Question for all:

In your opinion, which opcodes, in and of themselves, give the truest 
localization in stereo or quad?

Is the answer different for headphones and speakers?

Art Hunkins

----- Original Message ----- 
From: "luis jure" 
To: 
Sent: Monday, May 26, 2008 11:38 PM
Subject: [Csnd] Re: Re: Re: Re: spatialisation for headphones


El Mon, 26 May 2008 15:26:14 +0100
Victor Lazzarini  escribió:

> if you add the other cues (reverb, lowpass filtering,
> intensity changes), then you would have the full effect.
> The thing is that spat3d adds these things, but does not have the
> quality/precision of hrtf.
>
> Also, you might want to try babo. With it you can even do doppler (as
> a side-effect).

i only have a stereo system to do my tests (unfortunately), but
recently i began to make some tests using spat3d, and i found it very
good for moving sources, with good quality panning and realistic
doppler. first try it in free field mode (function table 0, see the
attached minimal example). i agree that the simulation of room
acoustics is not so good, at least using the values given in the
examples. but i began tweaking the table with the room characteristics,
and i think one might get eventually very good results. but for that you
need good knowledge of room acoustics, which i lack. it would be great
if someone with good knowledge in that area could contribute a nice set
of parameters for the room.

i should also revisit the babo opcode, i tested it only once and long
ago. from what i think to remember, the results were the opposite: very
nice reverberation but very poor localization.


best,

lj

Send bugs reports to this list.
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe 
csound" 

El Tue, 27 May 2008 09:27:40 -0400
"Art Hunkins"  escribió:

> In your opinion, which opcodes, in and of themselves, give the truest 
> localization in stereo or quad?

i haven't been able yet to set up a system beyond a simple two-channels
stereo, can't comment on quad. 

locsig gives good panning in stereo, and it can also generate quad
output. but it is strictly intensity panning, no doppler. good only for
fixed or slowly moving sources. the best from what i tested is spat3d,
because it uses a physical model of the movement, which gives a
realistic doppler effect, at least for my ears. another plus of spat3d
is that the ouput is in ambisonics format, and IMO that is the way to
go. once the signal is coded in ambisonics, you can decode it into any
speaker disposition you want. with so many multi-channel formats and
systems available today i think we should consider the different
possible set-ups we could have at our disposal to play our pieces.
that's why i think opcodes using ambisonics, like spat3d, are very
convenient. (disclaimer: i know _very_ little about the theory behind
ambisonics)

> Is the answer different for headphones and speakers?

certainly, but i'm not very interested in headphones so i never
experimented with that. perhaps the hrtf related opcodes?

best,

lj

Hector,

It would be really great to see and try your UDOs.  It would be great  
for
you to post them to the list and maybe you could include them in the  
UDO library too?

-dB


On May 28, 2008, at 12:26 AM, Hector Centeno wrote:

> Art,
>
> I've done multichannel spatialization for 6 and 8 channels using the
> vbap opcodes and they are very effective.  If you want to change the
> location and spread in realtime using k rate variables  you will have
> to change the source code or you can use the vbapmove version for
> predefined trajectories. (John ffitch changed the opcodes in CVS so
> they can take k-rate but still some extra changes are needed for the
> elevation and spread)
>
> Recently I've been using ambisonics and my own UDO's for
> encoding-decoding (so I can use the controlled opposites scalars by
> Richard Furse http://www.muse.demon.co.uk/ref/speakers.html ) and I'm
> finding the localization works pretty well and with a wide sweetspot.
> I can post the UDO's if someone is interested but anyone can easily
> make them using the information available at Furse's web site
> (recently I presented a 6 channel piece in a concert using ambisonics
> and created a binaural stereo "reduction" using the new hrtfstat
> opcode).
>
> Cheers,
>
>
> Hector
>
>
> On Tue, May 27, 2008 at 9:27 AM, Art Hunkins   
> wrote:
>> Question for all:
>>
>> In your opinion, which opcodes, in and of themselves, give the truest
>> localization in stereo or quad?
>>
>> Is the answer different for headphones and speakers?
>>
>> Art Hunkins
>>
>> ----- Original Message ----- From: "luis jure" 
>> To: 
>> Sent: Monday, May 26, 2008 11:38 PM
>> Subject: [Csnd] Re: Re: Re: Re: spatialisation for headphones
>>
>>
>> El Mon, 26 May 2008 15:26:14 +0100
>> Victor Lazzarini  escribió:
>>
>>> if you add the other cues (reverb, lowpass filtering,
>>> intensity changes), then you would have the full effect.
>>> The thing is that spat3d adds these things, but does not have the
>>> quality/precision of hrtf.
>>>
>>> Also, you might want to try babo. With it you can even do doppler  
>>> (as
>>> a side-effect).
>>
>> i only have a stereo system to do my tests (unfortunately), but
>> recently i began to make some tests using spat3d, and i found it very
>> good for moving sources, with good quality panning and realistic
>> doppler. first try it in free field mode (function table 0, see the
>> attached minimal example). i agree that the simulation of room
>> acoustics is not so good, at least using the values given in the
>> examples. but i began tweaking the table with the room  
>> characteristics,
>> and i think one might get eventually very good results. but for  
>> that you
>> need good knowledge of room acoustics, which i lack. it would be  
>> great
>> if someone with good knowledge in that area could contribute a  
>> nice set
>> of parameters for the room.
>>
>> i should also revisit the babo opcode, i tested it only once and long
>> ago. from what i think to remember, the results were the opposite:  
>> very
>> nice reverberation but very poor localization.
>>
>>
>> best,
>>
>> lj
>>
>> Send bugs reports to this list.
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>> "unsubscribe
>> csound"
>>
>>
>> Send bugs reports to this list.
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>> "unsubscribe
>> csound"
>>
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
> "unsubscribe csound"

