Discussion:
[Jamin] gtk3/clutter UI
Patrick Shirkey
2013-02-10 04:23:04 UTC
Permalink
Hi,

I ran a few tests of the various options for getting clutter and gtk3 into
the gtk2 UI and the results were not very good. So I spent some time now
(about 8 hours so far) migrating the existing code to gtk3 and adding a
basic clutter stage.

After a few hours of pain I have managed to get the interface to build and
run. Although I had to disable the hdeq, compressor graphs and meters as
well as make a few guesses about the layout of the grid columns which is
going to be a bit of a hassle to rebuild :-(

The good news is that it had to be done at some point if we want to keep
moving forward so it seems there is no time like the present.

The question now is should I spin up a new tree for this version or is it
ok to go ahead and update the code to the existing repo. Given that we
haven't been doing very much to jamin for the past few years I'm not
fussed if the old gui is completely replaced by a new shiny gtk3/clutter
version.

However this might be a good opportunity to move to svn too.

FYI, the benefits of clutter are opengl support, touch interface support,
UI animation support and that it frees us from the gtk h/vbox lock in by
allowing us to place any size or shape widgets at any location on the
x/y/z axis.

I have already been doing some work with it for the CMKeyboard project in
python. It's very powerful and a lot of fun too if you like UI dev :-)



Cheers

--
Patrick Shirkey
Boost Hardware Ltd
Jan Depner
2013-02-10 08:19:12 UTC
Permalink
Post by Patrick Shirkey
Hi,
I ran a few tests of the various options for getting clutter and gtk3 into
the gtk2 UI and the results were not very good. So I spent some time now
(about 8 hours so far) migrating the existing code to gtk3 and adding a
basic clutter stage.
After a few hours of pain I have managed to get the interface to build and
run. Although I had to disable the hdeq, compressor graphs and meters as
well as make a few guesses about the layout of the grid columns which is
going to be a bit of a hassle to rebuild :-(
The good news is that it had to be done at some point if we want to keep
moving forward so it seems there is no time like the present.
The question now is should I spin up a new tree for this version or is it
ok to go ahead and update the code to the existing repo. Given that we
haven't been doing very much to jamin for the past few years I'm not
fussed if the old gui is completely replaced by a new shiny gtk3/clutter
version.
However this might be a good opportunity to move to svn too.
FYI, the benefits of clutter are opengl support, touch interface support,
UI animation support and that it frees us from the gtk h/vbox lock in by
allowing us to place any size or shape widgets at any location on the
x/y/z axis.
I have already been doing some work with it for the CMKeyboard project in
python. It's very powerful and a lot of fun too if you like UI dev :-)
If you're going to go through all that agony why not just rewrite it
using Qt? Qt 4.8 is fine, I'm not so sure about 5.0. In either case
the documentation is about 100 times better than anything GTK. I've
personally ported about 30 applications from C/Motif/GTK/ANYGUI to C
++/Qt with no major problems. And Qt is now under LGPL!
Post by Patrick Shirkey
Cheers
--
Patrick Shirkey
Boost Hardware Ltd
------------------------------------------------------------------------------
Free Next-Gen Firewall Hardware Offer
Buy your Sophos next-gen firewall before the end March 2013
and get the hardware for free! Learn more.
http://p.sf.net/sfu/sophos-d2d-feb
_______________________________________________
Jamin-devel mailing list
https://lists.sourceforge.net/lists/listinfo/jamin-devel
Patrick Shirkey
2013-02-10 13:09:51 UTC
Permalink
Post by Jan Depner
Post by Patrick Shirkey
Hi,
I ran a few tests of the various options for getting clutter and gtk3 into
the gtk2 UI and the results were not very good. So I spent some time now
(about 8 hours so far) migrating the existing code to gtk3 and adding a
basic clutter stage.
After a few hours of pain I have managed to get the interface to build and
run. Although I had to disable the hdeq, compressor graphs and meters as
well as make a few guesses about the layout of the grid columns which is
going to be a bit of a hassle to rebuild :-(
The good news is that it had to be done at some point if we want to keep
moving forward so it seems there is no time like the present.
The question now is should I spin up a new tree for this version or is it
ok to go ahead and update the code to the existing repo. Given that we
haven't been doing very much to jamin for the past few years I'm not
fussed if the old gui is completely replaced by a new shiny gtk3/clutter
version.
However this might be a good opportunity to move to svn too.
FYI, the benefits of clutter are opengl support, touch interface support,
UI animation support and that it frees us from the gtk h/vbox lock in by
allowing us to place any size or shape widgets at any location on the
x/y/z axis.
I have already been doing some work with it for the CMKeyboard project in
python. It's very powerful and a lot of fun too if you like UI dev :-)
If you're going to go through all that agony why not just rewrite it
using Qt? Qt 4.8 is fine, I'm not so sure about 5.0. In either case
the documentation is about 100 times better than anything GTK. I've
personally ported about 30 applications from C/Motif/GTK/ANYGUI to C
++/Qt with no major problems. And Qt is now under LGPL!
QT is under the control of Microsoft and I just don't trust that company :-)

