Hi, to keep my patch tidy I'd like to use some sort of [receive] in a bunch of [expr] objects (mostly constants to configure the patch). I could use sub-patches and [send] the values to them there, is there an easier way? [expr] don't have names like arrays etc. - but maybe there's a trick?
Thanks, Michael.
-- nonconform? noiseconform: http://blauwurf.at http://soundcloud.com/noiseconformist
2011/8/25 Michael Zacherl. sdiy-mz01@blauwurf.info
Hi, to keep my patch tidy I'd like to use some sort of [receive] in a bunch of [expr] objects (mostly constants to configure the patch). I could use sub-patches and [send] the values to them there, is there an easier way? [expr] don't have names like arrays etc. - but maybe there's a trick?
something like [expr myvalue] and [v myvalue]?
gr, Tim
Thanks, Michael.
-- nonconform? noiseconform: http://blauwurf.at http://soundcloud.com/noiseconformist
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
On Thu, Aug 25, 2011 at 2:43 PM, tim vets timvets@gmail.com wrote:
something like [expr myvalue] and [v myvalue]? gr, Tim
I didn't know that one. Thanks!
[hsl] | [/ 127] | [v headphonesafe]
[noise~]
|
[*~ 999999]
|
[clip~ -1 1]
|
[expr~ $v1*headphonesafe]
|
|
[dac~]
--Funs
I'm liking the look of this to streamline a few patches. Only trouble is there doesn't seem to be an audio rate version. So Funs' patch will give you zipper noise. [value] doesn't seem to have an audio alternative. Which is fine as it's like a combination of [s] and [f], but it does mean you can't receive audio values in [expr~]... as far as I can tell. Andrew
Date: Thu, 25 Aug 2011 15:21:56 +0200 From: funsseelen@gmail.com To: timvets@gmail.com CC: pd-list@iem.at Subject: Re: [PD] receiving messages in [expr] ?
On Thu, Aug 25, 2011 at 2:43 PM, tim vets timvets@gmail.com wrote:
something like [expr myvalue] and [v myvalue]?gr,Tim I didn't know that one. Thanks!
[hsl]
| [/ 127] | [v headphonesafe]
[noise~]
|
[*~ 999999]
|
[clip~ -1 1]
|
[expr~ $v1*headphonesafe]
|
|
[dac~]
--Funs
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On 2011-08-25 17:32, Andrew Faraday wrote:
I'm liking the look of this to streamline a few patches. Only trouble is there doesn't seem to be an audio rate version. So Funs' patch will give you zipper noise. [value] doesn't seem to have an audio alternative. Which is fine as it's like a combination of [s] and [f], but it does mean you can't receive audio values in [expr~]... as far as I can tell.
tables?
pedagmode:ON generally i would suggest to avoid such hidden data sharing whenever possible. while it might remove a number of ugly connections, it will also remove a number of connections telling you where the data comes from.
connections are really Pd's strength, rather than it's weakness. pedagmode:OFF
fgmasdr IOhannes
On 25.8.2011, at 17:43 , IOhannes m zmoelnig wrote:
generally i would suggest to avoid such hidden data sharing whenever possible. while it might remove a number of ugly connections, it will also remove a number of connections telling you where the data comes from.
connections are really Pd's strength, rather than it's weakness.
I agree! I'm working on an abstraction so utilising [value] and trying to incorporate $0 might be clean but then again probably hard to read. Then again, curiosity if it might work is still there! ;-)
Michael.
-- noise chasers: http://blauwurf.at http://soundcloud.com/noiseconformist
On Thu, Aug 25, 2011 at 5:32 PM, Andrew Faraday jbturgid@hotmail.comwrote:
I'm liking the look of this to streamline a few patches. Only trouble is there doesn't seem to be an audio rate version. So Funs' patch will give you zipper noise. [value] doesn't seem to have an audio alternative. Which is fine as it's like a combination of [s] and [f], but it does mean you can't receive audio values in [expr~]... as far as I can tell.
Andrew
O, my patch was just a wink to the `headphone' topic. It seems to me that connecting [catch~ mysignal] to $v2 in [expr~] is the proper way, although I cannot distinguish zippernoise from clipped [noise~] in the example. Isn't an audio version of [value] an impossibility? How would you store a signal? --Funs
Nice obvious one there, Funs, you can't hear noise against noise :p
Date: Thu, 25 Aug 2011 18:17:16 +0200 Subject: Re: [PD] receiving messages in [expr] ? From: funsseelen@gmail.com To: jbturgid@hotmail.com; Pd-list@iem.at
On Thu, Aug 25, 2011 at 5:32 PM, Andrew Faraday jbturgid@hotmail.com wrote:
I'm liking the look of this to streamline a few patches. Only trouble is there doesn't seem to be an audio rate version. So Funs' patch will give you zipper noise. [value] doesn't seem to have an audio alternative. Which is fine as it's like a combination of [s] and [f], but it does mean you can't receive audio values in [expr~]... as far as I can tell.
