Problem connecting MiniLab on Linux OS

[acbakhit]

I am using version 1.1 of the MCC software:


Also, I'm not using wine, I'm using MCC on a Windows 10 PC
I've tried several different methods of installing the firmware: I've tried plugging in the MiniLab before launching and clicking upgrade in MCC, I've tried plugging in the MiniLab after launching MCC but before selecting upgrade, and I've tried plugging in the MiniLab while holding shift before launching MCC-- I haven't been able to find a method that allows me to select a *.led file

[Linuxmusician01]

I am using version 1.1 of the MCC software:
[...]
Also, I'm not using wine, I'm using MCC on a Windows 10 PC
I've tried several different methods of installing the firmware: I've tried plugging in the MiniLab before launching and clicking upgrade in MCC, I've tried plugging in the MiniLab after launching MCC but before selecting upgrade, and I've tried plugging in the MiniLab while holding shift before launching MCC-- I haven't been able to find a method that allows me to select a *.led file

That is a big bummer! Unbelievable! Is there a forum on the Arturia website? Can you contact customer support? This is ludicrous. If I' not mistaken they promise that the MiniLab is standard Midi compatible. I appears that without the firmware update it is not.

There's another thing that you might try which I would advise against greatly because it might brick your Minilab beyond repair. Rename the BIN file to a .LED file. But do that only if the Arturia people cannot help you and if you cannot sell your Minilab 2nd hand on the internet. After selling the thing you might want to "rebuy" it (if you really like it) but this time make sure that they can prove in the store before buying that it's a firmware 3 machine.

Arturia should be ashamed of themselves. Such a big name, such a weird problem that it's not really Midi compatible.

[acbakhit]

I actually tried selling it to a local store this weekend, they wouldn't have it-- they said they've been having issues with Arturia products lately... big suprise. I also tried renaming the *.led file to a *.bin file, and it got to about 93% of the update before it error'd out (it didn't brick the minilab though, if you hold shift and plug it back in you can load 2.1.0.2 back on it). I've also emailed support asking how I may update to 3.0.0.5 since this seems to be the version supporting Linux, and they responded as follows:
"As stated on the page, your Minilab can't be updated on the .led firmwares. It has to be on the 2.1.0.2
This depends of your serial like explained in the Minilab section on this page"
Arturia also has a support forum I've posted to with no luck:
http://forum.arturia.com/index.php?topic=87081.0
I've really used up all my resources here-- to be honest, I really regret buying this thing and only wish a could get a couple bucks to get rid of it. I won't ever be buying Arturia again.

[Linuxmusician01]

I actually tried selling it to a local store this weekend, they wouldn't have it-- they said they've been having issues with Arturia products lately... big suprise. I also tried renaming the *.led file to a *.bin file, and it got to about 93% of the update before it error'd out (it didn't brick the minilab though, if you hold shift and plug it back in you can load 2.1.0.2 back on it). I've also emailed support asking how I may update to 3.0.0.5 since this seems to be the version supporting Linux, and they responded as follows:
"As stated on the page, your Minilab can't be updated on the .led firmwares. It has to be on the 2.1.0.2
This depends of your serial like explained in the Minilab section on this page"
Arturia also has a support forum I've posted to with no luck:
http://forum.arturia.com/index.php?topic=87081.0
I've really used up all my resources here-- to be honest, I really regret buying this thing and only wish a could get a couple bucks to get rid of it. I won't ever be buying Arturia again.

I completely agree with you. It's a shame they bring out two different products under the same name in the same box. Only to discover on their website that full Midi compatibility depends on the letter N in the serial number! They should emphasize more clearly that this keyboard is a dongle for the classic synth emulation software (Analog Lab) that comes with it. Lets face it: the product one buys is 'Analog Lab', NOT the hardware keyboard they advertise.

When did you buy the keyboard? Because I'm wondering if they currently still sell this product with the wretched firmware that renders it useless as a real midi keyboard. I bought mine second hand an apparently I was lucky to get one w/ the right firmware already on it because my Minilab keyboard also has a serial without an N in the serial number. So even when you know the serial number then you still don't know what you've got.

