Re: snd-firewire-ctl-services: Digidesign Digi-002R - missing control for 'Optical mode'
Takashi Sakamoto <[email protected]>
| Newsgroups | org.alsa-project.alsa-devel |
|---|---|
| Message-ID | <[email protected]> |
On Tue, May 12, 2026 at 12:28:19AM -0400, [email protected] wrote: > On 2026-05-11 22:19, Takashi Sakamoto wrote: > > > In my understanding, when the Digi 002 detects S/PDIF signals at the > > optical input port, it automatically uses those signals for the packet > > and ignores the S/PDIF signals in the coaxial input port. In contrast, > > the Digi 003 allows the system software to select either the optical or > > coaxial input as the source of S/PDIF signals when both are available. > > > > However, I may have misunderstood the hardware behavior. Please let me > > know if you observe any behavior that differs from my understanding > > above. > > Thank you for a prompt response. > > I had based my understanding from a 'fuzzy' picture in the manual for the > Digi-002R. The screenshot appears to show a radial that the user can select, > to either send ADAT or S/PDIF data via the optical port(s). > > I was able to install the DigiDesign driver onto a Windows 7 machine, but I > don't have Pro-Tools and the driver does not install a 'setup application'. > Windows itself offers a 'properties' dialog, which lets me select between > 2/8 channels and 16/24bit operation. I don't have that either, so I never try it directly with the bundled software... > I have use 'bus hound' to record the coms to the device when the properties > dialog is opened (with previously set different values). There are some > slight changes in the data, maybe this makes sense to someone. I can attempt > to record more, given some guidance on what is useful... > > Yours, > Simon. Thanks for your effort to capturing packets. As a quick glance, they are for the isochronous communication in IEEE 1394 bus. While, the asynchronous transaction would be used to change the mode of optical interface. The destination address would be 0x'ffff'e000'011c. There are more offset addresses whose effects haven't been cleared yet[1]. [1] https://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound.git/tree/sound/firewire/digi00x/digi00x.h?h=v7.0#n63 Regards Takashi Sakamoto