Hello everyone,
I was wondering if anybody on this list would have an idea of how one can simulate sympathetic strings in pd, using any kind of audio input (especially NOT midi). I tried making a patch with 12 resonating filters (moog~), each tuned to a particular frequency belonging to the chromatic scale, but the result wasn't too good (it sounded more like a weird reverb). My problem is basically that of isolating certain frequencies in the input sound. I don't believe that objects like fiddle~ or sigmund~ would be very helpful in that matter, as they always start freaking out once you feed them with chords. I haven't found anything useful on the internet yet. Any idea?
Pierre
Hi Pierre,
I've found resonant comb filters to be the most useful in both simulating string sounds, and in getting sympathetic resonances. There's tons of stuff in the list archives on them, in particular a patch (maybe I even posted it) on calculating the correct delay length for produce a given pitch. Search for "comb filter".
Best Derek
Pierre Massat wrote:
Hello everyone,
I was wondering if anybody on this list would have an idea of how one can simulate sympathetic strings in pd, using any kind of audio input (especially NOT midi). I tried making a patch with 12 resonating filters (moog~), each tuned to a particular frequency belonging to the chromatic scale, but the result wasn't too good (it sounded more like a weird reverb). My problem is basically that of isolating certain frequencies in the input sound. I don't believe that objects like fiddle~ or sigmund~ would be very helpful in that matter, as they always start freaking out once you feed them with chords. I haven't found anything useful on the internet yet. Any idea?
Pierre
Hello, I'm not sure if I wrote that correctly-- I'd like to use an abstraction called t.pd in place of trigger. So I put the abstraction in my patch's folder but it won't alias [t] because I guess Pd only searches for abstractions after it searches for internals.
Basically I want to expirement with using a max-style [t], one in which you can use actual float values as args. But more importantly, if I wanted to use such an abstraction as an aid in converting a max patch to a pd patch, how could I force Pd to use the abstraction in place of the internal?
Thanks, Jonathan
This might be too much of an obvious answer, but try calling it something else? tt.pd?
Date: Thu, 11 Feb 2010 16:28:43 -0800 From: jancsika@yahoo.com To: pd-list@iem.at Subject: [PD] Alias an internal with an abstraction
Hello, I'm not sure if I wrote that correctly-- I'd like to use an abstraction called t.pd in place of trigger. So I put the abstraction in my patch's folder but it won't alias [t] because I guess Pd only searches for abstractions after it searches for internals.
Basically I want to expirement with using a max-style [t], one in which you can use actual float values as args. But more importantly, if I wanted to use such an abstraction as an aid in converting a max patch to a pd patch, how could I force Pd to use the abstraction in place of the internal?
Thanks, Jonathan
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
Got a cool Hotmail story? Tell us now http://clk.atdmt.com/UKM/go/195013117/direct/01/
Hi Andrew,
Thanks, but that doesn't help so much with with the max to pd conversion. If I made it an external, I would be able to use it to replace the internal, right?
-Jonathan
--- On Fri, 2/12/10, Andrew Faraday jbturgid@hotmail.com wrote:
From: Andrew Faraday jbturgid@hotmail.com Subject: RE: [PD] Alias an internal with an abstraction To: jancsika@yahoo.com, pd-list@iem.at Date: Friday, February 12, 2010, 1:41 AM
This might be too much of an obvious answer, but try calling it something else? tt.pd?
Date: Thu, 11 Feb 2010 16:28:43 -0800 From: jancsika@yahoo.com To: pd-list@iem.at Subject: [PD] Alias an internal with an abstraction
Hello, I'm not sure if I wrote that correctly--
I'd like to use an abstraction
called t.pd in place of trigger. So I put the
abstraction in my patch's
folder but it won't alias [t] because I guess Pd
only searches for
abstractions after it searches for internals.
Basically I want to expirement with using a max-style
[t], one in
which you can use actual float values as args. But
more importantly, if I
wanted to use such an abstraction as an aid in
converting a max patch to a
pd patch, how could I force Pd to use the abstraction
in place of the
internal?
Thanks, Jonathan
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management ->
http://lists.puredata.info/listinfo/pd-list
Got a cool Hotmail story? Tell us now
On Thu, 2010-02-11 at 18:39 -0800, Jonathan Wilkes wrote:
Hi Andrew,
Thanks, but that doesn't help so much with with the max to pd conversion. If I made it an external, I would be able to use it to replace the internal, right?
