preset interpolation/morphing

[miguel.negrao]

Hi

I've been casually investigating what people use to morph between
presets, and found an intersting application which is the metasurface of
audiomulch (paper: The Metasurface – Applying Natural Neighbour
Interpolation to Two-to-Many Mapping ;
http://www.audiomulch.com/~rossb/writings/nime2005_101.pdf).
   Has anyone developed something similar to this in supercollider, or
some other nice morphing scheme ?

thanks,
--
Miguel Negrão // ZLB
http://www.friendlyvirus.org/artists/zlb/

_______________________________________________
sc-users mailing list

info (subscription, etc.): http://www.beast.bham.ac.uk/research/sc_mailing_lists.shtml
archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
search: http://www.listarc.bham.ac.uk/lists/sc-users/search/

[nescivi]

On Saturday 18 April 2009 09:41:42 Miguel Negrao wrote:
> Hi
>
> I've been casually investigating what people use to morph between
> presets, and found an intersting application which is the metasurface of
> audiomulch (paper: The Metasurface – Applying Natural Neighbour
> Interpolation to Two-to-Many Mapping ;
> http://www.audiomulch.com/~rossb/writings/nime2005_101.pdf).
>    Has anyone developed something similar to this in supercollider, or
> some other nice morphing scheme ?

ParameterSpace in SenseWorld.

May not do the exact same morphing, but similar concept (I guess you could
change the function for the interpolation).

sincerely,
Marije


_______________________________________________
sc-users mailing list

info (subscription, etc.): http://www.beast.bham.ac.uk/research/sc_mailing_lists.shtml
archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
search: http://www.listarc.bham.ac.uk/lists/sc-users/search/

[Martin Vognsen-4]

Hi

This is sort of interesting for me to read, as I just finished a  
preset morphing system this very moment! Funny that I wasn't aware of  
ParameterSpace, I will be curious to check it out.

So far my system only interpolates between any two presets that you  
select from a bank, and it happens automatically at a time that you  
set manually. I'm storing values in arrays and morphing all synth  
parameters with a bunch of Lag's. So no gestures or multiple preset  
"zones". That would be an obvious next step. To be able to morph  
manually between multiple presets, you would need to approach some  
things differently. In fact, my own example is probably not one to  
follow, in terms of programming style. It's all in one patch which  
has grown to well over 4000 lines, eventhough I have tried to do as  
much nesting as I found practical... hm... let's talk about modular  
design another day! So it basically only works for my particular  
setup. But still, it isn't super difficult to do it this way, and it  
works.

In my case I have around 300 sliders on an OSC/MIDI controller that  
need to get updated depending on a specified morph time up to 30  
secs, and a similar number of buttons, so the update rate for the  
sliders isn't very high. All the sliders that control SC are OSC, and  
the morphing isn't so smart yet, so I am updating all objects always  
at the same rate (just one routine!), and I seem to get 5 updates per  
second safely this way. I don't mind the low resolution, as it's only  
to get the faders to where the synths are, all actuall sound morphing  
is at the resolution of the Lag's. Some of the faders control Ableton  
Live where I do basic mixing via SoundFlower (why re-invent the  
wheel...  actually I could think of a few reasons, but that will have  
to wait...). Of course Live means MIDI (one of the reasons!). I have  
made the MIDI morphing a little smarter than the OSC, as the update  
rate depends on the distance each seperate fader needs to travel, so  
if a fader doesn't need to move there is no need to update it and at  
small distances the rate is low. So the MIDI controllers in Live  
always move at full resolution.

I wanted any ongoing morphing to be interrupted when any controller  
is changed manually or a new preset selected, so that's just a  
question of stopping the routines that run the controller morphing,  
and changing the Lag times in the synths and setting the parameters.

So: A real programmer would probably frown at this approach, but it's  
a pretty managable and dirty way if you just want to get on with your  
life and make some noise. And that's what blows me away about SC,  
that musicians can build the kind of instrument they envision without  
too much special technical skill. Thanks to all the talented developers!

