I am seeking advice.

I sometimes write 2-D surround sound pieces, using ambisonics or HRTF,
but there is a use for a stereo down-mix, for example on the net,
soundcloud etc.  My question is how to do this.  I have produced the
ambisonic stereo but that is for speakers left and right rather than
the usual stereo setup (at least that was my interpretation)

I had thought of projecting each sound at position x degrees to the
front quadrant (not written the degree in 0-360 to ditto in -45 to +45
yet and that is a problem too).  Is there a sensible way to do this or
am I seeking a chimera?
==John ffitch

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

Did I misunderstand the cound decode to stereo from B format?  Looking at 
the code it seems the same as UHJ. 
==John ffitch

On Fri, 19 Apr 2019, Oscar Pablo Di Liscia wrote:

> I'm not sure of understanding your question.I suppose you already tried the
> UHJ stereo transcoding. To my knowledge, UHJ is very good both for speakers
> and headphones and it is full compatable with stereo systems. 
> In my opinion is the best option to make stereo versions of surround
> multichannel pieces. Even if the pieces are not originally Ambisonics, these
> may be transcoded to first order BF  Ambisonics (considering each loudspeaker
> as a virtual source) and further transcoding to stereo UHJ.
> Hope this helps.
> 
> El viernes, 19 de abril de 2019, John  escribió:
>       I am seeking advice.
>
>       I sometimes write 2-D surround sound pieces, using ambisonics or
>       HRTF,
>       but there is a use for a stereo down-mix, for example on the net,
>       soundcloud etc.  My question is how to do this.  I have produced
>       the
>       ambisonic stereo but that is for speakers left and right rather
>       than
>       the usual stereo setup (at least that was my interpretation)
>
>       I had thought of projecting each sound at position x degrees to
>       the
>       front quadrant (not written the degree in 0-360 to ditto in -45 to
>       +45
>       yet and that is a problem too).  Is there a sensible way to do
>       this or
>       am I seeking a chimera?
>       ==John ffitch
>

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 am using the abcode1 b format.  Never looked at spat3d.  
My orch code generates either the ten nonzero 2d ambisonic values or hrtf.  I generate 4 or 8 channels as a different pass and I am worrying about the 2 channel mapping.

As you might guess I am largely ignorant of the underlying mathematics.

