I recently started learning assembly for my TI-84 PCE and was testing out some sprite routines I made. I converted the calculator to 8bpp mode, but I can't figure out what values for my colors I should load into the mpLcdPalette. I don't have access to any windows / linux computers btw, so I can't use convPNG (unless I'm blind and just didn't see a mac version).

Could someone explain how the 2-byte colors you load into the color palette are formed (out of rgb components)? Also, do they work the same way as the 2-byte pixel colors when the calculator is in 16bpp mode?

I was testing some binary values by changing one bit at a time, but the results just confused me. When only one bit was on, the red, green, and blue values were all greater than zero (I was testing this using CEmu and checking the 3-byte pixel colors on my monitor). I was under the impression that the memory for each pixel on the calculator gave 5 bits to red, 6 to green, and 5 to blue, but now I have no clue.

Any help is appreciated!

(I also don't have much experience with Cemetech, so I apologize if this is in the wrong topic)

Palette colors are 1555. 1 is the least significant bit of green, 5 bits red, 5 upper bits of green, 5 bits blue. You can configure the lcd to swap the red and blue if you want. You can also use the C CE toolchain routine gfx_RGBTo1555() function to approximate a "close enough" value from the input 8 bit red, green, blue.

You can also get a mac version of convpng by downloading a mac version of the C CE toolchain.