Have you tried the outh opcode? http://www.csounds.com/manual/html/outh.html greetings, menno --- On Sat, 11/13/10, peiman khosravi <peimankhosravi@gmail.com> wrote: From: peiman khosravi <peimankhosravi@gmail.com> Subject: [Csnd] 'spatialization' opcodes To: csound@lists.bath.ac.uk Date: Saturday, November 13, 2010, 11:35 AM Hello, I've been looking at the opcodes available for distributing the signal among 6 speakers. The only one seems to be vbap8 (by just ignoring two outputs). But the problem is that this opcodes does not seem to allow arbitrary panning, meaning that for instance in order to move the source from the front left speaker to the rear  speaker one needs to pass through all the speakers in between. Is there an opcode that accepts x/y coordinates for 6-8 channel distribution? Thanks in advance Peiman

space


--
View this message in context: http://csound.1045644.n5.nabble.com/spatialization-opcodes-tp3263433p3263473.html
Sent from the Csound - General mailing list archive at Nabble.com.


Send bugs reports to the Sourceforge bug tracker
           https://sourceforge.net/tracker/?group_id=81968&atid=564599
Discussions of bugs and features can be posted here
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

It's actually a very difficult thing to do - placing sounds inside the speaker array (which implies a soundfield in effect radiating out ~towards~ the speakers in some directions) and having them pass (in effect) through the listener's head. Indeed, I am not sure to what extent it has yet been done in a technically rigorous way. You can do it "rhetorically" of course, by intensity panning from speaker to speaker across the circle, and if it is not too slow you may get an almost convincing result. What you hear when sitting other than in the centre is a very open question. One hears reports of technologies that create the sense of some phantom voice whispering in your ear, but ~if~ it works, I suspect the head has to be clamped immoveably in the sweet spot.

Richard Dobson

On 13/11/2010 11:46, peiman khosravi wrote:

Hi Menno,

Yes I am using the outh opcode but that doesn't offer any panning. I am
using vbap8 for panning but unfortunately vbap8 does not take x/y
parameters for positioning the source.

Best,

Peiman

Send bugs reports to the Sourceforge bug tracker
          https://sourceforge.net/tracker/?group_id=81968&atid=564599
Discussions of bugs and features can be posted here
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

 

On 13 November 2010 12:00, Richard Dobson <richarddobson@blueyonder.co.uk> wrote:

It's actually a very difficult thing to do - placing sounds inside the speaker array (which implies a soundfield in effect radiating out ~towards~ the speakers in some directions) and having them pass (in effect) through the listener's head. Indeed, I am not sure to what extent it has yet been done in a technically rigorous way. You can do it "rhetorically" of course, by intensity panning from speaker to speaker across the circle, and if it is not too slow you may get an almost convincing result. What you hear when sitting other than in the centre is a very open question. One hears reports of technologies that create the sense of some phantom voice whispering in your ear, but ~if~ it works, I suspect the head has to be clamped immoveably in the sweet spot.

Richard Dobson

On 13/11/2010 11:46, peiman khosravi wrote:

Hi Menno,

Yes I am using the outh opcode but that doesn't offer any panning. I am
using vbap8 for panning but unfortunately vbap8 does not take x/y
parameters for positioning the source.

Best,

Peiman

Send bugs reports to the Sourceforge bug tracker
          https://sourceforge.net/tracker/?group_id=81968&atid=564599
Discussions of bugs and features can be posted here
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Here is an example of how I tailor panning envelopes in the orc to being determined by cmask in the score. In the cmask file (see bottom of this email)  I stipulate choices from 0 to 1 (to one decimal point) and then from 2 to 18. In the instrument I stipulate that if the kbal pfield is less than or equal to 1 then use the exact number for kbal. That is a static pan... no movement. But for a value above 1 use one of the envelopes. This is a simple way for cmask to determine the panning. To understand what is happening, begin at the bottom of the following orc code excerpt.