Andrew
O, my patch was just a wink to the `headphone' topic. It seems to me that connecting [catch~ mysignal] to $v2 in [expr~] is the proper way, although I cannot distinguish zippernoise from clipped [noise~] in the example. Isn't an audio version of [value] an impossibility? How would you store a signal?
--Funs
Am 25.08.2011 um 19:40 schrieb Andrew Faraday:
you can't hear noise against noise :p
Interesting - but then again, what is "noise"? gr~~~
-- Thomas Grill http://grrrr.org
On Thu, Aug 25, 2011 at 11:20:47PM +0200, Thomas Grill wrote:
Am 25.08.2011 um 19:40 schrieb Andrew Faraday:
you can't hear noise against noise :p
Interesting - but then again, what is "noise"?
http://en.wikipedia.org/wiki/Wave_function_collapse
Chris.
Am 26.08.2011 um 04:06 schrieb Chris McCormick:
On Thu, Aug 25, 2011 at 11:20:47PM +0200, Thomas Grill wrote:
Am 25.08.2011 um 19:40 schrieb Andrew Faraday:
you can't hear noise against noise :p
Interesting - but then again, what is "noise"?
Hmmm, i guess your point with the reference is that when perceiving
one takes snapshots (measurements) of a statistical function (the wave
function).
However, there are widespread methods to tell one statistical
distribution ("noise") from another one ("noise").
See e.g. expectation maximization (EM) in (Gaussian) mixture models
(GMM), a popular technique in Music Information Retrieval.
http://en.wikipedia.org/wiki/Mixture_model#Expectation_maximization_.28EM.29
Why shouldn't human listeners be able to maximize their expectations
as well?
gr~~~
-- Thomas Grill http://grrrr.org +43 699 19715543
On Fri, 26 Aug 2011, Chris McCormick wrote:
On Thu, Aug 25, 2011 at 11:20:47PM +0200, Thomas Grill wrote:
Am 25.08.2011 um 19:40 schrieb Andrew Faraday:
you can't hear noise against noise :p
Interesting - but then again, what is "noise"?
Ears don't know what a wave function collapse is, and wouldn't differentiate quantum noise from a linear-congruential scrambler such as pd's [noise~].
You'll have to come up with a more phenomenological approach than picking whatever currently looks the most impredictable thing in the science book.
| Mathieu Bouchard ---- tél: +1.514.383.3801 ---- Villeray, Montréal, QC
Am 27.08.2011 um 21:59 schrieb Mathieu Bouchard:
Ears don't know what a wave function collapse is, and wouldn't
differentiate quantum noise from a linear-congruential scrambler
such as pd's [noise~].
i don't think that quantum noise is necessarily "white" in the audible
domain.
gr~~~
-- Thomas Grill http://grrrr.org
On Mon, 29 Aug 2011, Thomas Grill wrote:
Am 27.08.2011 um 21:59 schrieb Mathieu Bouchard:
Ears don't know what a wave function collapse is, and wouldn't differentiate quantum noise from a linear-congruential scrambler such as pd's [noise~].
i don't think that quantum noise is necessarily "white" in the audible domain.
Doesn't it depend on how it's recorded ?
Using a nonquantum example, if you record speaker membrane position as a function of time, store it in a table and then play it as if it had been recorded electrically, wouldn't it sound like it needs a big EQ ?
But then, if something quantum is gaussian, then its spectrum is gaussian too, and that needs quite trickier EQing in order to sound white.
In any case, once you EQ whichever noise source to sound like the other one, no-one can figure out which one is quantum just by ear. Thus, musically, it does not matter.
| Mathieu Bouchard ---- tél: +1.514.383.3801 ---- Villeray, Montréal, QC
On Sat, Aug 27, 2011 at 03:59:14PM -0400, Mathieu Bouchard wrote:
You'll have to come up with a more phenomenological approach than picking whatever currently looks the most impredictable thing in the science book.
Why? Randomness is epistemic.
Cheers,
Chris.
On Fri, 2 Sep 2011, Chris McCormick wrote:
On Sat, Aug 27, 2011 at 03:59:14PM -0400, Mathieu Bouchard wrote:
You'll have to come up with a more phenomenological approach than picking whatever currently looks the most impredictable thing in the science book.
Why? Randomness is epistemic.
What does that mean to you ?
| Mathieu Bouchard ---- tél: +1.514.383.3801 ---- Villeray, Montréal, QC
On Thu, Sep 01, 2011 at 07:38:32PM -0400, Mathieu Bouchard wrote:
On Fri, 2 Sep 2011, Chris McCormick wrote:
On Sat, Aug 27, 2011 at 03:59:14PM -0400, Mathieu Bouchard wrote:
You'll have to come up with a more phenomenological approach than picking whatever currently looks the most impredictable thing in the science book.
Why? Randomness is epistemic.
What does that mean to you ?
Epistemic: Adjective - "Of or relating to knowledge or to the degree of its validation."
