Well, I’ve used eqfil and it does what it says on the manual page, but it’s not the same
to say that it would emulate an arbitrary equaliser. 

========================
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 9 May 2016, at 20:54, Aaron Krister Johnson  wrote:
> 
> Hi, 
> 
> I want to emulate a graphical eq application, and apply cutting and boosting to 10 octave bands
> (basically, a 10-band graphical eq with 31, 63, 125, 250, 500, 1K, 2K, 4K, 8K, and 16K center frequencies)
> 
> I have been using the plugin for EQ in Linux's audacious, and I have the DB boost and cut parameters in decibels, but when I try and emulate what audacious is doing (and which I like very much for my mix); I'm not able to get the same sound. 
> 
> I suspect it's b/c the gain parameter isn't scaling correctly, even though I'm using ampdb to do a conversion for it. And, perhaps, the bandwidth isn't working as it is in the plugin, either.
> 
> Any tips? I'm thinking Victor Lazzarini would know, being the author of 'eqfil'.
> 
> Thanks,
> 
> 
> Aaron Krister Johnson
> http://www.untwelve.org
> 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'd suggest that you also try a serial bank, as phase cancelling issues might arise with parallel bank IIRs.

If the code is available, we could have a go at implementing the filter (a rainy Saturday job…) as an opcode. 
========================
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 10 May 2016, at 14:45, Aaron Krister Johnson  wrote:
> 
> Thanks Paul!
> 
> Does anyone who knows filter design recognize what the closest (or matching) filter type is coded in the Audacious code?
> 
> I'm thinking I can just do a parallel design with the right (closest) opcode and get the right color....
> 
> There's also this article: 
> 
> http://mymbs.mbs.net/~pfisher/FOV2-0010016C/FOV2-0010016E/FOV2-001001A3/chapters/28webman/index.html
> 
> I'll try the design there later. It's based on butterbp, which may be closer to what I need.
> 
> -AKJ
> 
> Aaron Krister Johnson
> http://www.untwelve.org
> 
> On Mon, May 9, 2016 at 11:32 PM, Paul Batchelor  wrote:
> There are countless ways to build something that fits the definition of an "equalizer", so it would make sense that the behavior won't be identical.
> 
> For anyone curious, here is the code for the audacious equalizer inner loop:
> 
> https://github.com/audacious-media-player/audacious/blob/master/src/libaudcore/equalizer.cc
> 
> And here is the code for eqfil:
> 
> https://github.com/csound/csound/blob/develop/Opcodes/eqfil.c
> 
> If the filter design didn't already exist in Csound as another opcode (it probably does), in theory it shouldn't be terribly difficult to implement the filter used in audacious as a UDO. 
> 
> -P
> 
> On Mon, May 9, 2016 at 1:32 PM, Victor Lazzarini  wrote:
> Well, I’ve used eqfil and it does what it says on the manual page, but it’s not the same
> to say that it would emulate an arbitrary equaliser.
> 
> ========================
> 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 9 May 2016, at 20:54, Aaron Krister Johnson  wrote:
> >
> > Hi,
> >
> > I want to emulate a graphical eq application, and apply cutting and boosting to 10 octave bands
> > (basically, a 10-band graphical eq with 31, 63, 125, 250, 500, 1K, 2K, 4K, 8K, and 16K center frequencies)
> >
> > I have been using the plugin for EQ in Linux's audacious, and I have the DB boost and cut parameters in decibels, but when I try and emulate what audacious is doing (and which I like very much for my mix); I'm not able to get the same sound.
> >
> > I suspect it's b/c the gain parameter isn't scaling correctly, even though I'm using ampdb to do a conversion for it. And, perhaps, the bandwidth isn't working as it is in the plugin, either.
> >
> > Any tips? I'm thinking Victor Lazzarini would know, being the author of 'eqfil'.
> >
> > Thanks,
> >
> >
> > Aaron Krister Johnson
> > http://www.untwelve.org
> > 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

I looked at the code for Audacious' EQ and it looks to me that the
signal is processed in series:

https://github.com/audacious-media-player/audacious/blob/master/src/libaudcore/equalizer.cc#L129-L148

as the yt information is first read in, reused through the loop for
each filter in the bank, then written to the output. (Please correct
me if I'm wrong though!)

