> I would say that advertising this pin correctly to the driver is the > first step. That's true. > Trying to make the auto-parser treat this pin better would > be a second step, separate from this patch. Definitely. But changing autoconfiguration behavior for this particular case requires extensive testing to prove that changes made for subwoofer code haven't affected any other systems with no subwoofer at all (and I don't have enough hardware to do it myself). As I've said, I know no reliable way to determine that particular pin is connected to a subwoofer - it appears like just another speaker pin. > Making a new model (if that > is what you call a "preset") would work as well, and might be easier for It's not mine :-) The struct to store all model parameters is called alc_config_preset. Anyway, I meant this very thing. > this particular case, while improving the auto-parser would be better in > the long run. That's true. In fact, they are not mutually exclusive: adding new models will have the benefits I've outlined in the last message, while having auto-parser treat subwoofer correctly means broader support for new machines. Valentine _______________________________________________ Alsa-devel mailing list Alsa-devel@xxxxxxxxxxxxxxxx http://mailman.alsa-project.org/mailman/listinfo/alsa-devel