You may notice that it incorporates an oscili for one of the panning choices. If I understood correctly you were asking about this in a previous post.

This approach is very simplistic yet its output is quite dynamic... and it can easily be expanded or generalized for many other uses.

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;panning algorithms (range 0 - 1 (prec 2), 2 - 18)
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

kenv101 linseg    0,    idur,    1
kenv102 linseg    1,    idur,    0

kenv103 linseg    0,    idur * .2,    1, idur * .8, 0
kenv104 linseg    1,    idur * .2,    .001, idur * .8, 1

kenv105 linseg    0,    idur * .15,    1,    idur * .15,    .2, idur * .3, .8, idur * .4, .4
kenv106 linseg    1,    idur * .15,    0,    idur * .15,    .8, idur * .3, .2, idur * .4, .6

kenv107 linseg    0,    idur * .5,    1,    idur * .15,    .2, idur * .15, .8, idur * .35, .6
kenv108 linseg    1,    idur * .5,    0,    idur * .15,    .8, idur * .15, .2, idur * .35, .4

kenv109 linseg    .4,    idur * .25,    .8,    idur * .75,    0
kenv110 linseg    .6,    idur * .25,    .2,    idur * .75,    1

kenv111 linseg    0,     idur * .8,    .15,    idur * .2,    1
kenv112 linseg    1,     idur * .8,    .85,    idur * .2,    0

kenv113 linseg    .25,     idur * .8,    .75,    idur * .2,    .5
kenv114 linseg    .75,     idur * .8,    .25,    idur * .2,    .5

kenv115 linseg    0, idur * .15, .6, idur * .15, .3, idur * .15, .8, idur * .15, .5, idur * .15, .9, idur * .25, 0
kenv116 linseg    1, idur * .15, .3, idur * .15, .6, idur * .15, .2, idur * .15, .5, idur * .15, .1, idur * .25, 1

kenv117 oscili    1,    kfreq,    1


kenv9      =    (kpan = 2   ? kenv102:kenv101)
kenv10    =    (kpan = 3   ? kenv103:kenv9)
kenv11    =    (kpan = 4   ? kenv104:kenv10)
kenv12    =    (kpan = 5   ? kenv105:kenv11)
kenv13    =    (kpan = 6   ? kenv106:kenv12)
kenv14    =    (kpan = 7   ? kenv107:kenv13)
kenv15    =    (kpan = 8   ? kenv108:kenv14)
kenv16    =    (kpan = 9   ? kenv109:kenv15)
kenv17    =    (kpan = 10  ? kenv109:kenv16)
kenv18    =    (kpan = 11  ? kenv110:kenv17)
kenv19    =    (kpan = 12  ? kenv111:kenv18)
kenv20    =    (kpan = 13  ? kenv112:kenv19)
kenv21    =    (kpan = 14  ? kenv113:kenv20)
kenv22    =    (kpan = 15  ? kenv114:kenv21)
kenv23    =    (kpan = 16  ? kenv115:kenv22)
kenv24    =    (kpan = 17  ? kenv116:kenv23)
kenv25    =    (kpan = 18  ? kenv117:kenv24)

kbal         =    (kpan <= 1  ? kpan:kenv25)



;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
; example cmask code for that pfield although any other mask could be used
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;


p6                    ;kpan (range 0 - 1, 2, 3, 4, 5, 6, 7, 8, 9, 10)
item random (0, .1, .2, .3, .4, .5, .6, .7, .8, .9, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18)



Send bugs reports to the Sourceforge bug tracker
          https://sourceforge.net/tracker/?group_id=81968&atid=564599
Discussions of bugs and features can be posted here
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

hi peiman,

questions of the efficacy of vbap aside, it sounds like you want to
convert cartesian coordinates to polar for use with vbap.  maybe this
will help.  i haven't looked it over in a while, so checking the math
may be in order...

