Dear list,
First of all i'd like to say that i'm very impressed by the potential of data structures in Pd. I've always kind of ignored this feature and it's a pity because it's really worth diving into it. That being said I think that help and example patches are far from sufficient for beginners, and if it wasn't for Chris McCormick's s-abstractions I would have been able to really figure out how to use them (stuff like how to make an entire polygon draggable, how to use GOP with proper scaling, etc.).
I m now stuck with a question. How can I identify the element which was just clicked ? I know that [struc] outputs the events, like click, selection and change, but I thought I could identify individual elements by their pointer id. It turns out that I get the same pointer for every element, although I created them sequentially (using [append]).
(I guess something must be escaping me about pointers... I've noticed that within the same template, I get different pointers for elements on different y-levels, but the same pointer for all the element on the same y-level regardless of their x.)
Cheers,
Pierre
Nevermind, it's working the way I expected, the y value was being output and I thought that was the pointer's id.
Cheers,
Pierre.
2014-03-05 11:24 GMT+01:00 Pierre Massat pimassat@gmail.com:
Dear list,
First of all i'd like to say that i'm very impressed by the potential of data structures in Pd. I've always kind of ignored this feature and it's a pity because it's really worth diving into it. That being said I think that help and example patches are far from sufficient for beginners, and if it wasn't for Chris McCormick's s-abstractions I would have been able to really figure out how to use them (stuff like how to make an entire polygon draggable, how to use GOP with proper scaling, etc.).
I m now stuck with a question. How can I identify the element which was just clicked ? I know that [struc] outputs the events, like click, selection and change, but I thought I could identify individual elements by their pointer id. It turns out that I get the same pointer for every element, although I created them sequentially (using [append]).
(I guess something must be escaping me about pointers... I've noticed that within the same template, I get different pointers for elements on different y-levels, but the same pointer for all the element on the same y-level regardless of their x.)
Cheers,
Pierre
On 03/05/2014 05:24 AM, Pierre Massat wrote:
Dear list,
First of all i'd like to say that i'm very impressed by the potential of data structures in Pd. I've always kind of ignored this feature and it's a pity because it's really worth diving into it. That being said I think that help and example patches are far from sufficient for beginners, and if it wasn't for Chris McCormick's s-abstractions I would have been able to really figure out how to use them (stuff like how to make an entire polygon draggable, how to use GOP with proper scaling, etc.).
It's not just the documentation, it's the interface. Having to walk linked-lists of graphically unlinked objects is bad. Having to use boilerplate to find the head of a glist just to create a scalar is bad.
I think Pd-l2ork is getting close to a release with my new data structure stuff in it. It's a first step at addressing some of these issues.
-Jonathan
I m now stuck with a question. How can I identify the element which was just clicked ? I know that [struc] outputs the events, like click, selection and change, but I thought I could identify individual elements by their pointer id. It turns out that I get the same pointer for every element, although I created them sequentially (using [append]).
(I guess something must be escaping me about pointers... I've noticed that within the same template, I get different pointers for elements on different y-levels, but the same pointer for all the element on the same y-level regardless of their x.)
Cheers,
Pierre
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
On 03/05/2014 05:24 AM, Pierre Massat wrote:
Dear list,
First of all i'd like to say that i'm very impressed by the potential
of data structures in Pd. I've always kind of ignored this feature and
it's a >>pity because it's really worth diving into it.That being said
I think that help and example patches are far from sufficient for
beginners, and if it wasn't for Chris McCormick's s->>abstractions I
would have been able to really figure out how to use them (stuff like
how to make an entire polygon draggable, how to use >>GOP with proper
scaling, etc.).It's not just the documentation, it's the interface. Having to walk
linked-lists of graphically unlinked objects is bad. Having to use
boilerplate to find the >head of a glist just to create a scalar is bad.I think Pd-l2ork is getting close to a release with my new data
structure stuff in it. It's a first step at addressing some of these
issues.
and any prospects of that stuff making it into vanilla or pd-ext, for the
non-unix users out there?
