4 Mar
2003
4 Mar
'03
5:08 a.m.
Greetings list,
Thanks again for your responses to my questions about iiwusynth, zynaddsubfx
and jack; you were all most helpfull. And now for something completely
different...
When creating music, there are various issues I run into. First, the way I
would like to work is to build tracks within something such as Rosegarden. I
use a hardware synth (alesis qs6), and other software tools. Since I have a
relatively slow computer, I would like to build the tracks, then record each
one separately, then composite them with something like ecasound.
My first problem is the recording of the various tracks. Right now, I use snd
because it has this record feature called "trigger". This feature is
extremely usefull. My question is do any other software packages support
this feature? For those who are not familiar with it (or cannot infer its
meaning), it essentially just waits until the audio level reaches a specified
threshold, then starts recording. I would love to know if ecasound can do
this.
My next question is one concerning synchronicity of my composited tracks, and
also of sample loops. I am not mathematically inclined, and i do in fact
suck at math. What I would like to know is if there is (im sure there is) a
formula for calculating audio segment lengths according to a specified BPM.
For example, say I have a loop of some recorded drums at 120 BPM. I would
like to know exactly how long the audio sample must be to match 120BPM,
rather then trial and error.
I'm sure I had more questions but I seem to have forgotten them at the moment,
but im sure this is enough for this particular post!
Thanks!
--
Levi Burton
http://www.puresimplicity.net/~ldb/
4 Mar
4 Mar
5:48 a.m.
For example, say I have a loop of some recorded drums
at 120 BPM. I would
like to know exactly how long the audio sample must be to match 120BPM,
rather then trial and error.
this is really easy. i would not even call this math ;-)
with samplerate R, 1 sec = R samples
at x BPM, 1 min = 60 sec = x beats. so the duration of 1 beat = 60/x sec
combined: 1 beat = R*60/x samples.
so at 120 BPM, 44100 Hz, 1 beat == 22050 samples. but you probably want your
loops to be 4 beats, so 88200 samples.
maarten
7:26 a.m.
Hi
it may be interesting:
beatmonkey
http://www.68k.org/~michael/beatmonkey/
On Tue, 4 Mar 2003 10:38:08 +0100
Maarten de Boer <mdeboer(a)iua.upf.es> wrote / a écrit:
Guy CLO~
#219055 counter.li.org
-------------------------------------
http://perso.wanadoo.fr/guy.clotilde/
-------------------------------------
For example,
say I have a loop of some recorded drums at 120 BPM. I would
like to know exactly how long the audio sample must be to match 120BPM,
rather then trial and error.
this is really easy. i would not even call this math ;-)
with samplerate R, 1 sec = R samples
at x BPM, 1 min = 60 sec = x beats. so the duration of 1 beat = 60/x sec
combined: 1 beat = R*60/x samples.
so at 120 BPM, 44100 Hz, 1 beat == 22050 samples. but you probably want your
loops to be 4 beats, so 88200 samples.
maarten 
8:50 a.m.
Semi OT side issue warning. Absolutely fascinating. This explains much about
why I often fail to understand explanations.
On Tuesday 04 March 2003 09:38, Maarten de Boer wrote:
this is really easy. i would not even call this math
;-)
with samplerate R, 1 sec = R samples
at x BPM, 1 min = 60 sec = x beats. so the duration of 1 beat = 60/x sec
combined: 1 beat = R*60/x samples.
so at 120 BPM, 44100 Hz, 1 beat == 22050 samples. but you probably want
your loops to be 4 beats, so 88200 samples.
With all due respect Maarten, You _have_ managed to make it look like rather
complex math. Either that, or I'm dyslexic. I would explain it thus:
each beat lasts 60/bpm seconds, multiply by the number of beats in the bar.
That will give you the length in seconds, you an convert that to length in Hz
as above if you need to. 4 beats to the bar _is_ the default setting, if you
like. but best to check otherwise you may end up with a loop twice or half
the length.
It's really ok to have odd numbers of beats to the bar if you want, but
that's just personal bias.
cheers
tim hall
8 Mar
8 Mar
6:14 a.m.
On Tue, 4 Mar 2003 12:38:32 +0000
Tim Hall <tim(a)glastonburymusic.org.uk> wrote / a écrit:
each beat lasts 60/bpm seconds,
I don't understand. Is it 60 bpm/s ?
