"Blinker!" Megabrite/Shifty VU LED Wall Project

I have some questions about performance and this is probably just my inexperience with the platform and with code optimization so hopefully it can be made faster.

  1. When I went from 48 lights to 64 lights I saw a huge decrease in performance. From ~ 25ms / loop ~ 45ms / loop. I probably just need to optimize my code but are there better platforms to be working on for this other than arduino? It would be nice to get refreshes down below 20ms/loop.

  2. My 4X20 green LCD screen needs to have delays added whenever a refresh is done otherwise it goes all wonky and this slows the whole light panel down when I'm changing parameters in the program because it's updating the LCD screen. I guess the same as question 1 – are there any platforms that are multi-threaded or is there a way to get around this "refresh" wait?

I eventually want to move up to 128 lights but I'm not sure how I'm going to do it with the current performance I'm getting. Is it possible to run arduino's in parallel maybe? I could hook up 2 or 4 and have them work together maybe.

I dunno. Let me know your thoughts.

Thanks,
R