Martin

_______________________________________________
sc-users mailing list

info (subscription, etc.): http://www.beast.bham.ac.uk/research/sc_mailing_lists.shtml
archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
search: http://www.listarc.bham.ac.uk/lists/sc-users/search/

[miguel.negrao]

nescivi escreveu:

> On Saturday 18 April 2009 09:41:42 Miguel Negrao wrote:
>> Hi
>>
>> I've been casually investigating what people use to morph between
>> presets, and found an intersting application which is the metasurface of
>> audiomulch (paper: The Metasurface – Applying Natural Neighbour
>> Interpolation to Two-to-Many Mapping ;
>> http://www.audiomulch.com/~rossb/writings/nime2005_101.pdf).
>>    Has anyone developed something similar to this in supercollider, or
>> some other nice morphing scheme ?
>
> ParameterSpace in SenseWorld.
>
> May not do the exact same morphing, but similar concept (I guess you could
> change the function for the interpolation).
>
> sincerely,
> Marije
>
>
> _______________________________________________
> sc-users mailing list
>
> info (subscription, etc.): http://www.beast.bham.ac.uk/research/sc_mailing_lists.shtml
> archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
> search: http://www.listarc.bham.ac.uk/lists/sc-users/search/
>
>

Very nice class. I was checking it and it might just do what I was
looking for.

thanks,
--
Miguel Negrão // ZLB
http://www.friendlyvirus.org/artists/zlb/

_______________________________________________
sc-users mailing list

info (subscription, etc.): http://www.beast.bham.ac.uk/research/sc_mailing_lists.shtml
archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
search: http://www.listarc.bham.ac.uk/lists/sc-users/search/

[Dan Stowell]

A few people have done parameter-morphing. At the SuperCollider
Symposium I saw Jan Trutzschler demo his neat approach using a
self-organising map (SOM) to create a morphable 2D map of presets;
also Ron Kuivila demoed the Conductor stuff (it's available from
quarks) which provides neat ways to map groups of presets and
crossfade them using a slider, among other things...

Dan

2009/4/19, Miguel Negrao <miguel.negrao@...>:

> nescivi escreveu:
>
>
> > On Saturday 18 April 2009 09:41:42 Miguel Negrao wrote:
> >
> > > Hi
> > >
> > > I've been casually investigating what people use to morph between
> > > presets, and found an intersting application which is the metasurface of
> > > audiomulch (paper: The Metasurface – Applying Natural Neighbour
> > > Interpolation to Two-to-Many Mapping ;
> > >
> http://www.audiomulch.com/~rossb/writings/nime2005_101.pdf).
> > >   Has anyone developed something similar to this in supercollider, or
> > > some other nice morphing scheme ?
> > >
> >
> > ParameterSpace in SenseWorld.
> >
> > May not do the exact same morphing, but similar concept (I guess you could
> change the function for the interpolation).
> >
> > sincerely,
> > Marije
> >
> >
> > _______________________________________________
> > sc-users mailing list
> >
> > info (subscription, etc.):
> http://www.beast.bham.ac.uk/research/sc_mailing_lists.shtml
> > archive:
> http://www.listarc.bham.ac.uk/marchives/sc-users/
> > search:
> http://www.listarc.bham.ac.uk/lists/sc-users/search/
> >
> >
> >
>  Very nice class. I was checking it and it might just do what I was looking
> for.
>
>  thanks,
>  --
>  Miguel Negrão // ZLB
>  http://www.friendlyvirus.org/artists/zlb/
>
>  _______________________________________________
>  sc-users mailing list
>
>  info (subscription, etc.):
> http://www.beast.bham.ac.uk/research/sc_mailing_lists.shtml
>  archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
>  search:
> http://www.listarc.bham.ac.uk/lists/sc-users/search/
>

--
http://www.mcld.co.uk

_______________________________________________
sc-users mailing list

info (subscription, etc.): http://www.beast.bham.ac.uk/research/sc_mailing_lists.shtml
archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
search: http://www.listarc.bham.ac.uk/lists/sc-users/search/

[miguel.negrao]

nescivi escreveu:

> On Saturday 18 April 2009 09:41:42 Miguel Negrao wrote:
>> Hi
>>
>> I've been casually investigating what people use to morph between
>> presets, and found an intersting application which is the metasurface of
>> audiomulch (paper: The Metasurface – Applying Natural Neighbour
>> Interpolation to Two-to-Many Mapping ;
>> http://www.audiomulch.com/~rossb/writings/nime2005_101.pdf).
>>    Has anyone developed something similar to this in supercollider, or
>> some other nice morphing scheme ?
>
> ParameterSpace in SenseWorld.
>
> May not do the exact same morphing, but similar concept (I guess you could
> change the function for the interpolation).
>
> sincerely,
> Marije
>
>
> _______________________________________________
> sc-users mailing list
>
> info (subscription, etc.): http://www.beast.bham.ac.uk/research/sc_mailing_lists.shtml
> archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
> search: http://www.listarc.bham.ac.uk/lists/sc-users/search/
>
>

