You are producing only odd harmonics, as far as I can see. The frequency seems to be a bit high to tell,
but when I brought it down to 400 Hz I could hear the odd spectrum.

Also the max NH should be calculated as sr/(2*f).
========================
Dr Victor Lazzarini
Dean of Arts, Celtic Studies and Philosophy,
Maynooth University,
Maynooth, Co Kildare, Ireland
Tel: 00 353 7086936
Fax: 00 353 1 7086952 

> On 23 Aug 2016, at 06:20, Michael Saunders  wrote:
> 
> I want to get gbuzz to produce a spectrum with only odd harmonics.
> 
> Dodge and Jerse (2nd ed., pp.170--171) describe a method that I implemented in Csound (minimal example below)---basically generating full and even spectra and subtracting.
> 
> The two buzz opcodes appear to do just what they should---the spectra of ao and ae are as expected---but when I subtract ae, the even harmonics are not cancelled out.  I get a full spectrum, much like ao.  What is wrong?
> 
> 
> 
> 
> 
> 
> sr = 96000
> kr = 96000
> ksmps = 1
> nchnls = 2
> 0dbfs=1
> 
> instr 1
> knho=((sr-50)/4000)-1;number of odd partials
> knhe=(.5*(sr-50)/8000)-1;number of even partials
> ao gbuzz .5, 4000, knho, 1, .5, giCos ;odd
> ae gbuzz .5, 8000, knhe, 1, (.5^2), giCos;even
> out ao-(knhe/knho)*ae
> endin
> 
> 
> 
> 
> giCos  ftgen 0, 0, 2^10+1, 11, 1
> 
> i1 0 1
> 
> e
>  
> 
> 
> 
> Csound mailing list Csound@listserv.heanet.ie https://listserv.heanet.ie/cgi-bin/wa?A0=CSOUND Send bugs reports to https://github.com/csound/csound/issues Discussions of bugs and features can be posted here

Csound mailing list
Csound@listserv.heanet.ie
https://listserv.heanet.ie/cgi-bin/wa?A0=CSOUND
Send bugs reports to
        https://github.com/csound/csound/issues
Discussions of bugs and features can be posted here

Thank you.  I corrected the line about the Nyquist and tried again.  When I look at the spectrum with my sound editor ("frequency analysis" on Audition), I see all the peaks, even and odd  (I had chosen a high frequency to make this easy to see).  At lower frequencies, I see the same thing.  

On Tue, Aug 23, 2016 at 7:41 AM, Victor Lazzarini <Victor.Lazzarini@nuim.ie> wrote:

You are producing only odd harmonics, as far as I can see. The frequency seems to be a bit high to tell,
but when I brought it down to 400 Hz I could hear the odd spectrum.

Also the max NH should be calculated as sr/(2*f).
========================
Dr Victor Lazzarini
Dean of Arts, Celtic Studies and Philosophy,
Maynooth University,
Maynooth, Co Kildare, Ireland
Tel: 00 353 7086936
Fax: 00 353 1 7086952

> On 23 Aug 2016, at 06:20, Michael Saunders <odradek5@gmail.com> wrote:
>
> I want to get gbuzz to produce a spectrum with only odd harmonics.
>
> Dodge and Jerse (2nd ed., pp.170--171) describe a method that I implemented in Csound (minimal example below)---basically generating full and even spectra and subtracting.
>
> The two buzz opcodes appear to do just what they should---the spectra of ao and ae are as expected---but when I subtract ae, the even harmonics are not cancelled out.  I get a full spectrum, much like ao.  What is wrong?
>
> <CsoundSynthesizer>
> <CsOptions>
> </CsOptions>
> <CsInstruments>
>
> sr = 96000
> kr = 96000
> ksmps = 1
> nchnls = 2
> 0dbfs=1
>
> instr 1
> knho=((sr-50)/4000)-1;number of odd partials
> knhe=(.5*(sr-50)/8000)-1;number of even partials
> ao gbuzz .5, 4000, knho, 1, .5, giCos ;odd
> ae gbuzz .5, 8000, knhe, 1, (.5^2), giCos;even
> out ao-(knhe/knho)*ae
> endin
>
> </CsInstruments>
> <CsScore>
>
> giCos  ftgen 0, 0, 2^10+1, 11, 1
>
> i1 0 1
>
> e
>  </CsScore>
> </CsoundSynthesizer>
>
>

> Csound mailing list Csound@listserv.heanet.ie https://listserv.heanet.ie/cgi-bin/wa?A0=CSOUND Send bugs reports to https://github.com/csound/csound/issues Discussions of bugs and features can be posted here

Csound mailing list
Csound@listserv.heanet.ie
https://listserv.heanet.ie/cgi-bin/wa?A0=CSOUND
Send bugs reports to
        https://github.com/csound/csound/issues
Discussions of bugs and features can be posted here

Csound mailing list Csound@listserv.heanet.ie https://listserv.heanet.ie/cgi-bin/wa?A0=CSOUND Send bugs reports to https://github.com/csound/csound/issues Discussions of bugs and features can be posted here

I had a look again. The even harmonics are only partially suppressed, because of the shape of the
gbuzz spectrum given by kmult.

