPU cycles
Remove sync receive
Disallow 2D effects on non-2D segments
Optimisations
Sync clarification
AR palettes
Return of 2 audio simulations
Bugfix in sync #3344
- remove excessive segments
- ignore inactive segments if not syncing bounds
- send UDP/WS on segment change
- pop_back() when removing last segment
Add pairing support for ESP-NOW sync
Reduce string RAM footprint
UDP parse optimisation
Make WizMote work with sync.
ESP-NOW wireless sync POC.
- caveat: devices have to be on the same channel
- clashes with WizMote handling ATM
this is a band-aid fix for random crashes when switching between presets with multiple segments - crossfade disabled.
!! adding type initializers fixed it for me on -S3, however I still see (less frequent) crashes on esp32, due to heap corruption.
It took me hours to get a meaningful stackdump:
assert failed: heap_caps_free heap_caps.c:360 (heap != NULL && "free() target pointer is outside heap areas")
Backtrace: 0x40084ee1:0x3ffb2570 0x4008e341:0x3ffb2590 0x40094709:0x3ffb25b0 0x4008534a:0x3ffb26e0 0x40094739:0x3ffb2700 0x400e9037:0x3ffb2720 0x400e917c:0x3ffb2740 0x400eaeeb:0x3ffb2760 0x40117ec5:0x3ffb27c0 0x401184ea:0x3ffb2800 0x4013509d:0x3ffb2820
#0 0x40084ee1:0x3ffb2570 in panic_abort at /Users/ficeto/Desktop/ESP32/ESP32S2/esp-idf-public/components/esp_system/panic.c:402
#1 0x4008e341:0x3ffb2590 in esp_system_abort at /Users/ficeto/Desktop/ESP32/ESP32S2/esp-idf-public/components/esp_system/esp_system.c:128
#2 0x40094709:0x3ffb25b0 in __assert_func at /Users/ficeto/Desktop/ESP32/ESP32S2/esp-idf-public/components/newlib/assert.c:85
#3 0x4008534a:0x3ffb26e0 in heap_caps_free at /Users/ficeto/Desktop/ESP32/ESP32S2/esp-idf-public/components/heap/heap_caps.c:360
(inlined by) heap_caps_free at /Users/ficeto/Desktop/ESP32/ESP32S2/esp-idf-public/components/heap/heap_caps.c:345
#4 0x40094739:0x3ffb2700 in free at /Users/ficeto/Desktop/ESP32/ESP32S2/esp-idf-public/components/newlib/heap.c:39
#5 0x400e9037:0x3ffb2720 in Segment::deallocateData() at wled00/FX_fcn.cpp:189
#6 0x400e917c:0x3ffb2740 in Segment::resetIfRequired() at wled00/FX_fcn.cpp:206
(inlined by) Segment::resetIfRequired() at wled00/FX_fcn.cpp:203
#7 0x400eaeeb:0x3ffb2760 in WS2812FX::service() at wled00/FX_fcn.cpp:1216 (discriminator 2)
#8 0x40117ec5:0x3ffb27c0 in WLED::loop() at wled00/wled.cpp:115 (discriminator 3)
#9 0x401184ea:0x3ffb2800 in loop() at C:/src/wled00/wled00.ino:20
#10 0x4013509d:0x3ffb2820 in loopTask(void*) at C:/Users/user/.platformio/packages/framework-arduinoespressif32/cores/esp32/main.cpp:50
ELF file SHA256: 18c20b536f4c6ef4
turns out that fastLED 3.6.0 has an explicit uint32_t operator that returns RGBA, however 3.5.0 does not have this and the conversion returned only the "red" component".
it seems that SEGMENT.blur() is the main bottleneck for many 2D effects.
This change optimizes performance of the function:
* avoid to re-write unchanged pixels
* early exit when blur_amount == 0 (=nothing to do)
* use _fast_ types where possible
I've seen up to 20% speedup with this change.
power estimation results from estimateCurrentAndLimitBri() were too low (example: estimated 1.3Amp, measured 1.6Amp). This change corrects the power calculation. Due to the changed behavior of getPixelColor, powerSum must be used as-is, not scaled down again by brightness.
this optimization avoids to apply brightness two times .
NeoPixelBusLg has already applied global brightness at sPC. Due to internal working of the Lg bus, its sufficient to only post-apply scaling, and set the new (scaled) brightness for the next frame.
I still see strange crashes in setPixelColor/GetpixelColor, which ssem to come from race conditions between async_tcp (change presets) and looptask (strip.service).
To make the situation better, its important that any global pointers are reset to NULL immediately after free().
* Avoid uint16 underflow in width() and height(): stop > start is possible, and means "inactive segment".
Without these checks, it was possible that width() and height() produce VERY large values due to underflow.
* adding an optional parameter to setBrightness(). ApplyPostAdjustments() will only be called if `immediate=true`. Only ABL will use immediate=true, to ensure electrical safety of equipment.
This allows some optimizations of performance, as ApplyPostAdjustments() is time consuming.
* busses.setBrightness(bri) --> applied to all future pixels (fast, lossless)
* busses.setBrightness(bri, true) --> applied directly to all previously set pixels (slower, lossy)
Brandon502