[ dev ]

4 views
Skip to first unread message

pi

unread,
Apr 16, 2009, 11:25:20 AM4/16/09
to Tagtool
hi,

as maki already opened up a thread for artists/users, this is meant to
discuss the tech stuff
http://groups.google.com/group/tagtool/browse_thread/thread/32264809f4e0352


i'm pi from austria, working at the tagtool headquarters in tulln
on concepts regarding the new release of tagtool software.
We all want the new version to work as a cross-plattform and
distributed developed one.

It's time to collect your ideas for frameworks and software design
of the new tagtool release. It will be a C++ application, basicly the
only decision so far.
You know the current tagtool application? Did you already have
ideas how to port it or rewrite it using OS - frameworks or libraries?
Do you want to implement spezific features or ideas in a distributed
development project?
Please let us know your thoughts and follow up in this thread.

With the Tagtool Think Tank
http://groups.google.com/group/tagtool/browse_thread/thread/32264809f4e0352
the brainstorming starts and we also want you
to participate remotely, if you can't join us in person.

cheers

pi

fan...@gmx.at

unread,
Apr 16, 2009, 1:01:45 PM4/16/09
to Tagtool
hi,

i know the current tagtool application.. and i think it should be
possible to rewrite the tagtool application with OpenFrameworks
(http://www.openframeworks.cc) for c++. It's a cross plattform
framework. For example there is also a addon to use OpenCV for
CameraTracking and stuff like that. I didn't have much experience with
that framework but i think you should take a look on it.

Another framework would be QT. With Qt i don't have any experience but
i know that it's also a cross plattform c++ framework. (http://
www.qtsoftware.com/products/library)

Maybe one of these frameworks can help you to build the new tagtool
application.

best,
CakaG

On 16 Apr., 17:25, pi <p...@modular-t.org> wrote:
> hi,
>
> as maki already opened up a thread for artists/users, this is meant to
> discuss the tech stuffhttp://groups.google.com/group/tagtool/browse_thread/thread/32264809f...
>
> i'm pi from austria, working at the tagtool headquarters in tulln
> on concepts regarding the new release of tagtool software.
> We all want the new version to work as a cross-plattform and
> distributed developed one.
>
> It's time to collect your ideas for frameworks and software design
> of the new tagtool release. It will be a C++ application, basicly the
> only decision so far.
> You know the current tagtool application? Did you already have
> ideas how to port it or rewrite it using OS - frameworks or libraries?
> Do you want to implement spezific features or ideas in a distributed
> development project?
> Please let us know your thoughts and follow up in this thread.
>
> With the Tagtool Think Tankhttp://groups.google.com/group/tagtool/browse_thread/thread/32264809f...

ii...@strukt.at

unread,
Apr 20, 2009, 11:48:52 AM4/20/09
to Tagtool
interesting ideas about using modules ..

http://www.vsxu.com/about.e

VSXu is a free tool to visually connect modules that perform rendering
and mathematical tasks to build graphical effects that react to music
in realtime or sequenced over time. The GUI is written from scratch
entirely in OpenGL (+GLUT) which enables a highly dynamic UI. VSXu is
also useful for learning OpenGL as it exposes many concepts in a very
comprehendable way.

GUI features:

* Hardware-accelerated smooth 3d GUI (zoom in out)
* Smooth controllers (knobs/sliders/spheres) to set parameters
* Project organizing features, macro abstraction
* Hypertree module browser
* Interactive help system (Luna)

Engine/module features:

* GLSL shader support
* particle systems
* mesh loading (OBJ) files
* TrueType vector font rendering (through FreeType)
* Render to texture, feedback loops
* PNG and JPEG loading
* Texture transformations
* Most of the basic OpenGL functions, cameras, texture transforms,
lighting and material parameters, backface culling
* A multitude of mathematical functions including boolean
operations
* Spline interpolation
* Sound module that produces FFT values that can control visuals
in realtime
* Sound playback module (using Fmod) that can play mp3's and
modules (xm, s3m, mod)

palmieri, ricardo

unread,
Apr 20, 2009, 1:09:03 PM4/20/09
to tag...@googlegroups.com
hi fellows... nice ideias, nice projects...


i talked with pi and maki when they was in brasil, to do somethng using PureData. its cross platform, the objects are written in C#, you can build modules using all the libraries of pd-exteded and using GEM to do output you will work with opengl.