Hector,

Thanks for making and posting these.
They are great and much appreciated.
If you could find the time, I would especially
love to see and use  the 4 channel version as well.

-dB


On Jun 8, 2008, at 12:12 AM, Hector Centeno wrote:

> Hello,
>
> Sorry for taking so long to reply to this. These are the UDOs for 2nd
> order ambisonic encoding and decoding to 8 channels (If there is
> interest in including them in the UDO library I can make the ones for
> 6, 5 and 4 channels, it's just about replacing the scalars available
> at Richard Furse's web site):
>
> 	opcode ambencode,aaaaaaaaa,akk
>
> asig, kAzim, kElev xin
>
> kchw	= 	0.707107
> kchx	=	cos(kAzim) * cos(kElev)
> kchy	=	sin(kAzim) * cos(kElev)
> kchz	=	sin(kElev)
> kchr	=	1.5 * sin(kElev) * sin(kElev) -0.5
> kchs	=	cos(kAzim) * sin(2 * kElev)
> kcht	=	sin(kAzim) * sin(2 * kElev)
> kchu	=	cos(2 * kAzim) * cos(kElev) * cos(kElev)
> kchv	=	sin(2 * kAzim) * cos(kElev) * cos(kElev)
>
>
> aw 	=	asig *	kchw
> ax 	=	asig *	kchx
> ay 	=	asig *	kchy
> az 	= 	asig *	kchz
> ar 	=	asig *	kchr
> as 	=	asig *	kchs
> at 	=	asig *	kcht
> au 	=	asig *	kchu
> av 	=	asig *	kchv
>
> xout aw, ax, ay, az, ar, as, at, au, av
>
> 	endop
>
>
>
>
> 	opcode ambdecode8,aaaaaaaa,aaaaaaaaa
>
> aw, ax, ay, az, ar, as, at, au, av xin
>
>
> ;Decode to octagon using controlled opposites scalars
>
>
> ; Speaker 1	<0.9239,0.3827,0.0000>
>  ach1 sum	aw *0.1768, ax *0.1591, ay * 0.0659, au *0.0342, av * 0.0342
>
> ; Speaker 2	<0.3827,0.9239,0.0000>
>  ach2 sum	aw *0.1768, ax *0.0659, ay * 0.1591, au *-0.0342, av *  
> 0.0342
>
> ; Speaker 3	<-0.3827,0.9239,0.0000>
>  ach3	sum	aw *0.1768, ax *-0.0659, ay * 0.1591, au *-0.0342, av *  
> -0.0342
>
> ; Speaker 4	<-0.9239,0.3827,0.0000>
>  ach4 sum	aw *0.1768, ax *-0.1591, ay * 0.0659, au *0.0342, av *  
> -0.0342
>
> ; Speaker 5 	<-0.9239,-0.3827,0.0000>
>  ach5	sum	aw *0.1768, ax *-0.1591, ay * -0.0659, au *0.0342, av *  
> 0.0342
>
> ; Speaker 6 	<-0.3827,-0.9239,0.0000>
>  ach6	sum	aw *0.1768, ax *-0.0659, ay * -0.1591, au *-0.0342,av *  
> 0.0342
>
> ; Speaker 7 	<0.3827,-0.9239,0.0000>
>  ach7	sum	aw *0.1768, ax *0.0659, ay * -0.1591, au *-0.0342, av *  
> -0.0342
>
> ; Speaker 8 	<0.9239,-0.3827,0.0000>
>  ach8 sum	aw *0.1768, ax *0.1591, ay * -0.0659, au *0.0342, av *  
> -0.0342
>
>
> xout ach1, ach2, ach3, ach4, ach5, ach6, ach7, ach8
>
> 	endop
>
>
> Hope they are helpful.
>
> Cheers!
>
> Hector
>
>
>
>
> On Sun, Jun 1, 2008 at 10:01 AM, Dr. Richard Boulanger
>  wrote:
>> Hector,
>>
>> It would be really great to see and try your UDOs.  It would be  
>> great for
>> you to post them to the list and maybe you could include them in  
>> the UDO
>> library too?
>>
>> -dB
>>
>>
>> On May 28, 2008, at 12:26 AM, Hector Centeno wrote:
>>
>>> Art,
>>>
>>> I've done multichannel spatialization for 6 and 8 channels using the
>>> vbap opcodes and they are very effective.  If you want to change the
>>> location and spread in realtime using k rate variables  you will  
>>> have
>>> to change the source code or you can use the vbapmove version for
>>> predefined trajectories. (John ffitch changed the opcodes in CVS so
>>> they can take k-rate but still some extra changes are needed for the
>>> elevation and spread)
>>>
