Replies: 6 comments 1 reply
-
Hi |
Beta Was this translation helpful? Give feedback.
-
Thanks for the quick reply. Here is the config: master...ConorIA:HyperSPI:master My total LEDs are 216. Top/Bottom = 70 When swapping between the ESP32 and the ESP32-S2, all I do is changes the baudrate. Otherwise the layout doesn't change at all. |
Beta Was this translation helpful? Give feedback.
-
Which baudrate? SPI speed? I generally do not recommend changing the SPI pinout because we may accidentally hit some important ESP32 "artery". Please activate Github Action and build your firmware on Github so I can upload it and test it on my esp32-s2 with the oscillator, although I do not promise when I will find the time for it. |
Beta Was this translation helpful? Give feedback.
-
Try also to compile firmware without using second segment and use for example GPIO4 (with still changed SPI for your connection). If that works its an I2S interference issue for the second pinout with this SPI configuration. Single segment is using different rendering method. |
Beta Was this translation helpful? Give feedback.
-
Thanks. Here is the running action: https://github.com/ConorIA/HyperSPI/actions/runs/12606377596/job/35136410940 I have to say though, unless you are truly curious, it might not be worth looking into further. I've got the WROOM all soldered up onto a project kit and I managed to increase the baudrate on the LED Hardward setup page to 20000000. I also realized that the serial monitor in PIO was just defaulting to the wrong baudrate - I get good output when I run with -b 115200 flag. I think this is good to go for me. |
Beta Was this translation helpful? Give feedback.
-
115200 is default baudrate https://github.com/awawa-dev/HyperSPI?tab=readme-ov-file#performancedebug-output but this speed is not important. because it is used only for debugging If your second board is working that's fine. esp32-s2 is a "stripped" version of the good old esp32 and although it performs its functions correctly with one core, I can't rule out that it suffers from some bad tempers with a changed configuration. I spent a lot of time to make I2S multichannel work correctly with the standard configuration because it behaved a bit differently from the classic esp32. |
Beta Was this translation helpful? Give feedback.
-
Hi,
I've been having fun and a lot of stress trying to get my system upgraded from a single segment running direct if a Pi Zero W (bad, I know) to a two-segment system with HyperSPI and a Pi 3B that I had kicking around.
Things are currently working pretty well with an old ESP32 Wroom. I have to keep the baudrate low (1Mhz, but it appears to be working just fine). The only bugaboo is that the serial monitor outputs garbage.
I had originally purchased two ESP32-S2 mini boards and this is the point of my question. I've found that no matter what I try, the only pin that wants to work is Pin 2. It doesn't matter if I call this -DDATA_PIN=2 or -DSECOND_SEGMENT_DATA_PIN=2. Pin two works for both either segment. However, no other pin appears to be working. I'm sure that the board isn't the problem, because it was working with HyperSerialWLED fork. Is there something in the HyperSPI code that is defining how the pins are defined?
Maybe unrelated, but I was also having trouble getting SPI communication working on the hard-coded pins, so I remapped the pins in main.cpp. SPI communication working great now and serial monitor outputting greeting info.
Is it worth pulling my hair out more to try to get the ESP32-S2 working, or is continuing with the ESP32-WROOM my best bet?
Beta Was this translation helpful? Give feedback.
All reactions