Does the WS2812b have to be clocked @ 800KHz or can it be clocked at a slower rate as long as the relativity of pulse widths is maintained? I realize faster is better … just curious …
It’s not about relativity of pulse widths. A 0 is a pulse that is high for around 250ns, a 1 is a pulse that is high for around 875ns, and the total length for a pulse needs to be at least 1.25us. However if the length of a pulse goes more than 4-5us, everything latches/resets.
Darn … Dan how in the world do you pull it off in software ?
I thought, because I have masochistic tendencies, I’d see if I could implement the timing via PWM on Zero … even w/ a clock speed of 48M … the ISR (with just one line of debug code currently) draws the platform to it’s knees such that supporting code (WIFI/MIDI) doesn’t get a chance to run …
@David_Brenchley i tried doing this using the Intel Galileo a while back… http://www.elec-tron.org/?p=663 allot of what i wrote in that article, in hindisght, is TOTALY wrong, but it has a logic to it. I was using the DATA pin of the SPI port driven at a specific frequency to construct data for the WS2812