Hello everyone,

I wonder if there is an opcode that is able to calculate the arithmetic mean
(or average math?) of a set of values (k or i).
Has anyone ever created a patch for this?

thanks,

e

kaver = (k1 + k2)/2

?

On 5 Feb 2011, at 18:04, Enrico Francioni wrote:

>
>
> Hello everyone,
>
> I wonder if there is an opcode that is able to calculate the  
> arithmetic mean
> (or average math?) of a set of values (k or i).
> Has anyone ever created a patch for this?
>
> thanks,
>
> e
> -- 
> View this message in context: http://csound.1045644.n5.nabble.com/arithmetic-mean-tp3372682p3372682.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"
>



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"

One of the vector objects I think calculates the average of all the
stored table values too.

P

On 5 February 2011 20:09, Victor Lazzarini  wrote:
> kaver = (k1 + k2)/2
>
> ?
>
> On 5 Feb 2011, at 18:04, Enrico Francioni wrote:
>
>>
>>
>> Hello everyone,
>>
>> I wonder if there is an opcode that is able to calculate the arithmetic
>> mean
>> (or average math?) of a set of values (k or i).
>> Has anyone ever created a patch for this?
>>
>> thanks,
>>
>> e
>> --
>> View this message in context:
>> http://csound.1045644.n5.nabble.com/arithmetic-mean-tp3372682p3372682.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"
>>
>
>
>
> 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"

For example:
I would like to calculate the arithmetic mean of kcps (in ptrack),
but only when kcps it holds for a certain time around the same values.

For example, around 440, then around 220, then around 390 ..., etc ...

e

Yeah, it works well to take out the spurious spikes in a pitch
tracking analysis signal.

I've included some example code below, it also includes a frequency
dependant method for auto-switching between analysis with different
hopsizes, so the example could have been simplified a bit to show
mediank more clearly.

best
Oeyvind

; Pitch and amp analysis

	kcps 		init 0	
	ihopsize1	= 4096
	ihopsize2	= 512
	iChangeover	= (sr/ihopsize2)*3

if kcps > iChangeover kgoto small
	kcps, kamp 	ptrack a1, ihopsize1
if kcps <= iChangeover kgoto nosmall
small:
	kcps, kamp 	ptrack a1, ihopsize2
nosmall:

; filter pitch tracking signal
	imedianSize	= int(0.09*kr)		; calc as (1/(ihopsize*2))*kr
	kcps		mediank	kcps, imedianSize, imedianSize




2011/2/6 Victor Lazzarini :
> you've beat me too it, Ian. I was looking for the name of the opcode...
> actually Oeyvind asked for this to use with pitch controller.
> Victor
> On 6 Feb 2011, at 11:00, Iain McCurdy wrote:
>
> Hi Enrico,
>
> You might like to take a look at the new mediank opcode (ver 5.13). It might
> provide you with the sort of data filtering you require here.
>
> Iain
>
>> Date: Sun, 6 Feb 2011 02:41:43 -0800
>> From: francioni61021@libero.it
>> To: csound@lists.bath.ac.uk
>> Subject: [Csnd] Re: …arithmetic mean?
>>
>>
>>
>> For example:
>> I would like to calculate the arithmetic mean of kcps (in ptrack),
>> but only when kcps it holds for a certain time around the same values.
>>
>> For example, around 440, then around 220, then around 390 ..., etc ...
>>
>> e
>> --
>> View this message in
>> context: http://csound.1045644.n5.nabble.com/arithmetic-mean-tp3372682p3373120.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"
>>
>
>


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"

many thanks, Oeyvind!

you can write a comment on your example?

... in fact, what is the opcode mediank?
how exactly is the performance of mediank?

e

All good.

