From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43) id 1NfeOL-0003iC-VT for qemu-devel@nongnu.org; Thu, 11 Feb 2010 14:07:57 -0500 Received: from [199.232.76.173] (port=50991 helo=monty-python.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1NfeOL-0003i4-JK for qemu-devel@nongnu.org; Thu, 11 Feb 2010 14:07:57 -0500 Received: from Debian-exim by monty-python.gnu.org with spam-scanned (Exim 4.60) (envelope-from ) id 1NfeOK-0003q9-FZ for qemu-devel@nongnu.org; Thu, 11 Feb 2010 14:07:57 -0500 Received: from mail-ew0-f224.google.com ([209.85.219.224]:63670) by monty-python.gnu.org with esmtp (Exim 4.60) (envelope-from ) id 1NfeOJ-0003pp-NG for qemu-devel@nongnu.org; Thu, 11 Feb 2010 14:07:56 -0500 Received: by ewy24 with SMTP id 24so1722200ewy.16 for ; Thu, 11 Feb 2010 11:07:54 -0800 (PST) MIME-Version: 1.0 In-Reply-To: References: <12d708831002101540m1061cc8ag2e5d31c6d73f31c0@mail.gmail.com> <12d708831002102247n3438bf09h502d1f4050723f3d@mail.gmail.com> <12d708831002102303t777e6f83g563311cc0f9cf1a1@mail.gmail.com> Date: Thu, 11 Feb 2010 19:07:54 +0000 Message-ID: <12d708831002111107n2c8f61c9j70ffd777979d0858@mail.gmail.com> Subject: Re: [Qemu-devel] Audio latency From: Alberich de megres Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: malc Cc: qemu-devel@nongnu.org On Thu, Feb 11, 2010 at 8:42 AM, malc wrote: > On Thu, 11 Feb 2010, Alberich de megres wrote: > >> On Thu, Feb 11, 2010 at 7:53 AM, malc wrote: >> > On Thu, 11 Feb 2010, Alberich de megres wrote: >> > >> >> I'm using the alsa driver. >> >> Even I tried to change some values using the export and some flags >> >> that -audio-help shows, but nothing significant. >> > >> > Please do not top post. >> >> Sorrry!! >> >> > >> >> >> >> >> >> On Thu, Feb 11, 2010 at 1:03 AM, malc wrote: >> >> > On Thu, 11 Feb 2010, Alberich de megres wrote: >> >> > >> >> >> Hi! >> >> >> >> >> >> I'm running qemu on suse, and i got a very high audio latency. For >> >> >> example, using windows latency is about 4s and audio gets distorte= d >> >> >> with little cuts. >> > >> > How do you measure it? >> With a program called dpc latency, it's for windows. >> >> > >> >> >> On the host side (suse), i changed the audio pci device latency to= 99 >> >> >> but this only improves a little bit. >> > >> > As in: with setpci? >> Yes. I also tired, on host, to low latency to VGA to 10. > > That's completely unrelated latency [1] > >> > >> >> >> Other thing i noticed is when lowering video resolution, the audio >> >> >> latency gets better and better. >> > >> > Lowering video resolution of what? Guest, host? >> >> Yes, it was on guest. Even for example if i change the desktop or >> switch to a text terminal when things works better what suggested me >> that could be something with painting the screen. But i'm not shure if >> its just only a side effect of a bad launch command, or something >> else. >> >> > >> >> >> I started played with source code, but i'm just a newbie on qemu. >> >> >> Anyone can point some light to me please? >> >> >> >> >> >> thanks guys!! >> >> > >> >> > -audio-help is your friend, from there you can deduce (at the very = least) >> >> > which audio driver your qemu uses. >> >> > >> > >> > It would also help to know which version of QEMU you run and exactly h= ow >> > you invoke it. >> >> You're right..sorry: >> >> qemu-kvm -cpu core2duo -smp 2 -m 1024 -hda /home/nine/imgs/hda1.img >> -vga std =A0-usbdevice tablet >> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0-name t1 =A0-net nic,macaddr=3D55:54:00:1= 2:34:56 -net >> tap,ifname=3Dtap1,script=3Dno >> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0-soundhw ac97 > > I can't help you with kvm and windows, that said here are the things > to try: > =A0a. Vanilla qemu > =A0b. No smp > The same thing. When playing sounds, i get little cuts. > And i'm not saying that those will help, just trying to collect more > data points. > > P.S. 4 seconds of latency is sortof insane, especially given that > =A0 =A0 default buffering for ALSA is nowhere near that, not by a > =A0 =A0 longest shot. > > [1] http://www.reric.net/linux/pci_latency.html > > -- > mailto:av1474@comtv.ru >