On Tue, May 10, 2016 at 1:39 PM, Aaron Krister Johnson
 wrote:
> Excellent!
>
> The article I posted mentioned that graphic equalizers tend to be parallel
> banks....and that this was the way to go. Thoughts?
>
> Aaron Krister Johnson
> http://www.untwelve.org
>
> On Tue, May 10, 2016 at 8:52 AM, Victor Lazzarini 
> wrote:
>>
>> I'd suggest that you also try a serial bank, as phase cancelling issues
>> might arise with parallel bank IIRs.
>>
>> If the code is available, we could have a go at implementing the filter (a
>> rainy Saturday job…) as an opcode.
>> ========================
>> 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 10 May 2016, at 14:45, Aaron Krister Johnson 
>> > wrote:
>> >
>> > Thanks Paul!
>> >
>> > Does anyone who knows filter design recognize what the closest (or
>> > matching) filter type is coded in the Audacious code?
>> >
>> > I'm thinking I can just do a parallel design with the right (closest)
>> > opcode and get the right color....
>> >
>> > There's also this article:
>> >
>> >
>> > http://mymbs.mbs.net/~pfisher/FOV2-0010016C/FOV2-0010016E/FOV2-001001A3/chapters/28webman/index.html
>> >
>> > I'll try the design there later. It's based on butterbp, which may be
>> > closer to what I need.
>> >
>> > -AKJ
>> >
>> > Aaron Krister Johnson
>> > http://www.untwelve.org
>> >
>> > On Mon, May 9, 2016 at 11:32 PM, Paul Batchelor
>> >  wrote:
>> > There are countless ways to build something that fits the definition of
>> > an "equalizer", so it would make sense that the behavior won't be identical.
>> >
>> > For anyone curious, here is the code for the audacious equalizer inner
>> > loop:
>> >
>> >
>> > https://github.com/audacious-media-player/audacious/blob/master/src/libaudcore/equalizer.cc
>> >
>> > And here is the code for eqfil:
>> >
>> > https://github.com/csound/csound/blob/develop/Opcodes/eqfil.c
>> >
>> > If the filter design didn't already exist in Csound as another opcode
>> > (it probably does), in theory it shouldn't be terribly difficult to
>> > implement the filter used in audacious as a UDO.
>> >
>> > -P
>> >
>> > On Mon, May 9, 2016 at 1:32 PM, Victor Lazzarini
>> >  wrote:
>> > Well, I’ve used eqfil and it does what it says on the manual page, but
>> > it’s not the same
>> > to say that it would emulate an arbitrary equaliser.
>> >
>> > ========================
>> > 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 9 May 2016, at 20:54, Aaron Krister Johnson 
>> > > wrote:
>> > >
>> > > Hi,
>> > >
>> > > I want to emulate a graphical eq application, and apply cutting and
>> > > boosting to 10 octave bands
>> > > (basically, a 10-band graphical eq with 31, 63, 125, 250, 500, 1K, 2K,
>> > > 4K, 8K, and 16K center frequencies)
>> > >
>> > > I have been using the plugin for EQ in Linux's audacious, and I have
>> > > the DB boost and cut parameters in decibels, but when I try and emulate what
>> > > audacious is doing (and which I like very much for my mix); I'm not able to
>> > > get the same sound.
>> > >
>> > > I suspect it's b/c the gain parameter isn't scaling correctly, even
>> > > though I'm using ampdb to do a conversion for it. And, perhaps, the
>> > > bandwidth isn't working as it is in the plugin, either.
>> > >
>> > > Any tips? I'm thinking Victor Lazzarini would know, being the author
>> > > of 'eqfil'.
>> > >
>> > > Thanks,
>> > >
>> > >
>> > > Aaron Krister Johnson
>> > > http://www.untwelve.org
>> > > 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
>
>
> 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 think the use of a bp filter to create an EQ is not appropriate for
the serial configuration.
If you think of each component in an EQ as merely boosting or
attenuating some part of the spectrum, the serial configuration will
be more intuitive.
Say, to create a boost at 1kHz, you would not use a bp at 1kHz, but
rather use a filter that boosts frequencies around 1kHz and ideally
leaving the rest of the spectrum untouched.
Then you could run this in serial with any other filter that boosts or
attenuates a specific part of the spectrum.