Here's a simpler version:

       ihopsize        = 1024
       kcps, kamp      ptrack a1, ihopsize
       imedianSize     = 129
       kcps            mediank kcps, imedianSize, imedianSize

The median filter is quite different from most audio filters, in that
the output is the middle value in it's buffer. The buffer size is  set
by imedianSize in the example above. You might want to experiment with
different sizes to see what best fits with your data.
For example for the input data
[7,1,8,57,2,7,8,9,8]
you can see that most of the activity is in the area around values 7,8 and 9,
with spurious peaks (probably noise) at 1,2 and 57.
If we sort this buffer of data in ascending order:
[1,2,7,7,8,8,8,9,57]
and take the middle member (number 5 from the left), we get 8 as the output.

For the next k-rate period, the data in the buffer is updated (adding
one member on the left side and discarding the one one the right)

all best
Oeyvind

2011/2/8 Enrico Francioni :
>
>
> many thanks, Oeyvind!
>
> you can write a comment on your example?
>
> ... in fact, what is the opcode mediank?
> how exactly is the performance of mediank?
>
> e
> --
> View this message in context: http://csound.1045644.n5.nabble.com/arithmetic-mean-tp3372682p3375500.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"
>
>


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,

Here's the manual page:
www.csounds.com/manual/html/mediank.html

Cheers,
Andres

On Tue, Feb 8, 2011 at 7:58 AM, Oeyvind Brandtsegg
 wrote:
> All good.
>
> Here's a simpler version:
>
>       ihopsize        = 1024
>       kcps, kamp      ptrack a1, ihopsize
>       imedianSize     = 129
>       kcps            mediank kcps, imedianSize, imedianSize
>
> The median filter is quite different from most audio filters, in that
> the output is the middle value in it's buffer. The buffer size is  set
> by imedianSize in the example above. You might want to experiment with
> different sizes to see what best fits with your data.
> For example for the input data
> [7,1,8,57,2,7,8,9,8]
> you can see that most of the activity is in the area around values 7,8 and 9,
> with spurious peaks (probably noise) at 1,2 and 57.
> If we sort this buffer of data in ascending order:
> [1,2,7,7,8,8,8,9,57]
> and take the middle member (number 5 from the left), we get 8 as the output.
>
> For the next k-rate period, the data in the buffer is updated (adding
> one member on the left side and discarding the one one the right)
>
> all best
> Oeyvind
>
> 2011/2/8 Enrico Francioni :
>>
>>
>> many thanks, Oeyvind!
>>
>> you can write a comment on your example?
>>
>> ... in fact, what is the opcode mediank?
>> how exactly is the performance of mediank?
>>
>> e
>> --
>> View this message in context: http://csound.1045644.n5.nabble.com/arithmetic-mean-tp3372682p3375500.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"
>>
>>
>
>
> 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 andrés -

can we add oeyvind's description at the page? i myself needed this
description to really understand what this very useful opcode does.

thanks -

	joachim



Am 08.02.2011 09:40, schrieb Andres Cabrera:
> Hi,
> 
> Here's the manual page:
> www.csounds.com/manual/html/mediank.html
> 
> Cheers,
> Andres
> 
> On Tue, Feb 8, 2011 at 7:58 AM, Oeyvind Brandtsegg
>  wrote:
>> All good.
>>
>> Here's a simpler version:
>>
>>       ihopsize        = 1024
>>       kcps, kamp      ptrack a1, ihopsize
>>       imedianSize     = 129
>>       kcps            mediank kcps, imedianSize, imedianSize
>>
>> The median filter is quite different from most audio filters, in that
>> the output is the middle value in it's buffer. The buffer size is  set
>> by imedianSize in the example above. You might want to experiment with
>> different sizes to see what best fits with your data.
>> For example for the input data
>> [7,1,8,57,2,7,8,9,8]
>> you can see that most of the activity is in the area around values 7,8 and 9,
>> with spurious peaks (probably noise) at 1,2 and 57.
>> If we sort this buffer of data in ascending order:
>> [1,2,7,7,8,8,8,9,57]
>> and take the middle member (number 5 from the left), we get 8 as the output.
>>
>> For the next k-rate period, the data in the buffer is updated (adding
>> one member on the left side and discarding the one one the right)
>>
>> all best
>> Oeyvind
>>
>> 2011/2/8 Enrico Francioni :
>>>
>>>
>>> many thanks, Oeyvind!
>>>
>>> you can write a comment on your example?
>>>
>>> ... in fact, what is the opcode mediank?
>>> how exactly is the performance of mediank?
>>>
>>> e
>>> --
>>> View this message in context: http://csound.1045644.n5.nabble.com/arithmetic-mean-tp3372682p3375500.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"
>>>
>>>
>>
>>
>> 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"

Would you contribute a UDO to the database with the other, switching  
version? I think it
should be very useful.

