[pulseaudio-discuss] ESI Juli@ inoptimal device.description

[Lennart Poettering]

On Thu, 07.01.10 19:51, David Henningsson (launchpad.web at epost.diwic.se) wrote:

> In the spirit of a recent thread here, I have a similar minor problem. I
> have an ESI Juli@ card, which PulseAudio names as "VT1720/24
> [Envy24PT/HT] PCI Multi-Channel Audio Controller". I'd prefer "ESI
> Juli@" as ALSA suggests, and I assume most other ESI Juli@ users would
> also prefer that.
> 
> I can provide a patch, if someone points me in the right direction.
> 
> Quoting relevant lines from "pactl list" and "lspci" below:
> 
> 		alsa.card_name = "ESI Juli@"
> 		device.product.name = "VT1720/24 [Envy24PT/HT] PCI Multi-Channel Audio
> Controller"
> 		device.description = "VT1720/24 [Envy24PT/HT] PCI Multi-Channel Audio
> Controller"
> 		module-udev-detect.discovered = "1"
> 
> 03:07.0 Multimedia audio controller [0401]: VIA Technologies Inc.
> VT1720/24 [Envy24PT/HT] PCI Multi-Channel Audio Controller [1412:1724]
> (rev 01)
>         Subsystem: Device [3031:4553]

Most of the time the product string from the usb.ids/pci.ids file is
more useful than the product string from the driver. A few times the
contrary is true.

Since we hardly can come up with an algorithm that automatically
figures out that if both strings are known which one is the nicer,
more useful one we have to statically prefer one over the
other. Naturally I decided to default to the usb.ids/pci.ids string
and use the hw string as fallback. In your case this actually made
things worse.

You can try to aks the pci.ids db guys to update its string so that it
is nicer to read.

Lennart

-- 
Lennart Poettering                        Red Hat, Inc.
lennart [at] poettering [dot] net
http://0pointer.net/lennart/           GnuPG 0x1A015CC4