Example of a basic RGB color blending function using floats. Takes two packed 32-bit RGB colors and a percentage byte as an input and returns a single 32-bit color as an output.

Linear RGB Color Blend.ino

uint32_t blendRGB(uint32_t c1, uint32_t c2, uint8_t p){
  /* Blends two RGB color values using a linear algorithm
   *  in the form of y = mx + b.
   *  
   * Where y is the resultant blended color (per channel), m
   *  is the slope of the color change divided by the number of steps,
   *  x is the current step, and b is the initial channel value.
   *  
   * This implementation uses a single byte for gradation, giving steps from 0-255.
   */
  
  float mR, mG, mB;
  uint8_t newR, newG, newB;

  mR = (float)(((uint8_t)(c2 >> 16)) - ((uint8_t)(c1 >> 16))) / 255;
  mG = (float)(((uint8_t)(c2 >> 8))  - ((uint8_t)(c1 >> 8)))  / 255;
  mB = (float)(((uint8_t)(c2))       - ((uint8_t)(c1)))       / 255;
  
  newR = (mR * p) + (uint8_t)(c1 >> 16);
  newG = (mG * p) + (uint8_t)(c1 >> 8);
  newB = (mB * p) + (uint8_t)(c1);

  return ((uint32_t)newR << 16) | ((uint32_t)newG <<  8) | newB;
}