I _think_ that this is not possible with Pd <= current versions. And if it would be possible to override internals, I'd strongly fight this ability for consistency reasons. It's already hard enough nowadays to create a patch working on a fair amount of systems with different OS' and different Pd flavours. The subset of internal object classes should behave reliably the same on _every_ Pd installation.
BTW: there is a library dedicated to making Pd behaving more like max/msp called cyclone. Just in case you didn't know it already (though from what I know it doesn't address the difference of [t]).
Roman
___________________________________________________________ Der frühe Vogel fängt den Wurm. Hier gelangen Sie zum neuen Yahoo! Mail: http://mail.yahoo.de
On Fri, 2010-02-12 at 09:40 +0100, Roman Haefeli wrote:
On Thu, 2010-02-11 at 18:39 -0800, Jonathan Wilkes wrote:
Hi Andrew,
Thanks, but that doesn't help so much with with the max to pd conversion. If I made it an external, I would be able to use it to replace the internal, right?
I _think_ that this is not possible with Pd <= current versions.
To correct myself: Since 0.42 or so there is way for externals to override internals. Personally I find it was wrong to implement this 'feature' without providing a mechanism for a patch to protect it from using externals that override internals. This really can make debugging hard and writing portable patches impossible.
IIRC, there is even a documented case on the list, where someone posted a patch, that was working only on some computers because a slight difference between Pd's [wrap] and zexy's [wrap] and only on *some* machines zexy was installed and did override Pd's internal [wrap].
Roman
Telefonate ohne weitere Kosten vom PC zum PC: http://messenger.yahoo.de
--- On Fri, 2/12/10, Roman Haefeli reduzierer@yahoo.de wrote:
From: Roman Haefeli reduzierer@yahoo.de Subject: Re: [PD] Alias an internal with an abstraction To: "Jonathan Wilkes" jancsika@yahoo.com Cc: pd-list@iem.at Date: Friday, February 12, 2010, 10:21 AM On Fri, 2010-02-12 at 09:40 +0100, Roman Haefeli wrote:
On Thu, 2010-02-11 at 18:39 -0800, Jonathan Wilkes
wrote:
Hi Andrew,
Thanks, but that doesn't help so much with with
the max to pd conversion. If I made it an external, I would be able to use it to
replace the internal, right?
I _think_ that this is not possible with Pd <=
current versions.
To correct myself: Since 0.42 or so there is way for externals to override internals. Personally I find it was wrong to implement this 'feature' without providing a mechanism for a patch to protect it from using externals that override internals. This really can make debugging hard and writing portable patches impossible.
IIRC, there is even a documented case on the list, where someone posted a patch, that was working only on some computers because a slight difference between Pd's [wrap] and zexy's [wrap] and only on *some* machines zexy was installed and did override Pd's internal [wrap].
Isn't this is a problem with the way libraries have loaded by default in Pd-ext (I think there's been a long thread about that but I haven't followed it)? I don't know much about zexy's [wrap], but this seems a documentation issue rather than a bug-- whether it's at load time or in the docs, at some point zexy needs to say, "Greetings, gentleman. All your [wrap] are belong to us." Or maybe Pd needs to say "All my [wrap] are belong to zexy."
If I have a bunch of objects in folder "maxwise-abs," and I tell Pd to add that folder to the path, it is extremely likely (99%) that I want the abstraction to take precedence over the internal. (Granted I currently only have one object in that folder but I just started...)
-Jonathan
Roman
___________________________________________________________
Telefonate ohne weitere Kosten vom PC zum PC: http://messenger.yahoo.de
Jonathan Wilkes a écrit :
--- On Fri, 2/12/10, Roman Haefeli reduzierer@yahoo.de wrote:
From: Roman Haefeli reduzierer@yahoo.de Subject: Re: [PD] Alias an internal with an abstraction To: "Jonathan Wilkes" jancsika@yahoo.com Cc: pd-list@iem.at Date: Friday, February 12, 2010, 10:21 AM On Fri, 2010-02-12 at 09:40 +0100, Roman Haefeli wrote:
On Thu, 2010-02-11 at 18:39 -0800, Jonathan Wilkes
wrote:
Hi Andrew,
Thanks, but that doesn't help so much with with
the max to pd conversion. If I made it an external, I would be able to use it to
replace the internal, right?
I _think_ that this is not possible with Pd <=
current versions.
To correct myself: Since 0.42 or so there is way for externals to override internals. Personally I find it was wrong to implement this 'feature' without providing a mechanism for a patch to protect it from using externals that override internals. This really can make debugging hard and writing portable patches impossible.
