Hi there, I'm planning to knock up a multi-band compressor, and I was
wondering if csound has any filters built for this job? My first
thought was to use butterworths, but I've just read that the combined
signal has a slight peak at the crossover point. Apparently
Linkwitz-Riley filters are better for the job. Do we have anything
like that?

Pete

I was thinking that too, and a big part of me wants to try and write
one, but I've already started 3 opcodes and never found the time to
finish any of them, so I should probably stop kidding myself that I'm
gonna get round to something like that in the near future, haha! :D

On Tue, Mar 1, 2016 at 8:50 PM, Dr. Richard Boulanger
 wrote:
> We could really use a multi-band compressor opcode!
>
> -dB
>
> _____________________________________________
> Dr. Richard Boulanger
> Professor of Electronic Production and Design
> Professional Writing and Music Technology Division
> Berklee College of Music
> 1140 Boylston Street, FB-75
> Boston, MA 02215-3693
> Office @ 161 Mass Ave - 4th Floor
> 617-747-2384 (office)  774-488-9166 (cell)
> ______________________________________________
> rboulanger@berklee.edu (email)
> https://www.facebook.com/richard.boulanger.58 (facebook)
> ______________________________________________
> President of Boulanger Labs - http://boulangerlabs.com
> Director of Csounds.com - http://csounds.com
> Author & Editor: The Csound Book - http://mitpress.mit.edu/books/csound-book
> Author & Editor: The Audio Programming Book -
> http://mitpress.mit.edu/books/audio-programming-book
> ______________________________________________
> about: http://www.boulangerlabs.com/about/richardboulanger/
> about: http://www.csounds.com/community/developers/dr-richard-boulanger/
> ______________________________________________
> music:
> http://www.csounds.com/community/developers/dr-richard-boulanger/dr-richard-boulanger-music/
>
> On Tue, Mar 1, 2016 at 3:15 PM, Peter Burgess
>  wrote:
>>
>> Hi there, I'm planning to knock up a multi-band compressor, and I was
>> wondering if csound has any filters built for this job? My first
>> thought was to use butterworths, but I've just read that the combined
>> signal has a slight peak at the crossover point. Apparently
>> Linkwitz-Riley filters are better for the job. Do we have anything
>> like that?
>>
>> Pete
>>
>> --
>> http://algorythmradio.com
>> https://soundcloud.com/algorythmradio
>>
>> 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

Why not just use the opcodes we have to create one?

That's fine, and is what I was intending, but does Csound have a
suitable filter for a crossover? As I have read and now also tested,
the butterworth isn't suitable

On Tue, Mar 1, 2016 at 11:30 PM, Rory Walsh  wrote:
> A suite of mastering UDOs would be a good start. Gleb proposed such a thing
> some time back, but I don't think much came of it.
>
> On 1 March 2016 at 23:28, Dr. Richard Boulanger 
> wrote:
>>
>> For sure Rory,
>>
>> That is what we do; but Csound could really use some "mastering" opcodes
>> not just "mastering instruments"  Dr.B.
>> _____________________________________________
>> On Tue, Mar 1, 2016 at 6:14 PM, Rory Walsh  wrote:
>>>
>>> Why not just use the opcodes we have to create one?
>>>
>>> On 1 Mar 2016 21:35, "Peter Burgess" 
>>> wrote:
>>>>
>>>> I was thinking that too, and a big part of me wants to try and write
>>>> one, but I've already started 3 opcodes and never found the time to
>>>> finish any of them, so I should probably stop kidding myself that I'm
>>>> gonna get round to something like that in the near future, haha! :D
>>>>
>>>> On Tue, Mar 1, 2016 at 8:50 PM, Dr. Richard Boulanger
>>>>  wrote:
>>>> > We could really use a multi-band compressor opcode!
>>>> >
>>>> > -dB
>>>> >
>>>> > _____________________________________________
>>>> > Dr. Richard Boulanger
>>>> > Professor of Electronic Production and Design
>>>> > Professional Writing and Music Technology Division
>>>> > Berklee College of Music
>>>> > 1140 Boylston Street, FB-75
>>>> > Boston, MA 02215-3693
>>>> > Office @ 161 Mass Ave - 4th Floor
>>>> > 617-747-2384 (office)  774-488-9166 (cell)
>>>> > ______________________________________________
>>>> > rboulanger@berklee.edu (email)
>>>> > https://www.facebook.com/richard.boulanger.58 (facebook)
>>>> > ______________________________________________
>>>> > President of Boulanger Labs - http://boulangerlabs.com
>>>> > Director of Csounds.com - http://csounds.com
>>>> > Author & Editor: The Csound Book -
>>>> > http://mitpress.mit.edu/books/csound-book
>>>> > Author & Editor: The Audio Programming Book -
>>>> > http://mitpress.mit.edu/books/audio-programming-book
>>>> > ______________________________________________
>>>> > about: http://www.boulangerlabs.com/about/richardboulanger/
>>>> > about:
>>>> > http://www.csounds.com/community/developers/dr-richard-boulanger/
>>>> > ______________________________________________
>>>> > music:
>>>> >
>>>> > http://www.csounds.com/community/developers/dr-richard-boulanger/dr-richard-boulanger-music/
>>>> >
>>>> > On Tue, Mar 1, 2016 at 3:15 PM, Peter Burgess
>>>> >  wrote:
>>>> >>
>>>> >> Hi there, I'm planning to knock up a multi-band compressor, and I was
>>>> >> wondering if csound has any filters built for this job? My first
>>>> >> thought was to use butterworths, but I've just read that the combined
>>>> >> signal has a slight peak at the crossover point. Apparently
>>>> >> Linkwitz-Riley filters are better for the job. Do we have anything
>>>> >> like that?
>>>> >>
>>>> >> Pete
>>>> >>
>>>> >> --
>>>> >> http://algorythmradio.com
>>>> >> https://soundcloud.com/algorythmradio
>>>> >>
>>>> >> 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
>>>>
>>>>
>>>>
>>>> --
>>>> http://algorythmradio.com
>>>> https://soundcloud.com/algorythmradio
>>>>
>>>> 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

 Ah hang on, a further read about crossovers says that the most
commonly used filters are 4th order Linkwitz–Riley filters, which
apparently just use two cascaded 2nd order butterworth filter.... I'm
no DSP expert, but doesn't that just make it a 4th order butterworth?

