Yahoo Tuning Groups Ultimate Backup tuning Partials for sound synthesis

Hi.

Here I have seen several discussions on matching temperaments, tunings and scales to certains harmonic intervals (up to a given limit).

What about the reverse approach: to design timbres based on the intervals of a given scale?

What could be a set of pseudo-harmonic partials for 34EDO, for instance?

Should the partials match actual degrees of the scale, or should they correspond to the mean of the frequencies of a degree and the nearest harmonic?

Have you thought about this? Do you know some approach to relate scale/timbre this way?

Regards,
Hudson

--
'-------------------------------------------------------------------.
Hudson Lacerda <http://geocities.yahoo.com.br/hfmlacerda/>
*Nï¿½o deixe seu voto sumir! http://www.votoseguro.org/
*Apï¿½ie o Manifesto: http://www.votoseguro.com/alertaprofessores/

== THE WAR IN IRAQ COSTS ==
http://nationalpriorities.org/index.php?option=com_wrapper&Itemid=182
.-------------------------------------------------------------------'
--

_______________________________________________________
Yahoo! Acesso Grï¿½tis - Internet rï¿½pida e grï¿½tis. Instale o discador agora!
http://br.acesso.yahoo.com

🔗Aaron Wolf <backfromthesilo@yahoo.com>

You need to read the new edition of Bill Sethares' Tuning Timbre Spectrum Scale. This is
all addressed along with examples on CD!

-Aaron

--- In tuning@yahoogroups.com, Hudson Lacerda <hfmlacerda@...> wrote:
>
> Hi.
>
> Here I have seen several discussions on matching temperaments, tunings
> and scales to certains harmonic intervals (up to a given limit).
>
> What about the reverse approach: to design timbres based on the
> intervals of a given scale?
>
> What could be a set of pseudo-harmonic partials for 34EDO, for instance?
>
> Should the partials match actual degrees of the scale, or should they
> correspond to the mean of the frequencies of a degree and the nearest
> harmonic?
>
> Have you thought about this? Do you know some approach to relate
> scale/timbre this way?
>
> Regards,
> Hudson
>
> --
> '-------------------------------------------------------------------.
> Hudson Lacerda <http://geocities.yahoo.com.br/hfmlacerda/>
> *Não deixe seu voto sumir! http://www.votoseguro.org/
> *Apóie o Manifesto: http://www.votoseguro.com/alertaprofessores/
>
> == THE WAR IN IRAQ COSTS ==
> http://nationalpriorities.org/index.php?option=com_wrapper&Itemid=182
> .-------------------------------------------------------------------'
> --
>
>
>
> _______________________________________________________
> Yahoo! Acesso Grátis - Internet rápida e grátis. Instale o discador agora!
> http://br.acesso.yahoo.com
>

🔗Hudson Lacerda <hfmlacerda@yahoo.com.br>

Hi Aaron.

Aaron Wolf escreveu:
> You need to read the new edition of Bill Sethares' Tuning Timbre
> Spectrum Scale. This is all addressed along with examples on CD!
> > -Aaron

As I've said, I found some online info:
http://eceserv0.ece.wisc.edu/~sethares/ttss.html
The approach is described here:
http://eceserv0.ece.wisc.edu/~sethares/consemi.html
and there are examples of code (in Matlab language -- but it works fine on GNU Octave):
http://eceserv0.ece.wisc.edu/~sethares/comprog.html

I've listen to some sound examples from:
http://eceserv0.ece.wisc.edu/~sethares/html/soundexamples.html

So far, my impression is that the synthesis application could be more interesting (new electronic timbres instead of conventional instruments imitations). Those inharmonic bass-like, guitar-like, flute-like tones sound strange, they resemble low quality instruments -- for example, the guitar/bass tone sounds like a guitar/bass with bad/old strings.

Another lacking (omited?) matter is the (difficult) problem of know the perceived pitch of an inharmonic tone. Listen to the very first example:
http://eceserv0.ece.wisc.edu/~sethares/mp3s/challoct.mp3
What are the perceived pitches? Do they correspond the the nominal ``fundamentals''?

Anyway, that approach to match spectra and scales is very interesting, and can be shown useful to compose with inharmonic sounds and to synthesize tones with minor ajustments for some temperaments. (BTW, the 19ET guitar example sounds really good, I think this is due to its small amount of inharmonicity.)

Thanks again for the useful reference.
Hudson

_______________________________________________________
Yahoo! Acesso Gr�tis - Internet r�pida e gr�tis. Instale o discador agora!
http://br.acesso.yahoo.com

🔗Aaron Wolf <backfromthesilo@yahoo.com>

Hudson,

Look through those links and see info about his book,
or look at Amazon. The book is Tuning Timbre Spectrum Scale.
While I have complaints, and some of the examples are hard
to listen to, it is overall excellent.