⁣Sent from TypeApp ​

On Apr 20, 2019, 01:42, at 01:42, Oscar Pablo Di Liscia  wrote:
>Which is the opcode you were looking at?
>Conventional stereo is pretty much like transcoding MS to LR:
>Having the four B-Format Order one signals: W X Y Z, conventional
>stereo
>would just be:
>L= W+Y
>R= W-Y
>But UHJ involves weighted sums of the W, X and Y signals with a 90
>degrees
>phase shift. In the "spat3d" Opcode documentation you can see the
>equation
>and the coefficients. The 90 degrees shift is achieved using the
>"hilbert"
>opcode.
>Now, there is a third format called "Super Stereo" that is pretty
>similar
>to UHJ but not the same and often confused with it. As a matter of fact
>UHJ
>is often called "Super stereo". But I don't know exactly its equations.
>In the Ambisonic Toolkit documentation you may see that the three
>formats
>are differenced, and you may take a look at the code to see the
>differences
>(it is written in Java).
>http://www.ambisonictoolkit.net/documentation/reaper/
>
>Hope this helps
>Pablo
>
>
>El viernes, 19 de abril de 2019, john  escribió:
>
>> Did I misunderstand the cound decode to stereo from B format? 
>Looking at
>> the code it seems the same as UHJ. ==John ffitch
>>
>> On Fri, 19 Apr 2019, Oscar Pablo Di Liscia wrote:
>>
>> I'm not sure of understanding your question.I suppose you already
>tried the
>>> UHJ stereo transcoding. To my knowledge, UHJ is very good both for
>>> speakers
>>> and headphones and it is full compatable with stereo systems.
>>> In my opinion is the best option to make stereo versions of surround
>>> multichannel pieces. Even if the pieces are not originally
>Ambisonics,
>>> these
>>> may be transcoded to first order BF  Ambisonics (considering each
>>> loudspeaker
>>> as a virtual source) and further transcoding to stereo UHJ.
>>> Hope this helps.
>>>
>>> El viernes, 19 de abril de 2019, John 
>escribió:
>>>       I am seeking advice.
>>>
>>>       I sometimes write 2-D surround sound pieces, using ambisonics
>or
>>>       HRTF,
>>>       but there is a use for a stereo down-mix, for example on the
>net,
>>>       soundcloud etc.  My question is how to do this.  I have
>produced
>>>       the
>>>       ambisonic stereo but that is for speakers left and right
>rather
>>>       than
>>>       the usual stereo setup (at least that was my interpretation)
>>>
>>>       I had thought of projecting each sound at position x degrees
>to
>>>       the
>>>       front quadrant (not written the degree in 0-360 to ditto in
>-45 to
>>>       +45
>>>       yet and that is a problem too).  Is there a sensible way to do
>>>       this or
>>>       am I seeking a chimera?
>>>       ==John ffitch
>>>
>>>
>> 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
>
>
>
>-- 
>Dr. Oscar Pablo Di Liscia
>Profesor Titular
>Director Programa de Investigación "Sistemas Temporales y Síntesis
>Espacial
>en el Arte Sonoro"
>http://stseas.web.unq.edu.ar/
>Director Colección Editorial "Música y Ciencia"
>Escuela Universitaria de Artes
>Universidad Nacional de Quilmes
>Argentina
>
>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 have to admit that the underliying math is too much for me also.
:)
So, you are generating a 2D B-Format 3rd order Ambisonics set of
signals and you want to see
which could be the best way to "translate" this to stereo?
The UHJ stereo encoding use just the W, X and Y signals. Higher order
components are lost in this
kind of encoding, because was initially developed for 2D B-Format 1st order.
I don´t know if at present a similar (stereo) option for higher orders
exists, but it seems very unlikely.
To my knowledge, the best option to monitor in stereo a High Order
Ambisonic stream is to use a Binaural Decoder.
What I do in my work is to generate the 3rd order 16 signals with
csound and, if I don´t have the needed array
of loudspeakers to monitor the mix, I use Reaper together with Ambix
Plugin Suite for High Order
Ambisonics to listen the binaural stereo withe headphones. This suite
has (among many other useful things) a
Binaural decoder for High Order Ambisonics, that is excellent, and it
works really well. See here.
http://www.matthiaskronlachner.com/?p=2015
So, if you pick a set of HRIRs that works well for you, you may have a
decent binaural listening of
your mix without having the needed array of loudspeakers. Of course,
the drawback of this strategy
is that an HRIR that works well for you may not work for other people.
And it´s exclusively to listen under
headphones.
BTW: we could gladly use a Binaural Decoder in Csound, if somebody is
reading this thread and is willing
to work on this. I understand that Luis Jure and his team are working
on High Order Encoders and Decoders.