Plus I have a long term commitment to GTK and now Clutter too. Clutter
development is partially funded by Intel and it is used for the new
Wayland desktop which is being rolled out to every distro over the coming
year.

It's 8 hours work so far. After it is done JAMin will once again be a
thoroughly modern interface showcasing the state of the art for
GTK3/Clutter and Linux Audio Apps.




--
Patrick Shirkey
Boost Hardware Ltd
Alexandre Prokoudine
2013-02-10 13:58:02 UTC
Permalink
Post by Patrick Shirkey
QT is under the control of Microsoft
What have you been smoking?

Alexandre Prokoudine
http://libregraphicsworld.org
John Rigg
2013-02-10 15:33:10 UTC
Permalink
Post by Alexandre Prokoudine
Post by Patrick Shirkey
QT is under the control of Microsoft
What have you been smoking?
Control is probably a stronger term than I would use, but Digia is a
Microsoft "partner".

John
Alexandre Prokoudine
2013-02-10 15:34:30 UTC
Permalink
Post by John Rigg
Post by Alexandre Prokoudine
Post by Patrick Shirkey
QT is under the control of Microsoft
What have you been smoking?
Control is probably a stronger term than I would use, but Digia is a
Microsoft "partner".
Well? They also have Oracle and IBM among partners.

I used to work for a corporation that had (and still has) Microsoft's
golden partner status. That didn't stop the corporation from growing a
strong Linux subdepartment. RHEL is quite OK whet in it comes to
margins :)

But seriously, of all the reasons to stick to GTK+ a conspiracy theory... :)

Alexandre Prokoudine
http://libregraphicsworld.org
Patrick Shirkey
2013-02-10 18:35:58 UTC
Permalink
Post by Alexandre Prokoudine
Post by John Rigg
Post by Alexandre Prokoudine
Post by Patrick Shirkey
QT is under the control of Microsoft
What have you been smoking?
Control is probably a stronger term than I would use, but Digia is a
Microsoft "partner".
Well? They also have Oracle and IBM among partners.
Paragons of Freedom
Post by Alexandre Prokoudine
I used to work for a corporation that had (and still has) Microsoft's
golden partner status. That didn't stop the corporation from growing a
strong Linux subdepartment. RHEL is quite OK whet in it comes to
margins :)
But seriously, of all the reasons to stick to GTK+ a conspiracy theory... :)
I was simply pointing out a basic fact.

There are plenty of other reasons to invest time and effort into GTK3 and
Clutter.

However if you want to make a QT version of jAMin and port it to C++ at
the same time there is nothing to stop you from doing so. But in that case
I would definitely suggest we spin up a new repo tree and host it there.

