After understanding the driver problem, I was certainly happy to see a matrix working, but I wondered how to make several of these matrices work together in harmony.
So I had to use the soldering iron to correct the design errors of the card manufacturer.
It is true that the brand is no longer very anonymous!
Due to various soldering problems, one card is no longer controllable and another has lost control of the three lower lines of the matrix.
Programming
Although the hardware gave me a bunch of various problems, programming the card was not much easier.The card being quite old, it has a system that takes the input data for itself, and as soon as the input data exceeds the size of the "buffer", it is emptied into the neighboring card.
Thus, if we want to change the color of the first pixel of the first matrix, we could send a single data, which would then be interpreted as the first pixel.
However, this would shift all the pixels of the entire matrix by one box.
In our example, to solve this, the only solution is to return all the pixel values \xc3\xa2\xc2\x80\xc2\x8b\xc3\xa2\xc2\x80\xc2\x8bby modifying that of the first.
I therefore had to design a system to manage an array of pixels and an update() function sending the values \xc3\xa2\xc2\x80\xc2\x8b\xc3\xa2\xc2\x80\xc2\x8bof each pixel.
I also undertook the creation of a setPixel function to calculate and store the index of a pixel to given X and Y coordinates.
After many failed tests, I finally get a success, writing scrolling text.
As you can see, a problem persists, because a kind of "flash" appears at each update.
I provide you with the source code of my program, if one day you need it, who knows?