opcode iXYCoor2Vbap, ii, ii
       ;X and Y from -1->1
       iX, iY  xin
       iDistance = ((iX^2)+(iY^2))^0.5
       iSpread = (sqrt(2)-iDistance)*100 ; get spread
       iPh  taninv2   iX, iY
       iDeg = ((iPh/(2*3.141))*360)  ; get angle
       xout iDeg, iSpread
       ;iDeg from 0->360; spread from 0->100
endop


—  ben

On Sat, Nov 13, 2010 at 2:03 PM, Michael Rhoades
<mrhoades@perceptionfactory.com> wrote:
> Hello Peiman and all,
>
> I have been working on this subject for many years now. Slowly I have been
> developing an instrument that provides a lot of versatility of
> spatialization to the point that now, spatialization is an important formal
> aspect in every composition. The instrument has been tailored to the use of
> cmask since I use that for all of my initial score generation... although I
> do sometimes modify scores after they have been generated using excel.
>
> My goal is to render 8 channel wav files. In the csound rendering sequence
> of events, by the time a sound gets to the spatialization algorithms it is
> stereo (two channels) and it has a panning envelope applied. From there I
> send it through what I call distance algorithms, which are a set of
> envelopes that are meant to emulate the qualities of distance movement that
> I think are pertinent. I do not apply Doppler effects because I do not like
> the cliche glissando sounds that are heard in so many E/A music pieces.
> Basically, at this point, I want the sound to pan left to right from near to
> far, far to near, near to near and far to far... etc. So I use amplitude
> envelopes and reverb envelopes to accomplish it. This can sound very
> convincing to my ear.
>
> Once I have applied those algorithms I then put the two channel sound into
> any combination of 8 speakers. So, for instance, the left sound might be on
> speakers 1, 3, 5 and the right sound might be on speaker 7. Then I provide
> additional panning envelopes so that as the left sound is panning to the
> right sound, it is also being panned across speakers 1, 3, 5. To the
> listener this provides for a beautiful curve to the panning. Then if it is,
> for instance, from very far away to near... well... you can imagine the
> possibilities.
>
> All of these choices are determined by cmask on an event by event basis in
> the score. I represent them with simple numbers and then let cmask choose
> them depending on the mask I am using for each individual parameter.
>
> The bottom line is that I do not think there is a quick easy way to
> accomplish what you are after but with a little creativity you can come up
> with some very flexible and unique ways to instantiate your dreams...
>
> I hope this helps,
>
> Michael
>
>
> Send bugs reports to the Sourceforge bug tracker
>           https://sourceforge.net/tracker/?group_id=81968&atid=564599
> Discussions of bugs and features can be posted here
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
> csound"
>
>


Send bugs reports to the Sourceforge bug tracker
           https://sourceforge.net/tracker/?group_id=81968&atid=564599
Discussions of bugs and features can be posted here
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Hi Peiman,

I haven't been following this email too closely as I was reading it on
my phone, but I thought I'd mention that Jan Jacob Hofmann and I had
done a presentation at the LAC a couple years ago that discusses using
Ambisonics with blue/csound, particularly using CMask and JMask
(blue's graphical version of CMask) to do spatialized granular
synthesis:

http://lac.linuxaudio.org/2009/cdm/Thursday/04_Hofmann/index.html

Jan Jacob has written pieces using this technique to have each sound
grain have their own spatial location (x/y/z, distance clues) via
Ambisonics in addition to using doppler, local/global reverb, etc.

I'm not sure if this is what you're looking for, but I thought it
worth mentioning.