I'll have to have a look and see what the ideas are... I don't know anything yet. Anyhow I think there are a couple of things that are higher priority: getting editing to be more user-friendly, and getting the IEM GUIs to behave better. And I'm afraid I can only write code at a fraction of the speed others can - so PD vanilla will always seem years behind everything else.
cheers Miller
On Sat, Mar 08, 2014 at 12:45:33AM +0100, João Pais wrote:
On 03/05/2014 05:24 AM, Pierre Massat wrote:
Dear list,
First of all i'd like to say that i'm very impressed by the potential of data structures in Pd. I've always kind of ignored this feature and it's a >>pity because it's really worth diving into it.That being said I think that help and example patches are far from sufficient for beginners, and if it wasn't for Chris McCormick's s->>abstractions I would have been able to really figure out how to use them (stuff like how to make an entire polygon draggable, how to use >>GOP with proper scaling, etc.).
It's not just the documentation, it's the interface. Having to walk linked-lists of graphically unlinked objects is bad. Having to use boilerplate to find the >head of a glist just to create a scalar is bad.
I think Pd-l2ork is getting close to a release with my new data structure stuff in it. It's a first step at addressing some of these issues.
and any prospects of that stuff making it into vanilla or pd-ext, for the non-unix users out there?
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
re: 'slow ass coders' usually takes me 15 years to get anything done, the things that are taking longer than 15 years will probably be wonderful. I will link to you something that will make you change your mind about 'being years behind everything else' 3:33:28 seconds of some /pd-0.45-4/bin$ ./pd
the pdBerlin has some nice examples of using Chris Mccormick's lfo datastructure editor, I was weary of it before along with some GOP scaling stuff, but think I might give it a try. something is weird with this new (old junk keyboard, whenever I hold down more than 2 or 3 keys too fast , jack is glitching like the mouse dropouts. I am used to it by now and not one bit of that dreaded static makes it inside the wave file.
I feel like I have crossed a hurdle yesterday #1 loadable filename list for scrolling through presets instead of searching with the dialog[ cant that dialog be used as a directory tool? I mean it allready does what everyone want's a directory listing to do, splits up the file name from the patch, etc... just make it an atom.
#2 I somehow miraculously figured out how to count up a binary sequence like that is the oputput of the euclidian function for slick beats [1 0 1 0 1 1 0( and convert it to something that can be used as a mode for scales [2 2 1 2( my method does not yet work for sequencs that start with 0 -rotations but that doesen't bother me one bit. I ould have done it so easy with c, assembly language or even machine code but doing that graphically had my panties in a wad. "***********************************************" A quotation by Hermann von Helmholtz Whoever in the pursuit of science, seeks after immediate practical utility may rest assured that he seeks in vain. Academic Discourse (Heidelberg 1862)
JOC/EFR February 2006
The URL of this page is: http://www-history.mcs.st-andrews.ac.uk/.../Helmholtz.htmlhttp://www-history.mcs.st-andrews.ac.uk/Quotations/Helmholtz.html
On Fri, Mar 7, 2014 at 6:55 PM, Miller Puckette msp@ucsd.edu wrote:
I'll have to have a look and see what the ideas are... I don't know anything yet. Anyhow I think there are a couple of things that are higher priority: getting editing to be more user-friendly, and getting the IEM GUIs to behave better. And I'm afraid I can only write code at a fraction of the speed others can - so PD vanilla will always seem years behind everything else.
cheers Miller
On Sat, Mar 08, 2014 at 12:45:33AM +0100, João Pais wrote:
On 03/05/2014 05:24 AM, Pierre Massat wrote:
Dear list,
First of all i'd like to say that i'm very impressed by the potential of data structures in Pd. I've always kind of ignored this feature and it's a >>pity because it's really worth diving into it.That being said I think that help and example patches are far from sufficient for beginners, and if it wasn't for Chris McCormick's s->>abstractions I would have been able to really figure out how to use them (stuff like how to make an entire polygon draggable, how to use >>GOP with proper scaling, etc.).