On Tue, Mar 1, 2016 at 11:36 PM, Peter Burgess
 wrote:
> That's fine, and is what I was intending, but does Csound have a
> suitable filter for a crossover? As I have read and now also tested,
> the butterworth isn't suitable
>
> On Tue, Mar 1, 2016 at 11:30 PM, Rory Walsh  wrote:
>> A suite of mastering UDOs would be a good start. Gleb proposed such a thing
>> some time back, but I don't think much came of it.
>>
>> On 1 March 2016 at 23:28, Dr. Richard Boulanger 
>> wrote:
>>>
>>> For sure Rory,
>>>
>>> That is what we do; but Csound could really use some "mastering" opcodes
>>> not just "mastering instruments"  Dr.B.
>>> _____________________________________________
>>> On Tue, Mar 1, 2016 at 6:14 PM, Rory Walsh  wrote:
>>>>
>>>> Why not just use the opcodes we have to create one?
>>>>
>>>> On 1 Mar 2016 21:35, "Peter Burgess" 
>>>> wrote:
>>>>>
>>>>> I was thinking that too, and a big part of me wants to try and write
>>>>> one, but I've already started 3 opcodes and never found the time to
>>>>> finish any of them, so I should probably stop kidding myself that I'm
>>>>> gonna get round to something like that in the near future, haha! :D
>>>>>
>>>>> On Tue, Mar 1, 2016 at 8:50 PM, Dr. Richard Boulanger
>>>>>  wrote:
>>>>> > We could really use a multi-band compressor opcode!
>>>>> >
>>>>> > -dB
>>>>> >
>>>>> > _____________________________________________
>>>>> > Dr. Richard Boulanger
>>>>> > Professor of Electronic Production and Design
>>>>> > Professional Writing and Music Technology Division
>>>>> > Berklee College of Music
>>>>> > 1140 Boylston Street, FB-75
>>>>> > Boston, MA 02215-3693
>>>>> > Office @ 161 Mass Ave - 4th Floor
>>>>> > 617-747-2384 (office)  774-488-9166 (cell)
>>>>> > ______________________________________________
>>>>> > rboulanger@berklee.edu (email)
>>>>> > https://www.facebook.com/richard.boulanger.58 (facebook)
>>>>> > ______________________________________________
>>>>> > President of Boulanger Labs - http://boulangerlabs.com
>>>>> > Director of Csounds.com - http://csounds.com
>>>>> > Author & Editor: The Csound Book -
>>>>> > http://mitpress.mit.edu/books/csound-book
>>>>> > Author & Editor: The Audio Programming Book -
>>>>> > http://mitpress.mit.edu/books/audio-programming-book
>>>>> > ______________________________________________
>>>>> > about: http://www.boulangerlabs.com/about/richardboulanger/
>>>>> > about:
>>>>> > http://www.csounds.com/community/developers/dr-richard-boulanger/
>>>>> > ______________________________________________
>>>>> > music:
>>>>> >
>>>>> > http://www.csounds.com/community/developers/dr-richard-boulanger/dr-richard-boulanger-music/
>>>>> >
>>>>> > On Tue, Mar 1, 2016 at 3:15 PM, Peter Burgess
>>>>> >  wrote:
>>>>> >>
>>>>> >> Hi there, I'm planning to knock up a multi-band compressor, and I was
>>>>> >> wondering if csound has any filters built for this job? My first
>>>>> >> thought was to use butterworths, but I've just read that the combined
>>>>> >> signal has a slight peak at the crossover point. Apparently
>>>>> >> Linkwitz-Riley filters are better for the job. Do we have anything
>>>>> >> like that?
>>>>> >>
>>>>> >> Pete
>>>>> >>
>>>>> >> --
>>>>> >> http://algorythmradio.com
>>>>> >> https://soundcloud.com/algorythmradio
>>>>> >>
>>>>> >> 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
>>>>>
>>>>>
>>>>>
>>>>> --
>>>>> http://algorythmradio.com
>>>>> https://soundcloud.com/algorythmradio
>>>>>
>>>>> 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
>
>
>
> --
> http://algorythmradio.com
> https://soundcloud.com/algorythmradio



-- 
http://algorythmradio.com
https://soundcloud.com/algorythmradio

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've done some 2 band crossover tests using these two butterworth
filter configurations:

icrossfreq = 60

; Crossover Test 1 (butxp opcodes)
alowsig1 butlp gacompsig, icrossfreq
alowsig2 butlp alowsig1, icrossfreq
ahisig1 buthp gacompsig, icrossfreq
ahisig2 buthp ahisig1, icrossfreq
out alowsig2 + ahisig2

; Crossover Test 2 (clfilt opcode)
inpol init 4
alowsig clfilt gacompsig, icrossfreq, 0, inpol
ahisig clfilt gacompsig, icrossfreq, 1, inpol
out alowsig + ahisig

I have tested these two crossovers with a rich kick sound, and with
pink noise, using various different crossover frequencies. When
watching the amplitudes in the terminal, it would appear that both
crossover setups produce an output with a higher peak somewhere,
compared to the unaffected sound.

For the kick, the unaltered sound has an amplitude of 0.49851.
With a crossover frequency of 60, Test 1 produces amplitudes of
between 0.72 and 0.86, while Test 2 produces amplitudes of between
0.75 and 0.94.

I couldn't hear a great deal of audible difference between the three,
so I spectrum analysed some samples. I have attached these spectrums,
as well as the audio samples.

The spectrum analysis shows a slight difference near the crossover
point for Test 1, but a much more noticeable difference around the
crossover for Test 2.

I am drawing the conclusion that Test 1's crossover is better, but are
there filters that will do even better still? Is it possible or easy
to get a totally level crossover? Or could I just limit the output of
the crossovers to the original signals level?

