Hartmut Z Noack &lt;symposion(a)onlinehome.de&gt; writes: Sorry. A better reference for low latency is... <http://www.linuxdj.com/audio/lad/resourceslatency.php3> There are two important kernel patches: low latency, and capabilities. You can avoid needing the capabilities patch by running as root (a poor solution, IMHO). But, there's no getting around needing the low latency patch. The stock kernels are just too badly broken from a latency perspective for most serious audio work. It's really not hard to build your own kernel. But, if you don't feel comfortable doing it, there are other options. Planet CCRMA[1] has properly patched kernel RPMs for most the Red Hat distributions. They also provide an excellent suite of audio applications. -- Jack O'Quin Austin, Texas, USA [1] <http://www-ccrma.stanford.edu/planetccrma/software/>

On Wed, 20 Aug 2003 10:41:38 -0700 "Mark Knecht" &lt;mknecht(a)controlnet.com&gt; wrote: Windowmaker has its own sound server, extensions like meters and mixers for the dock, etc, etc... It's reasonably easy to set up, massively stable, extensible, efficient, yadda, yadda... I think Gnome and Kde pretty nice but maybe a bit more oriented to office applications and home users. If you just want to run an application or a small group of applications at a time or if you want to set up your own environment you can't beat X. MWM, Flux/Open/Blackbox, FVWM ...Afterstep.

Hi Sorry for not posting to ardour list, but I know all of them are here too. I am trying to make my son use ardour for his recording and mixing. I have no prior experience in anything like this, I have always used ecasound for all my previous needs, but it may be too much for a kid. 1) Is there any walkthrough besides what is on the website. 2) I cannot record on more than 2 tracks, so what I do is start, then I create new session, add a mono track record it. Then I add another mono track and then record it. So now I have 2 tracks, once I add another track I cannot record the third track (the small "r" button would not press). Am I doing it the wrong way?? 3) I am using "Ardour/GTK 0.377.1 running with libardour 0.685.0" from CCRMA on linux-2.4.21 with low latency patch and "jackd 0.75.0" with Alsa 0.9.4. Is there a stability problem with this combination because after some use it stops working, after which it would not play or record a track anymore. Using scrub completely messes it up. Regards.

Hi Thanks for reply, but I am already through all this, I was talking about an ardour howto. Regards On Wed, 20 Aug 2003, Austin wrote:

Hi My son is 15 and he uses linux for everything (with my handholding, all my kids use it). He is a very gifted musician. I do have audacity and ecasound with a usable gui, but I think ardour has moch more potential than anything out there. I also have audio professionals that would help me figure it out. I usually do not let my kids touch anything that is easy to use (good brain excersice). I actually have been on ardour-dev for a long time and I think it is much more usable than it was a couple of years ago and I should have my kids try to break it. Regards On Wed, 20 Aug 2003, Frank Barknecht wrote:

This may be true, but at the moment it's mostly potentially rather than actually useful for the new user. Audacity is deep once you get into it, too - especially version 1.2.0-pre1. It works on both levels. They probably won't be able to give you much help with Ardour, unless they've been using it already. That's perverse - the brain excersize should be in making the music, not the basic operation of the tools! Cheers Daniel

I strongly suggest you start with Audacity 1.0, then move to Audacity 1.2.0 when it comes out. It's a far better program for beginners - I've found even people with no computer experience at all can start using it quickly. Cheers Daniel

Hallo, Mark Knecht hat gesagt: // Mark Knecht wrote: Then there is something wrong with the ebuild. I never had to run any recent ardour/jack combination as root thanks to jackstart. On Debian, though. ciao -- Frank Barknecht _ ______footils.org__

Frank, Following up, this is what I found in Gentoo Bugzilla <SNIP> Bug 21752 <SNIP> ------- Additional Comment #5 From J. Ellis 2003-05-29 07:27 EST ------- Heh torben, what's the issue with jack-start? Is this only for media-sound/j-a-c-k or media-sound/jack-cvs, too? jack-cvs seems to be building fine. Let me know. ------- Additional Comment #6 From J. Ellis 2003-05-29 09:05 EST ------- I'm not going to put the capabilities stuff back right now. The idea of having to hack kernel include files to get it to work seems 'distasteful'. I can only imagine the bug reports... <SNIP> I will still enter a request, but it looks like this is considered at least controversial by some at Gentoo. Is there any reason that the ebuild could not patch the kernel include file to add capabilities? It's like a two line change, isn't it? Cheers, Mark