It's not just the documentation, it's the interface. Having to walk linked-lists of graphically unlinked objects is bad. Having to use boilerplate to find the >head of a glist just to create a scalar is bad.
I think Pd-l2ork is getting close to a release with my new data structure stuff in it. It's a first step at addressing some of these issues.
and any prospects of that stuff making it into vanilla or pd-ext, for the non-unix users out there?
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management ->
http://lists.puredata.info/listinfo/pd-list
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
https://archive.org/details/isophi12moj
upgraded ubuntustudio from 13.04 to 13.10 this morning without any noticeable problems yet. how does jack2 handle pulse audio and alsa now? somehow it stopped working together before the upgrade. might have misconfigured jack with the old pasuspender -- use/bin/jackd or something
On Wed, Mar 12, 2014 at 3:57 PM, Billy Stiltner billy.stiltner@gmail.comwrote:
re: 'slow ass coders' usually takes me 15 years to get anything done, the things that are taking longer than 15 years will probably be wonderful. I will link to you something that will make you change your mind about 'being years behind everything else' 3:33:28 seconds of some /pd-0.45-4/bin$ ./pd
the pdBerlin has some nice examples of using Chris Mccormick's lfo datastructure editor, I was weary of it before along with some GOP scaling stuff, but think I might give it a try. something is weird with this new (old junk keyboard, whenever I hold down more than 2 or 3 keys too fast , jack is glitching like the mouse dropouts. I am used to it by now and not one bit of that dreaded static makes it inside the wave file.
I feel like I have crossed a hurdle yesterday #1 loadable filename list for scrolling through presets instead of searching with the dialog[ cant that dialog be used as a directory tool? I mean it allready does what everyone want's a directory listing to do, splits up the file name from the patch, etc... just make it an atom.
#2 I somehow miraculously figured out how to count up a binary sequence like that is the oputput of the euclidian function for slick beats [1 0 1 0 1 1 0( and convert it to something that can be used as a mode for scales [2 2 1 2( my method does not yet work for sequencs that start with 0 -rotations but that doesen't bother me one bit. I ould have done it so easy with c, assembly language or even machine code but doing that graphically had my panties in a wad. "***********************************************" A quotation by Hermann von Helmholtz Whoever in the pursuit of science, seeks after immediate practical utility may rest assured that he seeks in vain. Academic Discourse (Heidelberg 1862)
JOC/EFR February 2006
The URL of this page is: http://www-history.mcs.st-andrews.ac.uk/.../Helmholtz.htmlhttp://www-history.mcs.st-andrews.ac.uk/Quotations/Helmholtz.html
On Fri, Mar 7, 2014 at 6:55 PM, Miller Puckette msp@ucsd.edu wrote:
I'll have to have a look and see what the ideas are... I don't know anything yet. Anyhow I think there are a couple of things that are higher priority: getting editing to be more user-friendly, and getting the IEM GUIs to behave better. And I'm afraid I can only write code at a fraction of the speed others can - so PD vanilla will always seem years behind everything else.
cheers Miller
On Sat, Mar 08, 2014 at 12:45:33AM +0100, João Pais wrote:
On 03/05/2014 05:24 AM, Pierre Massat wrote:
Dear list,
First of all i'd like to say that i'm very impressed by the potential of data structures in Pd. I've always kind of ignored this feature and it's a >>pity because it's really worth diving into it.That being said I think that help and example patches are far from sufficient for beginners, and if it wasn't for Chris McCormick's s->>abstractions I would have been able to really figure out how to use them (stuff like how to make an entire polygon draggable, how to use >>GOP with proper scaling, etc.).
It's not just the documentation, it's the interface. Having to walk linked-lists of graphically unlinked objects is bad. Having to use boilerplate to find the >head of a glist just to create a scalar is bad.
I think Pd-l2ork is getting close to a release with my new data structure stuff in it. It's a first step at addressing some of these issues.
and any prospects of that stuff making it into vanilla or pd-ext, for the non-unix users out there?