On Wed, Mar 2, 2016 at 12:58 AM, Peter Burgess
 wrote:
>  Ah hang on, a further read about crossovers says that the most
> commonly used filters are 4th order Linkwitz–Riley filters, which
> apparently just use two cascaded 2nd order butterworth filter.... I'm
> no DSP expert, but doesn't that just make it a 4th order butterworth?
>
> On Tue, Mar 1, 2016 at 11:36 PM, Peter Burgess
>  wrote:
>> That's fine, and is what I was intending, but does Csound have a
>> suitable filter for a crossover? As I have read and now also tested,
>> the butterworth isn't suitable
>>
>> On Tue, Mar 1, 2016 at 11:30 PM, Rory Walsh  wrote:
>>> A suite of mastering UDOs would be a good start. Gleb proposed such a thing
>>> some time back, but I don't think much came of it.
>>>
>>> On 1 March 2016 at 23:28, Dr. Richard Boulanger 
>>> wrote:
>>>>
>>>> For sure Rory,
>>>>
>>>> That is what we do; but Csound could really use some "mastering" opcodes
>>>> not just "mastering instruments"  Dr.B.
>>>> _____________________________________________
>>>> On Tue, Mar 1, 2016 at 6:14 PM, Rory Walsh  wrote:
>>>>>
>>>>> Why not just use the opcodes we have to create one?
>>>>>
>>>>> On 1 Mar 2016 21:35, "Peter Burgess" 
>>>>> wrote:
>>>>>>
>>>>>> I was thinking that too, and a big part of me wants to try and write
>>>>>> one, but I've already started 3 opcodes and never found the time to
>>>>>> finish any of them, so I should probably stop kidding myself that I'm
>>>>>> gonna get round to something like that in the near future, haha! :D
>>>>>>
>>>>>> On Tue, Mar 1, 2016 at 8:50 PM, Dr. Richard Boulanger
>>>>>>  wrote:
>>>>>> > We could really use a multi-band compressor opcode!
>>>>>> >
>>>>>> > -dB
>>>>>> >
>>>>>> > _____________________________________________
>>>>>> > Dr. Richard Boulanger
>>>>>> > Professor of Electronic Production and Design
>>>>>> > Professional Writing and Music Technology Division
>>>>>> > Berklee College of Music
>>>>>> > 1140 Boylston Street, FB-75
>>>>>> > Boston, MA 02215-3693
>>>>>> > Office @ 161 Mass Ave - 4th Floor
>>>>>> > 617-747-2384 (office)  774-488-9166 (cell)
>>>>>> > ______________________________________________
>>>>>> > rboulanger@berklee.edu (email)
>>>>>> > https://www.facebook.com/richard.boulanger.58 (facebook)
>>>>>> > ______________________________________________
>>>>>> > President of Boulanger Labs - http://boulangerlabs.com
>>>>>> > Director of Csounds.com - http://csounds.com
>>>>>> > Author & Editor: The Csound Book -
>>>>>> > http://mitpress.mit.edu/books/csound-book
>>>>>> > Author & Editor: The Audio Programming Book -
>>>>>> > http://mitpress.mit.edu/books/audio-programming-book
>>>>>> > ______________________________________________
>>>>>> > about: http://www.boulangerlabs.com/about/richardboulanger/
>>>>>> > about:
>>>>>> > http://www.csounds.com/community/developers/dr-richard-boulanger/
>>>>>> > ______________________________________________
>>>>>> > music:
>>>>>> >
>>>>>> > http://www.csounds.com/community/developers/dr-richard-boulanger/dr-richard-boulanger-music/
>>>>>> >
>>>>>> > On Tue, Mar 1, 2016 at 3:15 PM, Peter Burgess
>>>>>> >  wrote:
>>>>>> >>
>>>>>> >> Hi there, I'm planning to knock up a multi-band compressor, and I was
>>>>>> >> wondering if csound has any filters built for this job? My first
>>>>>> >> thought was to use butterworths, but I've just read that the combined
>>>>>> >> signal has a slight peak at the crossover point. Apparently
>>>>>> >> Linkwitz-Riley filters are better for the job. Do we have anything
>>>>>> >> like that?
>>>>>> >>
>>>>>> >> Pete
>>>>>> >>
>>>>>> >> --
>>>>>> >> http://algorythmradio.com
>>>>>> >> https://soundcloud.com/algorythmradio
>>>>>> >>
>>>>>> >> 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
>>>>>>
>>>>>>
>>>>>>
>>>>>> --
>>>>>> http://algorythmradio.com
>>>>>> https://soundcloud.com/algorythmradio
>>>>>>
>>>>>> 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
>>
>>
>>
>> --
>> http://algorythmradio.com
>> https://soundcloud.com/algorythmradio
>
>
>
> --
> http://algorythmradio.com
> https://soundcloud.com/algorythmradio



-- 
http://algorythmradio.com
https://soundcloud.com/algorythmradio

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

Anyone have any ideas on this?

Basically I just need to know, is it generally acceptable to just live
with a bit of a peak in the crossover points, or should I aim for a
near flat response all round?

Are there any better filters for the job, or any tricks (like simply
using a limiter) that will help me overcome?