[might become new topic...]
But now for something important. Where might we find a cheap portable midi keyboard w/ a lot of knobs to use in classic synths in LMMS and Bristol? Most nice little k.brds. have about 8 knobs. That's not enough for me. Some expensive one's don't even have a lot. Arturia is out of the question, unless you're lucky enough to own one w/ the right firmware or a serial number that allows it's firmware to be upgraded.

[v.martyanov]

I have WORLDE MINI (I think it's a complete clone of Subject device) with the same VID:PID and the same problem. And I found a solution!

The problem is in buggy USB-configuration of the device: Configuration Descriptor contains 3 in it's iConfiguration field, but the device CAN'T return string with such ID! This string is a configuration name and, as far as I know, it's not necessary. The Windows doesn't ask the device for it, but the Linux does and the device is resetting after this request.

The simplest way to solve the issue is to comment a few lines in drivers/usb/core/message.c:
if (cp->string == NULL &&
!(dev->quirks & USB_QUIRK_CONFIG_INTF_STRINGS))
cp->string = usb_cache_string(dev, cp->desc.iConfiguration);

I did it and everything (I hope!) works fine. I think it would be better to check VID:PID before this code and avoid requesting iConfiguration...

[Linuxmusician01]

I have WORLDE MINI (I think it's a complete clone of Subject device) with the same VID:PID and the same problem. And I found a solution!

The problem is in buggy USB-configuration of the device: Configuration Descriptor contains 3 in it's iConfiguration field, but the device CAN'T return string with such ID! This string is a configuration name and, as far as I know, it's not necessary. The Windows doesn't ask the device for it, but the Linux does and the device is resetting after this request.

The simplest way to solve the issue is to comment a few lines in drivers/usb/core/message.c:
if (cp->string == NULL &&
!(dev->quirks & USB_QUIRK_CONFIG_INTF_STRINGS))
cp->string = usb_cache_string(dev, cp->desc.iConfiguration);

I did it and everything (I hope!) works fine. I think it would be better to check VID:PID before this code and avoid requesting iConfiguration...

First of all: thanks a lot for the debugging!! Second: the only location where I can find the sub directory drivers/usb/core/ is in the kernel source code. I cannot find the file message.c. Do you mean that we have to recompile the whole kernel for this keyboard to work? Can't we compile a kernel module (that is: driver)?

[dissonantmonk]

Well this is finally fixed!!
As of kernel 4.15 I'm able to connect with firmware 2.1.0.2
https://github.com/torvalds/linux/commi ... f4c1073cfa

I've had this keyboard for years and have been so frustrated with not being able to use it with linux. Refusing to return it, we've finally won!

Code: Select all

uname -r
4.15.7

dmsg

Code: Select all

[46042.968881] usb 2-1.2: new full-speed USB device number 5 using ehci-pci
[46043.050885] usb 2-1.2: New USB device found, idVendor=1c75, idProduct=0204
[46043.050892] usb 2-1.2: New USB device strings: Mfr=1, Product=2, SerialNumber=0
[46043.050897] usb 2-1.2: Product: Arturia MINILAB
[46043.050900] usb 2-1.2: Manufacturer: Arturia
[46043.161430] usbcore: registered new interface driver snd-usb-audio
[code]

[Linuxmusician01]

Well this is finally fixed!!
As of kernel 4.15 I'm able to connect with firmware 2.1.0.2
https://github.com/torvalds/linux/commi ... f4c1073cfa

I've had this keyboard for years and have been so frustrated with not being able to use it with linux. Refusing to return it, we've finally won!

Code: Select all

uname -r
4.15.7

dmsg

Code: Select all

[46042.968881] usb 2-1.2: new full-speed USB device number 5 using ehci-pci
[46043.050885] usb 2-1.2: New USB device found, idVendor=1c75, idProduct=0204
[46043.050892] usb 2-1.2: New USB device strings: Mfr=1, Product=2, SerialNumber=0
[46043.050897] usb 2-1.2: Product: Arturia MINILAB
[46043.050900] usb 2-1.2: Manufacturer: Arturia
[46043.161430] usbcore: registered new interface driver snd-usb-audio

That's great news!