One question: I have multiple synths running which are basically from
the same synthdef (same arg names); what is the best way to put the
parameters off all this synths in the ParameterSpace ?
   I could create an array of ParameterSpaces, one per synth and put the
values there, but then when I call start on all of them there are lots
of tasks running. would this be problematic ?
  the other option I can see is to append "-1", "-2", etc to the each
parameter name when storing and calling.
   From an organizational point of view the first options seems better
to me, but not as efficient as option 2. Any opinion ?


thanks,
--
Miguel Negrão // ZLB
http://www.friendlyvirus.org/artists/zlb/

_______________________________________________
sc-users mailing list

info (subscription, etc.): http://www.beast.bham.ac.uk/research/sc_mailing_lists.shtml
archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
search: http://www.listarc.bham.ac.uk/lists/sc-users/search/

[miguel.negrao]

Jan, are you planning on making this available ?

Dan, thanks.
Miguel

Dan Stowell escreveu:

> A few people have done parameter-morphing. At the SuperCollider
> Symposium I saw Jan Trutzschler demo his neat approach using a
> self-organising map (SOM) to create a morphable 2D map of presets;
> also Ron Kuivila demoed the Conductor stuff (it's available from
> quarks) which provides neat ways to map groups of presets and
> crossfade them using a slider, among other things...
>
> Dan
>
> 2009/4/19, Miguel Negrao <miguel.negrao@...>:
>> nescivi escreveu:
>>
>>
>>> On Saturday 18 April 2009 09:41:42 Miguel Negrao wrote:
>>>
>>>> Hi
>>>>
>>>> I've been casually investigating what people use to morph between
>>>> presets, and found an intersting application which is the metasurface of
>>>> audiomulch (paper: The Metasurface – Applying Natural Neighbour
>>>> Interpolation to Two-to-Many Mapping ;
>>>>
>> http://www.audiomulch.com/~rossb/writings/nime2005_101.pdf).
>>>>   Has anyone developed something similar to this in supercollider, or
>>>> some other nice morphing scheme ?
>>>>
>>> ParameterSpace in SenseWorld.
>>>
>>> May not do the exact same morphing, but similar concept (I guess you could
>> change the function for the interpolation).
>>> sincerely,
>>> Marije
>>>
>>>
>>> _______________________________________________
>>> sc-users mailing list
>>>
>>> info (subscription, etc.):
>> http://www.beast.bham.ac.uk/research/sc_mailing_lists.shtml
>>> archive:
>> http://www.listarc.bham.ac.uk/marchives/sc-users/
>>> search:
>> http://www.listarc.bham.ac.uk/lists/sc-users/search/
>>>
>>>
>>  Very nice class. I was checking it and it might just do what I was looking
>> for.
>>
>>  thanks,
>>  --
>>  Miguel Negrão // ZLB
>>  http://www.friendlyvirus.org/artists/zlb/
>>
>>  _______________________________________________
>>  sc-users mailing list
>>
>>  info (subscription, etc.):
>> http://www.beast.bham.ac.uk/research/sc_mailing_lists.shtml
>>  archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
>>  search:
>> http://www.listarc.bham.ac.uk/lists/sc-users/search/
>>
>
>

--
Miguel Negrão // ZLB
http://www.friendlyvirus.org/artists/zlb/

_______________________________________________
sc-users mailing list

info (subscription, etc.): http://www.beast.bham.ac.uk/research/sc_mailing_lists.shtml
archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
search: http://www.listarc.bham.ac.uk/lists/sc-users/search/

[Martin Vognsen-4]

Hello again

Sorry for being dense: I have SenseWorld, but don't see  
ParameterSpace anywhere. Um... It should be in the quark, right?

Thank you
Martin

Den 19/04/2009 kl. 14.35 skrev Miguel Negrao:

> nescivi escreveu:
>> On Saturday 18 April 2009 09:41:42 Miguel Negrao wrote:
>>> Hi
>>>
>>> I've been casually investigating what people use to morph between
>>> presets, and found an intersting application which is the  
>>> metasurface of
>>> audiomulch (paper: The Metasurface – Applying Natural Neighbour
>>> Interpolation to Two-to-Many Mapping ;
>>> http://www.audiomulch.com/~rossb/writings/nime2005_101.pdf).
>>>    Has anyone developed something similar to this in  
>>> supercollider, or
>>> some other nice morphing scheme ?
>> ParameterSpace in SenseWorld.
>> May not do the exact same morphing, but similar concept (I guess  
>> you could change the function for the interpolation).
>> sincerely,
>> Marije
>> _______________________________________________
>> sc-users mailing list
>> info (subscription, etc.): http://www.beast.bham.ac.uk/research/ 
>> sc_mailing_lists.shtml
>> archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
>> search: http://www.listarc.bham.ac.uk/lists/sc-users/search/
> Very nice class. I was checking it and it might just do what I was  
> looking for.
>
> thanks,
> --
> Miguel Negrão // ZLB
> http://www.friendlyvirus.org/artists/zlb/
>
> _______________________________________________
> sc-users mailing list
>
> info (subscription, etc.): http://www.beast.bham.ac.uk/research/ 
> sc_mailing_lists.shtml
> archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
> search: http://www.listarc.bham.ac.uk/lists/sc-users/search/

