Sunton LCD ESP32-4827S043 unexpected drift occurs (IDFGH-10421)

[Dadigno]

Answers checklist.

• [X] I have read the documentation ESP-IDF Programming Guide and the issue is not addressed there.
• [X] I have updated my IDF branch (master or release) to the latest version and checked that the issue is present there.
• [X] I have searched the issue tracker for a similar issue and not found a similar issue.

General issue report

IDF version v5.0

Operating System used. Windows

How did you build your project? idf.py build

If you are using Windows, please specify command line type. CMD

Development Kit. Sunton LCD ESP32-4827S043 4.3” 480X272-TN bare screen specs.pdf

What is the expected behavior? I started from rgb_lcd_example_main.c and managed to run lv_demo_widgets. It works, but drift issue show off if EXAMPLE_LCD_PIXEL_CLOCK_HZ is greater than (6*1000*1000). I would work with 9Mhz ( typical value in device specs )

https://github.com/espressif/esp-idf/assets/32482073/005dfe6e-2988-4e6e-8c00-b6b33dc196da

What I tried: I read all doc included with the dev kit ( schematics, arduino example ecc ) and here. Also, as explained here, I tried double buffering setting CONFIG_EXAMPLE_DOUBLE_FB = 1 and I'm using .fb_in_psram = true

Below my panel_config. I got pin timings and pin number from here and schematics.

.......
....
..
.
 esp_lcd_rgb_panel_config_t panel_config = {
        .clk_src = LCD_CLK_SRC_DEFAULT,
        .timings = {
            .pclk_hz = EXAMPLE_LCD_PIXEL_CLOCK_HZ,
            .h_res = LCD_H_RES,
            .v_res = LCD_V_RES,
            // The following parameters should refer to LCD spec
            .hsync_pulse_width = 4,
            .hsync_back_porch = 43,
            .hsync_front_porch = 8,
            .vsync_pulse_width = 4,
            .vsync_back_porch = 12,
            .vsync_front_porch = 8,
            .flags.pclk_active_neg = true,
        },
        .data_width = 16, // RGB565 in parallel mode, thus 16bit in width
        .num_fbs = 2,
         #if CONFIG_EXAMPLE_USE_BOUNCE_BUFFER
            .bounce_buffer_size_px = 10 * EXAMPLE_LCD_H_RES,
        #endif
        .psram_trans_align = 64,
        .hsync_gpio_num = EXAMPLE_PIN_NUM_HSYNC,
        .vsync_gpio_num = EXAMPLE_PIN_NUM_VSYNC,
        .de_gpio_num = EXAMPLE_PIN_NUM_DE,
        .pclk_gpio_num = EXAMPLE_PIN_NUM_PCLK,
        .disp_gpio_num = EXAMPLE_PIN_NUM_DISP_EN,
        .data_gpio_nums = {
            EXAMPLE_PIN_NUM_DATA0,
            EXAMPLE_PIN_NUM_DATA1,
            EXAMPLE_PIN_NUM_DATA2,
            EXAMPLE_PIN_NUM_DATA3,
            EXAMPLE_PIN_NUM_DATA4,
            EXAMPLE_PIN_NUM_DATA5,
            EXAMPLE_PIN_NUM_DATA6,
            EXAMPLE_PIN_NUM_DATA7,
            EXAMPLE_PIN_NUM_DATA8,
            EXAMPLE_PIN_NUM_DATA9,
            EXAMPLE_PIN_NUM_DATA10,
            EXAMPLE_PIN_NUM_DATA11,
            EXAMPLE_PIN_NUM_DATA12,
            EXAMPLE_PIN_NUM_DATA13,
            EXAMPLE_PIN_NUM_DATA14,
            EXAMPLE_PIN_NUM_DATA15,
        },
        .flags = { 
            .fb_in_psram = true // allocate frame buffer in PSRAM
        },
    };
    ESP_ERROR_CHECK(esp_lcd_new_rgb_panel(&panel_config, &panel_handle));

....
..
.

[suda-morris]

drift issue show off if EXAMPLE_LCD_PIXEL_CLOCK_HZ is greater than (610001000).

https://github.com/espressif/esp-idf/tree/master/examples/peripherals/lcd/rgb_panel#troubleshooting

[Dadigno]

Thank you for your response. I checked it out yet. As I said lowering frequency to 6MHz makes it work, but this value is out of LCD spec. I don't know what is depends on.

[suda-morris]

It's a bandwidth issue of the PSRAM. There're some instructions here: https://espressif-docs.readthedocs-hosted.com/projects/espressif-esp-faq/en/latest/software-framework/peripherals/lcd.html#why-do-i-get-drift-overall-drift-of-the-display-when-driving-an-rgb-lcd-screen

[Dadigno]

Okay. Sorry, I missed this page.

This: Reduce data_cache_line_size to 32 bytes makes it work at 9Mhz.