I did use those instructions to compile from source. Can I go back and run apt install serialosc on top of that?

Yeah, just run waf uninstall from both libmonome and serialosc source directories to avoid having 2 serialosc binaries in the system.

1 Like

i’ve added the user to the dialout group and have write access to the USB device and received the expected output from serialoscd confirming the monome is connected. However, when running grid-studies-1.py taken from : https://monome.org/docs/grid-studies/python/
no message appears and the monome appears inactive.

is serialosc running when you launch the grid-studies script?

Yes serialosc is running in the background before running the python script

Ok, so the shouldn’t be any output from the script and the grid should react to keypresses. If that doesn’t happen, can you check if pymonome is at the right version, i.e. what’s the output from command pip3 show pymonome?

The terminal reads the following:

Name: pymonome
Version: 0.9.0
Summery: a monôme serials client in python
Home-page: https://githib.com/artfwo/pymonome
Author: Artem Popov
Author-email: artfwo@gmail.com
License: MIT
Location: /home/pi/.local/libpython3.5/site-packages
Requires:aiosc

your setup is good. does the grid react to keypresses or are there any messages in the terminal when you push buttons?

The program runs but it looks like it just sits there as I press monome buttons before I use command interrupt.

when I plug the monome in it also says: [ 1429.033901] Under-voltage detected! (0x00050005). Does this affect anything?

Okay, can you try the following?

cat ~/.config/serialosc/*.conf
OR
cat /var/lib/serialosc/*.conf

use the latter command if you’ve installed the debian package. you should see the contents of serialosc grid state file, which looks like this:

server {
  port = 12988
}
application {
  osc_prefix = "/monome"
  host = "127.0.0.1"
  port = 58986
}
device {
  rotation = 0
}

make sure you have liblo-tools package installed. then run the following command replacing <port> with the server port value from the grid state file and <prefix> with the application osc_prefix option value:

oscsend localhost <port> /<prefix>/grid/led/set iii 0 0 1

for example:

oscsend localhost 12988 /monome/grid/led/set iii 0 0 1

the led should light up. if it doesn’t, but serialosc reports the grid as connected, something is not right.

additionally you can run oscdump with the application port as the argument, for example:

oscdump 58986

push some buttons, you should see /grid/key events in the terminal. if you do not - again, something is wrong with your serialosc setup, system configuration or power/hardware.

I’ve just tried using oscsend and the led does light up. However when I run the oscdump command, using the application port as its argument, I get nothing back from the terminal when pressing buttons, the program just hangs. So it seems that one way communication from host to device is working but the signals from the device are not getting back to osc application, is this more likely to be hardware or software issue?

more likely a hardware issue, serialosc is known to work well on the RPi. try following this troubleshooting guide. hope this helps!

I’ve since managed to make oscdump print the monome key presses, so now input and output are both functioning correctly. I then tried to run grid-studies-1.py but still the keys do not light up in response to key presses. The main function for the program as defined here https://monome.org/docs/grid-studies/python/, does not contain any calls to the grid_key functions, is there something which needs to be added to main whereby the grid_key function is passed to the event loop, such that it is called upon the event of a key press?

Running the script is definitely having some effect on the connected monome since if the script is run whilst also running oscdump, oscdump’s output changes from something like this “/monome/grid/key iii 12 4 1” to this "/sys/host s “127.0.0.1” or “/sys/port i 59898”

No need to add anything manually, the library connects the events to the handler methods under the hood. It’s strange this doesn’t work for you, because I’m using the same version with my RPi and the grid studies example runs well. Have you tried running the other examples? Or examples from the pymonome git repo?

Thanks for this thread i am grateful for this community, I am going to work on pd patches for the raspberry with a piSOund and a grids today so i am excited that this is here

TY!

1 Like

I have two versions of grid-studies-1.py the one from the zip file download and theres another one with variations from the zip download on the python grid studies page: https://monome.org/docs/grid-studies/python/
I’ve tried making both the files executables
when I execute the files like so: ./grid-studies-1.py
I get a couple or errors I hadn’t noticed before which say:

import: command not found
syntax error near unexpected token ‘(’
‘class GridStudies(monomer.App):’

when I run the files as:

python3 grid-studies-1.py

the file from the zip has an 'attributeError: module ‘monome’ has no attribute ‘Monome’

the file online doesn’t have this error of course where the code is adjusted and so runs but still with no result from the monome.

but does this have anything to do with the ‘import: command not found’ error that occurred using ./grid-study-1.py instead of python3 grid-study-1.py ?

I still haven’t got a response from oscdump. my also runs but just sits there. I noticed during the trouble shoot you’ve linked I got some errors that stood out?

ftdi_sio ttyUSB0: usb_serial_generic_read_bulk_callback - rub stopped: -32

ftdi_sio ttyUSB0: error from flowcontrol urb

I don’t know what they mean but perhaps they have something to do with oscdump not running?

…not sure how but my oscdump is working now

Success! the hello.py patch works, I’m yet to double check why the grid1 doesn’t

1 Like

dear all,
to my knowledge this problem is not completely unfamiliar, yet i haven’t found a solution anywhere.

i have serialosc (or better to say “serialoscd”) running on an rPi 3. tested unter jessy and strech, i’m encountering a persistent problem: it’s very unreliable. the only thing i can observe consistently is that on every first boot of the rPi, serialosc seems to be crashing. ssh’ing into the pi and running “top” states a CPU usage of ~100% (see screenshot).

does anybody know how to fix this? or even how to find out, what’s going on? i’m terrible (!) with linux :blush:
thanks :cat2: