Again, I asked ChatGPT if we could make a controller that could do what WLED does but without all the bloat. I ended up creating a cheap ESP8266 controller called ZIPP (Zero-Overhead Intelligent Pixel Protocol) that's actually better and faster than WLED.
Well, I kept asking more questions and one thing led to more. I've got enough in the works that I decided that I needed to start documenting before I started forgetting things. And what better place to document that the DIYChristmas Wiki?
You can read about what I've been working on here.
That is one heckuva strong project you're working on, Ernie! Take care of your health so you'll be able to finish it!
A little update on my ZIPP projects.
I've finished ZIPP-R (ZIPP-Relay) and am currently working on ZIPP-P (ZIPP-Pixel).
ZIPP-R is a wireless relay control system built around the inexpensive ESP-01S and optocoupled relay boards. It uses a Master/Slave arrangement to control conventional Christmas lights and other decorations over Wi-Fi.
The system was designed to support up to 254 individually addressable relay controllers per channel, with the possibility of using multiple Wi-Fi channels to expand well beyond that. Since relay control requires very little data, the network overhead is minimal.
ZIPP-P is my replacement for WLED, designed specifically for receiving DDP data and driving WS2811 pixels using the ESP8266 D1 Mini.
By eliminating unnecessary overhead and concentrating on efficient pixel processing, we've been able to substantially improve performance.
In our testing, WLED struggled around 400 pixels at 40 FPS, while ZIPP-P has successfully demonstrated 800 pixels at 40 FPS on the same inexpensive ESP8266 hardware.
We've also verified an actual output frame rate of 40.000 FPS, not simply a software setting or estimated frame rate. In one extended endurance test, the controller processed approximately 2.83 million complete DDP frames over nearly 20 hours without a single rejected packet, incomplete frame, or Wi-Fi disconnect.
We've also successfully tested Wi-Fi communication over approximately 1,100 feet, opening up some interesting possibilities for distributing controllers over larger areas.
Currently, I'm experimenting with recording and compressing DDP sequences directly into flash memory, with the goal of allowing controllers to play sequences locally without continuously receiving data from the show computer.
Some of the compression results have been surprisingly promising, and we're exploring several different methods to see just how much data we can eliminate.
Eventually, I'll update the Wiki with the technical details, firmware, and documentation. Unfortunately, between developing these controllers and getting my Halloween and Christmas displays ready, I haven't had much time for documentation.
For now, I just wanted to let everyone know that these projects are still very much alive and progressing.
And perhaps the most interesting part is that all of this is being accomplished with inexpensive ESP8266 controllers. Although Espressif has discontinued them, they're still widely available for around $1–$2 each.
It's amazing what these little controllers can do when you eliminate the unnecessary overhead. 🙂
Really interesting project, Ernie. I am no longer doing a display. But it is interesting to see what is happening in the world of light displays and controls!
beeiilll wrote:
... it is interesting to see what is happening in the world of light displays and controls!
I don't know if this is interesting to others. Some people have dismissed the old ESP8266 microcontrollers as old and obsolete. I've investigated the newer ESP32 controllers and though they're much faster, processing speed isn't the issue. It turns out that it just efficient use of the internal UART and I got that part working.
Currently, I'm working on minor data compression, currently about 2.339:1 or 57.25%, and storage.