Pete

On Wed, Mar 2, 2016 at 4:04 PM, Peter Burgess
 wrote:
> I've done some 2 band crossover tests using these two butterworth
> filter configurations:
>
> icrossfreq = 60
>
> ; Crossover Test 1 (butxp opcodes)
> alowsig1 butlp gacompsig, icrossfreq
> alowsig2 butlp alowsig1, icrossfreq
> ahisig1 buthp gacompsig, icrossfreq
> ahisig2 buthp ahisig1, icrossfreq
> out alowsig2 + ahisig2
>
> ; Crossover Test 2 (clfilt opcode)
> inpol init 4
> alowsig clfilt gacompsig, icrossfreq, 0, inpol
> ahisig clfilt gacompsig, icrossfreq, 1, inpol
> out alowsig + ahisig
>
> I have tested these two crossovers with a rich kick sound, and with
> pink noise, using various different crossover frequencies. When
> watching the amplitudes in the terminal, it would appear that both
> crossover setups produce an output with a higher peak somewhere,
> compared to the unaffected sound.
>
> For the kick, the unaltered sound has an amplitude of 0.49851.
> With a crossover frequency of 60, Test 1 produces amplitudes of
> between 0.72 and 0.86, while Test 2 produces amplitudes of between
> 0.75 and 0.94.
>
> I couldn't hear a great deal of audible difference between the three,
> so I spectrum analysed some samples. I have attached these spectrums,
> as well as the audio samples.
>
> The spectrum analysis shows a slight difference near the crossover
> point for Test 1, but a much more noticeable difference around the
> crossover for Test 2.
>
> I am drawing the conclusion that Test 1's crossover is better, but are
> there filters that will do even better still? Is it possible or easy
> to get a totally level crossover? Or could I just limit the output of
> the crossovers to the original signals level?
>
> On Wed, Mar 2, 2016 at 12:58 AM, Peter Burgess
>  wrote:
>>  Ah hang on, a further read about crossovers says that the most
>> commonly used filters are 4th order Linkwitz–Riley filters, which
>> apparently just use two cascaded 2nd order butterworth filter.... I'm
>> no DSP expert, but doesn't that just make it a 4th order butterworth?
>>
>> On Tue, Mar 1, 2016 at 11:36 PM, Peter Burgess
>>  wrote:
>>> That's fine, and is what I was intending, but does Csound have a
>>> suitable filter for a crossover? As I have read and now also tested,
>>> the butterworth isn't suitable
>>>
>>> On Tue, Mar 1, 2016 at 11:30 PM, Rory Walsh  wrote:
>>>> A suite of mastering UDOs would be a good start. Gleb proposed such a thing
>>>> some time back, but I don't think much came of it.
>>>>
>>>> On 1 March 2016 at 23:28, Dr. Richard Boulanger 
>>>> wrote:
>>>>>
>>>>> For sure Rory,
>>>>>
>>>>> That is what we do; but Csound could really use some "mastering" opcodes
>>>>> not just "mastering instruments"  Dr.B.
>>>>> _____________________________________________
>>>>> On Tue, Mar 1, 2016 at 6:14 PM, Rory Walsh  wrote:
>>>>>>
>>>>>> Why not just use the opcodes we have to create one?
>>>>>>
>>>>>> On 1 Mar 2016 21:35, "Peter Burgess" 
>>>>>> wrote:
>>>>>>>
>>>>>>> I was thinking that too, and a big part of me wants to try and write
>>>>>>> one, but I've already started 3 opcodes and never found the time to
>>>>>>> finish any of them, so I should probably stop kidding myself that I'm
>>>>>>> gonna get round to something like that in the near future, haha! :D
>>>>>>>
>>>>>>> On Tue, Mar 1, 2016 at 8:50 PM, Dr. Richard Boulanger
>>>>>>>  wrote:
>>>>>>> > We could really use a multi-band compressor opcode!
>>>>>>> >
>>>>>>> > -dB
>>>>>>> >
>>>>>>> > _____________________________________________
>>>>>>> > Dr. Richard Boulanger
>>>>>>> > Professor of Electronic Production and Design
>>>>>>> > Professional Writing and Music Technology Division
>>>>>>> > Berklee College of Music
>>>>>>> > 1140 Boylston Street, FB-75
>>>>>>> > Boston, MA 02215-3693
>>>>>>> > Office @ 161 Mass Ave - 4th Floor
>>>>>>> > 617-747-2384 (office)  774-488-9166 (cell)
>>>>>>> > ______________________________________________
>>>>>>> > rboulanger@berklee.edu (email)
>>>>>>> > https://www.facebook.com/richard.boulanger.58 (facebook)
>>>>>>> > ______________________________________________
>>>>>>> > President of Boulanger Labs - http://boulangerlabs.com
>>>>>>> > Director of Csounds.com - http://csounds.com
>>>>>>> > Author & Editor: The Csound Book -
>>>>>>> > http://mitpress.mit.edu/books/csound-book
>>>>>>> > Author & Editor: The Audio Programming Book -
>>>>>>> > http://mitpress.mit.edu/books/audio-programming-book
>>>>>>> > ______________________________________________
>>>>>>> > about: http://www.boulangerlabs.com/about/richardboulanger/
>>>>>>> > about:
>>>>>>> > http://www.csounds.com/community/developers/dr-richard-boulanger/
>>>>>>> > ______________________________________________
>>>>>>> > music:
>>>>>>> >
>>>>>>> > http://www.csounds.com/community/developers/dr-richard-boulanger/dr-richard-boulanger-music/
>>>>>>> >
>>>>>>> > On Tue, Mar 1, 2016 at 3:15 PM, Peter Burgess
>>>>>>> >  wrote:
>>>>>>> >>
>>>>>>> >> Hi there, I'm planning to knock up a multi-band compressor, and I was
>>>>>>> >> wondering if csound has any filters built for this job? My first
>>>>>>> >> thought was to use butterworths, but I've just read that the combined
>>>>>>> >> signal has a slight peak at the crossover point. Apparently
>>>>>>> >> Linkwitz-Riley filters are better for the job. Do we have anything
>>>>>>> >> like that?
>>>>>>> >>
>>>>>>> >> Pete
>>>>>>> >>
>>>>>>> >> --
>>>>>>> >> http://algorythmradio.com
>>>>>>> >> https://soundcloud.com/algorythmradio
>>>>>>> >>
>>>>>>> >> 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
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> --
>>>>>>> http://algorythmradio.com
>>>>>>> https://soundcloud.com/algorythmradio
>>>>>>>
>>>>>>> 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
>>>
>>>
>>>
>>> --
>>> http://algorythmradio.com
>>> https://soundcloud.com/algorythmradio
>>
>>
>>
>> --
>> http://algorythmradio.com
>> https://soundcloud.com/algorythmradio
>
>
>
> --
> http://algorythmradio.com
> https://soundcloud.com/algorythmradio



-- 
http://algorythmradio.com
https://soundcloud.com/algorythmradio

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

Hi Pete,

I've been interested in this thread and did some preliminary research
but haven't had any time to investigate further.  Out of curiosity,
have you tried measuring the bandpass signal returned from the
statevar filter?

