I am keen to try a vertical grid of 4 x 256 matrices, but I am not sure how best to wire them.
I have put two options below. The first is more logical, but the second (could be considered that it goes upside down every second matrix) would be cleaner wiring behind.
Would either option be more easy to map?
I’d like to be able to easily map every which way in the end…all the way down a single column, snaking down the columns row by row, etc…I’ve not really gotten into mapping, so not really sure if one wiring approach might be easier than the other.
Both are reasonable to write a map generator for. For the second one where the wiring is shorter, would you be taking the same kind of matrix and flipping it upside down?
Either way, the basic approach you’ll use is shown in one of the example map generators that’s included on a Pixelblaze. In the mapper, try loading the six-sided cube Multiple Panel Matrix example and see how it has one function for generating a matrix, then calls it once with different spatial orientations for each matrix being connected together. That’ll be the same approach you use.
Your first wiring approach is probably the simplest because you’ll call the same map generator function four times, but with a different vertical offset each time.
The only tiny possible downside to the second option is that the LED chips (die) will be in slightly different locations within the package between the upside down panels. If you have any kind of diffusion going on that won’t be an issue at all.
There’s a multi-matrix mapper example that is what I think Jeff was thinking of. It supports rotation as an argument too, so mapping either should be a snap.