Guy CLO~
#219055 counter.li.org
-------------------------------------
http://perso.wanadoo.fr/guy.clotilde/
-------------------------------------
6:28 a.m.
On Sat, 8 Mar 2003 14:01:32 +0400
Guy Daniel CLOTILDE <guy.clotilde(a)wanadoo.fr> wrote / a écrit:
On Tue, 4 Mar 2003 12:38:32 +0000
Tim Hall <tim(a)glastonburymusic.org.uk> wrote / a écrit:
Hum ok. I'm still sleeping.
each beats lasts 60/(the number of bpm) seconds.
I understood better Maarten's post. People's brains don't act similarly (even
the same brain at different times).
Cheers.
Guy CLO~
#219055 counter.li.org
-------------------------------------
http://perso.wanadoo.fr/guy.clotilde/
-------------------------------------
each beat lasts 60/bpm seconds,
I don't understand. Is it 60 bpm/s ?
10:29 a.m.
point taken
tim hall
On Saturday 08 March 2003 10:14, Guy Daniel CLOTILDE wrote:
On Sat, 8 Mar 2003 14:01:32 +0400
Guy Daniel CLOTILDE <guy.clotilde(a)wanadoo.fr> wrote / a écrit:
On Tue, 4 Mar 2003 12:38:32 +0000
Tim Hall <tim(a)glastonburymusic.org.uk> wrote / a écrit:
Hum ok. I'm still sleeping.
each beats lasts 60/(the number of bpm) seconds.
I understood better Maarten's post. People's brains don't act similarly
(even the same brain at different times).
Cheers.
Guy CLO~
#219055 counter.li.org
-------------------------------------
http://perso.wanadoo.fr/guy.clotilde/
------------------------------------- each beat lasts 60/bpm seconds,
I don't understand. Is it 60 bpm/s ? 
6:40 a.m.
el Sat, 8 Mar 2003 14:01:32 +0400
Guy Daniel CLOTILDE <guy.clotilde(a)wanadoo.fr> escribió:
On Tue, 4 Mar 2003 12:38:32 +0000
Tim Hall <tim(a)glastonburymusic.org.uk> wrote / a écrit:
no, it's 60/bpm
you know bpm means "beats per minute", so the duration of a beat is:
60 (seconds in a minute)
---
bpm (number of beats in a minute)
at a metronome marking of 60, each beats is a second long (60/60).
the higher the mm, the shorter each beat: for mm=120 each beat is 0.5
seconds (60/120); for lower mm, beats are longer, for mm=40 each beat is
1.5 seconds (60/40)
each beat lasts 60/bpm seconds,
I don't understand. Is it 60 bpm/s ?
10 Mar
10 Mar
12:41 p.m.
luis jure wrote:
el Sat, 8 Mar 2003 14:01:32 +0400
Guy Daniel CLOTILDE <guy.clotilde(a)wanadoo.fr> escribió:
The way I look at the word "per" is to just turn it into a divisor sign,
so that a 60 bpm song would be:
60 beats
--------
minute
a 120 bpm song is twice as fast, and hence has twice as many beats in a
minute, etc.
-Rich
explanatory PS: 60/bpm would equal 60/beats/minute == 60 minutes/beat
which would actually be 1/60 beat/minute , read as one sixtieth of a
beat per minute. At least that's the way I would read it...
On Tue, 4 Mar 2003 12:38:32 +0000
Tim Hall <tim(a)glastonburymusic.org.uk> wrote / a écrit:
no, it's 60/bpm
you know bpm means "beats per minute", so the duration of a beat is:
60 (seconds in a minute)
---
bpm (number of beats in a minute)
at a metronome marking of 60, each beats is a second long (60/60).
the higher the mm, the shorter each beat: for mm=120 each beat is 0.5
seconds (60/120); for lower mm, beats are longer, for mm=40 each beat is
1.5 seconds (60/40)
each beat lasts 60/bpm seconds,
I don't understand. Is it 60 bpm/s ?
12:58 p.m.
Richard Seymour wrote:
luis jure wrote:
replying to my own post... I think I completely misread the thread.
Don't mind my pedantry.