_______________________________________________
sc-users mailing list

info (subscription, etc.): http://www.beast.bham.ac.uk/research/sc_mailing_lists.shtml
archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
search: http://www.listarc.bham.ac.uk/lists/sc-users/search/

[miguel.negrao]

I got it from the quark...

Miguel

Martin Vognsen escreveu:

>
> Hello again
>
> Sorry for being dense: I have SenseWorld, but don't see ParameterSpace
> anywhere. Um... It should be in the quark, right?
>
> Thank you
> Martin
>
> Den 19/04/2009 kl. 14.35 skrev Miguel Negrao:
>
>> nescivi escreveu:
>>> On Saturday 18 April 2009 09:41:42 Miguel Negrao wrote:
>>>> Hi
>>>>
>>>> I've been casually investigating what people use to morph between
>>>> presets, and found an intersting application which is the
>>>> metasurface of
>>>> audiomulch (paper: The Metasurface – Applying Natural Neighbour
>>>> Interpolation to Two-to-Many Mapping ;
>>>> http://www.audiomulch.com/~rossb/writings/nime2005_101.pdf).
>>>>    Has anyone developed something similar to this in supercollider, or
>>>> some other nice morphing scheme ?
>>> ParameterSpace in SenseWorld.
>>> May not do the exact same morphing, but similar concept (I guess you
>>> could change the function for the interpolation).
>>> sincerely,
>>> Marije
>>> _______________________________________________
>>> sc-users mailing list
>>> info (subscription, etc.):
>>> http://www.beast.bham.ac.uk/research/sc_mailing_lists.shtml
>>> archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
>>> search: http://www.listarc.bham.ac.uk/lists/sc-users/search/
>> Very nice class. I was checking it and it might just do what I was
>> looking for.
>>
>> thanks,
>> --
>> Miguel Negrão // ZLB
>> http://www.friendlyvirus.org/artists/zlb/
>>
>> _______________________________________________
>> sc-users mailing list
>>
>> info (subscription, etc.):
>> http://www.beast.bham.ac.uk/research/sc_mailing_lists.shtml
>> archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
>> search: http://www.listarc.bham.ac.uk/lists/sc-users/search/
>
>
> _______________________________________________
> sc-users mailing list
>
> info (subscription, etc.):
> http://www.beast.bham.ac.uk/research/sc_mailing_lists.shtml
> archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
> search: http://www.listarc.bham.ac.uk/lists/sc-users/search/
>
>

--
Miguel Negrão // ZLB
http://www.friendlyvirus.org/artists/zlb/

_______________________________________________
sc-users mailing list

info (subscription, etc.): http://www.beast.bham.ac.uk/research/sc_mailing_lists.shtml
archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
search: http://www.listarc.bham.ac.uk/lists/sc-users/search/

[nescivi]

On Sunday 19 April 2009 09:49:48 Miguel Negrao wrote:
> One question: I have multiple synths running which are basically from
> the same synthdef (same arg names); what is the best way to put the
> parameters off all this synths in the ParameterSpace ?
>    I could create an array of ParameterSpaces, one per synth and put the
> values there, but then when I call start on all of them there are lots
> of tasks running. would this be problematic ?

You only need to use start, when you are using the moveFunc to move through
the space. So if you are not using that you do not have any task running.

>   the other option I can see is to append "-1", "-2", etc to the each
> parameter name when storing and calling.
>    From an organizational point of view the first options seems better
> to me, but not as efficient as option 2. Any opinion ?

I used the first option in a project, but the second one is equally valid.