It means I think that taking whatever currently looks the most impredictable thing in the science book is a good approach.
Don't run a lottery with Miller's [noise~] unless you like giving away money.
Cheers,
Chris.
On Fri, 2 Sep 2011, Chris McCormick wrote:
Epistemic: Adjective - "Of or relating to knowledge or to the degree of its validation." It means I think that taking whatever currently looks the most impredictable thing in the science book is a good approach. Don't run a lottery with Miller's [noise~] unless you like giving away money.
But in music, what is called noise has little to do with predictability of its samples, be it a snare drum sound, a white noise, or something from reality tv.
| Mathieu Bouchard ---- tél: +1.514.383.3801 ---- Villeray, Montréal, QC
On Thu, Sep 01, 2011 at 07:48:26PM -0400, Mathieu Bouchard wrote:
On Fri, 2 Sep 2011, Chris McCormick wrote:
Epistemic: Adjective - "Of or relating to knowledge or to the degree of its validation." It means I think that taking whatever currently looks
the most impredictable thing in the science book is a good approach.
Don't run a lottery with Miller's [noise~] unless you like giving away
money.But in music, what is called noise has little to do with predictability of its samples, be it a snare drum sound, a white noise, or something from reality tv.
Yes that's true.
Chris.
On 25.8.2011, at 14:43 , tim vets wrote:
something like [expr myvalue] and [v myvalue]? gr, Tim
great, thanks Tim!
any chance to get $0 working? [v $0-myvalue] is fine, [bang(--[expr $0-myvalue] (might be utterly wrong) delivers strange values, not even $0.
Michael.
-- noise chasers: http://blauwurf.at http://soundcloud.com/noiseconformist
[v $0-myvalue] = associate the symbol "$0-myvalue" with a float value [expr $0-myvalue] = subtract the float value associated with the symbol "myvalue" from $0
Therefore, [expr] should always return the value of $0 (unless you're assigning a value to "myvalue" somewhere)
[expr]'s parser seems only to recognize an atom as a symbol when the first character is not a number, so try the following workarounds:
[expr _$0myvalue] [expr myvalue$0]
Both are ugly. If using tables you can avoid this mess by using an $s variable.
-Jonathan
From: Michael Zacherl. sdiy-mz01@blauwurf.info To: PD list pd-list@iem.at Sent: Thursday, August 25, 2011 11:19 AM Subject: Re: [PD] receiving messages in [expr] ?
On 25.8.2011, at 14:43 , tim vets wrote:
something like [expr myvalue] and [v myvalue]? gr, Tim
great, thanks Tim!
any chance to get $0 working? [v $0-myvalue] is fine, [bang(--[expr $0-myvalue] (might be utterly wrong) delivers strange values, not even $0.
Michael.
-- noise chasers: http://blauwurf.at http://soundcloud.com/noiseconformist
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
Hi Jonathan, yeah, tables would work, I easily could keep them local any refer to the respective indices. But then again, despite what IOhannes pointed out (and I still agree with him) I like the idea of using [value]s since I could use self explaining names, which would make the mess, in this particular case, better readable.
So from this POV I don't find your solution that ugly.
Thanks, Michael.
On 25.8.2011, at 18:21 , Jonathan Wilkes wrote:
[v $0-myvalue] = associate the symbol "$0-myvalue" with a float value [expr $0-myvalue] = subtract the float value associated with the symbol "myvalue" from $0
Therefore, [expr] should always return the value of $0 (unless you're assigning a value to "myvalue" somewhere)
[expr]'s parser seems only to recognize an atom as a symbol when the first character is not a number, so try the following workarounds:
[expr _$0myvalue] [expr myvalue$0]
Both are ugly. If using tables you can avoid this mess by using an $s variable.
-Jonathan
From: Michael Zacherl. sdiy-mz01@blauwurf.info To: PD list pd-list@iem.at Sent: Thursday, August 25, 2011 11:19 AM Subject: Re: [PD] receiving messages in [expr] ?
On 25.8.2011, at 14:43 , tim vets wrote:
something like [expr myvalue] and [v myvalue]? gr, Tim
great, thanks Tim!
any chance to get $0 working? [v $0-myvalue] is fine, [bang(--[expr $0-myvalue] (might be utterly wrong) delivers strange values, not even $0.
Michael.
-- noise chasers: http://blauwurf.at http://soundcloud.com/noiseconformist
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
-- keep your ears open: http://blauwurf.at http://soundcloud.com/noiseconformist
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On 2011-08-31 16:11, Michael Zacherl. wrote:
despite what IOhannes pointed out (and I still agree with him)
i do not doubt that there are many use cases where it is indeed a better idea to avoid connections in favour of an accessing memory by name.
i only wanted to point out that it many cases it is not.
ghmsd IOhannes
-
Andrew Faraday -
Chris McCormick -
Funs Seelen -
IOhannes m zmoelnig -
Jonathan Wilkes -
Mathieu Bouchard -
Michael Zacherl. -
Thomas Grill -
tim vets