Risset's Drum Instrument
I've tried to create Risset's drum instrument based of of the example from Dodge and Reese's "Computer Music" on page 104. I think I have it all set, but I have a couple of questions on the patch I've written:
1.) I'm getting clicks every now and then when I "bang" the drum/instrument. I'm no sure if this is an interpolation issue, amplitude issue, etc. Any ideas?
2.) Right now I'm driving the amplitude envelopes through function objects. This works, but clicking/drawing in the function object really doesn't provide very precise control over the actual envelope. adsr~, on the other hand, only works with a few line segments, it seems. If someone has th Dodge and Reese, what would you suggest is the best way to implement amplitude envelopes such as those shown in the text? (Very close to f(x)=-log x)
3.) If you look in the inharm subpatch, you'll see how I've created a waveform for this portion of the sound. This waveform is composed of individual cosine waves at the following frequencies/amplitudes:
10th Partial: Amplitude = 1
16th Partial: Amplitude = 1.5
22nd Partial: Amplitude = 2
23rd Partial: Amplitude = 1.5
I have (possibly naively) calculated this waveform at 2048 points with an expr object and "peek~"ed those into a buffer, which I am reading through with a wave~ object. Nearly every computer music text I've read talks about doing this, but I didn't find it covered (very thoroughly, at least) in the MSP tutorials or documentation. The method I'm using here works, but is it the best method? Is there a better and/or more efficient way to do this?
Here is the patch:
Thanks for your help!
Best,
Brennon
As an aside, if anyone does have the Dodge and Reese, what is the equation for the two amplitude envelopes (F1 and F2) shown on page 104? I'm using the second edition, by the way.
Thanks again,
Brennon
The only clicks I hear seem to be because all the envelopes start
high, and because the things they are enveloping are free-running,
sometimes they will be at "unfortunate" points in their cycle.
Also, your "inharm" patch is not connected to the "/ 10." above it,
so all it produces is a click..
-C
--
Chris Muir | "There are many futures and only one status quo.
cbm@well.com | This is why conservatives mostly agree,
http://www.xfade.com | and radicals always argue." - Brian Eno
Quote: Chris Muir wrote on Sun, 06 January 2008 23:37
----------------------------------------------------
>
> Also, your "inharm" patch is not connected to the "/ 10." above it,
> so all it produces is a click..
>
Thanks, Chris. Any idea on generating those envelopes? I'm looking at a combination of expr objects and jasch's bspline--just seems like so much unnecessary work, though.
Thanks again,
Brennon
A simple solution would be to add a slide~ 100 100 after the envelopes.
Experiment with the arguments. Higher values will make the attacks softer.
My tl.envexp externals was built for dealing with envelopes with
exponential decays, but I unfortunately have not had the time (yet) to
do a UB port of it.
Best,
Trond
Brennon Bortz wrote:
> Quote: Chris Muir wrote on Sun, 06 January 2008 23:37
> ----------------------------------------------------
>> Also, your "inharm" patch is not connected to the "/ 10." above it,
>> so all it produces is a click..
>>
>
> Thanks, Chris. Any idea on generating those envelopes? I'm looking at a combination of expr objects and jasch's bspline--just seems like so much unnecessary work, though.
>
> Thanks again,
> Brennon
>
>
> --
> Brennon Bortz
> Composer / Engraver / Nutcase
>
Quote: Trond Lossius wrote on Mon, 07 January 2008 00:41
----------------------------------------------------
> A simple solution would be to add a slide~ 100 100 after the envelopes.
> Experiment with the arguments. Higher values will make the attacks softer.
>
This will certainly work, but still doesn't give the control I'd like to have.
> My tl.envexp externals was built for dealing with envelopes with
> exponential decays, but I unfortunately have not had the time (yet) to
> do a UB port of it.
>
This seems more interesting, however, and fortunately I'm still on a PowerPC-based Mac. Do you have the Dodge & Jerse? If so, I'd like to ask a couple of more specific questions relating to Risset's envelope specifications.
