I just got a new laptop. It's an Asus A6U with an AMD64 Turion ML-37,
which runs at 2G and has 1M L2 cache. The system also has 1G of
333Mhz RAM. The HD is a 5200 80G drive. I have installed linux
2.6.12, which comes with the RT-Limits patch and I patched PAM myself
to use it. If I run jackd without realtime priority it sounds like
shit but it works. If I try to make it sound better with -R then
ardour gets disconnected from jackd because it is "too slow". Here is
jackd's output:
subgraph starting at ardour timed out (subgraph_wait_fd=22, status =
0, state = Running)
**** alsa_pcm: xrun of at least 39.900 msecs
**** alsa_pcm: xrun of at least 10.455 msecs
**** alsa_pcm: xrun of at least 2.649 msecs
**** alsa_pcm: xrun of at least 0.028 msecs
Doing web searches on this tells me to update, but these are very old
messages. It isn't CPU frequency stepping because the CPU is running
at top speed via userspace setting. I suppose it could be acpi but
now I can't find the message that said how to turn that off...
Shouldn't this machine be more than adiquate for *playing* audio from ardour?!
What gives? Any help would be great.
I've recently tried pd on Suse 9.3 x86_64. When running the sample patch
'filter_floyd' which emulates the famous DSOTM vcs3 riff, I get a result
which is poorer than the same under windows or os x. I tried both
compiling latest version and installing a rpm for suse x86_64.
I think it can be some audio library's fault.
Any hint?
c.
Hi folks,
I try to find out if there is a way to split my midi input from a Piano
to record the left and the right hand seperatly in one shot.
My goal is to prepair notesheets for piano.
any ideas ?
best regards
potty
Greetings all,
I was just contacted by Erik de Castro Lopo who requested that I go edit
a bunch of my files. It appears that Erik is planning to do some sort of
comparison between what I've done and what he has done. This would be an
inappropriate response to my claim regarding the misuse of Smith's method
for bulk conversions at constant rates, so I'll take care of it right away:
1) If you INSIST on using something in Linux audio for resampling and you
don't really understand the issues involved, then you should definitely use
Erik's program(s) rather than anything I wrote because you're a lot less
likely to have very serious errors; however,
2) You really shouldn't use EITHER of these for anything as valuable as,
for example, an audio library. Once again, you'd probably be better off
acquiring a new library or re-recording the sources. If you cannot do
that, then you should probably get with a professional engineer who does
this all the time, then *thoroughly* test the results. Anything wrong
may suddenly pop up later on, and you may not consider that it could be
the resampling step that caused it.
You probably won't see anything more from me on resampling, because I'm
not really interested in it. My recent posts were merely a very belated reply
to abusive demands made long ago on this forum, and *only* because I started
feeling sorry for those who may be misled into damaging their audio files,
not because I felt obligated to the jackasses who heaped abuse on me.
Regards to all,
Dave.
P.S. LAU is on my spam list, so I haven't seen anything posted recently.
I will post announcements once in a while, but I'm really tired (just like
Noah) of abuse, personal attacks, and misrepresentation by the Illuminati
(not limited to Lee R) of what I regard as very much like a cult. Think
about the many parallels to one of those "spaceship" cults, and you'll see
why I say that. Certainly not all who read/post here buy into this cult
mentality, but a significant percentage have.
Been a very long time since i've witnessed any flaming amonst our number
here. I've seen Lee deal it out to those deserving of a slap, David was
on the warpath about something a while ago etc. All that is inevitable
and really quite healthy ( and in Lee's case much good from his way
comes..) but it is somewhat distressing to see this kind of unneccessary
shirt-fronting going on.
The Linux community is the most giving,involving,sharing & meaningful
groups of humainity I know... and I aint joking. The help, support and
services supplied without ANY requirement for reward is both abundant
and extraordinary. It is essential that everybody who takes part in or
belongs to this group, and Linux in general, remembers where they are
and to whom they are talking. Burning bridges only makes it harder for
some other traveller later on.
we're not on some "gamers" forum here.
very best,
geoff.
Hi all,
this is a pre-announcement to inform you that the 4th International
Linux Audio Conference (or LAC2006 for short) will take place on
Thursday, April 27th - Sunday, April 30th, 2006
at the ZKM in Karlsruhe, Germany
It will again be supported by the Institute for Music and Acoustics of
the ZKM, and will be (most probably) organized by Goetz Dipper of said
institute and myself.
We are only starting up the organizational work right now, so this is
not the formal Call for Papers/Music/Workshops yet - that will follow in
early October (and will look similar to that of last year).