Thanks!
steven

On Sat, Nov 13, 2010 at 7:36 AM, peiman khosravi
<peimankhosravi@gmail.com> wrote:
> Thanks Richard,
>
> Here's one without all the typos!!!
>
> Yes I agree with you. My intention is not to do this rigorously, of course I
> will be judging by ear and moving around the room during the composition to
> minimize the loss of spatial image for non-centered listeners. At the moment
> I am working with Cmask to create some 'spatial' textures (with
> grain-by-grain 'panning'). I want for instance to do the following:
>
> Use a Gaussian random generator with a 'mean' value corresponding with the
> centre of the frontal stereo pair and a deviation width to cover more or
> less all 6 or 8 speakers (I still haven't decided about the numbers). I
> experimented with this last night and it works. I had to have a weird
> virtual speaker set up to allow this though (vbaplsinit       2, 6, 120,
> 180, 60, 240, 0, 300) In this case a mean value of 150 (corresponding to 0.5
> if the degrees are scaled to fall within 0. and 1.) will be centered between
> the frontal stereo pair with this set-up:
>
>
>
>
>                           1           2
>
>                       3                     4
>
>                           5           6
>
>
> Now I would like to be able to move the center ('mean') towards the back of
> the room in a straight line so that the grain distribution is concentrated
> in between the differed stereo pairs in turn. At the same time I would like
> to have the ability to move the 'mean' in circles. I think with the right
> amount of spread among the speakers and the right textural density it should
> work. But the question is how do I realise the panning other than having to
> make my own UDO. I suspect a fair amount of maths is needed to make an 6
> channel signal distributor with x/y parameters.
>
> Thanks very much
>
>>
>>
>> On 13 November 2010 12:00, Richard Dobson <richarddobson@blueyonder.co.uk>
>> wrote:
>>>
>>> It's actually a very difficult thing to do - placing sounds inside the
>>> speaker array (which implies a soundfield in effect radiating out ~towards~
>>> the speakers in some directions) and having them pass (in effect) through
>>> the listener's head. Indeed, I am not sure to what extent it has yet been
>>> done in a technically rigorous way. You can do it "rhetorically" of course,
>>> by intensity panning from speaker to speaker across the circle, and if it is
>>> not too slow you may get an almost convincing result. What you hear when
>>> sitting other than in the centre is a very open question. One hears reports
>>> of technologies that create the sense of some phantom voice whispering in
>>> your ear, but ~if~ it works, I suspect the head has to be clamped immoveably
>>> in the sweet spot.
>>>
>>> Richard Dobson
>>>
>>> On 13/11/2010 11:46, peiman khosravi wrote:
>>>>
>>>> Hi Menno,
>>>>
>>>> Yes I am using the outh opcode but that doesn't offer any panning. I am
>>>> using vbap8 for panning but unfortunately vbap8 does not take x/y
>>>> parameters for positioning the source.
>>>>
>>>> Best,
>>>>
>>>> Peiman
>>>>
>>>
>>>
>>> Send bugs reports to the Sourceforge bug tracker
>>>           https://sourceforge.net/tracker/?group_id=81968&atid=564599
>>> Discussions of bugs and features can be posted here
>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
>>> csound"
>>>
>>
>
>


Send bugs reports to the Sourceforge bug tracker
           https://sourceforge.net/tracker/?group_id=81968&atid=564599
Discussions of bugs and features can be posted here
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

hi peiman,

questions of the efficacy of vbap aside, it sounds like you want to
convert cartesian coordinates to polar for use with vbap.  maybe this
will help.  i haven't looked it over in a while, so checking the math
may be in order...

opcode iXYCoor2Vbap, ii, ii
       ;X and Y from -1->1
       iX, iY  xin
       iDistance = ((iX^2)+(iY^2))^0.5
       iSpread = (sqrt(2)-iDistance)*100 ; get spread
       iPh  taninv2   iX, iY
       iDeg = ((iPh/(2*3.141))*360)  ; get angle
       xout iDeg, iSpread
       ;iDeg from 0->360; spread from 0->100
endop


