esp_rgb_matrix

#define RGB_DATA  D2
#define RGB_CLOCK D3
#define RGB_LATCH D4

#define SIZE_MATRIX 8
#define NO_MATRIX 3
#define BIT_COLOR_DEPTH 3

uint32_t framebuffer[SIZE_MATRIX][SIZE_MATRIX*NO_MATRIX];

uint8_t pwm_counter=0;

enum COLOR_T{CL_BLUE=0, CL_GREEN=1, CL_RED=2};

uint8_t extractColor(uint32_t *pixel, COLOR_T color){
  return (uint8_t)( ((*pixel)>>((uint8_t)color*8)) & 0xFF ) ;
}

uint8_t combineRow(uint32_t *fst, COLOR_T color, uint8_t treshold){
  uint8_t ret=0;
    for(uint8_t i=0;i<8;++i){
      if(extractColor(&fst[i], color) > treshold)
        ret += (1<<i);
    }

  return ret;
}

//spi transfer one byte, without latch control
void transfer(uint8_t a){

  for(volatile uint8_t ibit=0;ibit<8;++ibit){
    digitalWrite(RGB_DATA, a & (1<<ibit));
    digitalWrite(RGB_CLOCK, HIGH);
    digitalWrite(RGB_CLOCK, LOW);
  }

}

void display_task(void){
  static uint8_t displayed_row;

  displayed_row = (++displayed_row) % SIZE_MATRIX;
  //Serial.print("Sending value: ");
  //Serial.println(++a, HEX);

  for(pwm_counter=0;pwm_counter<(1<<(BIT_COLOR_DEPTH-1));++pwm_counter){
    transfer( combineRow(&framebuffer[displayed_row][0*SIZE_MATRIX], CL_BLUE, pwm_counter)); //blue row, columns 0-7
    transfer( combineRow(&framebuffer[displayed_row][0*SIZE_MATRIX], CL_GREEN, pwm_counter)); //green row, columns 0-7
    transfer( combineRow(&framebuffer[displayed_row][0*SIZE_MATRIX], CL_RED, pwm_counter));    //red row, columns 0-7
    transfer((1<<displayed_row));    //row selection :1
    transfer( combineRow(&framebuffer[displayed_row][1*SIZE_MATRIX], CL_BLUE, pwm_counter)); //blue row, columns 0-7
    transfer( combineRow(&framebuffer[displayed_row][1*SIZE_MATRIX], CL_GREEN, pwm_counter)); //green row, columns 0-7
    transfer( combineRow(&framebuffer[displayed_row][1*SIZE_MATRIX], CL_RED, pwm_counter));    //red row, columns 0-7
    transfer((1<<displayed_row));    //row selection :1
    transfer( combineRow(&framebuffer[displayed_row][2*SIZE_MATRIX], CL_BLUE, pwm_counter)); //blue row, columns 0-7
    transfer( combineRow(&framebuffer[displayed_row][2*SIZE_MATRIX], CL_GREEN, pwm_counter)); //green row, columns 0-7
    transfer( combineRow(&framebuffer[displayed_row][2*SIZE_MATRIX], CL_RED, pwm_counter));    //red row, columns 0-7
    transfer((1<<displayed_row));    //row selection :1
    digitalWrite(RGB_LATCH, HIGH);
    digitalWrite(RGB_LATCH, LOW);
    delayMicroseconds(1);
  }



}

void setup() {
  // put your setup code here, to run once:

  Serial.begin(9600);
  pinMode(RGB_DATA, OUTPUT);
  pinMode(RGB_CLOCK, OUTPUT);
  pinMode(RGB_LATCH, OUTPUT);
  digitalWrite(RGB_DATA, LOW);
  digitalWrite(RGB_CLOCK, LOW);
  digitalWrite(RGB_LATCH, LOW);


  //         ROWCOL     RRGGBB
  for(uint8_t i=0;i<8;++i){
    /*
    //Effect #1
    framebuffer[0][i]=i;
    framebuffer[0][i+8]=(i<<8);
    framebuffer[0][i+16]=(i<<16);
    */
    //Effect #2
    framebuffer[0][i]=    (i<<0)  +((8-i)<<8);
    framebuffer[0][i+8]=  (i<<8)  +((8-i)<<16);
    framebuffer[0][i+16]= (i<<16) +((8-i)<<0);
  }

}

void loop() {
  static uint8_t div;
  if((++div % 30) == 0)
  {
    uint32_t temp = framebuffer[0][0];
    for(uint8_t x=0;x<24;++x){
      framebuffer[0][x] = framebuffer[0][x+1 % 24];

    }
    framebuffer[0][23] = temp;
    for(uint8_t x=0;x<24;++x){
      framebuffer[1][x] = framebuffer[0][x+1 % 24];
      framebuffer[2][x] = framebuffer[0][x+2 % 24];
      framebuffer[3][x] = framebuffer[0][x+3 % 24];
      framebuffer[4][x] = framebuffer[0][x+4 % 24];
      framebuffer[5][x] = framebuffer[0][x+3 % 24];
      framebuffer[6][x] = framebuffer[0][x+2 % 24];
      framebuffer[7][x] = framebuffer[0][x+1 % 24];

    }

  }



  display_task();

}