Agreed. Since I need a simple point-to-point comms, that should be sufficient.Basically it's using the Pico W as a promiscuous Ethernet Packet Sniffer.
The thing is that I want to drive those HUB75 panels, which are daisy-chained. The Pimoroni boards I purchased for these panels come fully prepare to be used with these panels, including all the connectors. And it comes with the software to pass the data to up to 4 of these chained panels. I do not really have the time to build my own solution based on a different microprocessor board, though if it turns out that the PicoW is unreliable, I'll have to go down that path (but that'll cost me time I'd rather spend on the creative part generating the images on a Pi 5 using multiple cameras).If accessing the CYW43 proves to be more hassle than it's worth you could consider hardware which is RP2040-based but uses a different wireless chip; I believe Adafruit, WaveShare and WIZnet may have viable candidates. Alternatively you might be better off with something ESP32 or ESP8266 where all this is old hat.
Right, wired would indeed be better, since all components are installed next to each other. In fact, I might skip the network entirely and see if I can have the Pi5 send the data over SPI or something directly to the Picos. But since I have to use four Picos, I'd have to find 4 comms channels to talk to the 4 Picos. Don't know the Pi5 enough yet to know if that's possible. Any more suggestions are welcome!Does data need to be sent by Wi-Fi or would a wired-network be acceptable? A Pico and ENC28J60 would seem perfect
Yeah, that's what I am afraid of. I don't like to use such languages (with dynamic creation of objects) where I have little control over the dynamics behind it.Everything seemed to be working fine yesterday. Today I am getting lag of up to 80 seconds from sending a burst of four short 50 byte packets to the Pico W receiving them all.
Statistics: Posted by tem_pi — Fri Jul 26, 2024 2:59 pm