This is exactly what I have suggested yesterday and today. Gentoo does this other places, like with 3D acceleration. They provide ebuild that don't work until I compile my kernel to turn on certain things and edit my XF86config file to enable features. We'll see. I think we'll get there. A while back some of the kernel developers were a bit shocked we were using capabilities this way, but that's another story! ;-) - Mark

I think this is a known bug in the unstable versions. Set the number of tracks to record to 2 instead of 1 and it should be fine. If you don't use the second track (ie not stereo or 2 panned instruments) you can easily delete it with the cross on the individual track panel. Cheers Daniel

I'm running KDE under Gentoo and I've only had problems with one program and jack xruns: Swami and FluidSynth running a 119MB drum set patch. It began to produce so many xruns I could hit a crash cymbal without droupouts. But whatever, the developers say it's beta software. On the other hand I've run Rosegarden triggering Hydrogen, Zynaddsubfx and a MIDI synth with no droupouts and came up with a good tune, recorded it in Audacity. So it's workig out well. Swami and fluidsynth will be grand in due time... Also, I'm using jacko.py, a python script that gives a nice GUI to the jack server. It's been great, along with QTjackconnect and ALSA patch bay. This whole collection of windows take up a desktop, so I'm thinking of trying to put them into one interface...like a "start audio system" GUI. -l[e^2] ------------------------------ http://www.fallingforward.net/ people experimenting with music, art and technology On Wed, 20 Aug 2003, Mark Knecht wrote:

Jesse Chappell &lt;jesse(a)essej.net&gt; writes: Looks like a wonderful program. I want it. :-) Has anybody been able to build and run qjackctl on Debian woody? :-( -- Jack O'Quin Austin, Texas, USA

On Thu, Aug 21, 2003 at 11:32:46 +0900, Patrick Shirkey wrote: Its not neccesary though - I only got one IDE disk, with a plan to get another later, but I've not had problems so far (recording 8 or 10 tracks in ardour simultaneously). I probably will add one at some point, maybe external firewire, for the ability to move it about. But if you only use a few tracks it doesn't seem to be vital like it used to be. OT: If your using modern SCSI disks it can pay to turn down the disk elevator size with elvtune - otherwise the disks elevtator fights with the kernels and it can effect your throughput quite badly. - Steve

Daniel James wrote: Ok then I take it back. You can achieve multi track recording for a few tracks (< 10) with a single IDE. Go for more RAM. But if you want to take advantage of Jack then you will be doing more than 10 tracks pretty quickly. Although I'm not suggesting we all start taking advantage of Jack. -- Patrick Shirkey - Boost Hardware Ltd. Http://www.boosthardware.com Http://www.djcj.org - The Linux Audio Users guide ======================================== Being on stage with the band in front of crowds shouting, "Get off! No! We want normal music!", I think that was more like acting than anything I've ever done. Goldie, 8 Nov, 2002 The Scotsman

Am Fre, 2003-08-22 um 06.17 schrieb Patrick Shirkey: Indeed: i record basically somewhat "modern" Rockmusic with some Experimental/Noise-Addons and i need up to 30-34 Tracks for all the guits/Percs/vioces etc.

Hello, now that DVD recorders become affordable DVD audio might be an interesting alternative to audio CDs. I always feel a bit unhappy when I have to convert my 24/96 files down to 16/44.1. Is there any DVD audio authoring tool for Linux ? Are there any patent issues connected with DVD PCM audio ? And is there a software DVD audio player for Linux ? Matthias -- Dr. Matthias Nagorni SuSE Linux AG Deutschherrnstr. 15-19 phone: +49 911 74053375 D - 90429 Nuernberg fax : +49 911 74053483

--- Daniel James &lt;daniel(a)mondodesigno.com&gt; wrote: Can you offer any more details on this device (where to check out the specs, etc)? Kevin ===== "I should prefer this note not be read or, if skimmed, that it should be forgotten." - Mallarme __________________________________ Do you Yahoo!? Yahoo! SiteBuilder - Free, easy-to-use web site design software http://sitebuilder.yahoo.com

--- Daniel James &lt;daniel(a)mondodesigno.com&gt; wrote: Thanks, Daniel. It says it "employs a Linux-based operating system", I wonder if that means the code itself will be made available somewhere. The $4K pricetag is a bit high, so let's hope there are other options soon. In the studio I work in, we do DVD-A authoring under another OS with a standard $300 4x Pioneer DVD-R/RW and have done up to 8 channels of 24/96 using something called Meridian Lossless Packing (http://www.ambisonic.net/mlp.html). The software is rather spendy, but nothing near $4K. This is _the last_ thing we need any other OS for in our studio, and I would love to see linux able to support something like this functionality. Kevin __________________________________ Do you Yahoo!? Yahoo! SiteBuilder - Free, easy-to-use web site design software http://sitebuilder.yahoo.com

