On Tue, Mar 30, 2010 at 7:01 AM, Fritz Meissner <span dir="ltr"><<a href="mailto:meissner.fritz@gmail.com">meissner.fritz@gmail.com</a>></span> wrote:<br><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;"> <div class="im">On 30 March 2010 06:17, Brian David <<a href="mailto:beejunk@gmail.com">beejunk@gmail.com</a>> wrote:<br> > On Mon, Mar 29, 2010 at 2:49 PM, Fritz Meissner <<a href="mailto:meissner.fritz@gmail.com">meissner.fritz@gmail.com</a>><br> > wrote:<br> >><br> </div><div class="im">>> Secondly, I have recently seen rather exasperated comments (from<br> >> someone that I respect as an authority) that the mystique surrounding<br> >> "nice" settings is misplaced and that it has virtually nothing to do<br> >> with audio performance. If this is the case, then should that setting<br> >> not be removed from the Controls ?<br> <br> ><br> </div><div class="im">> I'd be interested to hear these comments, because from my understanding,<br> > niceness has a lot to do with audio performance. Or, more generally, it has<br> > a lot to with the performance of any process. By setting the niceness of<br> > programs run by users in the Audio group to -19, you are letting the kernel<br> > know that those programs take precedence over most others when deciding how<br> > to allocate CPU resources. How would this not help performance? If this is<br> > misguided, then it would definitely be good to know.<br> ><br> > --<br> > -Brian David<br> ><br> <br> </div>Brian,<br> <br> The comments I was remembering are in the following two posts on the<br> Linux Audio User mailing list :<br> <br> <a href="http://lists.linuxaudio.org/pipermail/linux-audio-user/2009-November/064820.html" target="_blank">http://lists.linuxaudio.org/pipermail/linux-audio-user/2009-November/064820.html</a><br> <a href="http://lists.linuxaudio.org/pipermail/linux-audio-user/2009-November/064828.html" target="_blank">http://lists.linuxaudio.org/pipermail/linux-audio-user/2009-November/064828.html</a><br> <br> I don't have the expertise to evaluate them, but there are enough<br> knowledgeable people on LAU that I would have expected the comments to<br> be shot down if they were out of line (they weren't).<br> <br> I noticed in those posts that they regard rtprio as the critical<br> setting - should this be included in Studio Controls ?<br> <div><div></div><div class="h5"><br> Fritz<br> <br> --<br> Ubuntu-Studio-devel mailing list<br> <a href="mailto:Ubuntu-Studio-devel@lists.ubuntu.com">Ubuntu-Studio-devel@lists.ubuntu.com</a><br> Modify settings or unsubscribe at: <a href="https://lists.ubuntu.com/mailman/listinfo/ubuntu-studio-devel" target="_blank">https://lists.ubuntu.com/mailman/listinfo/ubuntu-studio-devel</a><br> </div></div></blockquote></div><div><br></div>Hard to argue with Paul Davis. :)<div><br></div><div>Okay, so it's good to hear that Studio Control should eventually be able to configure the rtprio settings. Something that I am always needing to do on my Ubuntu Studio systems is manually setting the priorities of certain real-time threads (as described on this page: <a href="http://www.alsa-project.org/main/index.php/Low_latency_howto">http://www.alsa-project.org/main/index.php/Low_latency_howto</a>, down at the Thread Priorities section. Coincidentally one of the many pages that suggests adjusting the nice settings). Without doing this, I get an unusable amount of x-runs with the rt-kernel, and it works wonderfully after they are set. Can Ubuntu Studio Controls be made to set these thread priorities? That would be lovely.</div> <meta http-equiv="content-type" content="text/html; charset=utf-8"><div><br>-- <br>-Brian David<br> </div>