Comments on: Linux USB multiseat audio support

By: linux scripts - StartTags.com

linux scripts - StartTags.com — Tue, 02 Mar 2010 21:12:51 +0000

[...] rest 1-year or 3-year free updates, depending on the license you purchase No visible link to us …Linux USB multiseat audio support | PlugableHere's how to add audio support on top of the previous instructions for getting USB multiseat [...]

[WORDPRESS HASHCASH] The comment’s server IP (66.77.233.15) doesn’t match the comment’s URL host IP (74.52.116.226) and so is spam.

By: Bernie Thompson

Bernie Thompson — Sat, 20 Feb 2010 16:07:49 +0000

Oh – another question. In terms of not using the DVI/VGA output of the UD-160-A – is that because you’re thinking about using PCIe graphics cards?

You may want to get a unit and check out the performance of USB graphics on Linux first (with the udlfb driver). It still has the fundamental limitations of the 480Mbs USB 2.0 bus, but it’s surprisingly snappy — even more so than Windows.

And the plug-and-play advantages of USB for configuring/mantaining a multiseat configuration are pretty huge.

That said, it will only become really a big deal once this support is in the distros, and adding a terminal involves *no* configuration — and we’re not there yet.

Best wishes,
Bernie

By: Bernie Thompson

Bernie Thompson — Sat, 20 Feb 2010 15:55:10 +0000

Hi Dagg,

The sound portion should be portable to many USB audio devices that have Linux support, with a few modifications of the scripts. For the rest, the scripts here show everything — it’s currently designed around gnome and X, but definitely can do anything (that ultimately can sit on a kernel framebuffer).

What we’ve done with the UD-160-A that’s valuable is 1) Make sure all the components have underlying open source drivers 2) Have scripts that support its exact hardware configuration.

Because this is the terminal market, we’re always looking at ways to hit lower price points. What’s your “magic pricepoint” where a purchase here is an easy decision? We’re going to be launching a “basic” dock/terminal in the coming months to shave a little cost off …

Best wishes,
Bernie

[WORDPRESS HASHCASH] The poster sent us ’0 which is not a hashcash value.

By: dagg

dagg — Sat, 20 Feb 2010 09:38:22 +0000

hello, I’m in the process of researching to build a powerful multiseat.
as I don’t use gnome, I’ve wanted to know how portable is your code in regarding to the sound to other DEs? does it use alsa-usb-audio? am I right in assuming that the code splits the sound by users?
how does the code launches the multiple xserver sessions? does it uses nested xservers?
the reason I’m asking is because I’m not planning to use the plugable dock station because it’s price and the fact that I’m not going to use the dvi output.

By: Bernie Thompson

Bernie Thompson — Sat, 23 Jan 2010 23:42:16 +0000

Thanks for these notes, and the original work on this, Alexander! On the differences in udev rules – I assume (hope?) these will normalize with future versions of the distros.

By: Alexander Todorov

Alexander Todorov — Sat, 23 Jan 2010 09:35:12 +0000

It’s worth mentioning that for good audio experience there are some more things to try:

1) On some distros like RHEL and derived it is necessary to add the users to the audio auxiliary group

2) All sound aware applications need to be configured to use “ALSA Default” device so that they can pick up the device specified by ALSA_CARD. This is usually true for most applications but may not be true if you’re converting a previously used desktop machine to a multi-seat terminal.

3) Extra care must be taken when manipulating sound settings with application like system-config-soundcard, gnome-volume-control or alsamixer for example. Those applications are not multi-user aware, no access control is in place and if the user doesn’t know which is his SEAT_ID they may for example mute the sound for another seat by mistake.