Jesse Chappell wrote: Yes no xruns. It seems to be directly related to the amount of available physical memory. I have been talking with AthlonRob about how the Linux memory management works and it seems that the kernel will cache almost all the available memory up to between 10 and 4 MB before it starts swapping. It also appears that the kernel stores a certain amount of recently used data and I guess that by playing an audio stream it stores that to memory until something which needs it, like a gui app request the space. It will then transparently give up the memory it has cached giving the impression that no memory has been used to open an application. So It seems I am in a bit of a catch22. In order to use JACK with no sound degredation I have to run with no gui. However the kernel will eventually use up the memory space which will probably cause degredation. However I haven't tested this as I haven't had the time. From here it seems the kernel will cache memory at around 1MB per minute. Sometimes more, sometime less. -- Patrick Shirkey - Boost Hardware Ltd. For the discerning hardware connoisseur 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

Jesse Chappell wrote: Cleeecks and pawps :) It starts with the high volume areas getting distorted the more full the sound is. After a while (say 6 or 7 hours of heavy playback) everything becomes distorted clicks and pops. This is retained even if I stop and restart JACK. At first I thought it was a driver memory leak but the way that it happens seems to be standard behaviour for the kernel memory management process. -- Patrick Shirkey - Boost Hardware Ltd. For the discerning hardware connoisseur 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

Paul Winkler wrote: Yeah I need to look into this again. Last time I checked it seemed that everything was in order. The NVidia drivers don't give much scope for tweaking. Also I need high quality video. If I can't do both then I'm not happy. -- Patrick Shirkey - Boost Hardware Ltd. For the discerning hardware connoisseur 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

Patrick Shirkey wrote: Bizarre. If I use the nvagp driver I get clicks and pops with marginal distortion but if I use agpart (kernel agp driver) I get almost no clicks and pops but serious distortion on the high volume areas. I want a new ATI. -- Patrick Shirkey - Boost Hardware Ltd. For the discerning hardware connoisseur 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

On Sat, Nov 09, 2002 at 10:21:13AM +0900, Patrick Shirkey wrote: really? can't you just restart X? /me wishes there were a nice way to have multiple XF86Config's --PW -- Paul Winkler http://www.slinkp.com "Welcome to Muppet Labs, where the future is made - today!"

Patrick, I found your write up interesting. For the most part I agree with your conclusions. I don't think fluxbox 'sounds better' than Gnome, but more just that very few things are running so there is likely to be fewer conflicts. On the other hand, I wonder what part your USB sound card contributes to this problem simply because it's USB? USB transfers are pretty small. This causes the system to require more packets and packet headers to get the same amount of data moved. I wonder if the USB portion of Linux could be contributing to this problem? It may be that those of us using PCI based audio cards don't suffer this problem at all. I certainly haven't noticed any of the things your are talking about. Anyway, very interesting stuff. With best regards, Mark -----Original Message----- From: ardour-users-admin(a)lists.sourceforge.net [mailto:ardour-users-admin@lists.sourceforge.net]On Behalf Of Patrick Shirkey Sent: Friday, November 08, 2002 1:54 PM To: Jesse Chappell Cc: ardour-users(a)lists.sourceforge.net; linux-audio-user(a)music.columbia.edu Subject: Re: [Ardour-users] Re: [Alsa-devel] usb audio sound quality Jesse Chappell wrote: Yes no xruns. It seems to be directly related to the amount of available physical memory. I have been talking with AthlonRob about how the Linux memory management works and it seems that the kernel will cache almost all the available memory up to between 10 and 4 MB before it starts swapping. It also appears that the kernel stores a certain amount of recently used data and I guess that by playing an audio stream it stores that to memory until something which needs it, like a gui app request the space. It will then transparently give up the memory it has cached giving the impression that no memory has been used to open an application. So It seems I am in a bit of a catch22. In order to use JACK with no sound degredation I have to run with no gui. However the kernel will eventually use up the memory space which will probably cause degredation. However I haven't tested this as I haven't had the time. From here it seems the kernel will cache memory at around 1MB per minute. Sometimes more, sometime less. -- Patrick Shirkey - Boost Hardware Ltd. For the discerning hardware connoisseur 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 ------------------------------------------------------- This sf.net email is sponsored by: See the NEW Palm Tungsten T handheld. Power & Color in a compact size! http://ads.sourceforge.net/cgi-bin/redirect.pl?palm0001en _______________________________________________ Ardour-users mailing list Ardour-users(a)lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/ardour-users

On Sat, Nov 09, 2002 at 07:26:21 +0900, Patrick Shirkey wrote: Yes, it sounds to me like you just had too many interupts going on, running your AGP card at x4 will generate a lot of interupts and running your USB sounds device at low latency will generate a lot of interupts. So my guess is that you could have one of these things going flat out but not both. Do USB devices generate more interrupts per period than PCI? I think all my machines are running AGP x2 at most. - Steve

On Sun, Nov 10, 2002 at 06:15:19 +0900, Patrick Shirkey wrote: Thats wierd, my GF2go AGP 2x machine can play DVDs fine. - Steve

On Sun, Nov 10, 2002 at 06:30:26 +0900, Patrick Shirkey wrote: ... I tried to work out what I should do to make my current studio acoustically neutral. First I read some acoustics textbooks, then I used some modelling software, then I gave up and just moved stuff around until it sounded OK ;) Seriously, especially in a bar you just have to try it and see. The number of people in it will make a huge difference. IANAAE*, but I think the meat will counter out the problems from the hard floor. Ever done a soundcheck in an empty nightclub, they sound terrible. - Steve * I Am Not An Acoustics Engineer