Just a matter of where you want to split things, and if you want to control
the synths separately, you need to have two different spaces.
If not, then why not control a group with the settings, and put all synths in
that group?

sincerely,
Marije




_______________________________________________
sc-users mailing list

info (subscription, etc.): http://www.beast.bham.ac.uk/research/sc_mailing_lists.shtml
archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
search: http://www.listarc.bham.ac.uk/lists/sc-users/search/

[Martin Vognsen-4]

Thank you, Miguel, I got it, don't know what happened. Uninstalled  
and installed the quark a few times, but then erased the quark  
manually and reinstalled, and then all the classes were there. Cool,  
I will check it out!

Ah yes, Dan, and CVInterpolator is also a good idea, I think when Ron  
Kuivila presented Conductor at Bham a couple of years ago, I wasn't  
quite ready for it, but now seems like a perfect time. Except that 3  
weeks earlier it could probably have saved me the time of  
implementing morphing in my own patch, but hey, much fun was had! :-)

Thank you again,
Martin

Den 19/04/2009 kl. 16.12 skrev Miguel Negrao:

> I got it from the quark...
>
> Miguel
>
> Martin Vognsen escreveu:
>> Hello again
>> Sorry for being dense: I have SenseWorld, but don't see  
>> ParameterSpace anywhere. Um... It should be in the quark, right?
>> Thank you
>> Martin
>> Den 19/04/2009 kl. 14.35 skrev Miguel Negrao:
>>> nescivi escreveu:
>>>> On Saturday 18 April 2009 09:41:42 Miguel Negrao wrote:
>>>>> Hi
>>>>>
>>>>> I've been casually investigating what people use to morph between
>>>>> presets, and found an intersting application which is the  
>>>>> metasurface of
>>>>> audiomulch (paper: The Metasurface – Applying Natural Neighbour
>>>>> Interpolation to Two-to-Many Mapping ;
>>>>> http://www.audiomulch.com/~rossb/writings/nime2005_101.pdf).
>>>>>    Has anyone developed something similar to this in  
>>>>> supercollider, or
>>>>> some other nice morphing scheme ?
>>>> ParameterSpace in SenseWorld.
>>>> May not do the exact same morphing, but similar concept (I guess  
>>>> you could change the function for the interpolation).
>>>> sincerely,
>>>> Marije
>>>> _______________________________________________
>>>> sc-users mailing list
>>>> info (subscription, etc.): http://www.beast.bham.ac.uk/research/ 
>>>> sc_mailing_lists.shtml
>>>> archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
>>>> search: http://www.listarc.bham.ac.uk/lists/sc-users/search/
>>> Very nice class. I was checking it and it might just do what I  
>>> was looking for.
>>>
>>> thanks,
>>> --
>>> Miguel Negrão // ZLB
>>> http://www.friendlyvirus.org/artists/zlb/
>>>
>>> _______________________________________________
>>> sc-users mailing list
>>>
>>> info (subscription, etc.): http://www.beast.bham.ac.uk/research/ 
>>> sc_mailing_lists.shtml
>>> archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
>>> search: http://www.listarc.bham.ac.uk/lists/sc-users/search/
>> _______________________________________________
>> sc-users mailing list
>> info (subscription, etc.): http://www.beast.bham.ac.uk/research/ 
>> sc_mailing_lists.shtml
>> archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
>> search: http://www.listarc.bham.ac.uk/lists/sc-users/search/
>
>
> --
> Miguel Negrão // ZLB
> http://www.friendlyvirus.org/artists/zlb/
>
> _______________________________________________
> sc-users mailing list
>
> info (subscription, etc.): http://www.beast.bham.ac.uk/research/ 
> sc_mailing_lists.shtml
> archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
> search: http://www.listarc.bham.ac.uk/lists/sc-users/search/

_______________________________________________
sc-users mailing list

info (subscription, etc.): http://www.beast.bham.ac.uk/research/sc_mailing_lists.shtml
archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
search: http://www.listarc.bham.ac.uk/lists/sc-users/search/

[miguel.negrao]

nescivi escreveu:

> On Sunday 19 April 2009 09:49:48 Miguel Negrao wrote:
>> One question: I have multiple synths running which are basically from
>> the same synthdef (same arg names); what is the best way to put the
>> parameters off all this synths in the ParameterSpace ?
>>    I could create an array of ParameterSpaces, one per synth and put the
>> values there, but then when I call start on all of them there are lots
>> of tasks running. would this be problematic ?
>
> You only need to use start, when you are using the moveFunc to move through
> the space. So if you are not using that you do not have any task running.
>

indeed. The documentation is a bit misleading, it seems that calling
.start is necessary for the gui stuff. In fact there isn't any moveFunc
in the help file so it's not clear why .start is there.