I looked at his site for a while before getting the book.

The book is DEFINITELY the main complete idea, and I
highly recommend getting the book. Make sure it's
the second edition too.

Or if you can't spend any money on the book, get the local
library to purchase a copy. It really is worth it.

-Aaron

--- In tuning@yahoogroups.com, Hudson Lacerda <hfmlacerda@...> wrote:
>
> Hi Aaron.
>
> Aaron Wolf escreveu:
> > You need to read the new edition of Bill Sethares' Tuning Timbre
> > Spectrum Scale. This is all addressed along with examples on CD!
> >
> > -Aaron
>
> As I've said, I found some online info:
> http://eceserv0.ece.wisc.edu/~sethares/ttss.html
> The approach is described here:
> http://eceserv0.ece.wisc.edu/~sethares/consemi.html
> and there are examples of code (in Matlab language -- but it works
fine
> on GNU Octave):
> http://eceserv0.ece.wisc.edu/~sethares/comprog.html
>
> I've listen to some sound examples from:
> http://eceserv0.ece.wisc.edu/~sethares/html/soundexamples.html
>
> So far, my impression is that the synthesis application could be
more
> interesting (new electronic timbres instead of conventional
instruments
> imitations). Those inharmonic bass-like, guitar-like, flute-like
tones
> sound strange, they resemble low quality instruments -- for
example, the
> guitar/bass tone sounds like a guitar/bass with bad/old strings.
>
> Another lacking (omited?) matter is the (difficult) problem of know
the
> perceived pitch of an inharmonic tone. Listen to the very first
example:
> http://eceserv0.ece.wisc.edu/~sethares/mp3s/challoct.mp3
> What are the perceived pitches? Do they correspond the the nominal
> ``fundamentals''?
>
> Anyway, that approach to match spectra and scales is very
interesting,
> and can be shown useful to compose with inharmonic sounds and to
> synthesize tones with minor ajustments for some temperaments. (BTW,
the
> 19ET guitar example sounds really good, I think this is due to its
small
> amount of inharmonicity.)
>
> Thanks again for the useful reference.
> Hudson
>
>
>
> _______________________________________________________
> Yahoo! Acesso Grátis - Internet rápida e grátis. Instale o discador
agora!
> http://br.acesso.yahoo.com
>

🔗Jon Szanto <jszanto@cox.net>

Aaron, Hudson:

--- In tuning@yahoogroups.com, "Aaron Wolf" <backfromthesilo@...> wrote:
> The book is DEFINITELY the main complete idea, and I
> highly recommend getting the book. Make sure it's
> the second edition too.

Yes, second edition seconded. But also take a look (and listen) to
some of Bill's musical pieces (as opposed to just examples from the
book) from his CDs:

http://eceserv0.ece.wisc.edu/~sethares/exo.html
http://eceserv0.ece.wisc.edu/~sethares/xentone.html

Bill is one of the people that puts the theory into action and creates
some very new musical worlds. Very fun stuff, hope you enjoy it...

Cheers,
Jon

🔗Hudson Lacerda <hfmlacerda@yahoo.com.br>

*From timbre to scale*

I did an improved version of the code `dissontest.m' from:
http://eceserv0.ece.wisc.edu/~sethares/comprog.html
It also uses the function `dissmeasure.m'.

`disson.m' implements a function that plots the dissonance measurements, and outputs the intervals for a scale made of the local minima and/or maxima of dissonance.

It is written in GNU Octave <http://www.octave.org/> language; it is not difficult however make the adaptations for Matlab.

Cheers,
Hudson

###############################################################################
###############################################################################
###############################################################################

## function [intervals,dissonance] = disson(freq,amp,range,inc,minmax);
##
## 03/03/2006
##
## Calcula dissonancia com base em um espectro de 'freq' e 'amp'
##
## Plota o grï¿½fico de dissonï¿½ncia e retorna
## - intervalos para uso em escala musical (mï¿½nimos locais de dissonï¿½ncia)
## - e ainda a dissonï¿½ncia correspondente a cada intervalo
##
## minmax == -1, retorna os minimos locais de dissonancia (default)
## minmax == 1, retorna os maximos locais de dissonancia
## minmax == 0, retorna os minimos e maximos locais de dissonancia
##
## CODE ADAPTED/DEVELOPED BY HUDSON LACERDA FROM 'disstest.m':
## William A. Sethares
## http://eceserv0.ece.wisc.edu/~sethares/comprog.html

function [out1,out2] = disson(freq,amp,range,inc,minmax);

if(nargin==0 || isempty(freq))
freq=500*[1 2 3 4 5 6 7];
endif
if(nargin<2 || isempty(amp))
amp=ones(size(freq));
endif
if(nargin<4 || isempty(inc))
inc=0.01;
endif
if(nargin<3 || isempty(range))
range=2.0+(inc*2);
endif
if(nargin<5)
minmax=-1;
endif

alpha=1:inc:range;
diss=zeros(1,length(alpha));

