Re: [LAU] HDSPM inputs

A 03/02/2012, às 22:50, Florian Faber escreveu:

> Miguel,
>
>>> Once you have hdspmixer up and running, you should see the levels on the
>>> input row. If not, check if you have chosen the correct input (optic/coax).
>> Indeed, the input was not set to “optical”. Changing that setting to
>> optical made the inputs work. That saved my day !
>
> The ALSA init script should save that setting when shutting down the
> system. If not, make a custom boot script where you set everything to
> your needs.

Yes, it saves the setting.

>
>> Btw, currently the hdspm doesn’t work out in ubuntu 11.10. This is
>> fixed in the kernel git, so compiling the kernel from there, makes the
>> card functional again.
>
> I believe Adrian accidentally released a version some time ago that
> didn't recognize the PCI version of the card. I fixed exactly that about
> two years ago.

Well, all I can say is that in ubuntu 11.10 the card is not recognized. I was under the impression that it was this commit that fixed it :

commit c09403dcc5698abf214329fbbf3cf8dbb5558bea
Author: Adrian Knoth <adi@email-addr-hidden>
Date: Thu Oct 27 21:57:54 2011 +0200

    ALSA: hdspm - Enable all firmware ranges for PCI MADI/AES cards
    
    From the Windows INF file, we know the firmware ranges for all RME
    cards. For PCIe, a single revision ID per device (RayDAT, MADI, AIO,
    AES) is used. Contrary, the older PCI versions use ranges, that is,
    one revision ID per firmware version.
    
    Instead of listing all possible revisions individually, match the range.
    
    This commit enables all MADI and AES PCI versions ever shipped.
    
    Signed-off-by: Adrian Knoth <adi@email-addr-hidden>
    Signed-off-by: Takashi Iwai <tiwai@email-addr-hidden>

>
>> Also hdspmixer is not present in the
>> alsa-tools-gui package of ubuntu, it has to be compiled from source.
>
> I am no Ubuntu user, so no idea.

There is already a bug report on this.

>
>> thanks !
>
> You don't have to. It's always nice to meet other linux MADI users :)

Well, it’s always nice to get help when we desperately need it ! :-) There will be many happy users of two MADI soundcards in the Den Haag conservatory.

Btw, because we use the cards for live input processing in supercollider, I would like to use a small period in jack without getting xruns. Right now I have it set to 128samples. I have rebuilt the kernel setting the preemption model to low latency desktop (I didn’t see a setting for Real-Time) and the timer frequency to 1000Hz. I’ve configured jackd for realtime with 'dpkg-reconfigure -p high jackd’. I’m starting jackd with priority 70 (i.e.jackd -R -P 70 ... ). It is working mostly fine, I don’t hear any dropouts, and I don’t see XRUNs most of the time. If I start many synths in suprecollider at the same time, such that the overall cpu usage goes to 80% I do get some XRUNs just at the instance of starting the synths, but then no more. Are my settings correct or is there anything else I could do better ?

best,

Miguel Negrão

_______________________________________________
Linux-audio-user mailing list
Linux-audio-user@email-addr-hidden
http://lists.linuxaudio.org/listinfo/linux-audio-user
Received on Sat Feb 4 00:15:03 2012