—  ben

On Sat, Nov 13, 2010 at 2:03 PM, Michael Rhoades
<mrhoades@perceptionfactory.com> wrote:
> Hello Peiman and all,
>
> I have been working on this subject for many years now. Slowly I have been
> developing an instrument that provides a lot of versatility of
> spatialization to the point that now, spatialization is an important formal
> aspect in every composition. The instrument has been tailored to the use of
> cmask since I use that for all of my initial score generation... although I
> do sometimes modify scores after they have been generated using excel.
>
> My goal is to render 8 channel wav files. In the csound rendering sequence
> of events, by the time a sound gets to the spatialization algorithms it is
> stereo (two channels) and it has a panning envelope applied. From there I
> send it through what I call distance algorithms, which are a set of
> envelopes that are meant to emulate the qualities of distance movement that
> I think are pertinent. I do not apply Doppler effects because I do not like
> the cliche glissando sounds that are heard in so many E/A music pieces.
> Basically, at this point, I want the sound to pan left to right from near to
> far, far to near, near to near and far to far... etc. So I use amplitude
> envelopes and reverb envelopes to accomplish it. This can sound very
> convincing to my ear.
>
> Once I have applied those algorithms I then put the two channel sound into
> any combination of 8 speakers. So, for instance, the left sound might be on
> speakers 1, 3, 5 and the right sound might be on speaker 7. Then I provide
> additional panning envelopes so that as the left sound is panning to the
> right sound, it is also being panned across speakers 1, 3, 5. To the
> listener this provides for a beautiful curve to the panning. Then if it is,
> for instance, from very far away to near... well... you can imagine the
> possibilities.
>
> All of these choices are determined by cmask on an event by event basis in
> the score. I represent them with simple numbers and then let cmask choose
> them depending on the mask I am using for each individual parameter.
>
> The bottom line is that I do not think there is a quick easy way to
> accomplish what you are after but with a little creativity you can come up
> with some very flexible and unique ways to instantiate your dreams...
>
> I hope this helps,
>
> Michael
>
>
> Send bugs reports to the Sourceforge bug tracker
>           https://sourceforge.net/tracker/?group_id=81968&atid=564599
> Discussions of bugs and features can be posted here
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
> csound"
>
>


Send bugs reports to the Sourceforge bug tracker
           https://sourceforge.net/tracker/?group_id=81968&atid=564599
Discussions of bugs and features can be posted here
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

hi peiman,

questions of the efficacy of vbap aside, it sounds like you want to
convert cartesian coordinates to polar for use with vbap.  maybe this
will help.  i haven't looked it over in a while, so checking the math
may be in order...

opcode iXYCoor2Vbap, ii, ii
       ;X and Y from -1->1
       iX, iY  xin
       iDistance = ((iX^2)+(iY^2))^0.5
       iSpread = (sqrt(2)-iDistance)*100 ; get spread
       iPh  taninv2   iX, iY
       iDeg = ((iPh/(2*3.141))*360)  ; get angle
       xout iDeg, iSpread
       ;iDeg from 0->360; spread from 0->100
endop