%%
%% call function dissmeasure for each interval
%%
for i=2:length(alpha);
f=[freq, alpha(i)*freq];
a=[amp, amp];
d=dissmeasure(f, a);
diss(i)=d;
endfor
grid on
plot(alpha,diss);

%% local minima
x=find((diss(2:end-1)<diss(1:end-2))&(diss(2:end-1)<diss(3:end)))+1;
%% local maxima
y=find((diss(2:end-1)>diss(1:end-2))&(diss(2:end-1)>diss(3:end)))+1;

%% plot
hold on;
if(length(x)~=0)
plot(alpha(x),diss(x),'*b');
endif
if(length(y)~=0)
plot(alpha(y),diss(y),'*r');
endif
hold off;

%% output
if(nargout>0)
if(minmax==1)
x=y; % maxima
elseif(minmax==0)
x=[x,y]; % minima, maxima
endif
out1=[1,alpha(x)]';
if(nargout>1)
out2=[0,diss(x)]';
endif
endif

endfunction

_______________________________________________________
Yahoo! Acesso Grï¿½tis - Internet rï¿½pida e grï¿½tis. Instale o discador agora!
http://br.acesso.yahoo.com

🔗threesixesinarow <CACCOLA@NET1PLUS.COM>

🔗Carl Lumma <clumma@yahoo.com>

> > What about the reverse approach: to design timbres based on the
> > intervals of a given scale?
>
> An acoustic example but it doesn't elaborate the frequency part
> in the design.
>
> http://www.apentalpiano.co.uk/home.html

How do you find this stuff, Clark? Amazing!

I'm very curious to know exactly what the "Achromatic" design
elements are. I found the demonstration of octaves much more
impressive than that of the 2-note chords

http://www.apentalpiano.co.uk/hear.html

As for the arpegios and musical examples, well, there's clearly
a huge tone color difference between the instruments. The
apental piano sounds more like a fortepiano. The recording/
sampling quality isn't great, so it's hard to say more. I
think I'll buy the CD, though.

-Carl

🔗Carl Lumma <clumma@yahoo.com>

🔗Hudson Lacerda <hfmlacerda@yahoo.com.br>

Carl Lumma escreveu:
>>>Here I have seen several discussions on matching temperaments, >>>tunings and scales to certains harmonic intervals (up to a
>>>given limit).
>>>
>>>What about the reverse approach: to design timbres based on the >>>intervals of a given scale?
> > > It should be pointed out, and there has been much discussion on
> this in the past, that the Hammond organ has its partials in
> 12-tET. Its chords are indeed beat-free. Ironically, performers
> usually do everything they can to busy-up the sound, from
> crazy tremolo to broadcasting the sound out of a spinning
> speaker.

It seems that beating sounds are interesting. For synthesized sounds, this is indeed hold because static synthetic sounds are *very* boring.

Hudson

== THE WAR IN IRAQ COSTS ==
http://nationalpriorities.org/index.php?option=com_wrapper&Itemid=182
.-------------------------------------------------------------------'
--

_______________________________________________________
Yahoo! Acesso Grï¿½tis - Internet rï¿½pida e grï¿½tis. Instale o discador agora!
http://br.acesso.yahoo.com

🔗Hudson Lacerda <hfmlacerda@yahoo.com.br>

threesixesinarow escreveu:
> An acoustic example but it doesn't elaborate the frequency part in the > design.
> > http://www.apentalpiano.co.uk/home.html

There one can read:

<<<<
In the Apental Piano the striking point of the hammers has been moved to a point one fifth of the way along the string. It then produces no fifth harmonic, the beats are greatly reduced and the sound is sweeter to many listeners' ears. I have coined the term 'Apental' for this piano, suggesting "no fifth".
>>>>

The basic insight (greater 12EDO errors involve beats with a 5th harmonic in the interval: 4:5, 5:6, 5:3, 8:5) is interesting, and the approach (filter out the 5th harmonic) can be used for synthesized sounds. (For example, the bad fifths of 19EDO or 22EDO can be made smooth by filtering out -- or skipping -- the harmonics 3 and 9.)

