Excellent! What is the protocol? Hopefully something self-clocking. The WS2812 devices are a complete PITA because of the their fixed timings. Any spec sheets available? And what sort of price and availability?
-
-
-
Cheers Bunnie!
-
-
Many thanks for that
@drtune! I wished i’d seen this before writing tight assembler code to meet the timing criteria in the spec :-( Even so, i’d still love to see a self-clocked protocol, so that I can transmit NeoPixel data as fast/slow as I need to achieve desired frame rate. -
SPI or I2S interfaces are trivially easily used to drive WS strings btw. You just output 4 bits per data bit, at a clk of (I forget, 330ns I think). As long as you can keep the fifo full it works fine - I2S tends to be better under linux; required to be glitch-free unlike SPI
-
Once again
@drtune thank you so much for sharing your knowledge; VERY greatly appreciated.
End of conversation
New conversation -
-
-
CC
@LoialOtter you were asking about 2020 LEDs? I think you wanted just pure RGB non addressable ones but still you might find this interesting. ^.^ -
Tweet unavailable
-
Timing is not really a big issue here, as
@LoialOtter will be driving them from an FPGA… I recommend looking into those, they are fun :D -
Hi Piotr, yes absolutely. I bought myself and iCEstick a while back and recently started “programming” it. My purpose in getting it was to model hardware to map from I2C/SPI to NeoPixel devices. As someone with an EE background but long s/w experience, Verilog melted my mind.
-
You might look at my pygmyhdl tool that does simulation & synthesis in a Jupyter notebook. 20 minutes from sfw download to a working LED blinker on the iCEStick. No Verilog. https://xesscorp.github.io/pygmyhdl/docs/_build/singlehtml/index.html …
-
Ok Dave, you shamed me into actually reading (rather than bookmarking) your pgmyhdl docs; absolutely excellent. I learned so much, especially the hierarchy example with the comparison of gates used, and the PWM variants. Many thanks! I'll certainly recommend them to FPGA newbies.
-
Thanks! I learned shaming from my mother. One word of warning: pygmyhdl only works with Python 2.7 and 3.5. I haven't gotten around to porting it to 3.6 or 3.7. The underlying MyHDL works everywhere, however.
End of conversation
New conversation -
-
-
Wow, addressable 2020s? Can't wait to see matrices using those.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Please please please may we get some video of the fab/etc WorldSemi uses to make WS2812s etc. I expect
@strangepartscom would be keen to bring his camera (and respect secrecy). Thank you! - 1 more reply
New conversation -
-
-
Rad, meeting with the guy behind worldsemi sounds amazing, can't believe they managed a 2020 package. Bravo!pic.twitter.com/Ddl8cspQgg
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Very cool. I've used the APA102 2020 ICs, and they're very pretty. But the quiescent current is high 1.0-0.7mA, and furthermore unspec'ed.
-
The upcoming APA104 1515 are said to have 0.3mA quiescent current, but the APA102 1515 ICs don't have the Qc specified at all.
-
Overall, even 0.3mA is rather high for wearables, and I would like to see something below 0.1mA.
-
An alternative solution would be if the APA10X series of LEDs behaved better on power on, and didn't just light up in a random color, but rather waited for a valid signal.
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.