>>> Recently I've been using ambisonics and my own UDO's for
>>> encoding-decoding (so I can use the controlled opposites scalars by
>>> Richard Furse http://www.muse.demon.co.uk/ref/speakers.html ) and  
>>> I'm
>>> finding the localization works pretty well and with a wide  
>>> sweetspot.
>>> I can post the UDO's if someone is interested but anyone can easily
>>> make them using the information available at Furse's web site
>>> (recently I presented a 6 channel piece in a concert using  
>>> ambisonics
>>> and created a binaural stereo "reduction" using the new hrtfstat
>>> opcode).
>>>
>>> Cheers,
>>>
>>>
>>> Hector
>>>
>>>
>>> On Tue, May 27, 2008 at 9:27 AM, Art Hunkins   
>>> wrote:
>>>>
>>>> Question for all:
>>>>
>>>> In your opinion, which opcodes, in and of themselves, give the  
>>>> truest
>>>> localization in stereo or quad?
>>>>
>>>> Is the answer different for headphones and speakers?
>>>>
>>>> Art Hunkins
>>>>
>>>> ----- Original Message ----- From: "luis jure"  
>>>> 
>>>> To: 
>>>> Sent: Monday, May 26, 2008 11:38 PM
>>>> Subject: [Csnd] Re: Re: Re: Re: spatialisation for headphones
>>>>
>>>>
>>>> El Mon, 26 May 2008 15:26:14 +0100
>>>> Victor Lazzarini  escribió:
>>>>
>>>>> if you add the other cues (reverb, lowpass filtering,
>>>>> intensity changes), then you would have the full effect.
>>>>> The thing is that spat3d adds these things, but does not have the
>>>>> quality/precision of hrtf.
>>>>>
>>>>> Also, you might want to try babo. With it you can even do  
>>>>> doppler (as
>>>>> a side-effect).
>>>>
>>>> i only have a stereo system to do my tests (unfortunately), but
>>>> recently i began to make some tests using spat3d, and i found it  
>>>> very
>>>> good for moving sources, with good quality panning and realistic
>>>> doppler. first try it in free field mode (function table 0, see the
>>>> attached minimal example). i agree that the simulation of room
>>>> acoustics is not so good, at least using the values given in the
>>>> examples. but i began tweaking the table with the room  
>>>> characteristics,
>>>> and i think one might get eventually very good results. but for  
>>>> that you
>>>> need good knowledge of room acoustics, which i lack. it would be  
>>>> great
>>>> if someone with good knowledge in that area could contribute a  
>>>> nice set
>>>> of parameters for the room.
>>>>
>>>> i should also revisit the babo opcode, i tested it only once and  
>>>> long
>>>> ago. from what i think to remember, the results were the  
>>>> opposite: very
>>>> nice reverberation but very poor localization.
>>>>
>>>>
>>>> best,
>>>>
>>>> lj
>>>>
>>>> Send bugs reports to this list.
>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>>>> "unsubscribe
>>>> csound"
>>>>
>>>>
>>>> Send bugs reports to this list.
>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>>>> "unsubscribe
>>>> csound"
>>>>
>>>
>>>
>>> Send bugs reports to this list.
>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>>> "unsubscribe
>>> csound"
>>
>>
>>
>> Send bugs reports to this list.
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
>> "unsubscribe
>> csound"
>
>
> Send bugs reports to this list.
> To unsubscribe, send email sympa@lists.bath.ac.uk with body  
> "unsubscribe csound"