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management ->
http://lists.puredata.info/listinfo/pd-list
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
On 03/07/2014 06:55 PM, Miller Puckette wrote:
I'll have to have a look and see what the ideas are... I don't know anything yet.
Well there's the important stuff: https://jwilkes.nfshost.com/mm.webm
And then the less important stuff, like being able to patch 15 out of the 28 demos shown here: http://raphaeljs.com/
Essentially everything except the ones with gradients and text, but those features can be added later.
It's generally not as high level as Raphael-- for example I ported the Raphael code to subpatches for the easing demo, and I'm just using [line] to do the animation. However, it would not be too hard on the Pd side to add an "animate" method. In fact that'd be quite efficient as you'd only be sending a single message over the socket and letting the GUI take care of the details.
Also, I'm instantiating scalars inside object boxes. Since the user can send messages to update shape attributes straight to the parent draw command, this means he/she can do an end-run around pointers for prototyping. So essentially you have the ability to dynamically change visual attributes on the class level (i.e., the parentwidgetbehavior) and on the object level (for the specific scalar, as you can currently).
There are still lots of details to get right, like handling groups properly, but the basic stuff is there.
-Jonathan
Anyhow I think there are a couple of things that are higher priority: getting editing to be more user-friendly, and getting the IEM GUIs to behave better. And I'm afraid I can only write code at a fraction of the speed others can - so PD vanilla will always seem years behind everything else.
cheers Miller
On Sat, Mar 08, 2014 at 12:45:33AM +0100, João Pais wrote:
On 03/05/2014 05:24 AM, Pierre Massat wrote:
Dear list,
First of all i'd like to say that i'm very impressed by the potential of data structures in Pd. I've always kind of ignored this feature and it's a >>pity because it's really worth diving into it.That being said I think that help and example patches are far from sufficient for beginners, and if it wasn't for Chris McCormick's s->>abstractions I would have been able to really figure out how to use them (stuff like how to make an entire polygon draggable, how to use >>GOP with proper scaling, etc.).
It's not just the documentation, it's the interface. Having to walk linked-lists of graphically unlinked objects is bad. Having to use boilerplate to find the >head of a glist just to create a scalar is bad.
I think Pd-l2ork is getting close to a release with my new data structure stuff in it. It's a first step at addressing some of these issues.
and any prospects of that stuff making it into vanilla or pd-ext, for the non-unix users out there? _______________________________________________ Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
On 03/07/2014 06:45 PM, João Pais wrote:
On 03/05/2014 05:24 AM, Pierre Massat wrote:Dear list, First of all i'd like to say that i'm very impressed by the potential of data structures in Pd. I've always kind of ignored this feature and it's a pity because it's really worth diving into it. That being said I think that help and example patches are far from sufficient for beginners, and if it wasn't for Chris McCormick's s-abstractions I would have been able to really figure out how to use them (stuff like how to make an entire polygon draggable, how to use GOP with proper scaling, etc.).It's not just the documentation, it's the interface. Having to walk linked-lists of graphically unlinked objects is bad. Having to use boilerplate to find the head of a glist just to create a scalar is bad. I think Pd-l2ork is getting close to a release with my new data structure stuff in it. It's a first step at addressing some of these issues.and any prospects of that stuff making it into vanilla or pd-ext,
No. It requires a toolkit that has modern 2d features like affine transformations and opacity, etc. Pd-l2ork leverages Tkpath, a tcl/tk library. Other modern toolkits like Qt have their own 2d interfaces with the same features and could be used, but tcl/tk on its own does not.
for the non-unix users out there?
For OSX, one of the tcl/tk libraries-- Tkpath needs to be ported from Carbon to Cocoa.
I haven't investigated a Windows port yet but it's probably mostly a matter of setting up the proper compile environment more than anything else. Granted one would probably need to tweak pd.tk and L2ork's build script, but getting set up in Windows seems to be where most of the work is. (At least in my experience so far.)
-Jonathan
-
Billy Stiltner -
Jonathan Wilkes -
João Pais -
Miller Puckette -
Pierre Massat