2016-05-15 22:45 GMT+02:00 Aaron Krister Johnson :
> The more I think about this, the more it seems wrong that it could be
> filters in serial...if you do it for each band, you get energy cuts
> increasingly, such that by the time you get do your final spcectral band,
> there's nothing left to filter!
>
> Indeed, an empirical proof is that running my source material in serial
> through a bunch of butterbp filters at one octave spacings from 31hz to 16K
> hz gives an empty wave file...
>
> Am I misunderstanding what you guys mean by a serial setup, then?
>
>
> Aaron Krister Johnson
> http://www.untwelve.org
>
> On Wed, May 11, 2016 at 5:35 PM, Steven Yi  wrote:
>>
>> I looked at the code for Audacious' EQ and it looks to me that the
>> signal is processed in series:
>>
>>
>> https://github.com/audacious-media-player/audacious/blob/master/src/libaudcore/equalizer.cc#L129-L148
>>
>> as the yt information is first read in, reused through the loop for
>> each filter in the bank, then written to the output. (Please correct
>> me if I'm wrong though!)
>>
>> On Tue, May 10, 2016 at 1:39 PM, Aaron Krister Johnson
>>  wrote:
>> > Excellent!
>> >
>> > The article I posted mentioned that graphic equalizers tend to be
>> > parallel
>> > banks....and that this was the way to go. Thoughts?
>> >
>> > Aaron Krister Johnson
>> > http://www.untwelve.org
>> >
>> > On Tue, May 10, 2016 at 8:52 AM, Victor Lazzarini
>> > 
>> > wrote:
>> >>
>> >> I'd suggest that you also try a serial bank, as phase cancelling issues
>> >> might arise with parallel bank IIRs.
>> >>
>> >> If the code is available, we could have a go at implementing the filter
>> >> (a
>> >> rainy Saturday job…) as an opcode.
>> >> ========================
>> >> 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 10 May 2016, at 14:45, Aaron Krister Johnson 
>> >> > wrote:
>> >> >
>> >> > Thanks Paul!
>> >> >
>> >> > Does anyone who knows filter design recognize what the closest (or
>> >> > matching) filter type is coded in the Audacious code?
>> >> >
>> >> > I'm thinking I can just do a parallel design with the right (closest)
>> >> > opcode and get the right color....
>> >> >
>> >> > There's also this article:
>> >> >
>> >> >
>> >> >
>> >> > http://mymbs.mbs.net/~pfisher/FOV2-0010016C/FOV2-0010016E/FOV2-001001A3/chapters/28webman/index.html
>> >> >
>> >> > I'll try the design there later. It's based on butterbp, which may be
>> >> > closer to what I need.
>> >> >
>> >> > -AKJ
>> >> >
>> >> > Aaron Krister Johnson
