Trellis M4 open source grid controller

[effiks]

hi @okyeron, great news. I flashed the UF2 file on one of my Trellis M4 but Fates will not see it as Midi or Grid device (under System -> Devices). I use a normal (working) USB cable. Any idea?

[okyeron]

Right - you will need to do the diy device fix for Fates to see it as a grid. (with this firmware it’s grid only - no midi) Details here

[effiks]

Thanks @okyeron. With this firmware update it is now working and I slowly understand how Loom works.

Very cool. Now the Fates becomes even more useful and fun to play with.

@effiksmusic

[effiks]

@okyeron would your code also work with the Adafruit 8x8 NeoTrellis Frather M4 Pack?

The Neotrellis M4 is great for travelling and mobile use. But now that I know what to do with a Grid I like a bigger on of cause for my “home studio“.

[okyeron]

Yeah it should, but I don’t have a feather M4 to compile against right now to make a UF2 file.

To compile yourself you would need to change a couple core Arduino files for TinyUSB. I may also need to update the repo

Maybe I’ll get an M4 this week to test

[effiks]

@okyeron no hurry, I have to order the Adafruit 8x8 NeoTrellis Feather M4 Kit Pack from overseas and it is not in stock at adafruit.com at the moment. I wanted to know if it is possible before ordering. It can take weeks before I have the kit. Or I have to order the different parts from different shops in Germany.

I don’t mind to compile it myself, if the modified core Arduino files for TinyUSB are available in future. Then I cloud change also the color of the LEDs, I like colorful light on black hardware .

[okyeron]

So far the small M4 is a proof of concept, so the others should work. But if you want to be sure before ordering, give me a week and I’ll do some testing.

[effiks]

Great. As I said, no hurry (there are other eurorack modules waiting to be purchased before. No I don’t have GAS, I can stop buying tech and modular gear any time, really).

[jedgar]

I’ve created CircuitPython code for triggering sequences and samples on the 1010 music blackbox using the Trellis M4: https://github.com/jedgarpark/JEP_NeoTrellis_Blackbox_Triggers (also a version there for a 4x4 NeoTrellis driver + Feather M4).

[thynk]

Happy to report mine works really well with Fates.

It’s SO portable!

Thanks again for all of your hard work @okyeron!

[briangriffith]

Can you do a the same thing for a Norns Shield?

[okyeron]

EDIT - Please be aware this is a hack/workaround and is not officially supported.
Proceed at your own risk

See this message for details

DIY monome compatible grid w/ Adafruit NeoTrellis

OK, I remembered the steps that were missing - a couple of new packages were added, but are not in the shield kernel yet (these are already installed on fates): If you’re on norns shield with 200218 do the following (if you’re on another version, update first). cd ~/ sudo apt-get update sudo apt-get install libncurses5-dev libncursesw5-dev wget https://raw.githubusercontent.com/okyeron/fates/master/install/norns/files/device/device_monitor.c cd ~/norns git pull git submodule update --init sudo…

[briangriffith]

Ah cool, thank you! I tried it and it did something weird to the OS, particularly removing the “Audio>” in the “System>” so I’m just going to reflash the shield. I appreciate the quick response though!

[okyeron]

Could be something odd with the most recent norns code (?)

The steps above should put you on the “master” of the norns codebase as it is on github. In other words, this is ahead of a full release. Sometimes there are bugs.

Gonna test on my shield here shortly

EDIT - @briangriffith the AUDIO menu is removed/moved in the most recent code on github

[dan_derks]

got my m4 in the mail yesterday – it’s a sweet lil thing

unsure if folks have had this experience, but running into some interesting behavior with norns where about 100-150 seconds after it’s plugged in and running simple MIDI into a WIP script, all processes seem to come to a dead stop – they restore if i unplug the m4. running dmesg after, i’m seeing usb 1-1.3: urb status -32.

figured i’d put feelers out there to see if folks have run into similar weirdness . happy to assume it’s the WIP script, but this doesn’t happen with an OP-Z or similar controller, so would be good to hear any experiences.

[okyeron]

Which code are you running? I could do some testing later to see if that happens on fates/shield.

[dan_derks]

thanks steven!

able to repro with these steps:

• load the midi_keypad code onto the M4
• load molly_the_poly on norns
• plug in the M4, start playing notes, all’s well for a bit
• after 2 minutes-ish, the norns UI freezes but networking is still available
• unplug the M4, all returns to working

stock norns on 200712, super curious if this is reproducible on anyone else’s units. thanks so much!