Hudson Lacerda

== THE WAR IN IRAQ COSTS ==
http://nationalpriorities.org/index.php?option=com_wrapper&Itemid=182
.-------------------------------------------------------------------'
--

_______________________________________________________
Yahoo! Acesso Grï¿½tis - Internet rï¿½pida e grï¿½tis. Instale o discador agora!
http://br.acesso.yahoo.com

🔗Hudson Lacerda <hfmlacerda@yahoo.com.br>

Jon Szanto escreveu:
> Yes, second edition seconded. But also take a look (and listen) to
> some of Bill's musical pieces (as opposed to just examples from the
> book) from his CDs:
> > http://eceserv0.ece.wisc.edu/~sethares/exo.html
> http://eceserv0.ece.wisc.edu/~sethares/xentone.html

Hi all,

Has someone an implementation of the spectral mapping described here?
`Consonance Based Spectral Mappings'
http://eceserv0.ece.wisc.edu/~sethares/paperspdf/cmj98.pdf

Is there some Csound, puredata or C implementation? Is that mapping possible to do in real time from sounds of acoustic instruments?

Regards,
Hudson

_______________________________________________________
Yahoo! Acesso Gr�tis - Internet r�pida e gr�tis. Instale o discador agora!
http://br.acesso.yahoo.com

🔗Ozan Yarman <ozanyarman@ozanyarman.com>

I did not like the tone of the apental piano when the 5th partial was
eliminated. It hampers harmony to a great extent the way I hear it.

----- Original Message -----
From: "Hudson Lacerda" <hfmlacerda@yahoo.com.br>
To: <tuning@yahoogroups.com>
Sent: 07 Mart 2006 Salï¿½ 17:29
Subject: Re: [tuning] Re: Partials for sound synthesis

> threesixesinarow escreveu:
> > An acoustic example but it doesn't elaborate the frequency part in the
> > design.
> >
> > http://www.apentalpiano.co.uk/home.html
>
> There one can read:
>
> <<<<
> In the Apental Piano the striking point of the hammers has been moved to
> a point one fifth of the way along the string. It then produces no fifth
> harmonic, the beats are greatly reduced and the sound is sweeter to many
> listeners' ears. I have coined the term 'Apental' for this piano,
> suggesting "no fifth".
> >>>>
>
> The basic insight (greater 12EDO errors involve beats with a 5th
> harmonic in the interval: 4:5, 5:6, 5:3, 8:5) is interesting, and the
> approach (filter out the 5th harmonic) can be used for synthesized
> sounds. (For example, the bad fifths of 19EDO or 22EDO can be made
> smooth by filtering out -- or skipping -- the harmonics 3 and 9.)
>
> Hudson Lacerda
>
> --
> '-------------------------------------------------------------------.
> Hudson Lacerda <http://geocities.yahoo.com.br/hfmlacerda/>
> *Nï¿½o deixe seu voto sumir! http://www.votoseguro.org/
> *Apï¿½ie o Manifesto: http://www.votoseguro.com/alertaprofessores/
>

🔗Jon Szanto <jszanto@cox.net>

🔗Keenan Pepper <keenanpepper@gmail.com>

🔗Hudson Lacerda <hfmlacerda@yahoo.com.br>

Keenan Pepper escreveu:
> On 3/7/06, Hudson Lacerda <hfmlacerda@yahoo.com.br> wrote:
> [...]
> >>The basic insight (greater 12EDO errors involve beats with a 5th
>>harmonic in the interval: 4:5, 5:6, 5:3, 8:5) is interesting, and the
>>approach (filter out the 5th harmonic) can be used for synthesized
>>sounds. (For example, the bad fifths of 19EDO or 22EDO can be made
>>smooth by filtering out -- or skipping -- the harmonics 3 and 9.)
> > > And striking the string exactly in the middle would be perfect for
> Bohlen-Pierce, right?
> > Keenan

Exact!

This yet another conclusion, possibly useful even for acoustic instrumental music. (By the other hand, as Ozan pointed out, this causes a big and not ever desirable timbre change...)

For signal processing, one can try emphasize some harmonic relations by using Chebyschev distortion, for example.

Hudson

_______________________________________________________
Yahoo! Acesso Gr�tis - Internet r�pida e gr�tis. Instale o discador agora!
http://br.acesso.yahoo.com

🔗Gene Ward Smith <genewardsmith@coolgoose.com>

🔗Carl Lumma <clumma@yahoo.com>

🔗Yahya Abdal-Aziz <yahya@melbpc.org.au>

In [tuning], Carl Lumma wrote:

[snip]
> It should be pointed out, and there has been much discussion on
> this in the past, that the Hammond organ has its partials in
> 12-tET. Its chords are indeed beat-free. ...