Victor
On 8 Feb 2011, at 07:58, Oeyvind Brandtsegg wrote:

> All good.
>
> Here's a simpler version:
>
>       ihopsize        = 1024
>       kcps, kamp      ptrack a1, ihopsize
>       imedianSize     = 129
>       kcps            mediank kcps, imedianSize, imedianSize
>
> The median filter is quite different from most audio filters, in that
> the output is the middle value in it's buffer. The buffer size is  set
> by imedianSize in the example above. You might want to experiment with
> different sizes to see what best fits with your data.
> For example for the input data
> [7,1,8,57,2,7,8,9,8]
> you can see that most of the activity is in the area around values  
> 7,8 and 9,
> with spurious peaks (probably noise) at 1,2 and 57.
> If we sort this buffer of data in ascending order:
> [1,2,7,7,8,8,8,9,57]
> and take the middle member (number 5 from the left), we get 8 as the  
> output.
>
> For the next k-rate period, the data in the buffer is updated (adding
> one member on the left side and discarding the one one the right)
>
> all best
> Oeyvind
>
> 2011/2/8 Enrico Francioni :
>>
>>
>> many thanks, Oeyvind!
>>
>> you can write a comment on your example?
>>
>> ... in fact, what is the opcode mediank?
>> how exactly is the performance of mediank?
>>
>> e
>> --
>> View this message in context: http://csound.1045644.n5.nabble.com/arithmetic-mean-tp3372682p3375500.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"
>>
>>
>
>
> 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"

oh, I'll try :-)... it's a bit hectic for the next few weeks
Oeyvind

2011/2/8 Victor Lazzarini :
> Would you contribute a UDO to the database with the other, switching
> version? I think it
> should be very useful.
>
> Victor
> On 8 Feb 2011, at 07:58, Oeyvind Brandtsegg wrote:
>
>> All good.
>>
>> Here's a simpler version:
>>
>>      ihopsize        = 1024
>>      kcps, kamp      ptrack a1, ihopsize
>>      imedianSize     = 129
>>      kcps            mediank kcps, imedianSize, imedianSize
>>
>> The median filter is quite different from most audio filters, in that
>> the output is the middle value in it's buffer. The buffer size is  set
>> by imedianSize in the example above. You might want to experiment with
>> different sizes to see what best fits with your data.
>> For example for the input data
>> [7,1,8,57,2,7,8,9,8]
>> you can see that most of the activity is in the area around values 7,8 and
>> 9,
>> with spurious peaks (probably noise) at 1,2 and 57.
>> If we sort this buffer of data in ascending order:
>> [1,2,7,7,8,8,8,9,57]
>> and take the middle member (number 5 from the left), we get 8 as the
>> output.
>>
>> For the next k-rate period, the data in the buffer is updated (adding
>> one member on the left side and discarding the one one the right)
>>
>> all best
>> Oeyvind
>>
>> 2011/2/8 Enrico Francioni :
>>>
>>>
>>> many thanks, Oeyvind!
>>>
>>> you can write a comment on your example?
>>>
>>> ... in fact, what is the opcode mediank?
>>> how exactly is the performance of mediank?
>>>
>>> e
>>> --
>>> View this message in context:
>>> http://csound.1045644.n5.nabble.com/arithmetic-mean-tp3372682p3375500.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"
>>>
>>>
>>
>>
>> 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"

…unfortunately mediank does not work with MacCsound 1.5 ... (with Csound
5.13 UB) !