im not a deeple programmer, but im a patcher. i like the philosophie of patches in TT(nodekit), and i would like to help to rebuild the modules using PD.

lets rock guys... nice conversation.

cheers

palm

2009/4/20 <ii...@strukt.at>



--
ricardo palmieri
mobile# +551185833173
[palm.estudiolivre.org]
[myspace.com/livenoisetupi]
[skype:palmieriricardo]
[msn: ricardo...@bol.com.br]
[linux user # 392484]



:(){:|:&};:

"A renúncia é a libertação. Não querer é poder."
Fernando Pessoa

fan...@gmx.at

unread,
Apr 20, 2009, 2:31:40 PM4/20/09
to Tagtool
Hi,

i think with PD/GEM and some externals you are able to rebuild the
nodekit but i am also thinking about that some people want to use the
tagtool as simple as possible.. and in that case i think pd/gem is
maybe a bit to complicated.. I think something like the current
tagtool with only a hand full of nodeObjects would be easier to use
for people who don't have any experience in Programming. But therefore
would be great to have a good documentation of the nodeObjects.

on the other side it would be simpler to rebuild the tagtool with help
of pd and gem because you can process midi/audio/video/pictures.... as
you like and you are more flexible to change the behaivior how you
want..

so... i think for people who are familiar with PD/GEM would be great
to develop a "Tagtool Framework for PD/GEM" because you could extend
this framework for all the things you need..
for people who are not experienced with PD/GEM or Programming it would
be easier to use a very simple node based userinterface wich is well
documented.. i think the current userinterface of the tagtool is
alright, could be a bit simpler for users which are not familiar with
programming and a big thing what i am missing at the current tagtool
version is the documentation off the nodes.

best,
CakaG



On 20 Apr., 19:09, "palmieri, ricardo" <ricardopalmi...@gmail.com>
wrote:
> hi fellows... nice ideias, nice projects...
>
> i talked with pi and maki when they was in brasil, to do somethng using
> PureData. its cross platform, the objects are written in C#, you can build
> modules using all the libraries of pd-exteded and using GEM to do output you
> will work with opengl.
>
> im not a deeple programmer, but im a patcher. i like the philosophie of
> patches in TT(nodekit), and i would like to help to rebuild the modules
> using PD.
>
> lets rock guys... nice conversation.
>
> cheers
>
> palm
>
> 2009/4/20 <i...@strukt.at>
> [msn: ricardopalmi...@bol.com.br]

Lukas Gruber

unread,
Apr 21, 2009, 4:29:22 AM4/21/09
to tag...@googlegroups.com
Hi,

a few notes:

* refactor the core of tagtool (c or cpp) and create an easy to
integrate open project
- core: mapping and calibration of the wacom tablet
- core: handling of dynamic vertex buffers (v,n,t,c) plus layered
structure plus animation and make it accessible through simple interfaces
- core: maybe optionally ... some internal shader program system
(think of geometry shader :-) ) to speed up ... but .. will always
conflict with other frameworks
- not core: any kind of windowing or platform specific code

then let it out to the community ... should take one day to integrate it
then to pd/gem

* extend the notekid / tagtool .. and make it save for use .. artists
who are not developers would love it for its robustness
- add / finish / polish what is already there
- add sublayers for programming .. no visual programming without
sublayers
- maybe create a standalone mode, which does the basic drawing
- grab a designer to overwork the nodekits itself .. they take too
much visual space .. could be simpler (look at pd or max );
- add an extra layer hierarchy window like in photoshop, to quickly
add/delete/remove/select layers ...
- ... many more ideas ..

as far as I remember the nodekit code is not that bad ... and c-sharp
should emphasize rapid development too ... so do not throw it
away :-)

I think lots of artists would love to pay a certain amount of money to
get a STABLE and EASY to use drawing tool.
You could ask for the donation first ... and then hire a good software
developer :-)

~lux
--
MSc. Lukas Gruber
Institute for Computer Graphics and Vision,
Graz University of Technology,
Inffeldgasse 16a / 2. Floor, A-8010 Graz
Phone: +43 (0)316-873-5072
Mail : lgr...@icg.tugraz.at
http://studierstube.icg.tu-graz.ac.at/handheld_ar/

pi

unread,
Apr 23, 2009, 3:57:35 PM4/23/09
to tag...@googlegroups.com
hi all,

tomorrow the tag tool think tank [s]tarts.

if you have ideas or expectations either

technically or artistic ... write them down


# ideas for the new version

*) is it really useful to contruct tagtool
paintings as quad-stripes? shouldn't it be
a bitmap manipulation instead?

