On 2/11/08, Jonathan Woithe <jwoithe@physics.adelaide.edu.au> wrote:
AFAIK I'm the only FFADO developer equipped to investigate these protocols
(please correct me if I've missed anyone) and unfortunately I don't have
access to an 896HD.  This makes the deduction of the protocol rather awkward
at best.  To do useful tests yourself you need at the very least a
PCILynx-based firewire card, the nosy software, a knowledge of the firewire
protocol and a willingness to hack nosy to get the information you need.
Does this sound like the kind of thing you could do?

As far as I can tell, unibrain seems to be the only one making PCILynx firewire cards.  I have found these cards:




The second one is a Fireboard-800, and goes for $63.99 USD, where the first one is a Fireboard-400 and costs $249.  The Fireboard-400 is PCILynx.  On unibrain's site, the Fireboard-800 is spec'd as using the TI TSB82AA2 IEEE-1394b OHCI-Lynx Controller.  It has a single 1394a port and 2x1394b ports.  I can borrow an adapter cable from work to go from 1394b to 1394a.  I'm willing to buy the $64 card, but $249 is a bit much.  Any thoughts if this controller will work for the sniffing?  What exactly does PCILynx enable?