—  ben

On Sat, Nov 13, 2010 at 2:03 PM, Michael Rhoades
<mrhoades@perceptionfactory.com> wrote:
> Hello Peiman and all,
>
> I have been working on this subject for many years now. Slowly I have been
> developing an instrument that provides a lot of versatility of
> spatialization to the point that now, spatialization is an important formal
> aspect in every composition. The instrument has been tailored to the use of
> cmask since I use that for all of my initial score generation... although I
> do sometimes modify scores after they have been generated using excel.
>
> My goal is to render 8 channel wav files. In the csound rendering sequence
> of events, by the time a sound gets to the spatialization algorithms it is
> stereo (two channels) and it has a panning envelope applied. From there I
> send it through what I call distance algorithms, which are a set of
> envelopes that are meant to emulate the qualities of distance movement that
> I think are pertinent. I do not apply Doppler effects because I do not like
> the cliche glissando sounds that are heard in so many E/A music pieces.
> Basically, at this point, I want the sound to pan left to right from near to
> far, far to near, near to near and far to far... etc. So I use amplitude
> envelopes and reverb envelopes to accomplish it. This can sound very
> convincing to my ear.
>
> Once I have applied those algorithms I then put the two channel sound into
> any combination of 8 speakers. So, for instance, the left sound might be on
> speakers 1, 3, 5 and the right sound might be on speaker 7. Then I provide
> additional panning envelopes so that as the left sound is panning to the
> right sound, it is also being panned across speakers 1, 3, 5. To the
> listener this provides for a beautiful curve to the panning. Then if it is,
> for instance, from very far away to near... well... you can imagine the
> possibilities.
>
> All of these choices are determined by cmask on an event by event basis in
> the score. I represent them with simple numbers and then let cmask choose
> them depending on the mask I am using for each individual parameter.
>
> The bottom line is that I do not think there is a quick easy way to
> accomplish what you are after but with a little creativity you can come up
> with some very flexible and unique ways to instantiate your dreams...
>
> I hope this helps,
>
> Michael
>
>
> Send bugs reports to the Sourceforge bug tracker
>           https://sourceforge.net/tracker/?group_id=81968&atid=564599
> Discussions of bugs and features can be posted here
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
> csound"
>
>


Send bugs reports to the Sourceforge bug tracker
           https://sourceforge.net/tracker/?group_id=81968&atid=564599
Discussions of bugs and features can be posted here
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Here is an example of how I tailor panning envelopes in the orc to being determined by cmask in the score. In the cmask file (see bottom of this email)  I stipulate choices from 0 to 1 (to one decimal point) and then from 2 to 18. In the instrument I stipulate that if the kbal pfield is less than or equal to 1 then use the exact number for kbal. That is a static pan... no movement. But for a value above 1 use one of the envelopes. This is a simple way for cmask to determine the panning. To understand what is happening, begin at the bottom of the following orc code excerpt.

You may notice that it incorporates an oscili for one of the panning choices. If I understood correctly you were asking about this in a previous post.

This approach is very simplistic yet its output is quite dynamic... and it can easily be expanded or generalized for many other uses.

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;panning algorithms (range 0 - 1 (prec 2), 2 - 18)
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

kenv101 linseg    0,    idur,    1
kenv102 linseg    1,    idur,    0

kenv103 linseg    0,    idur * .2,    1, idur * .8, 0
kenv104 linseg    1,    idur * .2,    .001, idur * .8, 1

kenv105 linseg    0,    idur * .15,    1,    idur * .15,    .2, idur * .3, .8, idur * .4, .4
kenv106 linseg    1,    idur * .15,    0,    idur * .15,    .8, idur * .3, .2, idur * .4, .6

kenv107 linseg    0,    idur * .5,    1,    idur * .15,    .2, idur * .15, .8, idur * .35, .6
kenv108 linseg    1,    idur * .5,    0,    idur * .15,    .8, idur * .15, .2, idur * .35, .4

kenv109 linseg    .4,    idur * .25,    .8,    idur * .75,    0
kenv110 linseg    .6,    idur * .25,    .2,    idur * .75,    1

kenv111 linseg    0,     idur * .8,    .15,    idur * .2,    1
kenv112 linseg    1,     idur * .8,    .85,    idur * .2,    0

kenv113 linseg    .25,     idur * .8,    .75,    idur * .2,    .5
kenv114 linseg    .75,     idur * .8,    .25,    idur * .2,    .5

kenv115 linseg    0, idur * .15, .6, idur * .15, .3, idur * .15, .8, idur * .15, .5, idur * .15, .9, idur * .25, 0
kenv116 linseg    1, idur * .15, .3, idur * .15, .6, idur * .15, .2, idur * .15, .5, idur * .15, .1, idur * .25, 1