IIRC, there is even a documented case on the list, where someone posted a patch, that was working only on some computers because a slight difference between Pd's [wrap] and zexy's [wrap] and only on *some* machines zexy was installed and did override Pd's internal [wrap].
Isn't this is a problem with the way libraries have loaded by default in Pd-ext (I think there's been a long thread about that but I haven't followed it)? I don't know much about zexy's [wrap], but this seems a documentation issue rather than a bug-- whether it's at load time or in the docs, at some point zexy needs to say, "Greetings, gentleman. All your [wrap] are belong to us." Or maybe Pd needs to say "All my [wrap] are belong to zexy."
well, that is curently almost what happend. pd say somthing like :
warning: class 'wrap' overwritten; old one renamed 'wrap_aliased'
but i suspect that very few peoples care about what pd say at startup....
Cyrille
If I have a bunch of objects in folder "maxwise-abs," and I tell Pd to add that folder to the path, it is extremely likely (99%) that I want the abstraction to take precedence over the internal. (Granted I currently only have one object in that folder but I just started...)
-Jonathan
Roman
Telefonate ohne weitere Kosten vom PC zum PC: http://messenger.yahoo.de
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
On Feb 12, 2010, at 4:21 AM, Roman Haefeli wrote:
On Fri, 2010-02-12 at 09:40 +0100, Roman Haefeli wrote:
On Thu, 2010-02-11 at 18:39 -0800, Jonathan Wilkes wrote:
Hi Andrew,
Thanks, but that doesn't help so much with with the max to pd
conversion. If I made it an external, I would be able to use it to replace the internal, right?I _think_ that this is not possible with Pd <= current versions.
To correct myself: Since 0.42 or so there is way for externals to override internals. Personally I find it was wrong to implement this 'feature' without providing a mechanism for a patch to protect it from using externals that override internals. This really can make debugging hard and
writing portable patches impossible.IIRC, there is even a documented case on the list, where someone
posted a patch, that was working only on some computers because a slight difference between Pd's [wrap] and zexy's [wrap] and only on *some* machines zexy was installed and did override Pd's internal [wrap].
FYI: I agree with Roman. I already removed the 0.42 aliasing feature
from Pd-extended 0.42.5
.hc
I spent 33 years and four months in active military service and during
that period I spent most of my time as a high class muscle man for Big
Business, for Wall Street and the bankers. - General Smedley Butler
On 2010-02-13 05:54, Hans-Christoph Steiner wrote:
FYI: I agree with Roman. I already removed the 0.42 aliasing feature from Pd-extended 0.42.5
FYI: i agree with Roman, but i don't think it's a good idea to divert the behaviours of PdX and Pd in such cases: it will only make transition even nastier than it is.
fgmasdr IOhannes
On Feb 22, 2010, at 4:50 AM, IOhannes m zmoelnig wrote:
On 2010-02-13 05:54, Hans-Christoph Steiner wrote:
FYI: I agree with Roman. I already removed the 0.42 aliasing feature from Pd-extended 0.42.5
FYI: i agree with Roman, but i don't think it's a good idea to divert the behaviours of PdX and Pd in such cases: it will only make
transition even nastier than it is.
I see no way to support that 0.42 aliasing feature without mountains
of work to achieve poor results. Of course, anyone should feel free
to prove me wrong.
.hc
Computer science is no more related to the computer than astronomy is
related to the telescope. -Edsger Dykstra
On Wed, 2010-02-24 at 16:55 -0500, Hans-Christoph Steiner wrote:
On Feb 22, 2010, at 4:50 AM, IOhannes m zmoelnig wrote:
On 2010-02-13 05:54, Hans-Christoph Steiner wrote:
FYI: I agree with Roman. I already removed the 0.42 aliasing feature from Pd-extended 0.42.5
FYI: i agree with Roman, but i don't think it's a good idea to divert the behaviours of PdX and Pd in such cases: it will only make
transition even nastier than it is.I see no way to support that 0.42 aliasing feature without mountains
of work to achieve poor results. Of course, anyone should feel free
to prove me wrong.
Could that problem be solved by removing the aliasing feature in the next release of Pd-vanilla, so that both Pd-extended and Pd-vanilla would be in sync again? Is it because I am biased or are there really hardly any voices advocating that feature?
Roman
you're doing the conversion anyway with patches saves in text format,
right? you could do a search-replace to find for " t " and replace it with
" tt ". just make sure you don't replace more than you want.