>> >> > http://www.untwelve.org
>> >> >
>> >> > On Mon, May 9, 2016 at 11:32 PM, Paul Batchelor
>> >> >  wrote:
>> >> > There are countless ways to build something that fits the definition
>> >> > of
>> >> > an "equalizer", so it would make sense that the behavior won't be
>> >> > identical.
>> >> >
>> >> > For anyone curious, here is the code for the audacious equalizer
>> >> > inner
>> >> > loop:
>> >> >
>> >> >
>> >> >
>> >> > https://github.com/audacious-media-player/audacious/blob/master/src/libaudcore/equalizer.cc
>> >> >
>> >> > And here is the code for eqfil:
>> >> >
>> >> > https://github.com/csound/csound/blob/develop/Opcodes/eqfil.c
>> >> >
>> >> > If the filter design didn't already exist in Csound as another opcode
>> >> > (it probably does), in theory it shouldn't be terribly difficult to
>> >> > implement the filter used in audacious as a UDO.
>> >> >
>> >> > -P
>> >> >
>> >> > On Mon, May 9, 2016 at 1:32 PM, Victor Lazzarini
>> >> >  wrote:
>> >> > Well, I’ve used eqfil and it does what it says on the manual page,
>> >> > but
>> >> > it’s not the same
>> >> > to say that it would emulate an arbitrary equaliser.
>> >> >
>> >> > ========================
>> >> > 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 9 May 2016, at 20:54, Aaron Krister Johnson 
>> >> > > wrote:
>> >> > >
>> >> > > Hi,
>> >> > >
>> >> > > I want to emulate a graphical eq application, and apply cutting and
>> >> > > boosting to 10 octave bands
>> >> > > (basically, a 10-band graphical eq with 31, 63, 125, 250, 500, 1K,
>> >> > > 2K,
>> >> > > 4K, 8K, and 16K center frequencies)
>> >> > >
>> >> > > I have been using the plugin for EQ in Linux's audacious, and I
>> >> > > have
>> >> > > the DB boost and cut parameters in decibels, but when I try and
>> >> > > emulate what
>> >> > > audacious is doing (and which I like very much for my mix); I'm not
>> >> > > able to
>> >> > > get the same sound.
>> >> > >
>> >> > > I suspect it's b/c the gain parameter isn't scaling correctly, even
>> >> > > though I'm using ampdb to do a conversion for it. And, perhaps, the
>> >> > > bandwidth isn't working as it is in the plugin, either.
>> >> > >
>> >> > > Any tips? I'm thinking Victor Lazzarini would know, being the
>> >> > > author
>> >> > > of 'eqfil'.
>> >> > >
>> >> > > Thanks,
>> >> > >
>> >> > >
>> >> > > Aaron Krister Johnson
>> >> > > http://www.untwelve.org
>> >> > > 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
>> >
>> >
>> > 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