This mail is only to allow you to plan for this conference early and
think about possible topics that you might want to present.
Oh, and we will also need helping hands for all kinds of work - any
offers to work with us on planning and running the conference are most
welcome.
Questions about the conference can already be sent to our "OrgaTeam"
list address at lac2006 at zkm dot de. The web page for the conference
is again at http://lac.zkm.de - though there is no new content yet.
More details, as mentioned, in October.
Greetings,
Frank and Goetz
Just to make some tests I looped RME ADI-8 Pro input and output, connected
to a RME HDSP9652 (set as master), and I made some measurements:
http://michelespinolo.altervista.org/rme/meas-chain.pdf
this is the looped set-up with hdspmixer (i.e. total mix) at default volume.
Here the same is done:
http://michelespinolo.altervista.org/rme/meas-chain-gain.pdf
but some gain is given on input channel (no analog or digital clipping
occured).
The results in the latter case are a lot worse than first one, but I do not
know why.
My set-up use a Athlon XP 2500+ (Barton core) Asus A7N8X-X motherboard,
Fedora Core 3 (PlanetCCRMA distro), ALSA 1.0.9a.
Anyone has some idea why this occurs?
Thank you very much!
Michele
ce wrote:
>
>I have no problem offering the domain for it, as long as you do not
>expect me to be there too often ;-) .
>
I accept the offer! Thank's a lot!
However, I wonder wheter it is really useful and not non-productive way of giving information. I saw some forums which multiply the same infos ten times with ten times the same answers... then why writing again the stuff which got already an answer somewhere?! Moreover, linux + audio stuff doesn't involve a lot of people, so somehow separting them - us - is something which could be harmful to the audio-free-software-community!
But, it has also some advantage... I give example of the gentoo forums: Many wikis-howto started in a forum! (note that it could start also from a mailing list).
Anyway, one of the most important point at my eyes is that forums are in a certain way more "open" to the world: I feel that "newbie" have the tendancy to go to forums instead of mailing list ; no idea wheter this is true or not (it was for me). But if this is true, then it could bring more people to audio-open-source-way-of-making-music! (isn't that a consumerist/productivist attitude?!)
tim hall wrote:
>> sorry, but i think this is such a non-debate ...if
>there's an identifiable need and this much energy towards setting one up, go
>for it!
>
Ok, if the needs is here (I wait to see if it really exists), I'll manage to setup something. Unfortunately, I'm gonna move to Berlin for a year or so in some days, and I 'won't have time to do something before one or two month... to the one who wish a forum, you'll have to wait a while if you are waiting for me. Then my knowlege in forums are really basic, but I'm gonna dig deeper. What do we miss then? I think we've got all we need...
Does anybody here know of a language specifically designed to describe
routing of audio from one set of ports to another? In a sense something
that could be used to describe a jack graph, only it would need to
understand mono, stereo, maybe even n-channel routes.
I spose in a sense it would be a kind of directed graph, with attributes
for the edges.
Some facilities to do changing of routing on based on events would also
be nice.
thanks
John
> Message: 5
> Date: Fri, 26 Aug 2005 15:25:20 -0700
> From: Mike Jewell <mj405(a)oneupaudio.com>
> Subject: Re: [linux-audio-user] RE: I just want to EQ the quiet parts
> On Fri, 2005-08-26 at 23:51 +0100, philicorda wrote:
<snip>
> > If Jamin had inverse ratios for the compressor sections then you could
> > do some expansion. Perhaps if you ask the authors they will look into
> > it?
> >
>
> Now, that sounds very interesting. Are there any expander plugins? If
> one existed with the right thresholds, etc, I can imagine it could work
> like a proportional noise gate where it just starts attenuating
> gradually below a certain dB level. Could be a handy tool.
The TAP dynamics plugin has an expander setting. There is also the SWH
SE4.
Expanders are not quite the same as gates, as they boost signals above
the threshold by the ratio, rather than attenuating signals below it.
They tend to be a little finicky for full noise reduction on mixes in my
experience, as it's quite hard to get the settings bang on so the pumpy
artifacts of the expansion are less annoying than the noise. Then you
want to compress the result to get rid of the pumping and you are back
where you started. :) Can work great on single instruments though, or as
an effect to give things a little more punch.
It's all curves anyway. You can make your own ones with the TAP dynamics
somehow to get odd combinations like expander/compressor/gate.
I've never actually tried doing multiband expansion for noise reduction
on a mix to be honest, but if the noise is pretty steady, and limited to
a certain range of frequencies then it should work.
>
> Thanks,
>
> Mike
> Mike Jewell
> One-Up Audio