-Rich
el Sat, 8 Mar 2003 14:01:32 +0400
Guy Daniel CLOTILDE <guy.clotilde(a)wanadoo.fr> escribió:
The way I look at the word "per" is to just turn it into a divisor
sign, so that a 60 bpm song would be:
60 beats
--------
minute
a 120 bpm song is twice as fast, and hence has twice as many beats in
a minute, etc.
-Rich
explanatory PS: 60/bpm would equal 60/beats/minute == 60 minutes/beat
which would actually be 1/60 beat/minute , read as one sixtieth of a
beat per minute. At least that's the way I would read it...
On Tue, 4 Mar 2003 12:38:32 +0000
Tim Hall <tim(a)glastonburymusic.org.uk> wrote / a écrit:
no, it's 60/bpm
you know bpm means "beats per minute", so the duration of a beat is:
60 (seconds in a minute)
---
bpm (number of beats in a minute)
at a metronome marking of 60, each beats is a second long (60/60).
the higher the mm, the shorter each beat: for mm=120 each beat is 0.5
seconds (60/120); for lower mm, beats are longer, for mm=40 each beat is
1.5 seconds (60/40)
each beat lasts 60/bpm seconds,
I don't understand. Is it 60 bpm/s ?
4 Mar
4 Mar
5:50 a.m.
On Tue, Mar 04, 2003 at 02:57:48 -0600, Levi Burton wrote:
My next question is one concerning synchronicity of my
composited tracks, and
also of sample loops. I am not mathematically inclined, and i do in fact
suck at math. What I would like to know is if there is (im sure there is) a
formula for calculating audio segment lengths according to a specified BPM.
For example, say I have a loop of some recorded drums at 120 BPM. I would
like to know exactly how long the audio sample must be to match 120BPM,
rather then trial and error.
Yes, its easy
There are 4 beats in a bar (well, in 4/4 music)
so 60 seconds / 120 BPM means there is 1/2 second per beat
therefore there are 4 * 1/2 = 2 seconds per bar
in short: bar time = 240
---
bpm
I probably didn't explain that very well, but I expect someone will post a
better explanation.
Your triggered record thing sounds like a useful feature for TimeMachine
(http://plugin.org.uk/timemachine/) if I find time. It should be easy.
- Steve
2:22 p.m.
On Tue, 4 Mar 2003 02:57:48 -0600
Levi Burton <donburton(a)sbcglobal.net> wrote:
My first problem is the recording of the various
tracks. Right now, I use snd
because it has this record feature called "trigger". This feature is
extremely usefull. My question is do any other software packages support
this feature? For those who are not familiar with it (or cannot infer its
meaning), it essentially just waits until the audio level reaches a specified
threshold, then starts recording. I would love to know if ecasound can do
this.
I guess the closest thing in ecasound is this:
"""
-ge:open-threshold-%, close-thold-%,volume-mode
Threshold gate. Initially gate is closed. It is opened when volume goes over
'othreshold' percent. After this, if volume drops below 'cthold' percent,
gate is closed and won't be opened again. If 'value_mode' is
'rms', average RMS volume is used. Otherwise peak average is used. When
closed, passing audio buffers are trucated to zero length.
"""
janne
5:25 p.m.
On Tue, 4 Mar 2003, janne halttunen wrote:
Levi Burton <donburton(a)sbcglobal.net> wrote:
> extremely usefull. My question is do any other software packages support
> this feature? For those who are not familiar with it (or cannot infer its
> meaning), it essentially just waits until the audio level reaches a specified
> threshold, then starts recording. I would love to know if ecasound can do
[...]
I guess the closest thing in ecasound is this:
-ge:open-threshold-%, close-thold-%,volume-mode
Yup, it does just the right thing. As a bonus, you can also
stop recording based on a volume level trigger (the 'close-thold-%'
param).
Simple use (peak/rms):
ecasound -i alsa -o foo.wav -gc:5,0,0
ecasound -i alsa -o foo.wav -gc:5,0,rms
--
http://www.eca.cx
Audio software for Linux!
7931
days inactive
7937
days old
linux-audio-user@lists.linuxaudio.org
12 comments
9 participants
participants (9)
-
Guy Daniel CLOTILDE -
janne halttunen -
Kai Vehmanen -
Levi Burton -
luis jure -
Maarten de Boer -
Richard Seymour -
Steve Harris -
Tim Hall