-- 

Oeyvind Brandtsegg
Professor of Music Technology
NTNU
7491 Trondheim
Norway
Cell: +47 92 203 205

http://www.partikkelaudio.com/
http://soundcloud.com/brandtsegg
http://flyndresang.no/
http://soundcloud.com/t-emp

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

Indeed,
you might think of a filter that boosts a specific part of the
spectrum as a bp in parallel with the unprocessed signal, accounting
for phase issues relating to combining those two signals. And
simliarly for a cut.
Having this, and referring to it as one single component in the
processing chain, you would combine several of them in series to build
the EQ.

2016-05-15 23:56 GMT+02:00 Aaron Krister Johnson :
> Yeah, I see that...but also, BP filters in parallel seem to work fine (but
> obviously, not the same animal).
>
> I would use 'eqfil', but I'm preferring to use an intuitive bandwidth as
> outlined in the paper I linked earlier in this thread, where known eq types
> (1/3 oct, 1 oct, etc.) have Q values that you can use to figure you
> bandwidth for each EQ band by a simple division.
>
> Since the paper suggested parallel butterbp filters, and set out a known
> solution with code examples, I based my code on that, and it seems to
> reasonably emulate the mix I was getting through the Audacious equalizer.
>
> I'm open to exploring other solutions in the future, though :)
>
> -AKJ
>
>
> Aaron Krister Johnson
> http://www.untwelve.org
>
> On Sun, May 15, 2016 at 4:35 PM, Oeyvind Brandtsegg
>  wrote:
>>
>> I think the use of a bp filter to create an EQ is not appropriate for
>> the serial configuration.
>> If you think of each component in an EQ as merely boosting or
>> attenuating some part of the spectrum, the serial configuration will
>> be more intuitive.
>> Say, to create a boost at 1kHz, you would not use a bp at 1kHz, but
>> rather use a filter that boosts frequencies around 1kHz and ideally
>> leaving the rest of the spectrum untouched.
>> Then you could run this in serial with any other filter that boosts or
>> attenuates a specific part of the spectrum.
>>
>> 2016-05-15 22:45 GMT+02:00 Aaron Krister Johnson :
>> > The more I think about this, the more it seems wrong that it could be
>> > filters in serial...if you do it for each band, you get energy cuts
>> > increasingly, such that by the time you get do your final spcectral
>> > band,
>> > there's nothing left to filter!
>> >
>> > Indeed, an empirical proof is that running my source material in serial
>> > through a bunch of butterbp filters at one octave spacings from 31hz to
>> > 16K
>> > hz gives an empty wave file...
>> >
>> > Am I misunderstanding what you guys mean by a serial setup, then?
>> >
>> >
>> > Aaron Krister Johnson
>> > http://www.untwelve.org
>> >
>> > On Wed, May 11, 2016 at 5:35 PM, Steven Yi  wrote:
>> >>
>> >> I looked at the code for Audacious' EQ and it looks to me that the
>> >> signal is processed in series:
>> >>
>> >>
>> >>
>> >> https://github.com/audacious-media-player/audacious/blob/master/src/libaudcore/equalizer.cc#L129-L148
>> >>
>> >> as the yt information is first read in, reused through the loop for
>> >> each filter in the bank, then written to the output. (Please correct
>> >> me if I'm wrong though!)
>> >>
>> >> On Tue, May 10, 2016 at 1:39 PM, Aaron Krister Johnson
>> >>  wrote:
>> >> > Excellent!
>> >> >
>> >> > The article I posted mentioned that graphic equalizers tend to be
>> >> > parallel
>> >> > banks....and that this was the way to go. Thoughts?
>> >> >
>> >> > Aaron Krister Johnson
>> >> > http://www.untwelve.org
>> >> >
>> >> > On Tue, May 10, 2016 at 8:52 AM, Victor Lazzarini
>> >> > 
>> >> > wrote:
>> >> >>
>> >> >> I'd suggest that you also try a serial bank, as phase cancelling
>> >> >> issues
>> >> >> might arise with parallel bank IIRs.
>> >> >>
>> >> >> If the code is available, we could have a go at implementing the
>> >> >> filter
>> >> >> (a
>> >> >> rainy Saturday job…) as an opcode.
>> >> >> ========================
