Seems you are going a round in circles.....
Actually, I felt bad becuase I didn't post in a couple of weeks but wanted to say I had managed to make it work on another phone. yes it would be nice to have per pixel coverage without processing it first but it's just a project that I started years ago and managed to get it so much further than I ever thought possible.
i am interested in your sampling method, I honestly thought a pixel by pixel reproduction was going to be the simplest but in reality (and what has been stated), the core canvas aspect would need revisions and a long term plan. This project all started from some lights that I coded to write text, I designed my own font 8 pixels high. This year (and after playing with AI2), I learned I could read pixels from an image. thankfully, I have it mostly working on another phone and other things are letting me down (my poor BLE data sending code!)
Suffice to say, I've had a lot of fun then coding wit the camera to take a photo to reproduce in RGB LEDs. I convert to a canvas image of 64x64 but as you are all saying, it'd be best to sample with the limitations of the canvas functionality.
2 Likes
These photos are from a propeller display?
1 Like
I know it's not a proper question here but i'm just curious, does Kodular handle this the same way? Maybe @Anke or someone else who is familiar with that too can answer my question.
I just admit, I tried another free version of this and found the exact same problem (presumably they just built theirs atop of AI2 - AI2, in my opinion, is still the best)
My only solution without the excellent ideas listed (sampling areas) is to use another phone.
The photos are from LEDS mounted to a server rail lol!