Support of audio interfaces under Linux

[dsreyes1014]

It's strange as I'd have thought the 204 would be very similar to the 404.
Perhaps they aren't or perhaps something else is causing your misfortunes. There are so many variables it's hard to know where to start looking.

If you have a spare drive lying around you could try installing Mint (that's what I'm using) + lowlatency kernel + KXstudio and give it another go.

Alternatively, an easier test would be to boot up with AVLinux (or KXStudio?) from a USB stick and try again to see if that makes any difference. I'd imagine that you should be able to get decent sound with AVLinux + your hardware. If not, then perhaps there is some kind of issue with the soundcard as you mentioned.

(Check the URL in my signature for installation steps & tuning. It's not perfect but at least I'm getting reasonable performance from a UA-25 and UMC404 with it using LM17)

I came from Mint and I've troubleshooted on both AVLinux and KXStudio and I still get the clicks and pops. It works fine at 44.1khz duplex and quality is good but anything higher than that I get the clicks. It definitely has something to do with capture side of things just not entirely sure exactly what it is.

Thanks for your feedback and input by the way. It's really appreciated.

[asbak]

Yw, sorry I couldn't be of more help, guess you'll need to borrow some different hardware to test on.

[DepreTux]

I came from Mint and I've troubleshooted on both AVLinux and KXStudio and I still get the clicks and pops. It works fine at 44.1khz duplex and quality is good but anything higher than that I get the clicks. It definitely has something to do with capture side of things just not entirely sure exactly what it is.

Thanks for your feedback and input by the way. It's really appreciated.

I don't know if this will be of any help, but I've got a fast track pro IF (usb 1.1) and had a similar problem recently.

It had been working perfectly for months and all of the sudden, it would throw thousands of xruns then over 44.1kHz. It turned out to be I accidentally installed pulseaudio-x11 as a dependency to some package, and even though no daemon was visible to $ ps aux, a pulseaudio script was running at startup which set the card at 44.1kHz.
Uninstalling pulseaudio and wiping anything pulse in my harddrive restored normal functionality.

Pulseaudio should not be featured in any audio oriented distro, as far as I understand; it only adds another layer of problems.

[dsreyes1014]

I don't know if this will be of any help, but I've got a fast track pro IF (usb 1.1) and had a similar problem recently.

It had been working perfectly for months and all of the sudden, it would throw thousands of xruns then over 44.1kHz. It turned out to be I accidentally installed pulseaudio-x11 as a dependency to some package, and even though no daemon was visible to $ ps aux, a pulseaudio script was running at startup which set the card at 44.1kHz.
Uninstalling pulseaudio and wiping anything pulse in my harddrive restored normal functionality.

Pulseaudio should not be featured in any audio oriented distro, as far as I understand; it only adds another layer of problems.

That's interesting. I'll give that a look and see what I find. Thanks for pointing that out.

[asbak]

This probably doesn't apply to you but just in case, there is a firmware available which can be flashed to the unit (in Windows) which may or may not be of benefit.
http://www.behringer.com/EN/Products/UMC204.aspx

[frankenst3in]

UMC204 here
Finally, wasted two days trying to get the capture working on my Lubuntu 14.04. Neither it works in Ubuntu studio however.. looks like in KX-studio 14.04 it does (yey!). Can't help asking myself why.. Could it be that capture is not supported with generic kernel?

Perhaps worth mentioning I have upgraded the firmware to the one available on behringer page. Before it wasn't working in Windows either.

[asbak]

Another strange thing (not sure if it is due to my particular hardware or incorrect usage) I've experienced with the UMC404 under Linux is that it would periodically stop working, even when switching between different computers running different Linux distros.

When plugging it into a Windows computer it seems to "reset" itself somehow, after which it works again under Linux until things break again.

Anyway, my conclusion is that (for me), the UMC404 doesn't seem to work reliably under Linux.