>> >> >> 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 10 May 2016, at 14:45, Aaron Krister Johnson
>> >> >> > 
>> >> >> > wrote:
>> >> >> >
>> >> >> > Thanks Paul!
>> >> >> >
>> >> >> > Does anyone who knows filter design recognize what the closest (or
>> >> >> > matching) filter type is coded in the Audacious code?
>> >> >> >
>> >> >> > I'm thinking I can just do a parallel design with the right
>> >> >> > (closest)
>> >> >> > opcode and get the right color....
>> >> >> >
>> >> >> > There's also this article:
>> >> >> >
>> >> >> >
>> >> >> >
>> >> >> >
>> >> >> > http://mymbs.mbs.net/~pfisher/FOV2-0010016C/FOV2-0010016E/FOV2-001001A3/chapters/28webman/index.html
>> >> >> >
>> >> >> > I'll try the design there later. It's based on butterbp, which may
>> >> >> > be
>> >> >> > closer to what I need.
>> >> >> >
>> >> >> > -AKJ
>> >> >> >
>> >> >> > Aaron Krister Johnson
>> >> >> > http://www.untwelve.org
>> >> >> >
>> >> >> > On Mon, May 9, 2016 at 11:32 PM, Paul Batchelor
>> >> >> >  wrote:
>> >> >> > There are countless ways to build something that fits the
>> >> >> > definition
>> >> >> > of
>> >> >> > an "equalizer", so it would make sense that the behavior won't be
>> >> >> > identical.
>> >> >> >
>> >> >> > For anyone curious, here is the code for the audacious equalizer
>> >> >> > inner
>> >> >> > loop:
>> >> >> >
>> >> >> >
>> >> >> >
>> >> >> >
>> >> >> > https://github.com/audacious-media-player/audacious/blob/master/src/libaudcore/equalizer.cc
>> >> >> >
>> >> >> > And here is the code for eqfil:
>> >> >> >
>> >> >> > https://github.com/csound/csound/blob/develop/Opcodes/eqfil.c
>> >> >> >
>> >> >> > If the filter design didn't already exist in Csound as another
>> >> >> > opcode
>> >> >> > (it probably does), in theory it shouldn't be terribly difficult
>> >> >> > to
>> >> >> > implement the filter used in audacious as a UDO.
>> >> >> >
>> >> >> > -P
>> >> >> >
>> >> >> > On Mon, May 9, 2016 at 1:32 PM, Victor Lazzarini
>> >> >> >  wrote:
>> >> >> > Well, I’ve used eqfil and it does what it says on the manual page,
>> >> >> > but
>> >> >> > it’s not the same
>> >> >> > to say that it would emulate an arbitrary equaliser.
>> >> >> >
>> >> >> > ========================
>> >> >> > 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 9 May 2016, at 20:54, Aaron Krister Johnson
>> >> >> > > 
>> >> >> > > wrote:
>> >> >> > >
>> >> >> > > Hi,
>> >> >> > >
>> >> >> > > I want to emulate a graphical eq application, and apply cutting
>> >> >> > > and
>> >> >> > > boosting to 10 octave bands
>> >> >> > > (basically, a 10-band graphical eq with 31, 63, 125, 250, 500,
>> >> >> > > 1K,
>> >> >> > > 2K,
>> >> >> > > 4K, 8K, and 16K center frequencies)
>> >> >> > >
>> >> >> > > I have been using the plugin for EQ in Linux's audacious, and I
>> >> >> > > have
>> >> >> > > the DB boost and cut parameters in decibels, but when I try and
>> >> >> > > emulate what
>> >> >> > > audacious is doing (and which I like very much for my mix); I'm
>> >> >> > > not
>> >> >> > > able to
>> >> >> > > get the same sound.
>> >> >> > >
>> >> >> > > I suspect it's b/c the gain parameter isn't scaling correctly,
>> >> >> > > even
>> >> >> > > though I'm using ampdb to do a conversion for it. And, perhaps,
>> >> >> > > the
>> >> >> > > bandwidth isn't working as it is in the plugin, either.
>> >> >> > >
>> >> >> > > Any tips? I'm thinking Victor Lazzarini would know, being the
>> >> >> > > author
>> >> >> > > of 'eqfil'.
>> >> >> > >
>> >> >> > > Thanks,
>> >> >> > >
>> >> >> > >
>> >> >> > > Aaron Krister Johnson
>> >> >> > > http://www.untwelve.org
>> >> >> > > 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
>> >> >
>> >> >
>> >> > 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
>>
>>
>>
>> --
>>
>> Oeyvind Brandtsegg
>> Professor of Music Technology
>> NTNU
>> 7491 Trondheim
>> Norway
>> Cell: +47 92 203 205
>>
>> http://www.partikkelaudio.com/
>> http://soundcloud.com/brandtsegg
>> http://flyndresang.no/
>> http://soundcloud.com/t-emp
>>
>> 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



