HyperX Cloud III not detected

[Miepee]

Description

As the title says, the device isn't detected by headsetcontrol. I plugged it in via USB if that matters.

Headset Name

HyperX Cloud III

On which OS does the problem happen?

Linux

Device information

Detailed Device Information

Device Found
 VendorID: 0x03f0
ProductID: 0x089d
 path: /dev/hidraw3
 serial_number: 000000000000
 Manufacturer: HP, Inc
 Product:      HyperX Cloud III
 Interface:    3
 Usage-Page: 0xc Usageid: 0x1

Device Found
 VendorID: 0x03f0
ProductID: 0x089d
 path: /dev/hidraw3
 serial_number: 000000000000
 Manufacturer: HP, Inc
 Product:      HyperX Cloud III
 Interface:    3
 Usage-Page: 0x1c0 Usageid: 0x1

Device Found
 VendorID: 0x03f0
ProductID: 0x089d
 path: /dev/hidraw3
 serial_number: 000000000000
 Manufacturer: HP, Inc
 Product:      HyperX Cloud III
 Interface:    3
 Usage-Page: 0x170f Usageid: 0x1

Device Found
 VendorID: 0x03f0
ProductID: 0x089d
 path: /dev/hidraw3
 serial_number: 000000000000
 Manufacturer: HP, Inc
 Product:      HyperX Cloud III
 Interface:    3
 Usage-Page: 0xffc0 Usageid: 0x1

[Miepee]

I tried basic print debugging and for some reason, the headset is never searched for?
Its vendor id is 0x03f0 = 1008
Printing all the vendor ids that get searched for in find_device, I see 9610 four times and 1133 once.
patch in case i did something wrong:

    cur_dev   = devs;
    while (cur_dev) {
+       printf("Vendor ID: %d\n", cur_dev->vendor_id);
        found = get_device(device_found, cur_dev->vendor_id, cur_dev->product_id);

[nicola02nb]

A good way to find your device ID is to run:

headsetcontrol --dev -- --list

Then find your device in the list reported.
Example for Steelseries Arctics Nova 7:

Device Found
  VendorID: 0x1038
 ProductID: 0x2202
  path: 
  serial_number:
  Manufacturer: SteelSeries
  Product:      Arctis Nova 7
  Interface:    3
  Usage-Page: 

Then probably you should add your ProductId to the supported PRODUCT_IDS[] at:

HeadsetControl/src/devices/hyperx_cloud_3.c

Lines 8 to 13 in c96c3d9

	#define VENDOR_HYPERX_CLOUD 0x03f0
	#define ID_CLOUD3_WIRE 0x089d
	static struct device device_c3;
	static const uint16_t PRODUCT_IDS[] = { ID_CLOUD3_WIRE };

[Miepee]

A good way to find your device ID is to run:

See the output in the first message under "detailed device information".
You can see there that the vendor ID is 0x03f0 and the product ID is 0x089d - both values already exist in the snippet you linked!

Hence why I tried to do the print debugging (my second message) and found out that it doesn't even ever try to search for the vendor ID.

[nicola02nb]

See the output in the first message under "detailed device information".

Oh, didn't see that, sorry...

[nicola02nb]

I might have found the problem.
I should be there:

HeadsetControl/src/main.c

Lines 61 to 63 in c96c3d9

	if (found == 0) {
	break;
	}

Because it stops checking the rest of the devices... probably you have an already compatible device connected, and when it finds it, it stop searching for other compatible devices(Like your hyperx cloud 3).

Unplugging the device foud before your hyperx cloud 3 should work.

[nicola02nb]

It actually a problem of the entire software which was designed to support one device at a time...

[Miepee]

Unplugging the device foud before your hyperx cloud 3 should work.

I swear I tried it earlier and it didn't work. But now with the other device disconnected and several days/reboots later, it does work.

I wonder if maybe something internally caches something, when i have both plugged in, then plug out the other supported one and run headsetcontrol again. Not sure
Either way, thanks for finding that!

[nicola02nb]

Sond strange OS chaching usb devices lol.

Tho, no problem, you're welcome

[Sapd]

It actually a problem of the entire software which was designed to support one device at a time...

Yeah there is some work already in fixing that. For example the JSON/YAML output functionality already fully supports multiple devices. However main problem is still the question of specifying which to use for sending commands. With multiple devices a user would need to somehow specify which device, which I did not think about yet of how to do that in a user friendly way.

I wonder if maybe something internally caches something, when i have both plugged in

Should in theory not, as it simply uses the HID calls of the kernel to iterate (which in turn uses USB iteration). Maybe some weird kernel bug or edge case.

[Miepee]

It wouldn't surprise me if I had done a kernel upgrade that day and forgot to reboot.

[nicola02nb]

However main problem is still the question of specifying which to use for sending commands

I got an idea there #382