NO, I didn't check the polling script. I have problems in Zynthian with both Midi and the Encoder. Supposedly Zynthian have 4 encoders and a webconfig page to enable them. The first enc is linked to GPIO 100,101,102. The other three spaces are filled with -1 for not in use.Hi!
You can check your encoder with these instructions: https://palmopensource.com/hardware/io- ... rry-pi.php
No pullup needed, 2 pins connected to the MCP and the center pin to ground.
There are 4 more 23017 pins left for four switches I either do not use atm. The INTa and B are set to pins 36, 37, but the pins do not
corresponds from those indicated on the Slim page and those in the webconf. I don't know if this mismatch is preventing the enc to work.
Also doing pinctrl on pin 37 there is nothing in use, where there should be INTB, whilst pin 36 is used by something else.
I could resort to polling in extremis, I have nothing to lose, I might try that in the meanwhile, althought in reality I would never use it.
I would like to use interrupts because it's less cpu intensive.
Jumping back to the MIDI wagon, there has been other Zynthian users that after updating to Pi5 has seen DIN MIDI stop working, that so devs have rushed to fix some code, that has worked for some, but not for me...
So I'm back again with Pi5 Os Bookworm once again, after some days of no solutions with Zynthian Vangelis.
Apparently the guy who had MIDI DIN working again was using a pisound card that was getting midi through SPI--- and probably
a dedicated IC--- but it's just a guess...
I really would like to get DIN MIDI working first, I can only try to install a mod-ttymidi from Zynthian at this point to Pi Os, but I'm not that
hopeful it would work... we really need some guys at rpi to look at this...
The last time I checked the Midi RX pins on the Pi Side, my scope was dead at ground, as if the pin was actively decoupling anything going in...
I mean I brute forced a Midi signal that on the scope was having normal Midi bias and clock ticks just before connecting it to the Pi5 RX pin...
The signal completely went down to zero... that is not normal... like a protection mechanism... the RX pin should not have any pull down or other stuff going on but so it behaved on my scope... Idk if you have a Pi5 to check all of this, otherwise its really useless... you are just guessing this might work...
the I2C link you provided was edited in 2024, but here is a compelling line...
Here are the connections I've chosen, MCP to Raspberry PI 3 bus
.... that is... all of this was done on a Pi 3???
PS: Adafruit Blinka instructions also was last edited in 2018... that is prior to Pi5 introduction... I couldn't neither make that work...
I was trying to track down my procedure but when everything refused to work I discarded everything...
There's a lot more I tried but haven't written down, too much frustration..
I'm really near to completely give up on this project...
I'm sorry for your interface, it looks really good engineered, but it's like
some dark forces are preventing me to make any use of it...
Statistics: Posted by Zool64Pi — Sun Jun 15, 2025 8:03 am