-- 

Oeyvind Brandtsegg
Professor of Music Technology
NTNU
7491 Trondheim
Norway
Cell: +47 92 203 205

http://www.partikkelaudio.com/
http://soundcloud.com/brandtsegg
http://flyndresang.no/
http://soundcloud.com/t-emp

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 Aaron,

I got around to implementing Audacious' EQ using Csound UDO code.  I
put the audaciouseq UDO in my libsyi project here:

https://github.com/kunstmusik/libsyi/blob/master/audaciouseq.udo

A test file is available here:

https://github.com/kunstmusik/libsyi/blob/master/tests/audaciouseq.csd

I've only lightly tested and it seems to work.  The implementation
uses a single-sample audacious_bp2 UDO that does the filter for each
band.  The routing of the filters within the audaciouseq UDO shows the
serial routing.

Could you give it a try and let me know if it has the expected results
when compared to Audacious?

Thanks!
steven


On Sun, May 15, 2016 at 5:56 PM, Aaron Krister Johnson
 wrote:
> Yeah, I see that...but also, BP filters in parallel seem to work fine (but
> obviously, not the same animal).
>
> I would use 'eqfil', but I'm preferring to use an intuitive bandwidth as
> outlined in the paper I linked earlier in this thread, where known eq types
> (1/3 oct, 1 oct, etc.) have Q values that you can use to figure you
> bandwidth for each EQ band by a simple division.
>
> Since the paper suggested parallel butterbp filters, and set out a known
> solution with code examples, I based my code on that, and it seems to
> reasonably emulate the mix I was getting through the Audacious equalizer.
>
> I'm open to exploring other solutions in the future, though :)
>
> -AKJ
>
>
> Aaron Krister Johnson
> http://www.untwelve.org
>
> On Sun, May 15, 2016 at 4:35 PM, Oeyvind Brandtsegg
>  wrote:
>>
>> I think the use of a bp filter to create an EQ is not appropriate for
>> the serial configuration.
>> If you think of each component in an EQ as merely boosting or
>> attenuating some part of the spectrum, the serial configuration will
>> be more intuitive.
>> Say, to create a boost at 1kHz, you would not use a bp at 1kHz, but
>> rather use a filter that boosts frequencies around 1kHz and ideally
>> leaving the rest of the spectrum untouched.
>> Then you could run this in serial with any other filter that boosts or
>> attenuates a specific part of the spectrum.
>>
>> 2016-05-15 22:45 GMT+02:00 Aaron Krister Johnson :
>> > The more I think about this, the more it seems wrong that it could be
>> > filters in serial...if you do it for each band, you get energy cuts
>> > increasingly, such that by the time you get do your final spcectral
>> > band,
>> > there's nothing left to filter!
>> >
>> > Indeed, an empirical proof is that running my source material in serial
>> > through a bunch of butterbp filters at one octave spacings from 31hz to
>> > 16K
>> > hz gives an empty wave file...
>> >
>> > Am I misunderstanding what you guys mean by a serial setup, then?
>> >
>> >
>> > Aaron Krister Johnson
>> > http://www.untwelve.org
>> >
>> > On Wed, May 11, 2016 at 5:35 PM, Steven Yi  wrote:
>> >>
>> >> I looked at the code for Audacious' EQ and it looks to me that the
>> >> signal is processed in series:
>> >>
>> >>
>> >>
>> >> https://github.com/audacious-media-player/audacious/blob/master/src/libaudcore/equalizer.cc#L129-L148
>> >>
>> >> as the yt information is first read in, reused through the loop for
>> >> each filter in the bank, then written to the output. (Please correct
>> >> me if I'm wrong though!)
>> >>
>> >> On Tue, May 10, 2016 at 1:39 PM, Aaron Krister Johnson
>> >>  wrote:
>> >> > Excellent!
>> >> >
>> >> > The article I posted mentioned that graphic equalizers tend to be
>> >> > parallel
>> >> > banks....and that this was the way to go. Thoughts?
>> >> >
>> >> > Aaron Krister Johnson
>> >> > http://www.untwelve.org
>> >> >
>> >> > On Tue, May 10, 2016 at 8:52 AM, Victor Lazzarini
>> >> > 
>> >> > wrote:
>> >> >>
>> >> >> I'd suggest that you also try a serial bank, as phase cancelling
>> >> >> issues
>> >> >> might arise with parallel bank IIRs.
>> >> >>
>> >> >> If the code is available, we could have a go at implementing the
