Comment 49 for bug 994685

Awesome! I'll be awaiting the test kernel :) Thanks so much for your help on this problem.

It'd be interesting to also know which commit broke things between the Oneiric 3.0.0 kernel and Precise's 3.2 series, but the interest would be purely academical I think :)

This is the commit description:

commit 1c4a54b4513c175ba1a56d0aba8d9cf8f231d407
Author: Takashi Iwai <email address hidden>
Date: Thu Feb 16 16:45:59 2012 +0100

    ALSA: hda/realtek - Finer tuning of auto-parser with badness evaluation

    This patch improves the Realtek auto-parser for assigning the DACs and
    mixers in more suitable ways by evaluating the assignment with "badness"
    calculations.

    When assigning a DAC hinders the assignment of individual DACs for
    other pins, some badness point is given. Similarly, when it blocks the
    assignment of unique mixer controls, another badness point is added.
    Also, if no DAC, even shared DAC, can be assigned, more badness is
    pointed. Finally, comparing the accumulated badness, the best route is
    chosen among several trials.

    Signed-off-by: Takashi Iwai <email address hidden>