At Fri, 12 Sep 2003 07:00:11 -0700 (PDT), Tom Watson wrote: > > [1 <text/plain; us-ascii (7bit)>] > Well, I finally got this beast working in time for my demo tomorrow. > Turns out that the AK4355 codec has not one, not two, but THREE reset > bits in its registers. They ALL need to be 1 if you want the beast to > function.
are you sure that the last cvs version didn't work (which includes already the fix for this problem) ? i thought the last RSTN state is valid. > In my persuit of all of this I added some code to the > 'ice1724.c' to print out the codec registers. While it works for the > two codecs on the Revolution 7.1 card, it might not work for others (or > if there are multiple codecs of the same type). This printer (it goes > to the /proc interface works quite well. The big change was to > 'akm4xxx-adda.c' which I added the initialization bytes (in the array > for the AK4355). They were needed because the code (I thought it would > work) just to merge them in at the end of the 'snd_akm4xxx_reset' > routine dodn't work. For some reason that didn't work as well as I > thought. Oh, well. > > Bottom line: I got it working!! (and it sounds pretty good!) > > diffs enclosed (they will probably be in base64 because yahoo does it > that way, sorry!). well, these diffs revert many fixes... Takashi ------------------------------------------------------- This sf.net email is sponsored by:ThinkGeek Welcome to geek heaven. http://thinkgeek.com/sf _______________________________________________ Alsa-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/alsa-devel
