<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=iso-8859-1">
<META NAME="Generator" CONTENT="MS Exchange Server version 6.5.7638.1">
<TITLE>RE: [LAU] Phonic Helix Board 12 FireWireMKIIcompatibility(linux/ffado)</TITLE>
</HEAD>
<BODY>
<!-- Converted from text/plain format -->
<BR>
<BR>
<BR>
<P><FONT SIZE=2>-----Original Message-----<BR>
From: linux-audio-user-bounces@lists.linuxaudio.org on behalf of Simon W. Fielding<BR>
Sent: Thu 2/5/2009 12:33<BR>
To: linux-audio-user@lists.linuxaudio.org<BR>
Subject: Re: [LAU] Phonic Helix Board 12 FireWireMKIIcompatibility(linux/ffado)<BR>
<BR>
> From: Robert Jonsson [<A HREF="mailto:spamatica@gmail.com">mailto:spamatica@gmail.com</A>]<BR>
> Sent: 04 February 2009 13:07<BR>
> To: Simon W. Fielding<BR>
> Cc: linux-audio-user@lists.linuxaudio.org<BR>
> Subject: Re: [LAU] Phonic Helix Board 12 FireWire<BR>
MKIIcompatibility(linux/ffado)<BR>
<BR>
> That would be a good deed.<BR>
<BR>
> Probably Pete has more pressing needs than me, but still. The things<BR>
I'd like to<BR>
> know of the top of my head are basically:<BR>
> - if it's relatively stable.<BR>
> - no artifacts in the sound, clicks etc, recording for a while with<BR>
lowish latency should prove "something"<BR>
> - What kind of latency to expect, this might be a bit of work to<BR>
determine. Not sure myself how to do it<BR>
> other than tuning a complete system and bringing the buffers down to<BR>
the lowest that are usable.<BR>
<BR>
OK - I now have a working basic FC10 system - relevant details are :-<BR>
<BR>
Hardware: Toshiba Tecra M3<BR>
Kernel: 2.6.26.8-1.rt15.1.fc10.ccrma.i686.rt<BR>
Jackd: jackdmp 1.9.1<BR>
ffado: 2.0-0.6.rc1.fc10.ccrma (libffado 1.999.40)<BR>
<BR>
<BR>
Initial cursory testing shows (no recording load, just an idle mixer with nothing connected) :-<BR>
<BR>
jackd -R -dfirewire -r96000 -p16 -n3 ... jackd won't even start - uses too much CPU<BR>
jackd -R -dfirewire -r96000 -p32 -n3 ... jackd won't even start - uses too much CPU<BR>
<BR>
Both of the above seem to sometimes leave the 1394 bus in a strange state. A forced bus reset using<BR>
gscanbus seems to fix it.<BR>
<BR>
jackd -R -dfirewire -r96000 -p64 -n3 ... jackd starts - uses approx 70% CPU with xruns every 10-20 seconds<BR>
jackd -R -dfirewire -r96000 -p128 -n3 ... jackd starts - uses approx 50% CPU - no xruns<BR>
jackd -R -dfirewire -r96000 -p256 -n3 ... jackd starts - uses approx 45% CPU - no xruns<BR>
jackd -R -dfirewire -r96000 -p512 -n3 ... jackd starts - uses approx 25% CPU - no xruns<BR>
<BR>
This is all I have time for today - I'll try and do more later this week. If I've missed any useful<BR>
details, please ask and I'll try to get them posted.<BR>
<BR>
Cheers,<BR>
Simon<BR>
<BR>
</FONT>
</P>
</BODY>
</HTML>