Hope this helps.
Pablo

El sáb., 20 abr. 2019 a las 16:44, John ff () escribió:
>
> I am using the abcode1 b format.  Never looked at spat3d.
> My orch code generates either the ten nonzero 2d ambisonic values or hrtf.  I generate 4 or 8 channels as a different pass and I am worrying about the 2 channel mapping.
>
> As you might guess I am largely ignorant of the underlying mathematics.
>
> ⁣Sent from TypeApp
>
> On Apr 20, 2019, 01:42, at 01:42, Oscar Pablo Di Liscia  wrote:
> >Which is the opcode you were looking at?
> >Conventional stereo is pretty much like transcoding MS to LR:
> >Having the four B-Format Order one signals: W X Y Z, conventional
> >stereo
> >would just be:
> >L= W+Y
> >R= W-Y
> >But UHJ involves weighted sums of the W, X and Y signals with a 90
> >degrees
> >phase shift. In the "spat3d" Opcode documentation you can see the
> >equation
> >and the coefficients. The 90 degrees shift is achieved using the
> >"hilbert"
> >opcode.
> >Now, there is a third format called "Super Stereo" that is pretty
> >similar
> >to UHJ but not the same and often confused with it. As a matter of fact
> >UHJ
> >is often called "Super stereo". But I don't know exactly its equations.
> >In the Ambisonic Toolkit documentation you may see that the three
> >formats
> >are differenced, and you may take a look at the code to see the
> >differences
> >(it is written in Java).
> >http://www.ambisonictoolkit.net/documentation/reaper/
> >
> >Hope this helps
> >Pablo
> >
> >
> >El viernes, 19 de abril de 2019, john  escribió:
> >
> >> Did I misunderstand the cound decode to stereo from B format?
> >Looking at
> >> the code it seems the same as UHJ. ==John ffitch
> >>
> >> On Fri, 19 Apr 2019, Oscar Pablo Di Liscia wrote:
> >>
> >> I'm not sure of understanding your question.I suppose you already
> >tried the
> >>> UHJ stereo transcoding. To my knowledge, UHJ is very good both for
> >>> speakers
> >>> and headphones and it is full compatable with stereo systems.
> >>> In my opinion is the best option to make stereo versions of surround
> >>> multichannel pieces. Even if the pieces are not originally
> >Ambisonics,
> >>> these
> >>> may be transcoded to first order BF  Ambisonics (considering each
> >>> loudspeaker
> >>> as a virtual source) and further transcoding to stereo UHJ.
> >>> Hope this helps.
> >>>
> >>> El viernes, 19 de abril de 2019, John 
> >escribió:
> >>>       I am seeking advice.
> >>>
> >>>       I sometimes write 2-D surround sound pieces, using ambisonics
> >or
> >>>       HRTF,
> >>>       but there is a use for a stereo down-mix, for example on the
> >net,
> >>>       soundcloud etc.  My question is how to do this.  I have
> >produced
> >>>       the
> >>>       ambisonic stereo but that is for speakers left and right
> >rather
> >>>       than
> >>>       the usual stereo setup (at least that was my interpretation)
> >>>
> >>>       I had thought of projecting each sound at position x degrees
> >to
> >>>       the
> >>>       front quadrant (not written the degree in 0-360 to ditto in
> >-45 to
> >>>       +45
> >>>       yet and that is a problem too).  Is there a sensible way to do
> >>>       this or
> >>>       am I seeking a chimera?
> >>>       ==John ffitch
> >>>
> >>>
> >> 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
> >
> >
> >
> >--
> >Dr. Oscar Pablo Di Liscia
> >Profesor Titular
> >Director Programa de Investigación "Sistemas Temporales y Síntesis
> >Espacial
> >en el Arte Sonoro"
> >http://stseas.web.unq.edu.ar/
> >Director Colección Editorial "Música y Ciencia"
> >Escuela Universitaria de Artes
> >Universidad Nacional de Quilmes
> >Argentina
> >
> >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



-- 
Dr. Oscar Pablo Di Liscia
Profesor Titular
Director Programa de Investigación "Sistemas Temporales y Síntesis
Espacial en el Arte Sonoro"
http://stseas.web.unq.edu.ar/
Director Colección Editorial "Música y Ciencia"
Escuela Universitaria de Artes
Universidad Nacional de Quilmes
Argentina

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

For ambisonics to binaural there is the omnitone project by google that
could be ported to Csound.

But I also wonder if there is an update on this?