Thanks for your help,
Brennon
Yes I do, but it's at my office. I'll be able to access it tomorrow though.
Trond
> This seems more interesting, however, and fortunately I'm still on a PowerPC-based Mac. Do you have the Dodge & Jerse? If so, I'd like to ask a couple of more specific questions relating to Risset's envelope specifications.
I have the Dodge & Jerse, but it's at the office and I'm not...I'll
be going over there later, so might be able to help out then.
On Jan 7, 2008, at 11:25 AM, Brennon Bortz wrote:
>
> This seems more interesting, however, and fortunately I'm still on
> a PowerPC-based Mac. Do you have the Dodge & Jerse? If so, I'd
> like to ask a couple of more specific questions relating to
> Risset's envelope specifications.
----
Steven M. Miller
Professor, Contemporary Music Program
College of Santa Fe
Home
SFIFEM
Atrium Sound Space
OVOS
CMP
Quote: smill wrote on Mon, 07 January 2008 11:00
----------------------------------------------------
> I have the Dodge & Jerse, but it's at the office and I'm not...I'll
> be going over there later, so might be able to help out then.
>
Ahh, that's too bad. I was only now able to check the board and saw this post. I'm sure you're long gone from the office now!
When you do have a chance, I'm a bit confused on how to interpret the way Dodge & Jerse present Risset's equations for various amplitude envelopes, frequency envelopes, etc. For instance, on page 104 (2nd ed.), two functions are shown for two different amplitude anvelopes. They are obviously exponential functions, but Dodge and Jerse give no further insight. Each is labeled with 1 as the "peak" value of the function as 1. I'm assuming that .004 and .00012 are the values of the function for whatever magical value of x (or, time), is at the end of the graph of the function. I've estimated these by taking an exponential function in an expr object and driving the output of a line object through it for f(x)=1 down to f(x)=.004 and .00012, respectiively. I've tweaked the curve of the function slightly to compensate for the quicker falloff of F2. Am I approaching this correctly?
I assume that Risset provide a bit more depth in his Catalog. Have you ever worked with it? Does he actually describe his approach to things such as these, or is it left to the reader to approximate?
Thanks again for your help!
Best,
Brennon
> When you do have a chance, I'm a bit confused on how to interpret the way Dodge & Jerse present Risset's equations for various amplitude envelopes, frequency envelopes, etc. For instance, on page 104 (2nd ed.), two functions are shown for two different amplitude anvelopes. They are obviously exponential functions, but Dodge and Jerse give no further insight. Each is labeled with 1 as the "peak" value of the function as 1. I'm assuming that .004 and .00012 are the values of the function for whatever magical value of x (or, time), is at the end of the graph of the function. I've estimated these by taking an exponential function in an expr object and driving the output of a line object through it for f(x)=1 down to f(x)=.004 and .00012, respectiively. I've tweaked the curve of the function slightly to compensate for the quicker falloff of F2. Am I approaching this correctly?
>
> I assume that Risset provide a bit more depth in his Catalog. Have you ever worked with it? Does he actually describe his approach to things such as these, or is it left to the reader to approximate?
From the illustration I'm not sure, but you can find a Csound version
of his instrument here, complete with sound:
I'm convinced that all of his instruments has been available as Csound
instruments at a CD at some point, but I didn't managae to track it down
from a quick googling.
Best,
Trond
Quote: Trond Lossius wrote on Wed, 09 January 2008 05:42
----------------------------------------------------
> From the illustration I'm not sure, but you can find a Csound version
> of his instrument here, complete with sound:
>
> http://www.csounds.com/catalogfrom/risset/index.html
>
> I'm convinced that all of his instruments has been available as Csound
> instruments at a CD at some point, but I didn't managae to track it down
> from a quick googling.
>
Csound...my next endeavor. A bit backward, I know, but I'll be there soon, nonetheless.
Thanks Trond,
Brennon