--- Steve Harris &lt;S.W.Harris(a)ecs.soton.ac.uk&gt; wrote: The Pioneer doesn't do the real-time MLP encoding, it's done in software. An encoding app writes an .mlp file which is then burned from a separate tool. The fixed bandwidth starts to shrink _fast_ with 24/96 and multiple channels. 4 is usually safe, 6 can really be asking for it. If I remember correctly, the data rate for 6 channels of uncompressed 24/96 is close to 14 mbs, while DVD needs less than 9.6. Compromises have to made sometimes, say if you have a spot in a mix where all bits are on in all channels simultaneously, it won't work. If there is an error or situation that cannot be handled as-is, the encoder we use can "re-bit" if desired, including smart (even automatic) 'effective bit' reduction. So, for example, it can be set to take bits from the rear before the mains. Sometimes this is unacceptable and the track(s) must be remastered, but this is suprisingly infrequent. Most people in this studio are doing 4 channels anyway (ambisonic b-format, decoded as "square"). Regardless of MLP (highly unlikely, as you say, because of patents) or something like it, DVD-A support for linux would be a great thing. Kevin __________________________________ Do you Yahoo!? Yahoo! SiteBuilder - Free, easy-to-use web site design software http://sitebuilder.yahoo.com

Are you talking about your Pioneer DVD-R drive rather than the standalone Pioneer DVD recorder here, Kevin? Ogg Flac maybe? I guess in the future there will be very few pure hardware players - they will all be running some sort of upgradable software. So there'll be more scope for new formats. You could burn any format you wanted to a data DVD-R and include player software on the disc too. If that's what they are using in the standalone recorder. If that was the case, I expect Meridian might object unless they get a substantial cut. Cheers Daniel

How many minutes of audio can you get on a DVD-A then, assuming you use the maximum bandwidth available to get the highest sample rate for your number of channels? I can imagine this could work quite well without MLP encoding for stereo music, as opposed to surround mixes. As Mike Oldfield once put it, you only have two ears and your natural inclination is to face the thing you are listening to. I don't think he was very happy with some of the surround version done of his music, dating back to the quadraphonic vinyl days. A serious drawback I can see at the moment for 5.1 mixes of music is that if people are playing them back on systems designed for home cinema, the rear speakers are likely to be the those tiny plastic ones, really only intended for effects noises. I'm sure that two good speakers will sound better than six bad speakers. One factor that isn't really addressed by MLP is that DVDs are really cheap to make now. Plenty of films come with a second disc for the special features now - so is there any point in using an expensive compression system on DVD-Audio that requires hardware support? A two disc set is likely to retail for more than a single disc anyway - that's marketing for you... Cheers Daniel

--- Steve Harris &lt;S.W.Harris(a)ecs.soton.ac.uk&gt; wrote: Sounds silly, but I have no idea, never tried video or even looked at it :) The software (http://www.minnetonkaaudio.com) is audio-specific. Just to be clear, we own an internal DVD-R/RW Pioneer drive (sub $300), not the stand-alone Pioneer setup Daniel was referring to. The multi-channel mix (the music) is encoded in software to an mlp, and then burned with DVD-A authoring software. Anyone else have more information about this? Is there another way to handle high-resolution DVD-A? I'd be interested in having another look at this question. What can we do today with Linux for DVD Audio? Frankly, I remember being disappointed with Audio capabilities under the Video specs, which is why we spent the cash and bought the mlp stuff. How far (sampling/bit) does 9+ MB/s get us with, for example, 4 channels of audio (too lazy to do the math -- it's 5 am)? Kevin __________________________________ Do you Yahoo!? Yahoo! SiteBuilder - Free, easy-to-use web site design software http://sitebuilder.yahoo.com

It's probably an embedded variant with a lot of non-GPL'd software. But you'd expect that they must have made some changes to GPL'd code, in which case they should offer source code to anyone who buys the unit - since no-one else will have the binaries. I often get the impression that some embedded system vendors are testing the limits of the GPL though... I think that depends on the quality of the unit; I expect pro quality video equipment usually costs that much. Of course because we know it's really a Linux box, it looks expensive to us - but then Pioneer have done all the R&D and put together a complete system which they will support. On that basis, it could be a bargain for a busy video production firm. As much as I respect Meridian for their audio work, MLP is a real problem for free software. It's totally proprietary and it needs support in hardware players. So if it becomes the standard for digital audio, you might not be able to make finished musical projects on a free software system. Cheers Daniel