[glowrak guy]

I had that happen once (in three years) to my Fender Mustang usb amp,
and a couple of times on a usb modem, where even usb_modeswitch failed,
until being started in the dark ethers.
Addicted hardware needed that fix eventually

[miuzik]

I was looking after a new USB audio interface and found this thread, i read it from the beginning to the end and it was a nice reading and i think j_e_f_f_g is right. I also do own an Audigy 2 with an EMU10k2 chipset and yes, even the hardware does support more than 48 Khz the linux emu10k2 driver does not. But, i never noticed that, this thread made me attentive to this issue, thanks for that. But i also must say, that i used my Audigy 2 only for MIDI and not for audio, except typical consumer stuff like watching videos where 48 kHz is enough for me.

But after reading this thread i want to clearfiy something:

1. This posting here:

Have you notified the ALSA devs so they can include this improvement in the next version?

I'm subscribed to the ALSA dev list, and posted a patch to the list, but I don't think the email ever got through. Not sure what happened there. Here's what I added to quirks_table.c if you want to pass it on:

Code: Select all

/* Opcode devices */
{
   USB_DEVICE_VENDOR_SPEC(0x06f1, 0xb010),
   .driver_info = (unsigned long) & (const struct snd_usb_audio_quirk) {
   .vendor_name = "Opcode",
   .product_name = "MIDIport 32",
   .ifnum = QUIRK_ANY_INTERFACE,
   .type = QUIRK_MIDI_FIXED_ENDPOINT,
   .data = & (const struct snd_usb_midi_endpoint_info) {
      .out_cables = 0x0003,
      .in_cables = 0x0003
      }
   }
},

I just used grep over an actual kernel source looking for the USB Device Vendor String "0x06f1" and "0xb010" and to my surprise, nothing was found.

Code: Select all

grep --include=* -ir 0x06f1
grep --include=* -ir 0xb010

Which means, that your patch got never applied to the kernel source.
Then i tried to search for this string "0x06f1, 0xb010" with google and nothing was found, except this thread.
This means, that your eMail to the ALSA kernel-mailinglist never reached them. If it would have reached them, a google search would have definitely pointed to an entry in the ALSA kernel mailinglist archive, which is online everywhere and searched by a google search but nothing like this was found, so this is a clear evidence that your email never reached the alsa mailinglist in the past.
But this is not your fault, you already stated that the eMail you wrote probably never reached them.
But what is more disturbing is, that "raboof" complained and asked you, if you have sent the patch to the ALSA devs, but after you told him the above he had that knowledge too, but he obiously didn't bother.
If he had bothered, he would have noticed what i have noticed now, and he could had fixed this, but he didn't.

2. The ALSA wiki vs. a new wiki.

steevc wrote the following:

Do we know for certain that the PCIe models don't work under Linux?

Until/unless an enduser actually tries a particular audio device, no one knows for sure. Alsa devs don't document their work, the alsa database is inaccurate/ill-maintained, and lots of questionable assumptions are made (for example one would assume a Focusrite 18i20 works because the earlier 2i4 does... but that doesn't appear to be the case).

Trust only first-hand accounts. If taking a chance on an untested item, make sure you can return it.

This was what I was trying to improve upon in the old wiki. We could be linking to reports (blog/forum etc) of working hardware and build a useful reference. We need more than just 'it works' as some devices have extra features that may not be supported, or need extra configuration. These pages got migrated to the new wiki, but that broke it a bit. I see some updates, but a lot of the links were broken. I've just fixed most of those.

http://wiki.linuxaudio.org/wiki/hardware_matrix

It just needs people with working hardware to add anything that's missing. I count it as one of my contributions to open source

I'm a fan of wikis generally, but some people don't see the point when they can 'just Google it'. The down side of that is that you are just getting random results rather than building an actual reference.

I think this idea of adding another 1001+1 hardware wiki database to the internet for linux audio hardware support is a very stupid idea.
Instead of creating a new hw database wiki no user cares about and will get outdated soon it is be more wise to create an account at the alsa-wiki and add the information that is missing there.

http://www.alsa-project.org/main/index. ... =Main+Page

Everyone here complains about the status of the ALSA wiki, but everyone here could easily create an account at the ALSA wiki and fix what is not there.
So why do you think an new wiki is a better idea? It is a totally rubbish idea!

