[alsa-devel] Terratec EWS88D (ice1712) optical stuck on ADAT (no IEC958 optical i/o)

Alan Horstmann gineera at aspect135.co.uk
Mon Sep 13 22:35:26 CEST 2010


On Monday 13 September 2010 16:55, Niels Mayer wrote:
> I just scored a nice Terratec EWS88D for $20.00. :-)  (
> http://nielsmayer.com/envy24control/EWS88D_Board_L.jpg missing midi cable
> though (*))
> It works nicely with Linux and http://mudita24.googlecode.com ,  although
> it's missing some important controls to turn off the ADAT parts of the
> card. Alas, it wouldn't help to support them in mudita24, as I cannot get
> them to change values through standard ALSA tools like amixer(1) and
> alsamixer(1).
>
> Therefore, the question: is what is it about snd-ice1712 that prevents the
> mixer control values detailed below from changing, and is there anything I
> can do to fix it? (sometimes you want to use this card in it's dual IEC958
> optical + spdif I/O config, other times, you want to use the optical for
> ADAT and the spdif for IEC958, which is the only configuration I can get
> currently.)
>
> The results from amixer(1):
>   http://nielsmayer.com/envy24control/terratec-ews88d-amixer.txt
> Results of "cat /proc/asound/EWS88D/ice1712":
>   http://nielsmayer.com/envy24control/terratec-ews88d-ice1712.txt
>
> My attempts at changing optical port from ADAT to IEC958. Nothing happens:

<snip>

May be helpful to use alsamixer (command-line pseudo graphical) - that should 
present all the controls the driver provides.  Certainly I found that to be 
the most useful way of understanding/operating the underlying controls.  
(Which envy24control etc operate on top of).

I am fairly sure that the ice1712 driver has no run-time mechanism for 
enabling/disabling the paired digital streams that are ADAT on EWS88D, and go 
to/from the AK4524 codecs on most other versions.  With the DMX6fire, which 
normally has 3-pairs, I added a different register value used at start-up so 
that 2 channel are not active (and picking up 'static' noise), but the other 
ice1712's enable all 4 pairs always.

It is possible that the functions and controls have been coded from 
data-sheets and never debugged on real hardware, or at least not in every 
respect.  Perhaps others may know?

Hope this helps,

Alan



More information about the Alsa-devel mailing list