steven

On Fri, Mar 4, 2016 at 9:30 AM, Peter Burgess
 wrote:
> Anyone have any ideas on this?
>
> Basically I just need to know, is it generally acceptable to just live
> with a bit of a peak in the crossover points, or should I aim for a
> near flat response all round?
>
> Are there any better filters for the job, or any tricks (like simply
> using a limiter) that will help me overcome?
>
> Pete
>
> On Wed, Mar 2, 2016 at 4:04 PM, Peter Burgess
>  wrote:
>> I've done some 2 band crossover tests using these two butterworth
>> filter configurations:
>>
>> icrossfreq = 60
>>
>> ; Crossover Test 1 (butxp opcodes)
>> alowsig1 butlp gacompsig, icrossfreq
>> alowsig2 butlp alowsig1, icrossfreq
>> ahisig1 buthp gacompsig, icrossfreq
>> ahisig2 buthp ahisig1, icrossfreq
>> out alowsig2 + ahisig2
>>
>> ; Crossover Test 2 (clfilt opcode)
>> inpol init 4
>> alowsig clfilt gacompsig, icrossfreq, 0, inpol
>> ahisig clfilt gacompsig, icrossfreq, 1, inpol
>> out alowsig + ahisig
>>
>> I have tested these two crossovers with a rich kick sound, and with
>> pink noise, using various different crossover frequencies. When
>> watching the amplitudes in the terminal, it would appear that both
>> crossover setups produce an output with a higher peak somewhere,
>> compared to the unaffected sound.
>>
>> For the kick, the unaltered sound has an amplitude of 0.49851.
>> With a crossover frequency of 60, Test 1 produces amplitudes of
>> between 0.72 and 0.86, while Test 2 produces amplitudes of between
>> 0.75 and 0.94.
>>
>> I couldn't hear a great deal of audible difference between the three,
>> so I spectrum analysed some samples. I have attached these spectrums,
>> as well as the audio samples.
>>
>> The spectrum analysis shows a slight difference near the crossover
>> point for Test 1, but a much more noticeable difference around the
>> crossover for Test 2.
>>
>> I am drawing the conclusion that Test 1's crossover is better, but are
>> there filters that will do even better still? Is it possible or easy
>> to get a totally level crossover? Or could I just limit the output of
>> the crossovers to the original signals level?
>>
>> On Wed, Mar 2, 2016 at 12:58 AM, Peter Burgess
>>  wrote:
>>>  Ah hang on, a further read about crossovers says that the most
>>> commonly used filters are 4th order Linkwitz–Riley filters, which
>>> apparently just use two cascaded 2nd order butterworth filter.... I'm
>>> no DSP expert, but doesn't that just make it a 4th order butterworth?
>>>
>>> On Tue, Mar 1, 2016 at 11:36 PM, Peter Burgess
>>>  wrote:
>>>> That's fine, and is what I was intending, but does Csound have a
>>>> suitable filter for a crossover? As I have read and now also tested,
>>>> the butterworth isn't suitable
>>>>
>>>> On Tue, Mar 1, 2016 at 11:30 PM, Rory Walsh  wrote:
>>>>> A suite of mastering UDOs would be a good start. Gleb proposed such a thing
>>>>> some time back, but I don't think much came of it.
>>>>>
>>>>> On 1 March 2016 at 23:28, Dr. Richard Boulanger 
>>>>> wrote:
>>>>>>
>>>>>> For sure Rory,
>>>>>>
>>>>>> That is what we do; but Csound could really use some "mastering" opcodes
>>>>>> not just "mastering instruments"  Dr.B.
>>>>>> _____________________________________________
>>>>>> On Tue, Mar 1, 2016 at 6:14 PM, Rory Walsh  wrote:
>>>>>>>
>>>>>>> Why not just use the opcodes we have to create one?
>>>>>>>
>>>>>>> On 1 Mar 2016 21:35, "Peter Burgess" 
>>>>>>> wrote:
>>>>>>>>
>>>>>>>> I was thinking that too, and a big part of me wants to try and write
>>>>>>>> one, but I've already started 3 opcodes and never found the time to
>>>>>>>> finish any of them, so I should probably stop kidding myself that I'm
>>>>>>>> gonna get round to something like that in the near future, haha! :D
>>>>>>>>
>>>>>>>> On Tue, Mar 1, 2016 at 8:50 PM, Dr. Richard Boulanger
>>>>>>>>  wrote:
>>>>>>>> > We could really use a multi-band compressor opcode!
>>>>>>>> >
>>>>>>>> > -dB
>>>>>>>> >
>>>>>>>> > _____________________________________________
>>>>>>>> > Dr. Richard Boulanger
>>>>>>>> > Professor of Electronic Production and Design
>>>>>>>> > Professional Writing and Music Technology Division
>>>>>>>> > Berklee College of Music
>>>>>>>> > 1140 Boylston Street, FB-75
>>>>>>>> > Boston, MA 02215-3693
>>>>>>>> > Office @ 161 Mass Ave - 4th Floor
>>>>>>>> > 617-747-2384 (office)  774-488-9166 (cell)
>>>>>>>> > ______________________________________________
>>>>>>>> > rboulanger@berklee.edu (email)
>>>>>>>> > https://www.facebook.com/richard.boulanger.58 (facebook)
>>>>>>>> > ______________________________________________
>>>>>>>> > President of Boulanger Labs - http://boulangerlabs.com
>>>>>>>> > Director of Csounds.com - http://csounds.com
>>>>>>>> > Author & Editor: The Csound Book -
>>>>>>>> > http://mitpress.mit.edu/books/csound-book
>>>>>>>> > Author & Editor: The Audio Programming Book -
>>>>>>>> > http://mitpress.mit.edu/books/audio-programming-book
>>>>>>>> > ______________________________________________
>>>>>>>> > about: http://www.boulangerlabs.com/about/richardboulanger/
>>>>>>>> > about:
>>>>>>>> > http://www.csounds.com/community/developers/dr-richard-boulanger/
>>>>>>>> > ______________________________________________
>>>>>>>> > music:
>>>>>>>> >
>>>>>>>> > http://www.csounds.com/community/developers/dr-richard-boulanger/dr-richard-boulanger-music/
>>>>>>>> >
>>>>>>>> > On Tue, Mar 1, 2016 at 3:15 PM, Peter Burgess
>>>>>>>> >  wrote:
>>>>>>>> >>
>>>>>>>> >> Hi there, I'm planning to knock up a multi-band compressor, and I was
>>>>>>>> >> wondering if csound has any filters built for this job? My first
>>>>>>>> >> thought was to use butterworths, but I've just read that the combined
>>>>>>>> >> signal has a slight peak at the crossover point. Apparently
>>>>>>>> >> Linkwitz-Riley filters are better for the job. Do we have anything
>>>>>>>> >> like that?
>>>>>>>> >>
>>>>>>>> >> Pete
>>>>>>>> >>
>>>>>>>> >> --
>>>>>>>> >> http://algorythmradio.com
>>>>>>>> >> https://soundcloud.com/algorythmradio
>>>>>>>> >>
>>>>>>>> >> 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
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>> --
>>>>>>>> http://algorythmradio.com
>>>>>>>> https://soundcloud.com/algorythmradio
>>>>>>>>
>>>>>>>> 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
>>>>
>>>>
>>>>
>>>> --
>>>> http://algorythmradio.com
>>>> https://soundcloud.com/algorythmradio
>>>
>>>
>>>
>>> --
>>> http://algorythmradio.com
>>> https://soundcloud.com/algorythmradio
>>
>>
>>
>> --
>> http://algorythmradio.com
>> https://soundcloud.com/algorythmradio
>
>
>
> --
> http://algorythmradio.com
> https://soundcloud.com/algorythmradio
>
> 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

No actually, I've not gone beyond the butterworths yet. I will
certainly give that a try!

