From: Clemens Ladisch <clemens@la...> - 2011-08-08 09:30:43
Rolf Anderegg wrote:
> Hi there, I'm back on this issue (high samplerate, 176.4/192k, streams to DICE
> based devices) and have had no luck in debugging this. Has anyone yet had the
> chance or the interest to look into this?
192 kHz playback works just fine with my current snd-dice driver (not
yet pushed to a public repository). This is with special support for
the DICE dual-wire mode, which FFADO does not have.
> I'll attach my current ffado-diag and the less verbose output of my
> ffado-test-streaming, 96k works fine, 192k aborts.
> ... waitForPeriod: rather large TSP difference ... (260128, nom 131088)
This indeed looks as if FFADO gets timestamps labeled as 96 kHz where
it expects 192 kHz.
Apparently, FFADO's missing DICE dual-wire support is the only thing
that prevents high sample rates from working.
It shouldn't be too difficult to add this support (just pretend to have
a 96 kHz stream with twice the number of channels, and reorder the
samples a little), but I'm not very adept with FFADO's internals.
> 02:00.0 FireWire (IEEE 1394) [0c00]: Agere Systems FW643 PCI Express1394b Controller (PHY/Link) [11c1:5901] (rev 08)
> Subsystem: Santa Cruz Operation Device [1111:1111]
Just to rule out other problems: this controller is known to work
perfectly with audio devices and the new stack.
(And I didn't know that SCO now builds FW controller cards ... :-)
> Have you been able to contact Gina at TCAT for access to their SDK?
Yes, I've got it, thanks!
Get latest updates about Open Source Projects, Conferences and News.