kenv117 oscili    1,    kfreq,    1


kenv9      =    (kpan = 2   ? kenv102:kenv101)
kenv10    =    (kpan = 3   ? kenv103:kenv9)
kenv11    =    (kpan = 4   ? kenv104:kenv10)
kenv12    =    (kpan = 5   ? kenv105:kenv11)
kenv13    =    (kpan = 6   ? kenv106:kenv12)
kenv14    =    (kpan = 7   ? kenv107:kenv13)
kenv15    =    (kpan = 8   ? kenv108:kenv14)
kenv16    =    (kpan = 9   ? kenv109:kenv15)
kenv17    =    (kpan = 10  ? kenv109:kenv16)
kenv18    =    (kpan = 11  ? kenv110:kenv17)
kenv19    =    (kpan = 12  ? kenv111:kenv18)
kenv20    =    (kpan = 13  ? kenv112:kenv19)
kenv21    =    (kpan = 14  ? kenv113:kenv20)
kenv22    =    (kpan = 15  ? kenv114:kenv21)
kenv23    =    (kpan = 16  ? kenv115:kenv22)
kenv24    =    (kpan = 17  ? kenv116:kenv23)
kenv25    =    (kpan = 18  ? kenv117:kenv24)

kbal         =    (kpan <= 1  ? kpan:kenv25)



;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
; example cmask code for that pfield although any other mask could be used
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;


p6                    ;kpan (range 0 - 1, 2, 3, 4, 5, 6, 7, 8, 9, 10)
item random (0, .1, .2, .3, .4, .5, .6, .7, .8, .9, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18)



Send bugs reports to the Sourceforge bug tracker
          https://sourceforge.net/tracker/?group_id=81968&atid=564599
Discussions of bugs and features can be posted here
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

sounds like it outputs iDeg between -180 to 180 then.  it should still
work with vbap.  if not, try using the wrap opcode.