João
Hi Andrew,
Thanks, but that doesn't help so much with with the max to pd
conversion. If I made it an external, I would be able to use it to replace the internal, right?-Jonathan
--- On Fri, 2/12/10, Andrew Faraday jbturgid@hotmail.com wrote:
From: Andrew Faraday jbturgid@hotmail.com Subject: RE: [PD] Alias an internal with an abstraction To: jancsika@yahoo.com, pd-list@iem.at Date: Friday, February 12, 2010, 1:41 AM
This might be too much of an obvious answer, but try calling it something else? tt.pd?
Date: Thu, 11 Feb 2010 16:28:43 -0800 From: jancsika@yahoo.com To: pd-list@iem.at Subject: [PD] Alias an internal with an abstraction
Hello, I'm not sure if I wrote that correctly--
I'd like to use an abstraction
called t.pd in place of trigger. So I put the
abstraction in my patch's
folder but it won't alias [t] because I guess Pd
only searches for
abstractions after it searches for internals.
Basically I want to expirement with using a max-style
[t], one in
which you can use actual float values as args. But
more importantly, if I
wanted to use such an abstraction as an aid in
converting a max patch to a
pd patch, how could I force Pd to use the abstraction
in place of the
internal?
Thanks, Jonathan
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management ->
http://lists.puredata.info/listinfo/pd-list Got a cool Hotmail story? Tell us now
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management ->
http://lists.puredata.info/listinfo/pd-list
--- On Fri, 2/12/10, João Pais jmmmpais@googlemail.com wrote:
From: João Pais jmmmpais@googlemail.com Subject: Re: [PD] Alias an internal with an abstraction To: "Jonathan Wilkes" jancsika@yahoo.com, pd-list@iem.at, "Andrew Faraday" jbturgid@hotmail.com Date: Friday, February 12, 2010, 1:35 PM you're doing the conversion anyway with patches saves in text format, right? you could do a search-replace to find for " t " and replace it with " tt ". just make sure you don't replace more than you want.
Thanks, that's probably the best workaround-- then once it's working, change all the [tt]'s back to [t]'s and add msg boxes (if I want it to work without my abstraction).
-Jonathan
From: João Pais jmmmpais@googlemail.com Subject: Re: [PD] Alias an internal with an abstraction To: "Jonathan Wilkes" jancsika@yahoo.com, pd-list@iem.at, "Andrew
Faraday" jbturgid@hotmail.com Date: Friday, February 12, 2010, 1:35 PM you're doing the conversion anyway with patches saves in text format,right? you could do a search-replace
to find for " t " and replace it with" tt ". just make sure you don't replace more than you want.Thanks, that's probably the best workaround-- then once it's working, change all the [tt]'s back to [t]'s and add msg boxes (if I want it to work without my abstraction).
exactly. sometimes I prefer to edit pd files as text when there's too many
parameters to change - like changing the send/receive names of 20 gui
objects.
Dear Hans, I'm using [folder_list] on windows, and on 0.43.0-devel-20100126 when I create it, it prints the default path to the console, which is the path of the patch I'm working with (which is exactly what I want).
However, I'd like to use a relative path (../something/*) to access a folder from the parent, but when I try to use that relative path, it turns out that it's not relative to the working directory-- instead, it's relative to the path to Pd.
Any advice on getting [folder_list] to access that folder from the parent?
I'd rather not use an absolute path (especially since there are so many
spaces in windows paths).
Thanks, Jonathan
On Fri, Feb 12, 2010 at 8:38 PM, Jonathan Wilkes jancsika@yahoo.com wrote:
Dear Hans, I'm using [folder_list] on windows, and on 0.43.0-devel-20100126 when I create it, it prints the default path to the console, which is the path of the patch I'm working with (which is exactly what I want).
However, I'd like to use a relative path (../something/*) to access a folder from the parent, but when I try to use that relative path, it turns out that it's not relative to the working directory-- instead, it's relative to the path to Pd.
Any advice on getting [folder_list] to access that folder from the parent? I'd rather not use an absolute path (especially since there are so many spaces in windows paths).
Thanks, Jonathan
Hello Jonathan,
I'm not sure if it is the solution to your problem - i use [relativepath] and [absolutepath] by Antoine Rousseau. Andras
Thank you Derek! I'll give it a look. Best,
Pierre
2010/2/11 Derek Holzer derek@umatic.nl
Hi Pierre,
I've found resonant comb filters to be the most useful in both simulating string sounds, and in getting sympathetic resonances. There's tons of stuff in the list archives on them, in particular a patch (maybe I even posted it) on calculating the correct delay length for produce a given pitch. Search for "comb filter".
