tobozo

enclosing pushPixels between tft.startWrite() and tft.endWrite() sounds redundant https://github.com/lovyan03/LovyanGFX/blob/5438181440c71cf30bbdc347b0b1597ae3ebf77d/src/lgfx/v1/LGFXBase.hpp#L342-L344 what about pushImage ? ```cpp uint16_t testPixels[64]; for (int i = 0; i < 64; i++) { testPixels[i] = (rand() %...

same with pushImageDMA and pushPixelsDMA ? maybe now is a good time to play with cfg.spi_mode and other bus config values

this seems related ideas: - heap_cap_malloc(MALLOC_CAP_DMA), or ps_malloc() `testPixels` instead of using new() - try with sprite.setPsram(true) and sprite.pushSprite() -

bump, still looking for a solution thanks to your input the cause is confirmed to be DMA (ESP32P4 uses AXI_DMA) a quick performance-crippling workaround is to set `cfg.dma_channel` to 0....

hi, > One issue I have is with the shared i2c bus for the touch controller and the onboard IMU Please describe the actual symptoms. There's a config option for...

> My apolagies. Here is the correct link: https://github.com/FrostiFish/Waveshare-RP2350-Touch-LCD-1.28-Template If it wasn't for the typo I'd swear I'm talking to an AI 🤣

Joke apart, I suggest you have a look at [this thread](https://github.com/lovyan03/LovyanGFX/issues/749) where a similar conflict was solved by using lgfx::Bus_I2C to control the non-lgfx I2C device. This means you'll probably...

just a random idea but did you try to use the touch without interrupt? ```cpp cfg.pin_int = 21; //

more random things to test: - lower spi freq to 40MHz, see how it affects the timings - increase i2c freq, first to 500Khz, then up to 700KHz, see how...

Hi, From the [lvgl setup](https://github.com/Hack-a-Day/2025-Communicator_Badge/blob/main/firmware/badge/hardware/lvgl_setup.py) it looks like it's using a basic SPI config without miso, so Panel_ST7789 could be used as a boilerplate. If using QSPI there's a [Panel_NV3041A](https://github.com/lovyan03/LovyanGFX/blob/develop/src/lgfx/v1/panel/Panel_NV3041A.cpp)...