[and other opcodes do not work ...]
?

e
-- 
View this message in context: http://csound.1045644.n5.nabble.com/arithmetic-mean-tp3372682p3376205.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"

Use qutecsound?
On 8 Feb 2011, at 16:55, Enrico Francioni wrote:

>
>
>
> …unfortunately mediank does not work with MacCsound 1.5 ... (with  
> Csound
> 5.13 UB) !
>
> [and other opcodes do not work ...]
> ?
>
> e
> -- 
> View this message in context: http://csound.1045644.n5.nabble.com/arithmetic-mean-tp3372682p3376205.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"
>



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"

I think it's time to do it ...
[sig]

e

…anyone out there has created a patch to mediank alternative?

thanks…

e
-- 
View this message in context: http://csound.1045644.n5.nabble.com/arithmetic-mean-tp3372682p3376485.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"

what do you need exactely? in which situation?

Am 08.02.2011 20:40, schrieb Enrico Francioni:
> 
> 
> 
> …anyone out there has created a patch to mediank alternative?
> 
> thanks…
> 
> e


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 joachim!


I would like to have this situation:

; Pitch and amp analysis 

        kcps init 0	
        ihopsize1	= 4096 
        ihopsize2	= 512 
        iChangeover	= (sr/ihopsize2)*3 

if kcps > iChangeover kgoto small 
        kcps, kamp ptrack a1, ihopsize1 
if kcps <= iChangeover kgoto nosmall 
small: 
        kcps, kamp ptrack a1, ihopsize2 
nosmall: 

; filter pitch tracking signal 
        imedianSize	= int(0.09*kr)	 ; calc as (1/(ihopsize*2))*kr 
        kcps	 mediank	kcps, imedianSize, imedianSize 


but with an alternative patch;
I can not make use of the opcode mediank
(for use with Csound 5.13 1.5 MacCsound UB)


I would like to calculate the arithmetic mean of kcps (in ptrack), 
but only when kcps it holds for a certain time around the same values.

For example, around 440, then around 220, then around 390 ..., etc ... 

e

hi francesco -

there is an UDO "Stayed" which returns 1 if a k-value has not changed
since a certain time. you can download it here:
http://www.csounds.com/udo/displayOpcode.php?opcode_id=130

perhaps it can help. but i think, because your kcps will certainly
fluctuate continuously, you must first create a variable which is 1 when
kcps is in a certain range. for instance:

kinrange = (kcps > 400 && kcps < 480 ? 1 : 0) ;kcps between 400 and 480?

then you can apply Stayed:

kyes Stayed kinrange, .5 ;kcps between 400 and 480 for at least 0.5 seconds?

does this help?

	joachim


Am 09.02.2011 17:21, schrieb Enrico Francioni:
> 
> Hi joachim!
> 
> 
> I would like to have this situation:
> 
> ; Pitch and amp analysis 
> 
>         kcps init 0	
>         ihopsize1	= 4096 
>         ihopsize2	= 512 
>         iChangeover	= (sr/ihopsize2)*3 
> 
> if kcps > iChangeover kgoto small 
>         kcps, kamp ptrack a1, ihopsize1 
> if kcps <= iChangeover kgoto nosmall 
> small: 
>         kcps, kamp ptrack a1, ihopsize2 
> nosmall: 
> 
> ; filter pitch tracking signal 
>         imedianSize	= int(0.09*kr)	 ; calc as (1/(ihopsize*2))*kr 
>         kcps	 mediank	kcps, imedianSize, imedianSize 
> 
> 
> but with an alternative patch;
> I can not make use of the opcode mediank
> (for use with Csound 5.13 1.5 MacCsound UB)
> 
> 
> I would like to calculate the arithmetic mean of kcps (in ptrack), 
> but only when kcps it holds for a certain time around the same values.
> 
> For example, around 440, then around 220, then around 390 ..., etc ... 
> 
> e