*) instead of shaders a "brush-engine" could
provide more aesthetical possibilities and
would be quiet low-cost, seen from cpu side.
in this module a scripting language, giving
users the possibility to generate their own
set of brushes would be great. maybe python like:
http://www.goghproject.com/

*) wouldn't it be great if you fire up tagtool
like a console game? complex configuration's
could happen in xml style tagtoolrc's. writing
a stable messaging system is much work, especially
integrating all the interfaces needed and there
is already such nice software available. but
compressing the tagtool application on it's core
functions and compiling it as a pd external would
make perfectly sense. anyway, full osc/midi output
for all messages happening on the interface side
is crucial.

these are just some thoughts and highly RFC!

cheers

pi

Lukas Gruber

unread,
Apr 23, 2009, 4:20:49 PM4/23/09
to tag...@googlegroups.com
pi wrote:
> hi all,
>
> tomorrow the tag tool think tank [s]tarts.
>
> if you have ideas or expectations either
>
> technically or artistic ... write them down
>
>
> # ideas for the new version
>
> *) is it really useful to contruct tagtool
> paintings as quad-stripes? shouldn't it be
> a bitmap manipulation instead?
>
>
maybe both? how would you animate the bitmap then? you could
map the brush on the quad :-)
> *) instead of shaders a "brush-engine" could
> provide more aesthetical possibilities and
> would be quiet low-cost, seen from cpu side.
>
shaders are gpu .. so cpu is more or less free

> in this module a scripting language, giving
> users the possibility to generate their own
> set of brushes would be great. maybe python like:
> http://www.goghproject.com/
>
>
hm .. in a later step .. you can always draw on a fullscreen quad to have
the canvas for the "brush-engine"
> *) wouldn't it be great if you fire up tagtool
> like a console game?
even console games are unstable :-)
as far as I noticed .. creating complex scenes needs a bit of administration
so some layer windows visualizing your graphics (selectable ) would be a
good choice

> complex configuration's
> could happen in xml style tagtoolrc's.

> writing
> a stable messaging system is much work, especially
> integrating all the interfaces needed and there
> is already such nice software available. but
> compressing the tagtool application on it's core
> functions and compiling it as a pd external would
> make perfectly sense.

full ack
> anyway, full osc/midi output
> for all messages happening on the interface side
> is crucial.
>
>
> these are just some thoughts and highly RFC!
>
> cheers
>
> pi
>
>
>
>
~lux

samth

unread,
Apr 23, 2009, 4:55:11 PM4/23/09
to Tagtool
Hi,
Amazing!
It's already two years since we have this Tagtool Group!
And you guys is moving forward to the next step, the next generation
Tagtool Application.

I agreed with Lukas regarding:

>>as far as I remember the nodekit code is not that bad ... and c-sharp
>>should emphasize rapid development too ... so do not throw it
>>away :-)

And also CaKaG's comments:
>>I think something like the current
>>tagtool with only a hand full of nodeObjects would be easier to use
>>for people who don't have any experience in Programming. But therefore
>>would be great to have a good documentation of the nodeObjects.

I don't know much about the Software Development, to give you guys the
idea or comments.
I can only write the code to add my custom Node to the Nodekit.

Any how, there is only one wish that I would like to see in the Next
Generation Nodekit is
the capability of record the animation into the movie file format i.e
*.wav, mpeg, etc. directly,
without having to put all the series of still images into the program
like Premiere Pro
or something similar to produce the movie file.

Looking forward to hearing about the outcome of the Tagtool Think Tank
09,
and loonig forward to joining the next meeting,

Cheers,

gelerli

unread,
Apr 28, 2009, 5:50:46 AM4/28/09
to Tagtool
do you guys think that merging layers is gonna be possible? in our
experiments with tagtool i kinda felt this need to merge some of the
layers to make the moving across the layers easier....
Reply all
Reply to author
Forward
0 new messages