We could even have a competition to see which tree grows the fastest and
offers the most powerful interface to the end user :-)

It's all good but I'll put my time into the GTK3/Clutter interface.





--
Patrick Shirkey
Boost Hardware Ltd
Jan Depner
2013-02-11 11:45:20 UTC
Permalink
Post by Patrick Shirkey
Post by Jan Depner
Post by Patrick Shirkey
Hi,
I ran a few tests of the various options for getting clutter and gtk3 into
the gtk2 UI and the results were not very good. So I spent some time now
(about 8 hours so far) migrating the existing code to gtk3 and adding a
basic clutter stage.
After a few hours of pain I have managed to get the interface to build and
run. Although I had to disable the hdeq, compressor graphs and meters as
well as make a few guesses about the layout of the grid columns which is
going to be a bit of a hassle to rebuild :-(
The good news is that it had to be done at some point if we want to keep
moving forward so it seems there is no time like the present.
The question now is should I spin up a new tree for this version or is it
ok to go ahead and update the code to the existing repo. Given that we
haven't been doing very much to jamin for the past few years I'm not
fussed if the old gui is completely replaced by a new shiny gtk3/clutter
version.
However this might be a good opportunity to move to svn too.
FYI, the benefits of clutter are opengl support, touch interface support,
UI animation support and that it frees us from the gtk h/vbox lock in by
allowing us to place any size or shape widgets at any location on the
x/y/z axis.
I have already been doing some work with it for the CMKeyboard project in
python. It's very powerful and a lot of fun too if you like UI dev :-)
If you're going to go through all that agony why not just rewrite it
using Qt? Qt 4.8 is fine, I'm not so sure about 5.0. In either case
the documentation is about 100 times better than anything GTK. I've
personally ported about 30 applications from C/Motif/GTK/ANYGUI to C
++/Qt with no major problems. And Qt is now under LGPL!
QT is under the control of Microsoft and I just don't trust that company :-)
Qt is not under Microsoft's control. It is now being developed by
Digia but, since it is LGPL, it is not under anyone's control. It can
be forked anytime anyone wants to fork it.
Post by Patrick Shirkey
Plus I have a long term commitment to GTK and now Clutter too. Clutter
development is partially funded by Intel and it is used for the new
Wayland desktop which is being rolled out to every distro over the coming
year.
It's 8 hours work so far. After it is done JAMin will once again be a
thoroughly modern interface showcasing the state of the art for
GTK3/Clutter and Linux Audio Apps.
--
Patrick Shirkey
Boost Hardware Ltd
------------------------------------------------------------------------------
Free Next-Gen Firewall Hardware Offer
Buy your Sophos next-gen firewall before the end March 2013
and get the hardware for free! Learn more.
http://p.sf.net/sfu/sophos-d2d-feb
_______________________________________________
Jamin-devel mailing list
https://lists.sourceforge.net/lists/listinfo/jamin-devel
Patrick Shirkey
2013-02-11 16:19:52 UTC
Permalink
Post by Jan Depner
Post by Patrick Shirkey
Post by Jan Depner
Post by Patrick Shirkey
Hi,
I ran a few tests of the various options for getting clutter and gtk3 into
the gtk2 UI and the results were not very good. So I spent some time
now
Post by Jan Depner
Post by Patrick Shirkey
(about 8 hours so far) migrating the existing code to gtk3 and adding
a
Post by Jan Depner
Post by Patrick Shirkey
basic clutter stage.
After a few hours of pain I have managed to get the interface to
build
Post by Jan Depner
Post by Patrick Shirkey
and
run. Although I had to disable the hdeq, compressor graphs and meters
as
Post by Jan Depner
Post by Patrick Shirkey
well as make a few guesses about the layout of the grid columns which
is
Post by Jan Depner
Post by Patrick Shirkey
going to be a bit of a hassle to rebuild :-(
The good news is that it had to be done at some point if we want to
keep
Post by Jan Depner
Post by Patrick Shirkey
moving forward so it seems there is no time like the present.
The question now is should I spin up a new tree for this version or
is
Post by Jan Depner
Post by Patrick Shirkey
it
ok to go ahead and update the code to the existing repo. Given that
we
Post by Jan Depner
Post by Patrick Shirkey
haven't been doing very much to jamin for the past few years I'm not
fussed if the old gui is completely replaced by a new shiny
gtk3/clutter
Post by Jan Depner
Post by Patrick Shirkey
version.
However this might be a good opportunity to move to svn too.
FYI, the benefits of clutter are opengl support, touch interface support,
UI animation support and that it frees us from the gtk h/vbox lock in
by
Post by Jan Depner
Post by Patrick Shirkey
allowing us to place any size or shape widgets at any location on the
x/y/z axis.
I have already been doing some work with it for the CMKeyboard
project
Post by Jan Depner
Post by Patrick Shirkey
in
python. It's very powerful and a lot of fun too if you like UI dev
:-)
Post by Jan Depner
If you're going to go through all that agony why not just rewrite
it
Post by Jan Depner
using Qt? Qt 4.8 is fine, I'm not so sure about 5.0. In either case
the documentation is about 100 times better than anything GTK. I've
personally ported about 30 applications from C/Motif/GTK/ANYGUI to C
++/Qt with no major problems. And Qt is now under LGPL!
QT is under the control of Microsoft and I just don't trust that company :-)
Qt is not under Microsoft's control. It is now being developed by
Digia but, since it is LGPL, it is not under anyone's control. It can
be forked anytime anyone wants to fork it.
Realistically who is going to do that when we have GTK?

