Hi all,
Check out http://www.soundclick.com/bands/8/eastsidemilitiamusic.htm
After getting Specimen to an acceptable state, I threw together this
proper demo. It has no vocals since me and my larynx don't get along,
but I encourage anyone who so desires to rectify that deficiency.
I used Seq24 for MIDI sequencing, Specimen for sampling, and Ardour
for recording. Except for the guitars, everything in this song was
produced by Specimen+Seq24. Someone on this list asked me to keep a
detailed journal of the process by which I created this (unfortunately
I can't remember who). Well, as far as the Linux side of things goes,
it's not very interesting. I used the aforementioned tools in a
typical fashion. For general recording, any account I have to write
would be duplication of effort. Instead, check out:
http://www.tweakheadz.com/guide.htm
Oh, and give the song a listen beyond the cheesy introduction. That
part is meant to sound cheesy. If the rest of the song sounds cheesy,
then I'm a failure, but have patience with the intro ;-)
[pb]
I added lines that were in modules.conf to modprobe.conf.
Now and mpu401 driven devices is active!
The USB still is not, though snd-usb-audio is registering on boot and
deregistering on shutdown.
Both are listed in lsmod, similar to listing in 2.4.22:
# lsmod | grep mpu401
snd_mpu401 4224 0
snd_mpu401_uart 8352 2 snd_mpu401,snd_es1968
snd_rawmidi 25536 2 snd_usb_audio,snd_mpu401_uart
snd 55012 10
snd_mixer_oss,snd_mpu401,snd_usb_audio,snd_es19
68,snd_ac97_codec,snd_pcm,snd_timer,snd_mpu401_uart,snd_rawmidi,snd_seq_device
# lsmod | grep usb
snd_usb_audio 63968 0
usbcore 114652 1 snd_usb_audio
snd_pcm 104640 2 snd_usb_audio,snd_es1968
snd_rawmidi 25536 2 snd_usb_audio,snd_mpu401_uart
snd 55012 10
snd_mixer_oss,snd_mpu401,snd_usb_audio,snd_es19
68,snd_ac97_codec,snd_pcm,snd_timer,snd_mpu401_uart,snd_rawmidi,snd_seq_device
# lsmod | grep es1968
snd_es1968 34020 0
snd_ac97_codec 55556 1 snd_es1968
snd_pcm 104640 2 snd_usb_audio,snd_es1968
snd_page_alloc 12100 2 snd_es1968,snd_pcm
snd_mpu401_uart 8352 2 snd_mpu401,snd_es1968
snd 55012 10
snd_mixer_oss,snd_mpu401,snd_usb_audio,snd_es1968,snd_ac97_codec,snd_pcm,snd_timer,snd_mpu401_uart,snd_rawmidi,snd_seq_device
The snd-es1968 and snd_ac97_codec are similarly listed, even though this card
is also not active (the card has a AGOPO chip but different codecand there is
presently no ALSA driver for it).
Only the mpu401 is listed in cat /proc/asound/cards. Both the mpu401 and the
USB interface worked in 2.4.22.
Comments (I haven't listened to it yet) indicate why, alak and alas, I still
use Windows for music production.
1. DX and VST effects. VST has coming, now to linux. I am more in the DX arena
with Cakewalk software, however. The DX effects I do use have VST equivalents
or twins so I have no objection to going over to VST under Linux. It is, I
understand, a much easier interface to program. ...but...
2. MIDI post-processing. I use Ntonyx software to process finished MIDI
sequences. This inserts controllers for more realistic playing style and the
results can be very good. Nothing of this ilk on Linux and not likely to
be--we are very accostumed to OpenSource and Gnu. These folks want to sell
licenses. The cost of this software is very reasonable for professional use.
The cost and development and supporting a Linux port is nothing to
snicker-at--Windows is a larger market still and one where folks fork up the
dough. (I have been employed in this business and understand the issue too
well.)
3. MIDI arranging tools. I use Jammer but Ntonyx's Onyx is an up-and-comer and
Band in a Box is an old mainstay. Again, these folks want to sell licenses
and ...
I can send mp3's on request (with usual copyright understanding) to illustrate
my point.
Running Ntonyx software under wine? Yes, I have tried.
Style enhancer version 3 will only run from the login profile where it was
installed. I have complained about this but it has not been fixed to my
knowledge. So one get's the message: You need to reinstall the styles.
Style enhancer version 2.* will attempt to run. Programs that need to load
something else under wine will rarely work. This one needs to load either a
16bit or 32bit version of a sequencer component (apparently not recognized as
a local dll). This fails. Wine needs a PIF file for each application to
judicate such things and also to resolve the symbol table once instead of
every load. Hint-hint, winemakers!
Onyx did not work if I recall correctly.
Other software:
Jammer, as opposed to a few other sequecing and MIDI playering apps, did mesh
with the active MIDI devices emulated from ALSA through wine's OSS. (Wine's
ALSA interface has not worked for me). The program died in neverland.
Audio stuff I cannot try as of yet, but since this become RT-critical, I would
say not to bother. A VST enabled Adour will be as good as some Windows DAW
programs.