The ALSA wiki is old, yes, but it is the central and main starting point for everyone that is interessed in knowing if a specific audio hardware does work under linux.
And there is no alternative to ALSA, ALSA is the defacto standard audio driver architecture on Linux, so every information that is know about a specific hardware, should also go to the official alsa wiki.
Everyone who knows something, that is not in the ALSA wiki should go to the ALSA wiki, create an account and add the information he knows. It's simple as that.

I am not a ALSA dev, but at least, i DO have an account at the ALSA wiki and i did add some informations to that wiki others could might need. I did NOT start to create my own personal linux audio hardware database wiki no one cares for.


3. About PCI and new mainboards that do mainly have PCIe.

Some suggest to buy mainboards, that do also support the old PCI standard.
I say: "Don't!"
PCI does not work very well on PCIe mainboards. Even if you choose chipsets with PCI support. It is not a good idea.
Instead buy a new PCIe ONLY mainboard wihtout the old PCI slots and buy new PCIe cards. This will work much more reliable.

4. U-PHORIA UMC204 problems mentioned above
There is a new version of the UMC204 out, it is a HD version and supports 192 kHz @ 24 Bit. Maybe this is the version, that doesn't work properly as of now.
The older none UMC204 version should work, that's what i have heared here and there.
The newer UMC204HD is VERY new. It's only on the marked since a couple of weeks ago. So i think we should give the alsa devs a little time.

5. Does anyone know if the Steinberg's UR22 audio interface midi ports do work under linux?
A log in the linux kernel commits says that audio works with a quirks patch, this patch is already in the vanilla kernel sources, but the committer said, that he didn't test MIDI.
So if MIDI does work is uncertain.
It is also not clear what kind of audio mode is supported under Linux.
Does "audio works" mean only that 48 kHz is working? Or does it mean, that 192 kHz is working?
This is undefined too.
If there is someone who owns a UR22 and could clarify this, this would be great. And it would be much greater when this information would be also added to the official alsa wiki.

6. Because i am looking for a new USB audio interface with MIC and MIDI support i was thinking about one of these 3 audio interfaces:

- Focusrite Scarlett 2i4
- Focusrite Saffire 6 USB
- Steinberg UR22

Which one would you recommend?

[wolftune]

USB Audio Class 1.0 supports more than 16 bits!

I didn't bother reading all this. I said ages ago that this was the case. Whoever says otherwise has been wrong forever. I was using USB 1 class-compliant interface, the Tascam US-428 around 15 years ago before USB 2 even existed and it supported 24-bit audio. I've also used the current Alesis io2 and io4 series which are USB 1 class compliant and support 24 bits. I don't care where the wrong idea comes from, it's been clearly wrong from the beginning.

[miuzik]

@wolftune

You obviously don't know what the posting title is for and why it is has a "Re:" in front of it. Go back 3 pages and do your homework!
For the rest based on your posting i do a big LOL.

[wolftune]

Oh, sorry, I see I was rash here. Not only is your post not really about 24-bits at all (you didn't even talk about that), it seems you were maybe trying to reply to some earlier post. This is why I strongly prefer threaded discussions. These flat forums make discourse much more confusing. Sorry for my sloppiness here. No comment about the actual content of your post…

[ChaoticGuitars]

So... After reading 5 pages of posts... Windows or MAC is the better choice for recording music??? I guess that the best thing for me (Us????) is to use Linux for our every day use, and have a PC/MAC setup for recording music???

[tatch]

So... After reading 5 pages of posts... Windows or MAC is the better choice for recording music??? I guess that the best thing for me (Us????) is to use Linux for our every day use, and have a PC/MAC setup for recording music???

short answer: yes

slightly longer answer: depending on what you want to do and how much effort you're willing to put into it, linux audio can either suit your needs pretty satisfactorily or be a huge pain in the ass

[raboof]

[quote="tatch"linux audio can either suit your needs pretty satisfactorily or be a huge pain in the ass[/quote]
... or open up a level of freedom and collaboration not available anywhere else .

As for all interesting questions in life: "it depends" .