Try this instead

instr 1
kh linseg 0,p3,1
ih = sr/(2*p5)
af1 gbuzz p4,p5,ih,1,0.9,1
af2 gbuzz p4,2*p5,ih/2,1,0.9^2,1
  out af1-(kh/2)*af2
  dispfft af1-(kh/2)*af2,0.5,2048
endin

If you use buzz, as described in the book, it will work better.

========================
Dr Victor Lazzarini
Dean of Arts, Celtic Studies and Philosophy,
Maynooth University,
Maynooth, Co Kildare, Ireland
Tel: 00 353 7086936
Fax: 00 353 1 7086952 

> On 23 Aug 2016, at 15:16, Victor Lazzarini  wrote:
> 
> Look again. You can use dispfft for that. Sweep the odd partials gain and see the effect. 
> 
> Victor Lazzarini
> Dean of Arts, Celtic Studies, and Philosophy
> Maynooth University
> Ireland
> 
> On 23 Aug 2016, at 15:00, Michael Saunders  wrote:
> 
>> Thank you.  I corrected the line about the Nyquist and tried again.  When I look at the spectrum with my sound editor ("frequency analysis" on Audition), I see all the peaks, even and odd  (I had chosen a high frequency to make this easy to see).  At lower frequencies, I see the same thing.  
>> 
>> On Tue, Aug 23, 2016 at 7:41 AM, Victor Lazzarini  wrote:
>> You are producing only odd harmonics, as far as I can see. The frequency seems to be a bit high to tell,
>> but when I brought it down to 400 Hz I could hear the odd spectrum.
>> 
>> Also the max NH should be calculated as sr/(2*f).
>> ========================
>> Dr Victor Lazzarini
>> Dean of Arts, Celtic Studies and Philosophy,
>> Maynooth University,
>> Maynooth, Co Kildare, Ireland
>> Tel: 00 353 7086936
>> Fax: 00 353 1 7086952
>> 
>> > On 23 Aug 2016, at 06:20, Michael Saunders  wrote:
>> >
>> > I want to get gbuzz to produce a spectrum with only odd harmonics.
>> >
>> > Dodge and Jerse (2nd ed., pp.170--171) describe a method that I implemented in Csound (minimal example below)---basically generating full and even spectra and subtracting.
>> >
>> > The two buzz opcodes appear to do just what they should---the spectra of ao and ae are as expected---but when I subtract ae, the even harmonics are not cancelled out.  I get a full spectrum, much like ao.  What is wrong?
>> >
>> > 
>> > 
>> > 
>> > 
>> >
>> > sr = 96000
>> > kr = 96000
>> > ksmps = 1
>> > nchnls = 2
>> > 0dbfs=1
>> >
>> > instr 1
>> > knho=((sr-50)/4000)-1;number of odd partials
>> > knhe=(.5*(sr-50)/8000)-1;number of even partials
>> > ao gbuzz .5, 4000, knho, 1, .5, giCos ;odd
>> > ae gbuzz .5, 8000, knhe, 1, (.5^2), giCos;even
>> > out ao-(knhe/knho)*ae
>> > endin
>> >
>> > 
>> > 
>> >
>> > giCos  ftgen 0, 0, 2^10+1, 11, 1
>> >
>> > i1 0 1
>> >
>> > e
>> >  
>> > 
>> >
>> >
>> > Csound mailing list Csound@listserv.heanet.ie https://listserv.heanet.ie/cgi-bin/wa?A0=CSOUND Send bugs reports to https://github.com/csound/csound/issues Discussions of bugs and features can be posted here
>> 
>> Csound mailing list
>> Csound@listserv.heanet.ie
>> https://listserv.heanet.ie/cgi-bin/wa?A0=CSOUND
>> Send bugs reports to
>>         https://github.com/csound/csound/issues
>> Discussions of bugs and features can be posted here
>> 
>> Csound mailing list Csound@listserv.heanet.ie https://listserv.heanet.ie/cgi-bin/wa?A0=CSOUND Send bugs reports to https://github.com/csound/csound/issues Discussions of bugs and features can be posted here
> Csound mailing list Csound@listserv.heanet.ie https://listserv.heanet.ie/cgi-bin/wa?A0=CSOUND Send bugs reports to https://github.com/csound/csound/issues Discussions of bugs and features can be posted here

Csound mailing list
Csound@listserv.heanet.ie
https://listserv.heanet.ie/cgi-bin/wa?A0=CSOUND
Send bugs reports to
        https://github.com/csound/csound/issues
Discussions of bugs and features can be posted here