<html><head><style type="text/css"></style></head><body><div style="font-family:times new roman, new york, times, serif;font-size:12pt"><DIV style="FONT-SIZE: 12pt; FONT-FAMILY: times new roman, new york, times, serif">I missed a few of the responses.  My questions came faster than my registration.  Anyway, I don't think DRC is the answer in my case.  I am working on an embedded system and expect pulseaudio to route and mix as needed.  The driver I have now is quite limited and not conducive to mixing streams.  I envision applications sending data at full resolution and since they would not have knowledge of other streams there is no way for them to decrease their dynamic range in anticipation of the final mix.  It seems that the simplest means for limiting would be to divide the output pf the mixer by the number of streams present inside pulse.  I haven't fully thought out what this might mean when some of the streams are close to silence, but as a first stab it might make sense.  I'm a little afraid of the artifacts from rapid volume change in each stream, but ramping the changes would help.</DIV> <DIV style="FONT-SIZE: 12pt; FONT-FAMILY: times new roman, new york, times, serif"> </DIV> <DIV style="FONT-SIZE: 12pt; FONT-FAMILY: times new roman, new york, times, serif">I'll play with it a bit.</DIV> <DIV style="FONT-SIZE: 12pt; FONT-FAMILY: times new roman, new york, times, serif"> </DIV> <DIV style="FONT-SIZE: 12pt; FONT-FAMILY: times new roman, new york, times, serif">Dennis Fleming<BR><BR></DIV> <DIV style="FONT-SIZE: 12pt; FONT-FAMILY: times new roman, new york, times, serif">----- Original Message ----<BR>From: Tanu Kaskinen <tanuk@iki.fi><BR>To: pulseaudio-discuss@mail.0pointer.de<BR>Sent: Thursday, November 15, 2007 10:58:52 AM<BR>Subject: Re: [pulseaudio-discuss] Mixing streams<BR><BR>On Thu, Nov 15, 2007 at 02:56:47PM +0100, Lennart Poettering wrote:<BR>> I think normalizing a mixed stream is perfectly ligitimate. Don't<BR>> forget that normalization through DRC would only take effect when<BR>> otherwise we'd clip. So the option you have is not DRC vs non-DRC. But<BR>> it is clipping vs. DRC. And in this case DRC is certainly the smaller<BR>> evil.<BR><BR>I wasn't considering DRC in my answer, and I'm not strictly<BR>against it. I hope that it will be optional, though (and<BR>judging from other features, I believe it would be optional<BR>even if no-one didn't specifically ask for it).<BR><BR>Personally, if the sound is clipping, I'd like to know it's<BR>clipping instead of the sound system pretending everything's<BR>ok, and then increase the headroom (lower the stream<BR>volumes, that is). I believe I'm in the minority with this<BR>wish.<BR><BR>> Uh. That way you practically decrease the bit depth of your audio<BR>> output to 15 bit. (Unless you happen to have a 24 bit soundcard) Not<BR>> a very good idea for most people.<BR><BR>I haven't compared 15 bit signal to 16 bit. I have a<BR>24-bit-only sound card and don't feel like setting up any<BR>elaborate listening tests, so I'll take your word that<BR>there's a perceivable difference.<BR><BR>-- <BR>Tanu Kaskinen<BR></DIV> <DIV style="FONT-SIZE: 12pt; FONT-FAMILY: times new roman, new york, times, serif"><BR></DIV></div></body></html>