—  ben

On Sat, Nov 13, 2010 at 3:38 PM, peiman khosravi
<peimankhosravi@gmail.com> wrote:
> Hi Ben,
>
> Just to follow up. I don't quite understand the math in your code but I get
> the idea.
>
> However the udo is not returning what I expect it to return. For an x value
> of -1 and a y value of -1 I get iDeg = -135. Should there be any negative
> degrees? (csd bellow)
>
> Thanks again
>
> Peiman
>
> <CsoundSynthesizer>
>
> <CsOptions>
> -W -o dac
> </CsOptions>
>
> <CsInstruments>
>
>
> sr     = 96000
> ksmps  = 16
> nchnls = 1
> 0dbfs  = 1
>
>
> ;    vbaplsinit          2, 8, 0, 45, 90, 135, 180, 225, 270, 315
>
>
> opcode iXYCoor2Vbap, ii, ii
>        ;X and Y from -1->1
>        iX, iY  xin
>        iDistance = ((iX^2)+(iY^2))^0.5
>        iSpread = (sqrt(2)-iDistance)*100 ; get spread
>        iPh  taninv2   iX, iY
>        iDeg = ((iPh/(2*3.141))*360)  ; get angle
>        xout iDeg, iSpread
>        ;iDeg from 0->360; spread from 0->100
> endop
>
>
>
>     instr 1
>
> ix = -1
> iy = -1
>
> iDeg, iSpread    iXYCoor2Vbap    ix, iy
>
> print iDeg, iSpread
>
> ;ar1, ar2, ar3, ar4, ar5, ar6, ar7, ar8 vbap8 a1, iDeg, 0, iSpread
>
>
> ;    outo ar1, ar2, ar3, ar4, ar5, ar6, ar7, ar8
> endin
>
> </CsInstruments>
>
> <CsScore>
>
> i1    0    1
>
> </CsScore>
>
> </CsoundSynthesizer>
>
> On 13 November 2010 13:36, ben hackbarth <hackbarth@gmail.com> wrote:
>>
>> hi peiman,
>>
>> questions of the efficacy of vbap aside, it sounds like you want to
>> convert cartesian coordinates to polar for use with vbap.  maybe this
>> will help.  i haven't looked it over in a while, so checking the math
>> may be in order...
>>
>> opcode iXYCoor2Vbap, ii, ii
>>        ;X and Y from -1->1
>>        iX, iY  xin
>>        iDistance = ((iX^2)+(iY^2))^0.5
>>        iSpread = (sqrt(2)-iDistance)*100 ; get spread
>>        iPh  taninv2   iX, iY
>>        iDeg = ((iPh/(2*3.141))*360)  ; get angle
>>        xout iDeg, iSpread
>>        ;iDeg from 0->360; spread from 0->100
>> endop
>>
>>
>> —  ben
>>
>>
>>
>> On Sat, Nov 13, 2010 at 2:03 PM, Michael Rhoades
>> <mrhoades@perceptionfactory.com> wrote:
>> > Hello Peiman and all,
>> >
>> > I have been working on this subject for many years now. Slowly I have
>> > been
>> > developing an instrument that provides a lot of versatility of
>> > spatialization to the point that now, spatialization is an important
>> > formal
>> > aspect in every composition. The instrument has been tailored to the use
>> > of
>> > cmask since I use that for all of my initial score generation...
>> > although I
>> > do sometimes modify scores after they have been generated using excel.
>> >
>> > My goal is to render 8 channel wav files. In the csound rendering
>> > sequence
>> > of events, by the time a sound gets to the spatialization algorithms it
>> > is
>> > stereo (two channels) and it has a panning envelope applied. >From there
>> > I
>> > send it through what I call distance algorithms, which are a set of
>> > envelopes that are meant to emulate the qualities of distance movement
>> > that
>> > I think are pertinent. I do not apply Doppler effects because I do not
>> > like
>> > the cliche glissando sounds that are heard in so many E/A music pieces.
>> > Basically, at this point, I want the sound to pan left to right from
>> > near to
>> > far, far to near, near to near and far to far... etc. So I use amplitude
>> > envelopes and reverb envelopes to accomplish it. This can sound very
>> > convincing to my ear.
>> >
>> > Once I have applied those algorithms I then put the two channel sound
>> > into
>> > any combination of 8 speakers. So, for instance, the left sound might be
>> > on
>> > speakers 1, 3, 5 and the right sound might be on speaker 7. Then I
>> > provide
>> > additional panning envelopes so that as the left sound is panning to the
>> > right sound, it is also being panned across speakers 1, 3, 5. To the
>> > listener this provides for a beautiful curve to the panning. Then if it
>> > is,
>> > for instance, from very far away to near... well... you can imagine the
>> > possibilities.
>> >
>> > All of these choices are determined by cmask on an event by event basis
>> > in
>> > the score. I represent them with simple numbers and then let cmask
>> > choose
>> > them depending on the mask I am using for each individual parameter.
>> >
>> > The bottom line is that I do not think there is a quick easy way to
>> > accomplish what you are after but with a little creativity you can come
>> > up
>> > with some very flexible and unique ways to instantiate your dreams...
>> >
>> > I hope this helps,
>> >
>> > Michael
>> >
>> >
>> > Send bugs reports to the Sourceforge bug tracker
>> >           https://sourceforge.net/tracker/?group_id=81968&atid=564599
>> > Discussions of bugs and features can be posted here
>> > To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
>> > csound"
>> >
>> >
>>
>>
>> Send bugs reports to the Sourceforge bug tracker
>>            https://sourceforge.net/tracker/?group_id=81968&atid=564599
>> Discussions of bugs and features can be posted here
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
>> csound"
>>
>
>


Send bugs reports to the Sourceforge bug tracker
           https://sourceforge.net/tracker/?group_id=81968&atid=564599
Discussions of bugs and features can be posted here
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"