I never used SPDIF, but at least the UA-25 can be run @ 24/96 (but then not in full duplex mode) and it features digital I/O (not sure if it is SPDIF). Ask Google for details. Here's a link to the device (no, I'm not paid by Thomann ;-) : http://www.thomann.de/thoiw2_edirol_ua25_usb_audio_prodinfo.html Best regards ce

OK, but I guess it will be hard to find a simpler card with 24/96. I once used an el cheapo Terratec Aureon USB which offers optical I/O but at 16/48 only. Best regards ce

John Anderson wrote: I tried more than an USB 1.1 soundcard on different operating systems, and I think you can't go beyond 5.6ms, setting -p64 -n4 in jackd (this is what I use with my UA-25). But even that is less latency that I used to get with my old PCI soundcard (EgoSys Waveterminal 24/96) on WinXP. This was an expensive professional soundcard when I bought it six years ago. It's just my experience, but with gnu/Linux, jack and the alsa-usb-audio driver you get far lower latencies than when using other operating systems. Ciao, c. www.cesaremarilungo.com

On Sun, 18 Dec 2005 16:45:59 +0100 Christoph Eckert <ce(a)christeck.de> wrote: Are you sure you meant microseconds? Not milliseconds? That sounds 'kin fast to me! -- F

On Sun, 2005-12-18 at 16:21 +0200, John Anderson wrote: just for the record, Sound On Sound's latest issue has an interesting roundtable on firewire, usb, usb 2.0, mLAN etc. featuring various "luminaries" from RME, Yamaha, MOTU, ESI etc etc. lots of *very* different opinions voiced, particularly about USB2.0. --p

Christoph We did use snd-usb-usx2y. But there's also a firmware upload to do for this device, so you first have to find that one, and then we were missing several packages (to 4 levels "deep" in the dependencies) to be able to do this upload. That's what I meant by 'bit of a hassle"... Hey, till a very short time ago I wandered about on Mac and WinXP with ProTools, Logic and Cubase ! ;-) Rgds, ivo ----- Original Message ----- From: "Christoph Eckert" <ce(a)christeck.de> To: "A list for linux audio users" <linux-audio-user(a)music.columbia.edu> Sent: Sunday, December 18, 2005 6:02 PM Subject: Re: [linux-audio-user] Looking for soundcard.

Hi all, 2005/12/18, Christoph Eckert <ce(a)christeck.de>de>: Yeah, I can read you! I don't have internet during the weekend... I am still using my tascam and I am back in a state where I am praising it! I had some troubles with it in combination with my new turion64 laptop, but some testing showed that its not a problem with the tascam or its driver, but with the hd being to slow. Because if I record into ram or on usb-stick, all is well. So I think I will save some money next year and get a usb-disk for audio... But currently all my money is spent, I got the monitors I was dreaming about on linuxtag! Have a nice week, Arnold -- visit http://dillenburg.dyndns.org/~arnold/ --- Wenn man mit Raubkopien Bands wie Brosis oder Britney Spears wirklich verhindern könnte, würde ich mir noch heute einen Stapel Brenner und einen Sack Rohlinge kaufen.

2005/12/19, Christoph Eckert <ce(a)christeck.de>de>: Got a girlfriend now. No kidding... Yep, midi even works before the firmware is loaded (if I remember right). The next weekend I am planning to use my laptop and the tascam as my only synth, no external soundmoduls this time... So for the first time I am heavily dependent on it. Arnold -- visit http://dillenburg.dyndns.org/~arnold/ --- Wenn man mit Raubkopien Bands wie Brosis oder Britney Spears wirklich verhindern könnte, würde ich mir noch heute einen Stapel Brenner und einen Sack Rohlinge kaufen.

2005/12/20, lilli chiffon <lilli.chiffon(a)free.fr>fr>: Hey, I really got a girlfriend now! Even almost since my move to a new flat without internet... Given the state of FreeBob, you should really get a usb-card if you want to use it reliable. Otherwise if you want to do development, get a bebop-firewire-device... Arnold -- visit http://dillenburg.dyndns.org/~arnold/ --- Wenn man mit Raubkopien Bands wie Brosis oder Britney Spears wirklich verhindern könnte, würde ich mir noch heute einen Stapel Brenner und einen Sack Rohlinge kaufen.

Thanks for your answers. OK I forget firewire device. Edirol UA 25 looks good for my needs. Tascam is cheaper, good ratio quality/price. Merry christmas. P'tit Louis.