An aside, XGEdit will run under wine, play to and control my sw60xg throught
the mpu401 driver. Attempting to read the usb-audio (which is working under
kernel 2.4.22) hangs.
ABCmuse will play through the mpu401 interface just fine.
How any of this stuff would fair under a more intensive VM such as VMWare or
WIn4Linux -- has anyone tried? I do not have the money to buy them and now
trial is available.
simsam-0.1.7 was released. Changes include:
- multiple instruments
- multiple outputs (JACK only)
- config loading (at last)
- some fixes & cleanups
- some new bugs ;-)
Source tarball and i386 .deb for Debian/unstable are available at simsam's SourceForge download page.
http://sourceforge.net/project/showfiles.php?group_id=65022http://simsam.sourceforge.net/
cheers,
Christian Henz
I would also vote for multi sample support **BUT**, and I may be
in the minority here, I dig the current simplicity of specimen. I don't
need filters or LFO (I can roll my own in any number of programs). Fast
amplitude envelopes would be nice (so I can actually create my own quick
percussion sounds from bits of found noise) but outside of that I like
how it is now. I love the easy of use and simplicity.
If I need something more complex (like say CC control over loop
points, or filters or lfo) I would just build it in PD. Its nice to just
be able to quickly start up an instance or 2 of specimen and create fun
little or synth patterns.
m.
> -----Original Message-----
> From: linux-audio-user-bounces(a)music.columbia.edu [mailto:linux-audio-
> user-bounces(a)music.columbia.edu] On Behalf Of Pete Bessman
>
> > (Just another todo item for your undoubtedly long list!)
>
> I think it's in the tarball if you care to see what I'm working
> on. I'm pretty sure I can knock out most of the high-priority items
> this weekend and make another release. Filters, envelopes, LFOs,
> portamento, and a whole-9-yards approach to JACK are must-haves, and
> will really chang things for the better IMHO.
>
_________________________________________________
Scanned on 19 Feb 2004 23:50:38
Scanning by http://erado.com
Followed the tips everyone offered on my "Latency, Mandrake and Dell
post...missed the emails because I was on "digest" delivery and had only
signed up that day...
Anyway, here's what I have so far for using my Latitude CPXj 650 as a
live fx unit...
I have found that the "Freeverb" LADSPA plugin is about the best. Does
not seem to be a resource hog and has very clean and flexible digital
reverbs. I was using Ecamegapedal to host the plugins but it seems a bit
lacking. For example, I can only run one plug at a time. If I want to eq
the reverb I dont think I can.(?)
So here's what I settled on (The Ardour guys will probably not be
impressed) I'm using Ardour as the host because the layout is killer and
the flexibility is there (in Mixer window) to add plugins as if it were
on a real console. I can add the Reverb in pre fader and a parametric
post fade, etc and have complete control. The latency is probably
running about 10ms in real world terms. It's not real noticeable in this
application anyway. This is an extreme underuseage of Ardour as it was
intended, but for now it's what I want. Good news is the realtime
latency and sound of the reverbs, etc, is running better than Sonar XL
on Win XP so thats a good thing!
I am still unable to run Jack at anything less than 1024 without having
Xrun hell. I set the HD to DMA "on" I had apm as opposed to acpi
running...turned it off and found no difference. I set /tmp up as tmpfs
in fstab per Austin's suggestion, I'm running in IceWM instead of KDE
because it seems to use less resources. But I have not been able to
improve the overall performance that much with these tweaks...
I notice that running Ardour (Beta .9beta 9 RPM from Thacs RPMs), after
I have fired up Jackstart or Jackd-realtime from Qjackctl with -d alsa
hw:0 1024 2 as root and then fire up Ardour and add a track, the pcm
chain is connected (also visible in Qjackctl "Connect")and I have sound
but the pc starts running slow and the Hard drive light is constantly
on. I'm not recording anything...is this Jack writing to disc? I only
have 128 megs in this machine...is the virtual memory trying to
substitute for Jack writing to ram? It is still useable but if there was
a solution for this one item, I'm in digital fx heaven ready for my next
live gig running this as my PA fx.
TIA
I'm searching for a way to resample audio while preserving the highest
quality possible. One possible target would be the 96 --> 44.1
conversion, for obvious reasons, but other ratios are not excluded.
I'm speaking mostly from an "audiophile" perspective; raw numbers are
good ("while using this tool, you preserve such-and-such signal to noise
ratio and such-and-such bandwidth...") but i'd also like to hear about
musical tests, if anyone performed anything like that.
What would be the Linux tools that are supposed to provide the
highest-quality resampling (according to the above definition of
"quality")?
libsamplerate?
Something else?
--
Florin Andrei
http://florin.myip.org/
Hi,
The aim is: to capture some source using JACK. So, latency
is not significant here.
The question is: which case to prefer for jack server settings -
2048 frames/period and 4 periods/buffer,
or
4096 frames/period and 2 periods/buffer?
Or - are these cases equivalent (buffer size has the same size)?
Andrew