>> >> >> filter
>> >> >> (a
>> >> >> rainy Saturday job…) as an opcode.
>> >> >> ========================
>> >> >> 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 10 May 2016, at 14:45, Aaron Krister Johnson
>> >> >> > 
>> >> >> > wrote:
>> >> >> >
>> >> >> > Thanks Paul!
>> >> >> >
>> >> >> > Does anyone who knows filter design recognize what the closest (or
>> >> >> > matching) filter type is coded in the Audacious code?
>> >> >> >
>> >> >> > I'm thinking I can just do a parallel design with the right
>> >> >> > (closest)
>> >> >> > opcode and get the right color....
>> >> >> >
>> >> >> > There's also this article:
>> >> >> >
>> >> >> >
>> >> >> >
>> >> >> >
>> >> >> > http://mymbs.mbs.net/~pfisher/FOV2-0010016C/FOV2-0010016E/FOV2-001001A3/chapters/28webman/index.html
>> >> >> >
>> >> >> > I'll try the design there later. It's based on butterbp, which may
>> >> >> > be
>> >> >> > closer to what I need.
>> >> >> >
>> >> >> > -AKJ
>> >> >> >
>> >> >> > Aaron Krister Johnson
>> >> >> > http://www.untwelve.org
>> >> >> >
>> >> >> > On Mon, May 9, 2016 at 11:32 PM, Paul Batchelor
>> >> >> >  wrote:
>> >> >> > There are countless ways to build something that fits the
>> >> >> > definition
>> >> >> > of
>> >> >> > an "equalizer", so it would make sense that the behavior won't be
>> >> >> > identical.
>> >> >> >
>> >> >> > For anyone curious, here is the code for the audacious equalizer
>> >> >> > inner
>> >> >> > loop:
>> >> >> >
>> >> >> >
>> >> >> >
>> >> >> >
>> >> >> > https://github.com/audacious-media-player/audacious/blob/master/src/libaudcore/equalizer.cc
>> >> >> >
>> >> >> > And here is the code for eqfil:
>> >> >> >
>> >> >> > https://github.com/csound/csound/blob/develop/Opcodes/eqfil.c
>> >> >> >
>> >> >> > If the filter design didn't already exist in Csound as another
>> >> >> > opcode
>> >> >> > (it probably does), in theory it shouldn't be terribly difficult
>> >> >> > to
>> >> >> > implement the filter used in audacious as a UDO.
>> >> >> >
>> >> >> > -P
>> >> >> >
>> >> >> > On Mon, May 9, 2016 at 1:32 PM, Victor Lazzarini
>> >> >> >  wrote:
>> >> >> > Well, I’ve used eqfil and it does what it says on the manual page,
>> >> >> > but
>> >> >> > it’s not the same
>> >> >> > to say that it would emulate an arbitrary equaliser.
>> >> >> >
>> >> >> > ========================
>> >> >> > 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 9 May 2016, at 20:54, Aaron Krister Johnson
>> >> >> > > 
>> >> >> > > wrote:
>> >> >> > >
>> >> >> > > Hi,
>> >> >> > >
>> >> >> > > I want to emulate a graphical eq application, and apply cutting
>> >> >> > > and
>> >> >> > > boosting to 10 octave bands
>> >> >> > > (basically, a 10-band graphical eq with 31, 63, 125, 250, 500,
>> >> >> > > 1K,
>> >> >> > > 2K,
>> >> >> > > 4K, 8K, and 16K center frequencies)
>> >> >> > >
>> >> >> > > I have been using the plugin for EQ in Linux's audacious, and I
>> >> >> > > have
>> >> >> > > the DB boost and cut parameters in decibels, but when I try and
>> >> >> > > emulate what
>> >> >> > > audacious is doing (and which I like very much for my mix); I'm
>> >> >> > > not
>> >> >> > > able to
>> >> >> > > get the same sound.
>> >> >> > >
>> >> >> > > I suspect it's b/c the gain parameter isn't scaling correctly,
>> >> >> > > even
>> >> >> > > though I'm using ampdb to do a conversion for it. And, perhaps,
>> >> >> > > the
>> >> >> > > bandwidth isn't working as it is in the plugin, either.
>> >> >> > >
>> >> >> > > Any tips? I'm thinking Victor Lazzarini would know, being the
>> >> >> > > author
>> >> >> > > of 'eqfil'.
>> >> >> > >
>> >> >> > > Thanks,
>> >> >> > >
>> >> >> > >
>> >> >> > > Aaron Krister Johnson
>> >> >> > > http://www.untwelve.org
>> >> >> > > 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
>> >> >
>> >> >
>> >> > 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
>>
>>
>>
>> --
>>
>> Oeyvind Brandtsegg
>> Professor of Music Technology
>> NTNU
>> 7491 Trondheim
>> Norway
>> Cell: +47 92 203 205
>>
>> http://www.partikkelaudio.com/
>> http://soundcloud.com/brandtsegg
>> http://flyndresang.no/
>> http://soundcloud.com/t-emp
>>
>> 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