Digia get a significant amount of funding from Microsoft and are therefore
beholden to their corporate agenda. The same happens to AMD among several
other companies. Over the years Microsoft has paid alot of people to
(attempt to) sabotage various projects that they could get their claws
into. This is well documented and not a surprise, theory or figment of my
imagination.

I choose not to support Microsoft or Microsoft partners so as I have a
choice between QT and GTK, I choose GTK. Everyone else is free do do what
they want and there is certainly no reason that JAMin cannot be ported to
c++ and QT by anyone who has the motivation to do so. Just not me :-)

Currently I have all the default widgets ported and displaying correctly.
I am going to port the fastmeter widget from ardour/guitarix to gtk3
tonight and then I will get stuck into the hdeq/spectrum/comp curves code.
Yesterday I disabled all the ui specific code in the hdeq.c/h and the
files are building successfully.

Probably will all be done by the end of the week if I don't get distracted.

The question is should I create a new repo/tree/branch or just update the
existing cvs with the new codebase? It a lot easier to update the existing
codebase but if anyone thinks they would like to continue developing the
gtk2 interface then we should keep it as a separate project.
Post by Jan Depner
Post by Patrick Shirkey
Plus I have a long term commitment to GTK and now Clutter too. Clutter
development is partially funded by Intel and it is used for the new
Wayland desktop which is being rolled out to every distro over the coming
year.
It's 8 hours work so far. After it is done JAMin will once again be a
thoroughly modern interface showcasing the state of the art for
GTK3/Clutter and Linux Audio Apps.
--
Patrick Shirkey
Boost Hardware Ltd
Alexandre Prokoudine
2013-02-11 16:30:35 UTC
Permalink
Post by Patrick Shirkey
Digia get a significant amount of funding from Microsoft and are therefore
beholden to their corporate agenda. The same happens to AMD among several
other companies. Over the years Microsoft has paid alot of people to
(attempt to) sabotage various projects that they could get their claws
into. This is well documented and not a surprise, theory or figment of my
imagination.
Noooo, of course it's not :D
Post by Patrick Shirkey
The question is should I create a new repo/tree/branch or just update the
existing cvs with the new codebase?
CVS was supposed to be disabled on SourceForge eventually. If you
don't really need CVS all that much, and you don't wish to fiddle with
upgrading to e.g. SVN (the procedure is crazy), maybe you could just
drop the revision history and start a clean SVN/Git/whatever repo?