https://www.youtube.com/watch?v=UXi6CYpOZDo


On 21/4/19 17:07, Oscar Pablo Di Liscia wrote:
> I have to admit that the underliying math is too much for me also.
> :)
> So, you are generating a 2D B-Format 3rd order Ambisonics set of
> signals and you want to see
> which could be the best way to "translate" this to stereo?
> The UHJ stereo encoding use just the W, X and Y signals. Higher order
> components are lost in this
> kind of encoding, because was initially developed for 2D B-Format 1st order.
> I don´t know if at present a similar (stereo) option for higher orders
> exists, but it seems very unlikely.
> To my knowledge, the best option to monitor in stereo a High Order
> Ambisonic stream is to use a Binaural Decoder.
> What I do in my work is to generate the 3rd order 16 signals with
> csound and, if I don´t have the needed array
> of loudspeakers to monitor the mix, I use Reaper together with Ambix
> Plugin Suite for High Order
> Ambisonics to listen the binaural stereo withe headphones. This suite
> has (among many other useful things) a
> Binaural decoder for High Order Ambisonics, that is excellent, and it
> works really well. See here.
> http://www.matthiaskronlachner.com/?p=2015
> So, if you pick a set of HRIRs that works well for you, you may have a
> decent binaural listening of
> your mix without having the needed array of loudspeakers. Of course,
> the drawback of this strategy
> is that an HRIR that works well for you may not work for other people.
> And it´s exclusively to listen under
> headphones.
> BTW: we could gladly use a Binaural Decoder in Csound, if somebody is
> reading this thread and is willing
> to work on this. I understand that Luis Jure and his team are working
> on High Order Encoders and Decoders.
>
> Hope this helps.
> Pablo
>
> El sáb., 20 abr. 2019 a las 16:44, John ff () escribió:
>> I am using the abcode1 b format.  Never looked at spat3d.
>> My orch code generates either the ten nonzero 2d ambisonic values or hrtf.  I generate 4 or 8 channels as a different pass and I am worrying about the 2 channel mapping.
>>
>> As you might guess I am largely ignorant of the underlying mathematics.
>>
>> ⁣Sent from TypeApp
>>
>> On Apr 20, 2019, 01:42, at 01:42, Oscar Pablo Di Liscia  wrote:
>>> Which is the opcode you were looking at?
>>> Conventional stereo is pretty much like transcoding MS to LR:
>>> Having the four B-Format Order one signals: W X Y Z, conventional
>>> stereo
>>> would just be:
>>> L= W+Y
>>> R= W-Y
>>> But UHJ involves weighted sums of the W, X and Y signals with a 90
>>> degrees
>>> phase shift. In the "spat3d" Opcode documentation you can see the
>>> equation
>>> and the coefficients. The 90 degrees shift is achieved using the
>>> "hilbert"
>>> opcode.
>>> Now, there is a third format called "Super Stereo" that is pretty
>>> similar
>>> to UHJ but not the same and often confused with it. As a matter of fact
>>> UHJ
>>> is often called "Super stereo". But I don't know exactly its equations.
>>> In the Ambisonic Toolkit documentation you may see that the three
>>> formats
>>> are differenced, and you may take a look at the code to see the
>>> differences
>>> (it is written in Java).
>>> http://www.ambisonictoolkit.net/documentation/reaper/
>>>
>>> Hope this helps
>>> Pablo
>>>
>>>
>>> El viernes, 19 de abril de 2019, john  escribió:
>>>
>>>> Did I misunderstand the cound decode to stereo from B format?
>>> Looking at
>>>> the code it seems the same as UHJ. ==John ffitch
>>>>
>>>> On Fri, 19 Apr 2019, Oscar Pablo Di Liscia wrote:
>>>>
>>>> I'm not sure of understanding your question.I suppose you already
>>> tried the
>>>>> UHJ stereo transcoding. To my knowledge, UHJ is very good both for
>>>>> speakers
>>>>> and headphones and it is full compatable with stereo systems.
>>>>> In my opinion is the best option to make stereo versions of surround
>>>>> multichannel pieces. Even if the pieces are not originally
>>> Ambisonics,
>>>>> these
>>>>> may be transcoded to first order BF  Ambisonics (considering each
>>>>> loudspeaker
>>>>> as a virtual source) and further transcoding to stereo UHJ.
>>>>> Hope this helps.
>>>>>
>>>>> El viernes, 19 de abril de 2019, John 
>>> escribió:
>>>>>       I am seeking advice.
>>>>>
>>>>>       I sometimes write 2-D surround sound pieces, using ambisonics
>>> or
>>>>>       HRTF,
>>>>>       but there is a use for a stereo down-mix, for example on the
>>> net,
>>>>>       soundcloud etc.  My question is how to do this.  I have
>>> produced
>>>>>       the
>>>>>       ambisonic stereo but that is for speakers left and right
>>> rather
>>>>>       than
>>>>>       the usual stereo setup (at least that was my interpretation)
>>>>>
>>>>>       I had thought of projecting each sound at position x degrees
>>> to
>>>>>       the
>>>>>       front quadrant (not written the degree in 0-360 to ditto in
>>> -45 to
>>>>>       +45
>>>>>       yet and that is a problem too).  Is there a sensible way to do
>>>>>       this or
>>>>>       am I seeking a chimera?
>>>>>       ==John ffitch
>>>>>
>>>>>
>>>> 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
>>>
>>>
>>> --
>>> Dr. Oscar Pablo Di Liscia
>>> Profesor Titular
>>> Director Programa de Investigación "Sistemas Temporales y Síntesis
>>> Espacial
>>> en el Arte Sonoro"
>>> http://stseas.web.unq.edu.ar/
>>> Director Colección Editorial "Música y Ciencia"
>>> Escuela Universitaria de Artes
>>> Universidad Nacional de Quilmes
>>> Argentina
>>>
>>> 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