Note: This behaviour isn't visible in the simulator (perhaps because the simulator is only showing the number of LEDs configured by the create() call). In the real world this would only be noticed if neopixel.create() was called with a lower number of pixels than actually contained on the physical strip.
Expected Behaviour (on real strip): Should see pixels 0,1,2 all red, only
Actual Behaviour (on real strip): As expected except pixel 3 is lit green and stays on after the clear()
Notes: I've tested with various configured lengths, whether 3, 15, or 60 pixels, the library call will always light this extra (last + 1) LED green. This discounts the possibility that the issue is simply a bad pixel in my hardware.
Simplest repeatable failing test: https://makecode.microbit.org/_YmYf9i6AzAxA
Note: This behaviour isn't visible in the simulator (perhaps because the simulator is only showing the number of LEDs configured by the create() call). In the real world this would only be noticed if neopixel.create() was called with a lower number of pixels than actually contained on the physical strip.
Expected Behaviour (on real strip): Should see pixels 0,1,2 all red, only
Actual Behaviour (on real strip): As expected except pixel 3 is lit green and stays on after the clear()
Notes: I've tested with various configured lengths, whether 3, 15, or 60 pixels, the library call will always light this extra (last + 1) LED green. This discounts the possibility that the issue is simply a bad pixel in my hardware.