That's really a *great* improvement compared to the talk of Daniel Wagner at LAC 2005 where he told that the latency still is at 50msecs. Great to see that firewire audio support improves! Best regards ce

On Monday 19 December 2005 10:50, Dmitry Baikov wrote: 1) If you've not already done so, recompile your kernel with the PREEMPT flag turned on (i personally use 2.6.11 and 2.6.14 vanilla kernels). Otherwise you may experience too many xruns, e.g., when switching windows or running tons of apps etc. I'm still testing a 2.6.14 kernel further patched with Ingo Molnar's low-latency patches. I'll keep you posted. 2) change the PCI bus latency of your IEEE controller (let me insist: this may be a +real+ bottleneck; an evidence that you've got such a bottleneck there is the occurence of tons of "IEC61883C: possible raw1394 error" messages when jackd is running...). First, fetch the ID of your IEEE controller: root@stooges:/audio > lspci -vv | grep IEEE 0000:02:04.0 FireWire (IEEE 1394): Texas Instruments TSB43AB21 IEEE-1394a-2000 Mine is 0000:02:04.0. Replace 02:04.0 with YOUR number in the following: root@stooges:/audio > setpci -v -s 0000:02:04.0 latency_timer=f0 (0xf0=240 sounds ok for me, but the rule is simply that it should be set to the maximum allowed value, so that the IEEE controller can hold the bus for a sufficiently long period for the data bursts to complete on time). 3) run freebob, then jackd with the following parameters. I personally prefer to run them with startproc, and stop them with "killall" (i actually put all the following in a rcjack start|stop script in /etc/init.d; ask for it if you're interested). startproc -t 3 -l /audio/freebob.log /usr/bin/freebob && echo "freebob running" JACKD_XML="-o osc.udp://localhost:31000" JACKD_OPTS="-p 64 -b 40 -i 8 -t 16 -s 4" startproc -t 3 -n -20 -l /audio/jackd.log /usr/bin/jackd -R -d iec61883 $JACKD_XML $JACKD_OPTS && echo "jackd running" ("-n -20" sets the lower possible nice value for jackd). This leads to ~11ms as reported by jdelay. And if you're interested, I can send you my edirolFA66.xml; this avoids running the freebob daemon (in this case, JACKD_XML="-f /audio/bin/edirolFA66.xml"). Minor changes shoud be needed for the FA100. You can build one by hand by retrieving all the necessary info from the output of "libfreebobctl/examples/test-freebobctl-osc osc.udp://localhost:31000" Note also that i run everything as root. Surely, this is not good practice, but ok for testing purpose... Otherwise, patch your kernel with the rt-lsm patch, or -even safer- use rlimits. As an ultimate side-note, i also shutdown all services that are not mandatory when doing audio recording, e.g., rcnetwork, rcpowersaved, rcpccardd, (and with SuSE: suseplugger, susewatcher,.... ) Syd

Lee, Thank you for your clear explanation. This is a point i'd totally overlooked, and leaving the "-n -20" flag in the command line was just "so-mechanical" (i.e., a remnant of the time where i'd run jackd w/o the -R flag, and where i'd notice slightly better perf ...) I've just redone a couple of tests without the -n -20 option and with my 2.6.11-preempt kernel this doesn't seem to change anything indeed (wooow, thanks god, there are no serious bugs ;-))) Now, looking back on this, I must admit i also got a bit muddled up over "nice" vs "priority" issues when i recently started working with a 2.6.14-rt kernel. I could read in a previous post on this list that -to sum up- one should have jackd running at a priority "higher than all irq handlers besides the soundcard irq" (where soundcard=ieee1394 controller in my case): as i take it, running jackd with, e.g., -P -70, simply increases the RT scheduling priority of jack then (irrespective of the nice value of the non-RT part of jackd if i understand what you said). Am i right? (btw, jackd -P anynegativenumber triggers an error with my 2.6.14-rt kernel, i get a "cannot use real-time scheduling (FIFO at priority -70) etc."; probably the subject for another thread though) Regards, Syd

On Tuesday 20 December 2005 12:03, Salvatore Di Pietro wrote: grrrr.... i'm floored. That was THAT obvious and i just couldn't see it ;-) Thanks. Syd

On Tuesday 20 December 2005 01:34, Dmitry Baikov wrote: You'll find my freebob XML "configuration" file for the Edirol FA-66 here: http://www-reynal.ensea.fr/linux-audio/edirolFA66.xml.gz (no other stuff there so far ;-). Regards, Syd