I'm currently working on an opcode that relates to the crossover
experiments, so I will get back onto the filters again tomorrow I
think, or as soon as I have my opcode fully operational.

Let me know if you find anything out :D

On Fri, Mar 4, 2016 at 3:30 PM, Steven Yi  wrote:
> Hi Pete,
>
> I've been interested in this thread and did some preliminary research
> but haven't had any time to investigate further.  Out of curiosity,
> have you tried measuring the bandpass signal returned from the
> statevar filter?
>
> steven
>
> On Fri, Mar 4, 2016 at 9:30 AM, Peter Burgess
>  wrote:
>> Anyone have any ideas on this?
>>
>> Basically I just need to know, is it generally acceptable to just live
>> with a bit of a peak in the crossover points, or should I aim for a
>> near flat response all round?
>>
>> Are there any better filters for the job, or any tricks (like simply
>> using a limiter) that will help me overcome?
>>
>> Pete
>>
>> On Wed, Mar 2, 2016 at 4:04 PM, Peter Burgess
>>  wrote:
>>> I've done some 2 band crossover tests using these two butterworth
>>> filter configurations:
>>>
>>> icrossfreq = 60
>>>
>>> ; Crossover Test 1 (butxp opcodes)
>>> alowsig1 butlp gacompsig, icrossfreq
>>> alowsig2 butlp alowsig1, icrossfreq
>>> ahisig1 buthp gacompsig, icrossfreq
>>> ahisig2 buthp ahisig1, icrossfreq
>>> out alowsig2 + ahisig2
>>>
>>> ; Crossover Test 2 (clfilt opcode)
>>> inpol init 4
>>> alowsig clfilt gacompsig, icrossfreq, 0, inpol
>>> ahisig clfilt gacompsig, icrossfreq, 1, inpol
>>> out alowsig + ahisig
>>>
>>> I have tested these two crossovers with a rich kick sound, and with
>>> pink noise, using various different crossover frequencies. When
>>> watching the amplitudes in the terminal, it would appear that both
>>> crossover setups produce an output with a higher peak somewhere,
>>> compared to the unaffected sound.
>>>
>>> For the kick, the unaltered sound has an amplitude of 0.49851.
>>> With a crossover frequency of 60, Test 1 produces amplitudes of
>>> between 0.72 and 0.86, while Test 2 produces amplitudes of between
>>> 0.75 and 0.94.
>>>
>>> I couldn't hear a great deal of audible difference between the three,
>>> so I spectrum analysed some samples. I have attached these spectrums,
>>> as well as the audio samples.
>>>
>>> The spectrum analysis shows a slight difference near the crossover
>>> point for Test 1, but a much more noticeable difference around the
>>> crossover for Test 2.
>>>
>>> I am drawing the conclusion that Test 1's crossover is better, but are
>>> there filters that will do even better still? Is it possible or easy
>>> to get a totally level crossover? Or could I just limit the output of
>>> the crossovers to the original signals level?
>>>
>>> On Wed, Mar 2, 2016 at 12:58 AM, Peter Burgess
>>>  wrote:
>>>>  Ah hang on, a further read about crossovers says that the most
>>>> commonly used filters are 4th order Linkwitz–Riley filters, which
>>>> apparently just use two cascaded 2nd order butterworth filter.... I'm
>>>> no DSP expert, but doesn't that just make it a 4th order butterworth?
>>>>
>>>> On Tue, Mar 1, 2016 at 11:36 PM, Peter Burgess
>>>>  wrote:
>>>>> That's fine, and is what I was intending, but does Csound have a
>>>>> suitable filter for a crossover? As I have read and now also tested,
>>>>> the butterworth isn't suitable
>>>>>
>>>>> On Tue, Mar 1, 2016 at 11:30 PM, Rory Walsh  wrote:
>>>>>> A suite of mastering UDOs would be a good start. Gleb proposed such a thing
>>>>>> some time back, but I don't think much came of it.
>>>>>>
>>>>>> On 1 March 2016 at 23:28, Dr. Richard Boulanger 
>>>>>> wrote:
>>>>>>>
>>>>>>> For sure Rory,
>>>>>>>
>>>>>>> That is what we do; but Csound could really use some "mastering" opcodes
>>>>>>> not just "mastering instruments"  Dr.B.
>>>>>>> _____________________________________________
>>>>>>> On Tue, Mar 1, 2016 at 6:14 PM, Rory Walsh  wrote:
>>>>>>>>
>>>>>>>> Why not just use the opcodes we have to create one?
>>>>>>>>
>>>>>>>> On 1 Mar 2016 21:35, "Peter Burgess" 
>>>>>>>> wrote:
>>>>>>>>>
>>>>>>>>> I was thinking that too, and a big part of me wants to try and write
>>>>>>>>> one, but I've already started 3 opcodes and never found the time to
>>>>>>>>> finish any of them, so I should probably stop kidding myself that I'm
>>>>>>>>> gonna get round to something like that in the near future, haha! :D
>>>>>>>>>
>>>>>>>>> On Tue, Mar 1, 2016 at 8:50 PM, Dr. Richard Boulanger
>>>>>>>>>  wrote:
>>>>>>>>> > We could really use a multi-band compressor opcode!
>>>>>>>>> >
>>>>>>>>> > -dB
>>>>>>>>> >
>>>>>>>>> > _____________________________________________
>>>>>>>>> > Dr. Richard Boulanger
>>>>>>>>> > Professor of Electronic Production and Design
>>>>>>>>> > Professional Writing and Music Technology Division
>>>>>>>>> > Berklee College of Music
>>>>>>>>> > 1140 Boylston Street, FB-75
>>>>>>>>> > Boston, MA 02215-3693
>>>>>>>>> > Office @ 161 Mass Ave - 4th Floor
>>>>>>>>> > 617-747-2384 (office)  774-488-9166 (cell)
>>>>>>>>> > ______________________________________________
>>>>>>>>> > rboulanger@berklee.edu (email)
>>>>>>>>> > https://www.facebook.com/richard.boulanger.58 (facebook)
>>>>>>>>> > ______________________________________________
>>>>>>>>> > President of Boulanger Labs - http://boulangerlabs.com
>>>>>>>>> > Director of Csounds.com - http://csounds.com
>>>>>>>>> > Author & Editor: The Csound Book -
>>>>>>>>> > http://mitpress.mit.edu/books/csound-book
>>>>>>>>> > Author & Editor: The Audio Programming Book -
>>>>>>>>> > http://mitpress.mit.edu/books/audio-programming-book
>>>>>>>>> > ______________________________________________
>>>>>>>>> > about: http://www.boulangerlabs.com/about/richardboulanger/
>>>>>>>>> > about:
>>>>>>>>> > http://www.csounds.com/community/developers/dr-richard-boulanger/
>>>>>>>>> > ______________________________________________
>>>>>>>>> > music:
>>>>>>>>> >
>>>>>>>>> > http://www.csounds.com/community/developers/dr-richard-boulanger/dr-richard-boulanger-music/
>>>>>>>>> >
>>>>>>>>> > On Tue, Mar 1, 2016 at 3:15 PM, Peter Burgess
>>>>>>>>> >  wrote:
>>>>>>>>> >>
>>>>>>>>> >> Hi there, I'm planning to knock up a multi-band compressor, and I was
>>>>>>>>> >> wondering if csound has any filters built for this job? My first
>>>>>>>>> >> thought was to use butterworths, but I've just read that the combined
>>>>>>>>> >> signal has a slight peak at the crossover point. Apparently
>>>>>>>>> >> Linkwitz-Riley filters are better for the job. Do we have anything
>>>>>>>>> >> like that?
>>>>>>>>> >>
>>>>>>>>> >> Pete
>>>>>>>>> >>
>>>>>>>>> >> --
>>>>>>>>> >> http://algorythmradio.com
>>>>>>>>> >> https://soundcloud.com/algorythmradio
>>>>>>>>> >>
>>>>>>>>> >> 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
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> --
>>>>>>>>> http://algorythmradio.com
>>>>>>>>> https://soundcloud.com/algorythmradio
>>>>>>>>>
>>>>>>>>> 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
>>>>>
>>>>>
>>>>>
>>>>> --
>>>>> http://algorythmradio.com
>>>>> https://soundcloud.com/algorythmradio
>>>>
>>>>
>>>>
>>>> --
>>>> http://algorythmradio.com
>>>> https://soundcloud.com/algorythmradio
>>>
>>>
>>>
>>> --
>>> http://algorythmradio.com
>>> https://soundcloud.com/algorythmradio
>>
>>
>>
>> --
>> http://algorythmradio.com
>> https://soundcloud.com/algorythmradio
>>
>> 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



