On Thu, May 15, 2008 at 03:24:26PM +0100, Colin Guthrie wrote: > There has been a lot of good information on this mailing list over the last > couple weeks relating to wine and why it doesn't work with pulse. I've been kind of following this discusion. The patch is the result. > From what I can tell two of them are valid and the third is invalid and > relates to improper use of the ALSA api in wine which needs to be fixed. > This relates to the assumption that snd_pcm_delay() will return 0 if there > are no samples left to be played. This may never reach 0 with pulse as > network delay/latency is also taken into consideration here. I've taken a > very, very cursory look at your patch and I see you've commented out a call > to snd_pcm_delay() so this ties in. > I didn't simply commented out the call. Look carefully. What I did was guess where would be the correct point for doing the call that snd_pcm_delay was setting condition for and simply it that call there. Now the sound seems to work correctly, but I'd like to hear a second opinion from somebody, who actually know this stuff or at least is better at hearing delays/crackling/other stuff then I am. > In that case, I would strongly recommend using Gmane. (http://gmane.org) If > you have an NNTP reader (e.g. Thunderbird) it's ace for keeping up with > lots of projects and posting messages to mailing lists without ever > officially "subscribing" in the classic sense... I use it for lot's of open > sourcey mailing lists. > That's not the point. The point is I have to subscribe to post, but have no need for the most of other discussion. And CC me in a way that doesn't remove mailing list address, so I can simply hit reply.