Wow! That's a BIG claim!!! ;-)

When you say chords, Carl, do you mean just the usual diatonic
triads? Including diminished? Or do you include various 7th
chords and others using 2nds and 6ths or clusters?

> ... Ironically, performers
> usually do everything they can to busy-up the sound, from
> crazy tremolo to broadcasting the sound out of a spinning
> speaker.

Yeah, purity _is_ hard to take! ;-)

This would go some way to explaining why I find the majority
of electronic organs, and much synthesised sound, DEEPLY
unsatisfying. OTOH, I love the typical beating of piano tone.

Regards,
Yahya

--
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.1.375 / Virus Database: 268.2.0/275 - Release Date: 6/3/06

🔗Carl Lumma <clumma@yahoo.com>

> [snip]
> > It should be pointed out, and there has been much discussion on
> > this in the past, that the Hammond organ has its partials in
> > 12-tET. Its chords are indeed beat-free. ...
>
> Wow! That's a BIG claim!!! ;-)
>
> When you say chords, Carl, do you mean just the usual diatonic
> triads? Including diminished? Or do you include various 7th
> chords and others using 2nds and 6ths or clusters?

Any chord whose fundamentals and lower-order partials (aside
from the fact that the Hammond's tonewheels do not produce
perfectly pure sine waves, they don't employ all that many
partials) nowhere coincide in a critical band, will be
essentially beat-free.

That's true of any instrument, of course, so one still has to
do the math. The neat thing is the Hammond is essentially
in JI in all 12 keys AND benefits from a relative lack of
higher partials. In the middle of the keyboard, all chords
without half-steps are basically beat-free.

> > ... Ironically, performers
> > usually do everything they can to busy-up the sound, from
> > crazy tremolo to broadcasting the sound out of a spinning
> > speaker.
>
> Yeah, purity _is_ hard to take! ;-)
>
> This would go some way to explaining why I find the majority
> of electronic organs, and much synthesised sound, DEEPLY
> unsatisfying. OTOH, I love the typical beating of piano tone.

Let's not get ahead of ourselves. Motion is attractive,
but the motion of an oboe is not beating, it's motion as I
described recently on MMM (in reply to Hudson). And purity
can be desirable as well.

Until recently, most "hammond" sounds on synths did not
replicate the tempered partials of the original instrument.
Other synth sounds will beat more or less like any acoustic
instrument (though for some reason it is often less
pronouced).

Pipe organs, though they have a "chiff" at the attack and some
variance due to sharing an air supply among many pipes, more
or less produce a motionless sound. Traditionally, reverberant
spaces were used to create thick textures of swirling
harmonics. But around the turn of the last century, theater
organs became popular, with their tremulant sound. So this
sort of thing isn't unique to the Hammond.

-Carl

🔗threesixesinarow <CACCOLA@NET1PLUS.COM>

--- In tuning@yahoogroups.com, "Carl Lumma" <clumma@...> wrote:
>
> > I did not like the tone of the apental piano when the 5th partial
> > was eliminated. It hampers harmony to a great extent the way I
> > hear it.
>
> I'm not sure if the difference was due to the lack of those
> harmonics, or some other design trait. (Also, there was
> apparently a difference in mic position or something between
> the recordings.) In fact, from the photos it looks as if the
> multiples of 5 are only being attenuated in the bass of this
> instrument. Anyone agree? Do you know any more, Clark?
>
> -Carl
>

It's easier to move the string so it acts like the hammer's softer.

🔗Carl Lumma <clumma@yahoo.com>

🔗threesixesinarow <CACCOLA@NET1PLUS.COM>

--- In tuning@yahoogroups.com, "Carl Lumma" <clumma@...> wrote:
>
> > > I'm not sure if the difference was due to the lack of those
> > > harmonics, or some other design trait. (Also, there was
> > > apparently a difference in mic position or something between
> > > the recordings.) In fact, from the photos it looks as if the
> > > multiples of 5 are only being attenuated in the bass of this
> > > instrument. Anyone agree? Do you know any more, Clark?
> > >
> > > -Carl
> > >
> >
> > It's easier to move the string so it acts like the hammer's
softer.
>
> Sorry, I don't follow. Could you explain? I super-interested
> in this instrument.
>
> -Carl
>
The hammer strikes about 40 mm from the end of a 400mm trichord unison
but here it's 80mm, it might take a little longer for the hammer to
bounce off of them.

Müller Ditanaklasis has funny striking points, I don't know how
systematic.

http://www.accademiacristofori.it/strumenti/F07.htm