>>   the other option I can see is to append "-1", "-2", etc to the each
>> parameter name when storing and calling.
>>    From an organizational point of view the first options seems better
>> to me, but not as efficient as option 2. Any opinion ?
>
> I used the first option in a project, but the second one is equally valid.
>
> Just a matter of where you want to split things, and if you want to control
> the synths separately, you need to have two different spaces.
> If not, then why not control a group with the settings, and put all synths in
> that group?

I want to keep each synth with it's own values, so I will be taking the
first option. I got it working now, and I made a little gui window to
control things but each time I drag the mouse over it supercollider app
goes to 90% cpu . I'm using some 50 knobs (7 synths each with 7 args)
which are updated in realtime each time the settings are being changed
plus the xy window itself. The gui code I'm using:

(
                var scale = 400/10;
                w = Window.new("presets",Rect(10,10,400,400));
               
                v = UserView(w,Rect(0,0,400,400));
                v.relativeOrigin = true;
                v.background_(Color.white);
               
                v.drawFunc = {
                       
                        ~spaces[0].points.do{ |point|
                                // set the Color
                                Pen.color = Color.blue(0.5, 0.5);
                                Pen.addWedge(point*scale, 10, 0, 2pi);
                                Pen.perform( \fill);
                        };
                       
                        ~spaces[0].curloc.do{ |point|
                                // set the Color
                                Pen.color = Color.red(0.5, 0.5);
                                Pen.addWedge(point*scale, 5, 0, 2pi);
                                Pen.perform( \fill);
                        }
                       
                };
               
       
                v.mouseMoveAction_({ arg view,x,y;
               
                        if(v.bounds.contains(Point(x,y))){
                        ~spaces.do(_.set(Point(x/scale,y/scale)));
                        v.refresh;
                        }
                        });
                       
                w.front;
)

~spaces is an array of ParameterSpace's. All of them have the same
curloc and the same points (just different values for the keys).

Could this because of this window above, because of the knob updates or
because of something else ?

thanks,
--
Miguel Negrão // ZLB
http://www.friendlyvirus.org/artists/zlb/

_______________________________________________
sc-users mailing list

info (subscription, etc.): http://www.beast.bham.ac.uk/research/sc_mailing_lists.shtml
archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
search: http://www.listarc.bham.ac.uk/lists/sc-users/search/

[nescivi]

On Sunday 19 April 2009 15:24:38 Miguel Negrao wrote:

> nescivi escreveu:
> > On Sunday 19 April 2009 09:49:48 Miguel Negrao wrote:
> >> One question: I have multiple synths running which are basically from
> >> the same synthdef (same arg names); what is the best way to put the
> >> parameters off all this synths in the ParameterSpace ?
> >>    I could create an array of ParameterSpaces, one per synth and put the
> >> values there, but then when I call start on all of them there are lots
> >> of tasks running. would this be problematic ?
> >
> > You only need to use start, when you are using the moveFunc to move
> > through the space. So if you are not using that you do not have any task
> > running.
>
> indeed. The documentation is a bit misleading, it seems that calling
> .start is necessary for the gui stuff. In fact there isn't any moveFunc
> in the help file so it's not clear why .start is there.

hmm..
ok. I made the help file rather quick, I guess, porting code from a
composition where I used it.

> >>   the other option I can see is to append "-1", "-2", etc to the each
> >> parameter name when storing and calling.
> >>    From an organizational point of view the first options seems better
> >> to me, but not as efficient as option 2. Any opinion ?
> >
> > I used the first option in a project, but the second one is equally
> > valid.
> >
> > Just a matter of where you want to split things, and if you want to
> > control the synths separately, you need to have two different spaces.
> > If not, then why not control a group with the settings, and put all
> > synths in that group?
>
> I want to keep each synth with it's own values, so I will be taking the
> first option. I got it working now, and I made a little gui window to
> control things but each time I drag the mouse over it supercollider app
> goes to 90% cpu . I'm using some 50 knobs (7 synths each with 7 args)
> which are updated in realtime each time the settings are being changed
> plus the xy window itself. The gui code I'm using:
>
> (
> var scale = 400/10;
> w = Window.new("presets",Rect(10,10,400,400));
>
> v = UserView(w,Rect(0,0,400,400));
> v.relativeOrigin = true;
> v.background_(Color.white);
>
> v.drawFunc = {
>
> ~spaces[0].points.do{ |point|
> // set the Color
> Pen.color = Color.blue(0.5, 0.5);
> Pen.addWedge(point*scale, 10, 0, 2pi);
> Pen.perform( \fill);
> };
>
> ~spaces[0].curloc.do{ |point|
> // set the Color
> Pen.color = Color.red(0.5, 0.5);
> Pen.addWedge(point*scale, 5, 0, 2pi);
> Pen.perform( \fill);
> }
>
> };
>
>
> v.mouseMoveAction_({ arg view,x,y;
>
> if(v.bounds.contains(Point(x,y))){
> ~spaces.do(_.set(Point(x/scale,y/scale)));
> v.refresh;
> }
> });
>
> w.front;
> )
>
> ~spaces is an array of ParameterSpace's. All of them have the same
> curloc and the same points (just different values for the keys).
>
> Could this because of this window above, because of the knob updates or
> because of something else ?