-- 
http://algorythmradio.com
https://soundcloud.com/algorythmradio

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've had a go with svfilter, and I've tried both one instance, and 2
instances in series (actually 3, the second for lowpass, and the third
for highpass).

I've had a go with statevar aswell as you suggested, but only with a
single instance so far. I'm not sure if I'm using it in the best way
for the task though. I am using default oversampling (3 times), and a
Q of 1. Will that give the best roll-off for a crossover? I'm not
entirely sure what figure to use for Q for no resonance with statevar.

It just occured to me though, that I can simply use 2x2 pole LP
butterworths for the low band, and simply take that away from the
original signal for the high band, thus achieving the exact oposite of
the lowband curve. Am I right in remembering that that's how you make
a highpass filter anyway?

This might not give a symetrical curve, but it shoudn't be too far
off, and it's probably more important to have a flat frequency
response than symetrical curves. What dyou think?

On Fri, Mar 4, 2016 at 3:58 PM, Peter Burgess
 wrote:
> No actually, I've not gone beyond the butterworths yet. I will
> certainly give that a try!
>
> I'm currently working on an opcode that relates to the crossover
> experiments, so I will get back onto the filters again tomorrow I
> think, or as soon as I have my opcode fully operational.
>
> Let me know if you find anything out :D
>
> On Fri, Mar 4, 2016 at 3:30 PM, Steven Yi  wrote:
>> Hi Pete,
>>
>> I've been interested in this thread and did some preliminary research
>> but haven't had any time to investigate further.  Out of curiosity,
>> have you tried measuring the bandpass signal returned from the
>> statevar filter?
>>
>> steven
>>
>> On Fri, Mar 4, 2016 at 9:30 AM, Peter Burgess
>>  wrote:
>>> Anyone have any ideas on this?
>>>
>>> Basically I just need to know, is it generally acceptable to just live
>>> with a bit of a peak in the crossover points, or should I aim for a
>>> near flat response all round?
>>>
>>> Are there any better filters for the job, or any tricks (like simply
>>> using a limiter) that will help me overcome?
>>>
>>> Pete
>>>
>>> On Wed, Mar 2, 2016 at 4:04 PM, Peter Burgess
>>>  wrote:
>>>> I've done some 2 band crossover tests using these two butterworth
>>>> filter configurations:
>>>>
>>>> icrossfreq = 60
>>>>
>>>> ; Crossover Test 1 (butxp opcodes)
>>>> alowsig1 butlp gacompsig, icrossfreq
>>>> alowsig2 butlp alowsig1, icrossfreq
>>>> ahisig1 buthp gacompsig, icrossfreq
>>>> ahisig2 buthp ahisig1, icrossfreq
>>>> out alowsig2 + ahisig2
>>>>
>>>> ; Crossover Test 2 (clfilt opcode)
>>>> inpol init 4
>>>> alowsig clfilt gacompsig, icrossfreq, 0, inpol
>>>> ahisig clfilt gacompsig, icrossfreq, 1, inpol
>>>> out alowsig + ahisig
>>>>
>>>> I have tested these two crossovers with a rich kick sound, and with
>>>> pink noise, using various different crossover frequencies. When
>>>> watching the amplitudes in the terminal, it would appear that both
>>>> crossover setups produce an output with a higher peak somewhere,
>>>> compared to the unaffected sound.
>>>>
>>>> For the kick, the unaltered sound has an amplitude of 0.49851.
>>>> With a crossover frequency of 60, Test 1 produces amplitudes of
>>>> between 0.72 and 0.86, while Test 2 produces amplitudes of between
>>>> 0.75 and 0.94.
>>>>
>>>> I couldn't hear a great deal of audible difference between the three,
>>>> so I spectrum analysed some samples. I have attached these spectrums,
>>>> as well as the audio samples.
>>>>
>>>> The spectrum analysis shows a slight difference near the crossover
>>>> point for Test 1, but a much more noticeable difference around the
>>>> crossover for Test 2.
>>>>
>>>> I am drawing the conclusion that Test 1's crossover is better, but are
>>>> there filters that will do even better still? Is it possible or easy
>>>> to get a totally level crossover? Or could I just limit the output of
>>>> the crossovers to the original signals level?
>>>>
>>>> On Wed, Mar 2, 2016 at 12:58 AM, Peter Burgess
>>>>  wrote:
>>>>>  Ah hang on, a further read about crossovers says that the most
>>>>> commonly used filters are 4th order Linkwitz–Riley filters, which
>>>>> apparently just use two cascaded 2nd order butterworth filter.... I'm
>>>>> no DSP expert, but doesn't that just make it a 4th order butterworth?
>>>>>
>>>>> On Tue, Mar 1, 2016 at 11:36 PM, Peter Burgess
>>>>>  wrote:
>>>>>> That's fine, and is what I was intending, but does Csound have a
>>>>>> suitable filter for a crossover? As I have read and now also tested,
>>>>>> the butterworth isn't suitable
>>>>>>
>>>>>> On Tue, Mar 1, 2016 at 11:30 PM, Rory Walsh  wrote:
>>>>>>> A suite of mastering UDOs would be a good start. Gleb proposed such a thing
>>>>>>> some time back, but I don't think much came of it.
>>>>>>>
>>>>>>> On 1 March 2016 at 23:28, Dr. Richard Boulanger 
>>>>>>> wrote:
>>>>>>>>
>>>>>>>> For sure Rory,
>>>>>>>>
>>>>>>>> That is what we do; but Csound could really use some "mastering" opcodes
>>>>>>>> not just "mastering instruments"  Dr.B.
>>>>>>>> _____________________________________________
>>>>>>>> On Tue, Mar 1, 2016 at 6:14 PM, Rory Walsh  wrote:
>>>>>>>>>
>>>>>>>>> Why not just use the opcodes we have to create one?
>>>>>>>>>
>>>>>>>>> On 1 Mar 2016 21:35, "Peter Burgess" 
>>>>>>>>> wrote:
>>>>>>>>>>
>>>>>>>>>> I was thinking that too, and a big part of me wants to try and write
>>>>>>>>>> one, but I've already started 3 opcodes and never found the time to
>>>>>>>>>> finish any of them, so I should probably stop kidding myself that I'm
>>>>>>>>>> gonna get round to something like that in the near future, haha! :D
>>>>>>>>>>
>>>>>>>>>> On Tue, Mar 1, 2016 at 8:50 PM, Dr. Richard Boulanger
>>>>>>>>>>  wrote:
>>>>>>>>>> > We could really use a multi-band compressor opcode!
>>>>>>>>>> >
>>>>>>>>>> > -dB
>>>>>>>>>> >
>>>>>>>>>> > _____________________________________________
>>>>>>>>>> > Dr. Richard Boulanger
>>>>>>>>>> > Professor of Electronic Production and Design
>>>>>>>>>> > Professional Writing and Music Technology Division
>>>>>>>>>> > Berklee College of Music
>>>>>>>>>> > 1140 Boylston Street, FB-75
>>>>>>>>>> > Boston, MA 02215-3693
>>>>>>>>>> > Office @ 161 Mass Ave - 4th Floor
>>>>>>>>>> > 617-747-2384 (office)  774-488-9166 (cell)
>>>>>>>>>> > ______________________________________________
>>>>>>>>>> > rboulanger@berklee.edu (email)
>>>>>>>>>> > https://www.facebook.com/richard.boulanger.58 (facebook)
>>>>>>>>>> > ______________________________________________
>>>>>>>>>> > President of Boulanger Labs - http://boulangerlabs.com
>>>>>>>>>> > Director of Csounds.com - http://csounds.com
>>>>>>>>>> > Author & Editor: The Csound Book -
>>>>>>>>>> > http://mitpress.mit.edu/books/csound-book
>>>>>>>>>> > Author & Editor: The Audio Programming Book -
>>>>>>>>>> > http://mitpress.mit.edu/books/audio-programming-book
>>>>>>>>>> > ______________________________________________
>>>>>>>>>> > about: http://www.boulangerlabs.com/about/richardboulanger/
>>>>>>>>>> > about:
>>>>>>>>>> > http://www.csounds.com/community/developers/dr-richard-boulanger/
>>>>>>>>>> > ______________________________________________
>>>>>>>>>> > music:
>>>>>>>>>> >
>>>>>>>>>> > http://www.csounds.com/community/developers/dr-richard-boulanger/dr-richard-boulanger-music/
>>>>>>>>>> >
>>>>>>>>>> > On Tue, Mar 1, 2016 at 3:15 PM, Peter Burgess
>>>>>>>>>> >  wrote:
>>>>>>>>>> >>
>>>>>>>>>> >> Hi there, I'm planning to knock up a multi-band compressor, and I was
>>>>>>>>>> >> wondering if csound has any filters built for this job? My first
>>>>>>>>>> >> thought was to use butterworths, but I've just read that the combined
>>>>>>>>>> >> signal has a slight peak at the crossover point. Apparently
>>>>>>>>>> >> Linkwitz-Riley filters are better for the job. Do we have anything
>>>>>>>>>> >> like that?
>>>>>>>>>> >>
>>>>>>>>>> >> Pete
>>>>>>>>>> >>
>>>>>>>>>> >> --
>>>>>>>>>> >> http://algorythmradio.com
>>>>>>>>>> >> https://soundcloud.com/algorythmradio
>>>>>>>>>> >>
>>>>>>>>>> >> 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
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> --
>>>>>>>>>> http://algorythmradio.com
>>>>>>>>>> https://soundcloud.com/algorythmradio
>>>>>>>>>>
>>>>>>>>>> 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
>>>>>>
>>>>>>
>>>>>>
>>>>>> --
>>>>>> http://algorythmradio.com
>>>>>> https://soundcloud.com/algorythmradio
>>>>>
>>>>>
>>>>>
>>>>> --
>>>>> http://algorythmradio.com
>>>>> https://soundcloud.com/algorythmradio
>>>>
>>>>
>>>>
>>>> --
>>>> http://algorythmradio.com
>>>> https://soundcloud.com/algorythmradio
>>>
>>>
>>>
>>> --
>>> http://algorythmradio.com
>>> https://soundcloud.com/algorythmradio
>>>
>>> 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
>
>
>
> --
> http://algorythmradio.com
> https://soundcloud.com/algorythmradio



-- 
http://algorythmradio.com
https://soundcloud.com/algorythmradio

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