Hi Daniel, Friday 22 August 2003 11:13 skrev Daniel James: I feel I have to try and clear up somethings, I think there may be some miss understandings here about what a input and what a track is. - In a digital world there is a big difference between the number of external inputs/outputs you have and the number of internal tracks you utilize. Many external inputs are mainly needed if you _need_ to record many sources at the same time, ANALOG sources. Like if you are recording a symphony orchestra you may wish to use lots of inputs to capture a room fully. If you record a 4-piece rock band - live - and wish to have great control of the result, you probably need between 8-16 inputs. If you don't record live, but rather take it one step at the time you need nowhere near that amount of inputs. Another reason for having lots of inputs (and outputs) is that you may have external effects that you wish to route the signal through. This is quickly going out of fashion, atleast in my world, hobby musican. I work entirely inside the computer once I got the analog sources recorded. -- Internal tracks on the other hand exist only inside the computer, normally you mix these inside the computer before sending it to a output (e.g. monitors). These can normally be created and utilized almost unlimitedly. It is true you can't record an unlimited amount of channels, but the limits are normally not a problem. Of the top of my head I'd say that 100 playback tracks should NOT be a problem. (_Should_ not be a problem, I imagine few people are trying the limits right now in Linux audio, programs are probably not fully optimized yet) Working in a computer you are not bound to a physical number of tracks that you must "fit" your music in. Rather you would probably start creating tracks as needed, group them as needed. Later in the composition if you for instance needed to add a distorted Oboe to get those wierd tones that define your speed metal recording, you would not reuse a track that wasn't utilized at that point. Rather you would create a NEW track to keep the composition well structured. (reusing tracks is probably not applicable to pro recordings, I imagine they would NEVER reuse tracks, when there are no more left, there are no more left. In a home-recording situation, e.g. 4-track or 8-track it is quite common to reuse tracks though.) Conclusion: in a fully digital recording studio there are no real limits to the number of tracks you can utilize. You tend to use MORE tracks than in an similar analog situation. Just because you are allowed to structure things much better. This provides a much better basis for realizing your compositions, you have less constraints to take into account. The number of external inputs and outputs you need is defined by the number of concurrent audio sources you wish to record/listen to. If you record step by step and use artificial drums, like me, you don't need more than say 4 inputs and 2 outputs. The more advanced you get you probably need more inputs and outputs, but very few people need 24 inputs in a digital studio. /Robert ps. I really don't know these things, I just pretend I do. ds.

Daniel James wrote: I was referencing Jack because once you start using it as it is intended to be used you will be working with more than 10 tracks at one time. You may only be recording one track but you could be streaming 20 to get that one track sounding right. In that case a lot of RAM is very important so you don't fill up the tmp dir constantly and two HDDs will make the whole process more likely to succeed. Flip a coin... -- Patrick Shirkey - Boost Hardware Ltd. Http://www.boosthardware.com Http://www.djcj.org - The Linux Audio Users guide ======================================== Being on stage with the band in front of crowds shouting, "Get off! No! We want normal music!", I think that was more like acting than anything I've ever done. Goldie, 8 Nov, 2002 The Scotsman

Am Fre, 2003-08-22 um 11.13 schrieb Daniel James: I must admit, that there is some truth in that. At the other hand: the Opportunity of using virtually as much tracks as you want allows to work faster and to concentrate on the music itself. I recorded an Album with my former Band on a Fostex 12-Track Tapemachine and wanted just one Song to be stuffed with dubbed Gits, VoiceFX and some Extrapercussion. We stopped the Project because to bounce all the tracks to allow us to do more Overdubs took so much Time that we would have need twice the Budget we had for the entire Album to do just one Song. (In the End we just "polished" a Demo of that Track that i recorded at home using a simple Mixer and ping-ponging with 2 HIFI-VHS-Recorders...) To have lots of tracks does not make the impossible happen, it just helps to smoothen the Workflow. That is one of the reasons, why i will not touch Steinbergs Nuendo again. 10 minutes Recording music : 2 houres turning fancy knobs and trying to get "the best setting". ARDOUR appeares quite more basic and once you set up the Mixerstrips OK, you only stop playing to light a cigarrette ;-)