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"

thanks joachim

but with UDO Stayed I can only assess if the frequency is changed (1 or 0).
I am unable to make a mean value of kcps...

e

yes this is the next step. i think you can do this with a k-rate lowpass
filter, as one solution. like this:



instr 1
kmntm = .1 ;meantime in seconds
kcps random 400, 480
kmean tonek kcps, 1/kmntm
printk .1, kmean
endin


i 1 0 10



when you change the meantime from 0.1 to 1, you get a much smoother output.

one problem is that the tonek filter always starts from 0, so the filter
needs some time to get around the mean values. i don't know if there is
a good solution, or if anyone else knows a better way of calculating the
mean values (it could be done by writing the last x kcps values in a
table and calculating the mean then, but this is slightly more work).

hope this helps -

	joachim

Am 11.02.2011 10:14, schrieb Enrico Francioni:
> 
> 
> thanks joachim
> 
> but with UDO Stayed I can only assess if the frequency is changed (1 or 0).
> I am unable to make a mean value of kcps...
> 
> e


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"

…to solve the problem,
a friend advised me to apply this patch.
I added some modifications.
What do you think?

;===============================

instr 1

asig inch 1
iSize	pow	2, 10
gkcps,kA 	ptrack 	aSig, iSize 

kindex	init	-1
kfreq		init	0	; ...first pitch-detector

kindex = kindex + 1
kfreq = kfreq + gkcps

;printk gitime, gkfreq, 80

if kindex == 32 then
kindex = 0
gkfreq = kfreq / 32
kfreq	  = 0
endif

printk .001, gkcps
printk .001, gkfreq, 10

endin

;===============================


e
-- 
View this message in context: http://csound.1045644.n5.nabble.com/arithmetic-mean-tp3372682p3381605.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"

hmm, to me it looks like you would be better off using a lowpass
filter like tonek or portk.
If I'm not mistaken, your solution will give you the average of 32
k-rate values, and it's not a "moving average" (not updated on every
k-rate pass), but it's rather a "windowed average" (someone please
correct me if I'm using the wrong term) in that it gives the average
of the first 32 values, then it needs to fill up 32 new values before
outputting the average of those (value 33 to 64) and then progresses
in chunks of 32.

best
Oeyvind

2011/2/11 Enrico Francioni :
>
>
> …to solve the problem,
> a friend advised me to apply this patch.
> I added some modifications.
> What do you think?
>
> ;===============================
>
> instr 1
>
> asig inch 1
> iSize   pow     2, 10
> gkcps,kA        ptrack  aSig, iSize
>
> kindex  init    -1
> kfreq           init    0       ; ...first pitch-detector
>
> kindex = kindex + 1
> kfreq = kfreq + gkcps
>
> ;printk gitime, gkfreq, 80
>
> if kindex == 32 then
> kindex = 0
> gkfreq = kfreq / 32
> kfreq     = 0
> endif
>
> printk .001, gkcps
> printk .001, gkfreq, 10
>
> endin
>
> ;===============================
>
>
> e
> --
> View this message in context: http://csound.1045644.n5.nabble.com/arithmetic-mean-tp3372682p3381605.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"
>
>


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"

Yes, Oeyvind

but it is also true that the buffer is a few milliseconds...

The road opened by joachim does not seem to be discarded ...
In Csound can take different paths with each other to achieve the same
results, no?

e

p.s.
I also go into the patch joachim

Yes, indeed, you can achieve your goal in many different ways. And it
seems you search and experiment very fruitfully, there is no right or
wrong, so keep it up.
all best
Oeyvind

2011/2/11 Enrico Francioni :
>
>
>
> Yes, Oeyvind
>
> but it is also true that the buffer is a few milliseconds...
>
> The road opened by joachim does not seem to be discarded ...
> In Csound can take different paths with each other to achieve the same
> results, no?
>
> e
>
> p.s.
> I also go into the patch joachim
> --
> View this message in context: http://csound.1045644.n5.nabble.com/arithmetic-mean-tp3372682p3381977.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"
>
>


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"