I forgot: sometimes it does not appear at top, scroll down to see if something appears, even with grey status (not connected). In past I had problems with Keystation 32 similar to yours.

 

Try unplugging all your USB devices and in the Device Manager, uninstall your USB Root-Hubs then restart.

 

Test it again on your old system or another pc just to check if your usb cable is broken. If it works on the other pc...
First read this:
http://www.acorn-instruments.com/s.nl/ctype.KB/it.I/id.837/KB.732/.f
Seems like logitech can cause conflict. Also they explain how to troubleshoot

Also read this:
https://www.noterepeat.com/articles...ting-class-compliant-usb-computer-peripherals

 

Last edited: Oct 24, 2019

Thanks again for all replies, guys. You're awesome!

Anyway, I kindda solved it, but it's still weird...

I'd already tried different cables, but then I've decided to try another one. It worked.

It turned out the other two (USB A/B) cables I tried previously were 3m long. The one who worked was 1,8m.

The funny thing is I'd tested both 3m cables (with the same device, of course) on Ubuntu and they worked just fine. Not on Windows, though, for some reason.

Maybe Windows 10 has issues sending/receiving MIDI data through longer USB cables?

I have no idea.

Anyway, it's solved guys.

Thanks again.

 

I'll admit I only read your original post so if this is repeated that is why. I've had this issue with USB ports before.

If you didn't get it to reset, try this as a temp fix:

1. Select the item showing the error in your device manager.
2. Right click over it and select "update driver software"
3. Click on "browse my computer for driver software"
4. Click "let me pick from a list"
5. Click through the rest of the "next" options and it should reset it. Meaning you don't actually need to change or update the driver.

 

Have you tried to suppress phantom devices?
Phantom devices are devices (mostly usb) not connected anymore, but still present as non-existent devices. This is a big windows problem of all eternity (since win95!), i struggled myself with USB in the past, here is a last thing to try:

1°) In your device manager, go to the third menu from the left and choose "Show hidden devices".

2°) Suppress everything that is faded (looks like a ghost, everything that says that the device is disconnected or not present anymore), especially things with "USB" or "HID" in their name, but don't stop there.

Suppress really any "ghost" that you can find. A healthy system is a system that haven't any ghost nor yellow triangle in the device list (the only exception: in "Audio and game controllers", 2 things called "Microsoft T converter" (ksfilter.inf) and another named "Microsoft audio proxy" or something like that (I can't be more precise in English, my system is in French, sorry!) You can let them in peace). Perhaps you'll have to (re-)install missing drivers.

Perhaps you'll find IRQ conflicts? This is really a pain in the ass to solve! You have to uninstall everything that conflicts, then try to reinstall in a certain order, but it is impossible to predict what is the best order, it is only empiric trial and error.

3°) Reboot, then try again. Works?

 

WIndows OS have and interesting console command "SFC".
Use "SFC /verifyonly" to veryfy if some files are corrupted.

More info about it here

 

usually 3m would work but with bad signal. If you need 2m or longer buy a active cable like for example this:
https://www.amazon.co.uk/UGREEN-Pri...le+type+b+5&qid=1571952888&s=computers&sr=1-3
or buy something like this:
https://www.startech.com/uk/Cables/...s/USB-2-Active-Extension-Cable-5m~USB2AAEXT5M
I have an 5m long active USB cable that works perfectly. (I first tried out 4m long USB cable that was not active and the signal was very bad).

 

Last edited: Oct 24, 2019

Glad you got it going! That's why I love this forum, lots of super helpful people here!
Now if we can just stop arguing about mac vs PC...

 

hmm so basically sort of what I pointed out earlier by "any other cause for insufficient voltage in usb bus chain" - longer cable causes bigger voltage loss which may result in such issues, my general recommendation of ANY cable type is to use shortest which you can if possible

 

Yes, it's totally related to what you've said earlier.

What puzzles me is why the same cables worked fine on Linux. Like I said, I've tested them on my Ubuntu notebook before; That's why initially I had ruled the "damaged cables" variables out.

Any ideas?

 

When you say the cable works fine on Linux do you mean the midi Keyboard worked fine on Linux with the same cable that did not work on Windows? A midi keyboard needs more "power" than for example a game controller. So if a game controller works with the same cable does not mean that the midi keyboard would work. Has that midi Keyboard even a Linux driver? If not how do you know that the cable works on Linux? Read my comment about active USB cable.
Of course it could be some differences in the voltage on the USB port on the different motherboards. But that is more a bios settings thing than a different Operating System difference.

 

Last edited: Oct 25, 2019

it's not easy to explain, English isn't my native language either,
problem is, Windows kernel communicates with all the hardware and software, what you see in Windows user environment is just a tip of the iceberg,
when you plug a usb device, usb bus controllers need to figure out IF there is a device connected, what kind of device, what kind of prequisitions and requirements kernel is supposed to passthrough into software (like device driver) level, and only if this all 2-way communication going through physical usb busses and motherboard, its BIOS and stuff, possible reboots etc.., "creates" accessible devices for user to see - as you can see this is done by every operating system bit different, regardless of hardware - this explains why you can have no problems on Linux, even if that Linux is installed on the very same computer;
I can think of similar example, perhaps more illustrative - let's say you plug a monitor screen to PC, monitor sends data about its resolution and specs, and computer sends "hopefully" matching video signal - in case of longer or lower quality cable, interference or signal loss can happen and monitor may fail to show any image, or not even show up as monitor in Windows device manager at all

 

Yes, exactly.
I didn't have to mess up with drivers. Just installed VMPK, which is a virtual piano, and it simply worked.
Anyway, this could be about the quality of the hardware components, although my Linux notebook is way older than my DAW PC.

Yes, it makes sense indeed.

 

Sounds more like a bios USB voltage settings issue than a OS issue. Of course it could be that Windows OS "demands" a little more USB voltage for the midi Keyboard to be "found" than in a Linux OS. But my guess is that if you had installed Linux on that problem pc the same issue had occurred. Maybe all you need is to update the bios on the motherboard where the midi Keyboard is not found

 

Last edited: Oct 25, 2019

This completely fixed my problem. I've used Nirsoft for years with many of their other excellent utils. In my case I have Computer Workstation that is used for pro audio (studio recording) and I have 8 USB based MIDI controllers. One by one, they kept failing to register in Device Manager; I was literally down to my last two controllers. With this handy little utility, you can uninstall these failed instances. Once I rebooted, all my controllers registered correctly with no errors. Excellent suggestion, so I wanted to share my experience here.
Windows 10 (fully updated)