Best Derek
Pierre Massat wrote:
Hello everyone,
I was wondering if anybody on this list would have an idea of how one can simulate sympathetic strings in pd, using any kind of audio input (especially NOT midi). I tried making a patch with 12 resonating filters (moog~), each tuned to a particular frequency belonging to the chromatic scale, but the result wasn't too good (it sounded more like a weird reverb). My problem is basically that of isolating certain frequencies in the input sound. I don't believe that objects like fiddle~ or sigmund~ would be very helpful in that matter, as they always start freaking out once you feed them with chords. I haven't found anything useful on the internet yet. Any idea?
Pierre
-- ::: derek holzer ::: http://macumbista.net ::: ---Oblique Strategy # 130: "Question the heroic"
Hi, I have tried your patch called karplus-strong. This is just amazing! This solved two problems at once : getting a sitar-like sound plus a sympathetic strings simulation. Thank a lot!
Although i think i should open a new thread about this, i was a little shocked by the time fiddle~ takes to output a pitch (i used it to control a resonator with my guitar). I think it was way above 10 ms. I haven't tried sigmund~ yet, though. Do you know what latency i should expect? I think 5 ms should do. Anyway, thanks again for introducing me to the Karplus-Strong model!
Pierre
2010/2/12 Pierre Massat pimassat@gmail.com
Thank you Derek! I'll give it a look. Best,
Pierre
2010/2/11 Derek Holzer derek@umatic.nl
Hi Pierre,
I've found resonant comb filters to be the most useful in both simulating string sounds, and in getting sympathetic resonances. There's tons of stuff in the list archives on them, in particular a patch (maybe I even posted it) on calculating the correct delay length for produce a given pitch. Search for "comb filter".
Best Derek
Pierre Massat wrote:
Hello everyone,
I was wondering if anybody on this list would have an idea of how one can simulate sympathetic strings in pd, using any kind of audio input (especially NOT midi). I tried making a patch with 12 resonating filters (moog~), each tuned to a particular frequency belonging to the chromatic scale, but the result wasn't too good (it sounded more like a weird reverb). My problem is basically that of isolating certain frequencies in the input sound. I don't believe that objects like fiddle~ or sigmund~ would be very helpful in that matter, as they always start freaking out once you feed them with chords. I haven't found anything useful on the internet yet. Any idea?
Pierre
-- ::: derek holzer ::: http://macumbista.net ::: ---Oblique Strategy # 130: "Question the heroic"
Hi Pierre,
ah yes, Karplus-Strong, I forgot that keyword!
As for the analysis time of fiddle~, remember that computers aren't really real-time systems ;-) Besides that, you can change the analysis window size. Smaller windows equal faster analysis but less precision for lower frequencies. But yes, maybe new thread....
Good luck! Derek
Pierre Massat wrote:
Hi, I have tried your patch called karplus-strong. This is just amazing! This solved two problems at once : getting a sitar-like sound plus a sympathetic strings simulation. Thank a lot!
Although i think i should open a new thread about this, i was a little shocked by the time fiddle~ takes to output a pitch (i used it to control a resonator with my guitar). I think it was way above 10 ms. I haven't tried sigmund~ yet, though. Do you know what latency i should expect? I think 5 ms should do. Anyway, thanks again for introducing me to the Karplus-Strong model!
Pierre
2010/2/12 Pierre Massat <pimassat@gmail.com mailto:pimassat@gmail.com>
Thank you Derek! I'll give it a look. Best, Pierre 2010/2/11 Derek Holzer <derek@umatic.nl <mailto:derek@umatic.nl>> Hi Pierre, I've found resonant comb filters to be the most useful in both simulating string sounds, and in getting sympathetic resonances. There's tons of stuff in the list archives on them, in particular a patch (maybe I even posted it) on calculating the correct delay length for produce a given pitch. Search for "comb filter". Best Derek Pierre Massat wrote: Hello everyone, I was wondering if anybody on this list would have an idea of how one can simulate sympathetic strings in pd, using any kind of audio input (especially NOT midi). I tried making a patch with 12 resonating filters (moog~), each tuned to a particular frequency belonging to the chromatic scale, but the result wasn't too good (it sounded more like a weird reverb). My problem is basically that of isolating certain frequencies in the input sound. I don't believe that objects like fiddle~ or sigmund~ would be very helpful in that matter, as they always start freaking out once you feed them with chords. I haven't found anything useful on the internet yet. Any idea? Pierre -- ::: derek holzer ::: http://macumbista.net ::: ---Oblique Strategy # 130: "Question the heroic"