Date: Thu, 2 Dec 2004 21:41:03 +0100 (CET) From: Barry Bouwsma <freebsd-misuser@remove-NOSPAM-to-reply.NOSPAM.dyndns.dk> To: freebsd-usb@freebsd.org Subject: Re: USB uaudio soundcards question Message-ID: <200412022041.iB2Kf3l00921@Mail.NOSPAM.DynDNS.dK> References: <200410242008.i9OK8jd10485@Mail.NOSPAM.DynDNS.dK> <200412010929.iB19TLW08470@Mail.NOSPAM.DynDNS.dK>
next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, 1 Dec 2004 10:29:22 +0100 (CET) I wrote to freebsd-multimedia something about one external USB-attached sound device (sound card they call it, but it sure doesn't look like a card, though it has the functions of one) I thought I'd clarify/update my post, should anyone be interested. The original can be found in multimedia@ around the above time. See my References: header. (oops, this isn't usenet, sorry) > > in shops too, as well as a Hercules Muse knob that supposedly > > also works on a Mac as only 2-channel, so may well work on BSD... > There are issues with my OHCI source code that cause playback to > stop after 10 to 13 minutes; this is not an issue with UHCI and > playback for several hours -- nor did I notice this problem with > OHCI when booted into NetBSD. This still happens with the latest 4.x OHCI code downloaded this morning, with negligible hacks (probably due to having leftovers from Ian Dowse's patchsets somewhere else). My USB stack at present is rather close to 4-stable. When I did the test this time, I was in single-user, no swap active. I noted the following from the kernel after about 11 minutes of playback: swap_pager_getswapspace: failed Playing a file (waveplay) was the only thing I was doing. Then something like half a minute later, the usual usb0: isoc TD alloc failed (x2) pcm1:play:0: play interrupt timeout, channel dead And that was that. Interesting? > I didn't think to check this with NetBSD, as all my audio samples > are at 48kHz samplerate, but attempts to use a lower rate for > playback with FreeBSD did not work well at all: the pitch tries > to remain the same as the samples are stretched to an extended Argh, forgot to check this when booted into netbsd > I noticed that when playing over UHCI, the first fraction of a > second would be repeated at the beginning of playback. Ah, this > also affects OHCI. I'll need to try NetBSD again. Perhaps this This is *NOT* a problem with NetBSD. The audio starts and stops cleanly, with no stuttering or repeated bits. I suspect this is not USB-related, but has to do with FreeBSD 4.x audio. (hmmm, can I build uaudio into 3.x, which if I remember, plays audio much more cleanly than 4.x ? hmmm) thanks barry bouwsma
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200412022041.iB2Kf3l00921>