Try testing updating parameters without the GUI, so you can see whether that
is the culprit.

I used ParameterSpace with control via a gamepad joystick, so that is why
maybe the control (with the moveFunc) is a bit strange for your use.
I would use a 2Dslider though with this, rather than a Pen based gui.

sincerely,
Marije



_______________________________________________
sc-users mailing list

info (subscription, etc.): http://www.beast.bham.ac.uk/research/sc_mailing_lists.shtml
archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
search: http://www.listarc.bham.ac.uk/lists/sc-users/search/

[Martin .]

Hi,

i started making a basic version of Colorblobs (http://tirl.org/software/colorblobs/) for SC a while ago. Its got bugs, is unfinished and is not tested in Cocoa, but im happy to share the code with anyone interested.

cheers,
martin

On Sat, Apr 18, 2009 at 3:41 PM, Miguel Negrao <miguel.negrao@...> wrote:

Hi

I've been casually investigating what people use to morph between presets, and found an intersting application which is the metasurface of audiomulch (paper: The Metasurface – Applying Natural Neighbour
Interpolation to Two-to-Many Mapping ; http://www.audiomulch.com/~rossb/writings/nime2005_101.pdf).
 Has anyone developed something similar to this in supercollider, or some other nice morphing scheme ?

thanks,
--
Miguel Negrão // ZLB
http://www.friendlyvirus.org/artists/zlb/

_______________________________________________
sc-users mailing list

info (subscription, etc.): http://www.beast.bham.ac.uk/research/sc_mailing_lists.shtml
archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
search: http://www.listarc.bham.ac.uk/lists/sc-users/search/

[nescivi]

On Monday 20 April 2009 03:54:47 Martin . wrote:
> Hi,
>
> i started making a basic version of Colorblobs (
> http://tirl.org/software/colorblobs/) for SC a while ago. Its got bugs, is
> unfinished and is not tested in Cocoa, but im happy to share the code with
> anyone interested.

That would be interesting!

sincerely,
Marije

> cheers,
> martin
>
> On Sat, Apr 18, 2009 at 3:41 PM, Miguel Negrao <
>
> miguel.negrao@...> wrote:
> > Hi
> >
> > I've been casually investigating what people use to morph between
> > presets, and found an intersting application which is the metasurface of
> > audiomulch (paper: The Metasurface – Applying Natural Neighbour
> > Interpolation to Two-to-Many Mapping ;
> > http://www.audiomulch.com/~rossb/writings/nime2005_101.pdf<http://www.aud
> >iomulch.com/%7Erossb/writings/nime2005_101.pdf> ).
> >  Has anyone developed something similar to this in supercollider, or some
> > other nice morphing scheme ?
> >
> > thanks,
> > --
> > Miguel Negrão // ZLB
> > http://www.friendlyvirus.org/artists/zlb/
> >
> > _______________________________________________
> > sc-users mailing list
> >
> > info (subscription, etc.):
> > http://www.beast.bham.ac.uk/research/sc_mailing_lists.shtml
> > archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
> > search: http://www.listarc.bham.ac.uk/lists/sc-users/search/

_______________________________________________
sc-users mailing list

info (subscription, etc.): http://www.beast.bham.ac.uk/research/sc_mailing_lists.shtml
archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
search: http://www.listarc.bham.ac.uk/lists/sc-users/search/

[Martin Vognsen-4]

Hi Marije

I also had a chance to play around with ParameterSpace now, it's very  
useful, so thank you for that!
(and sorry for semi-hijacking this thread, btw... all I can add is  
thanks and praise, really. hopefully some day I will get around to  
personally thanking each individual person who wrote a class that I  
use...!)