here's the output from dmesg

[   62.972210] usb 1-1.1: new full-speed USB device number 4 using dwc_otg
[   63.112307] usb 1-1.1: New USB device found, idVendor=239a, idProduct=802f
[   63.112315] usb 1-1.1: New USB device strings: Mfr=1, Product=2, SerialNumber=3
[   63.112320] usb 1-1.1: Product: Adafruit Trellis M4
[   63.112324] usb 1-1.1: Manufacturer: Adafruit LLC
[   63.112328] usb 1-1.1: SerialNumber: D01B51C85338533335202020FF132720
[   63.138445] cdc_acm 1-1.1:1.0: ttyACM0: USB ACM device
[   63.140656] usbcore: registered new interface driver cdc_acm
[   63.140665] cdc_acm: USB Abstract Control Model driver for USB modems and ISDN adapters
[   63.165882] usbcore: registered new interface driver snd-usb-audio
[  165.415942] usb 1-1.1: urb status -32
[  165.421935] usb 1-1.1: urb status -32
[  165.424954] usb 1-1.1: urb status -32
[  165.427965] usb 1-1.1: urb status -32
[  165.430939] usb 1-1.1: urb status -32
[  165.433952] usb 1-1.1: urb status -32
[  165.439961] usb 1-1.1: urb status -32
[  165.442953] usb 1-1.1: urb status -32
[  165.445939] usb 1-1.1: urb status -32
[  165.448954] usb 1-1.1: urb status -32
[  165.451957] usb 1-1.1: urb status -32
[  165.454953] usb 1-1.1: urb status -32
[  165.457963] usb 1-1.1: urb status -32
[  165.460955] usb 1-1.1: urb status -32
[  165.463963] usb 1-1.1: urb status -32
[  165.469951] usb 1-1.1: urb status -32
[  165.472957] usb 1-1.1: urb status -32
[  165.478949] usb 1-1.1: urb status -32
[  165.481941] usb 1-1.1: urb status -32
[  165.484943] usb 1-1.1: urb status -32
[  165.487949] usb 1-1.1: urb status -32
[  165.493950] usb 1-1.1: urb status -32
[  165.496940] usb 1-1.1: urb status -32
[  165.499943] usb 1-1.1: urb status -32
[  165.502950] usb 1-1.1: urb status -32
[  165.505952] usb 1-1.1: urb status -32
[  165.508950] usb 1-1.1: urb status -32
[  165.511951] usb 1-1.1: urb status -32
[  165.514955] usb 1-1.1: urb status -32
[  165.517958] usb 1-1.1: urb status -32
[  165.520953] usb 1-1.1: urb status -32
[  165.523954] usb 1-1.1: urb status -32
[  165.526950] usb 1-1.1: urb status -32
[  165.529955] usb 1-1.1: urb status -32
[  165.532952] usb 1-1.1: urb status -32
[  165.535945] usb 1-1.1: urb status -32
[  165.538941] usb 1-1.1: urb status -32
[  165.541942] usb 1-1.1: urb status -32
[  165.547955] usb 1-1.1: urb status -32
[  165.550940] usb 1-1.1: urb status -32
[  165.553951] usb 1-1.1: urb status -32
[  165.559951] usb 1-1.1: urb status -32
[  165.562950] usb 1-1.1: urb status -32
[  165.565941] usb 1-1.1: urb status -32
[  165.571943] usb 1-1.1: urb status -32
[  165.574948] usb 1-1.1: urb status -32
[  165.577940] usb 1-1.1: urb status -32
[  165.580950] usb 1-1.1: urb status -32
[  165.583980] usb 1-1.1: urb status -32
[  165.584929] usb 1-1.1: USB disconnect, device number 4
[  175.842650] sound midiC1D0: rawmidi drain error (avail = 431, buffer_size = 4096)

[okyeron]

hey dan…

Looking at this now and I think the dmesg output might be a red herring.

It’s likely to do with the M4 showing up as both a ACM device and as a MIDI device.

Do you see this error in Maiden - about 6-10 seconds after connect the M4:

>> ttyACM found, but not a crow
dev_list_add: error allocating device data

?

EDIT - testing now on shield with molly the polly and I’m not seeing a UI freeze (although - I was just on an older version of molly… updated and still not seeing UI freeze, but did get a random stuck note or two.)

[dan_derks]

thank you so much for helping debug — i’m curious if this isn’t due to kernel differences between stock / shield / fates, as some offlines convos suggest.

if you’re having a slow night and have insights from your experience, the arduino code is my first dip into c — in the “expressive midi” scene it seems that the hardware state is polled (and a message is sent?) every 10ms rather than solely executing code when hardware is actually engaged. am i reading this totally wrong?

long story short, i’m wondering if things are getting clogged

[okyeron]

@dan_derks

i’m curious if this isn’t due to kernel differences between stock / shield

in theory the kernel is the same between OG norns and shield.

@dan_derks

in the “expressive midi” scene

Where is this sketch? Do you mean the midi_keypad.ino discussed on the Adafruit guide?

if I’m reading the library right, it’s only sending data if there’s something pending to be sent - not constantly flooding info every 10ms

void Adafruit_NeoTrellisM4::sendMIDI(void) {
  if (_midi_usb && _pending_midi) {
    MidiUSB.flush();
    _pending_midi = false;
  }
}

_pending_midi is only set to true when an event happens (noteOn, etc)

---

Additional thought - have you updated all the Arduino libraries for the Neotrellis M4?

FWIW - from your original dmesg output the only thing that looks odd to me is the [ 175.842650] sound midiC1D0: rawmidi drain error (avail = 431, buffer_size = 4096) error.