WS2812 LED string demo with attiny10

[ccrause]

Bought an addressable RGB LED string (WS2812 compatible) and wanted to test whether an attiny10 can manage bit-banging the protocol and generate some effects. Demo code here. Attached gif shows the t10 in action driving the LED string.

Edit: updated link.

[ackarwow]

Many thanks for your excellent pascal library for AVRs! However it seems, that this link is broken...

[dseligo]

Many thanks for your excellent pascal library for AVRs! However it seems, that this link is broken...

He changed it a little, here is new link: https://github.com/ccrause/fpc-avr/tree/master/src/examples/ws2812-led

To ccrause: this looks great, I am looking forward to try this code.

[ccrause]

Many thanks for your excellent pascal library for AVRs! However it seems, that this link is broken...

He changed it a little, here is new link: https://github.com/ccrause/fpc-avr/tree/master/src/examples/ws2812-led

To ccrause: this looks great, I am looking forward to try this code.

Thanks for posting the correct link dseligo. I targeted the attiny10 initially, then refactored the code a bit to make it more general.

It is a lot of fun playing with these LED strings.

[MarkMLl]

There's a couple of issues with those LED strings, plus one with potential users :-)

(I'm sure everybody here present is aware of that, but I'm writing more for people who arrive here via Google.)

The first string issue is that while the WS2812 is a 5V device, the WS2811 ** operates at 12V which is a substantial improvement for long runs.

The second string issue is that the WS2813/WS2815 ** has an additional bypass connection so that if an LED goes dud it only affects that one pixel, rather than killing everything else downstream.

The user issue is that a recurrent question in various foramina is "I've got 1,000 pixels hooked up to an Arduino: how do I get a 60Hz frame update?".

The bottom line is that you're best off connecting a single row of pixels to each controller chip, and having an additional chip select and RLL download protocol to control the entire display.

I think the project below is about the limit, so while it might be scalable physically it will need more controllers if it gets much bigger.

https://hackaday.com/2021/06/07/portable-digital-scoreboard-goes-anywhere/

** Numbers largely from memory.

MarkMLl