Martin


Den 20/04/2009 kl. 16.23 skrev nescivi:

> On Monday 20 April 2009 03:54:47 Martin . wrote:
>> Hi,
>>
>> i started making a basic version of Colorblobs (
>> http://tirl.org/software/colorblobs/) for SC a while ago. Its got  
>> bugs, is
>> unfinished and is not tested in Cocoa, but im happy to share the  
>> code with
>> anyone interested.
>
> That would be interesting!
>
> sincerely,
> Marije
>
>> cheers,
>> martin
>>
>> On Sat, Apr 18, 2009 at 3:41 PM, Miguel Negrao <
>>
>> miguel.negrao@...> wrote:
>>> Hi
>>>
>>> I've been casually investigating what people use to morph between
>>> presets, and found an intersting application which is the  
>>> metasurface of
>>> audiomulch (paper: The Metasurface – Applying Natural Neighbour
>>> Interpolation to Two-to-Many Mapping ;
>>> http://www.audiomulch.com/~rossb/writings/nime2005_101.pdf<http://
>>> www.aud
>>> iomulch.com/%7Erossb/writings/nime2005_101.pdf> ).
>>>  Has anyone developed something similar to this in supercollider,  
>>> or some
>>> other nice morphing scheme ?
>>>
>>> thanks,
>>> --
>>> Miguel Negrão // ZLB
>>> http://www.friendlyvirus.org/artists/zlb/
>>>
>>> _______________________________________________
>>> sc-users mailing list
>>>
>>> info (subscription, etc.):
>>> http://www.beast.bham.ac.uk/research/sc_mailing_lists.shtml
>>> archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
>>> search: http://www.listarc.bham.ac.uk/lists/sc-users/search/
>
>
>
> _______________________________________________
> sc-users mailing list
>
> info (subscription, etc.): http://www.beast.bham.ac.uk/research/ 
> sc_mailing_lists.shtml
> archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
> search: http://www.listarc.bham.ac.uk/lists/sc-users/search/

_______________________________________________
sc-users mailing list

info (subscription, etc.): http://www.beast.bham.ac.uk/research/sc_mailing_lists.shtml
archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
search: http://www.listarc.bham.ac.uk/lists/sc-users/search/

[Martin Vognsen-4]

Hi again Marije, just one question if you have time:

The distance of each point from the edge must be >=1, else the  
defined values in each point start to change. Is that a correct  
observation? Meaning that a situation where you have the "un-morphed"  
values in the extreme corners is not possible, is that true?

Thanks
Martin


Den 20/04/2009 kl. 16.23 skrev nescivi:

> On Monday 20 April 2009 03:54:47 Martin . wrote:
>> Hi,
>>
>> i started making a basic version of Colorblobs (
>> http://tirl.org/software/colorblobs/) for SC a while ago. Its got  
>> bugs, is
>> unfinished and is not tested in Cocoa, but im happy to share the  
>> code with
>> anyone interested.
>
> That would be interesting!
>
> sincerely,
> Marije
>
>> cheers,
>> martin
>>
>> On Sat, Apr 18, 2009 at 3:41 PM, Miguel Negrao <
>>
>> miguel.negrao@...> wrote:
>>> Hi
>>>
>>> I've been casually investigating what people use to morph between
>>> presets, and found an intersting application which is the  
>>> metasurface of
>>> audiomulch (paper: The Metasurface – Applying Natural Neighbour
>>> Interpolation to Two-to-Many Mapping ;
>>> http://www.audiomulch.com/~rossb/writings/nime2005_101.pdf<http://
>>> www.aud
>>> iomulch.com/%7Erossb/writings/nime2005_101.pdf> ).
>>>  Has anyone developed something similar to this in supercollider,  
>>> or some
>>> other nice morphing scheme ?
>>>
>>> thanks,
>>> --
>>> Miguel Negrão // ZLB
>>> http://www.friendlyvirus.org/artists/zlb/
>>>
>>> _______________________________________________
>>> sc-users mailing list
>>>
>>> info (subscription, etc.):
>>> http://www.beast.bham.ac.uk/research/sc_mailing_lists.shtml
>>> archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
>>> search: http://www.listarc.bham.ac.uk/lists/sc-users/search/
>
>
>
> _______________________________________________
> sc-users mailing list
>
> info (subscription, etc.): http://www.beast.bham.ac.uk/research/ 
> sc_mailing_lists.shtml
> archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
> search: http://www.listarc.bham.ac.uk/lists/sc-users/search/

_______________________________________________
sc-users mailing list

info (subscription, etc.): http://www.beast.bham.ac.uk/research/sc_mailing_lists.shtml
archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
search: http://www.listarc.bham.ac.uk/lists/sc-users/search/