Alexandre Prokoudine
http://libregraphicsworld.org
Patrick Shirkey
2013-02-11 16:54:18 UTC
Permalink
Post by Alexandre Prokoudine
Post by Patrick Shirkey
Digia get a significant amount of funding from Microsoft and are therefore
beholden to their corporate agenda. The same happens to AMD among several
other companies. Over the years Microsoft has paid alot of people to
(attempt to) sabotage various projects that they could get their claws
into. This is well documented and not a surprise, theory or figment of my
imagination.
Noooo, of course it's not :D
Just do some basic research and you will find that your beloved Microsoft
has done plenty of things to attack Linux and Open Source over the years.
They pay people a lot of money to work on their behalf and "poison the
well" at every opportunity. For example, who do you think the SCO case was
funded by? What is the real point of UEFI? AMD's inability to cooperate
with the open source multimedia community. Nokia dropping all open source
development in favour of Windows 8/RT...

These are just a few things that spring to mind. The list is long and
pretty frustrating so if you are not interested in it then don't worry
your pretty little head about it.
Post by Alexandre Prokoudine
Post by Patrick Shirkey
The question is should I create a new repo/tree/branch or just update the
existing cvs with the new codebase?
CVS was supposed to be disabled on SourceForge eventually. If you
don't really need CVS all that much, and you don't wish to fiddle with
upgrading to e.g. SVN (the procedure is crazy), maybe you could just
drop the revision history and start a clean SVN/Git/whatever repo?
If no one else has any opinion then I will spin up a new svn or git repo.
SF is offering git these days too.



--
Patrick Shirkey
Boost Hardware Ltd
Alexandre Prokoudine
2013-02-11 17:02:26 UTC
Permalink
Post by Patrick Shirkey
Just do some basic research and you will find that your beloved Microsoft
Low blow, Patrick. Low blow. It's as much mine as yours.
Post by Patrick Shirkey
has done plenty of things to attack Linux and Open Source over the years.
I wasn't talking about "over the years", I was merely referring to Digia.

Although it's quite fun to see you attempting to patronize me :) Very cute :)
Post by Patrick Shirkey
Post by Alexandre Prokoudine
CVS was supposed to be disabled on SourceForge eventually. If you
don't really need CVS all that much, and you don't wish to fiddle with
upgrading to e.g. SVN (the procedure is crazy), maybe you could just
drop the revision history and start a clean SVN/Git/whatever repo?
If no one else has any opinion then I will spin up a new svn or git repo.
SF is offering git these days too.
Jolly good. Although since this is about Git now, you probably would
want using something like Github.

Alexandre Prokoudine
http://libregraphicsworld.org
Patrick Shirkey
2013-02-11 17:05:47 UTC
Permalink
Post by Alexandre Prokoudine
Post by Patrick Shirkey
Just do some basic research and you will find that your beloved Microsoft
Low blow, Patrick. Low blow. It's as much mine as yours.
Post by Patrick Shirkey
has done plenty of things to attack Linux and Open Source over the years.
I wasn't talking about "over the years", I was merely referring to Digia.
Although it's quite fun to see you attempting to patronize me :) Very cute :)
Glad you liked it.
Post by Alexandre Prokoudine
Post by Patrick Shirkey
Post by Alexandre Prokoudine
CVS was supposed to be disabled on SourceForge eventually. If you
don't really need CVS all that much, and you don't wish to fiddle with
upgrading to e.g. SVN (the procedure is crazy), maybe you could just
drop the revision history and start a clean SVN/Git/whatever repo?
If no one else has any opinion then I will spin up a new svn or git repo.
SF is offering git these days too.
Jolly good. Although since this is about Git now, you probably would
want using something like Github.
I don't think that is necessary. We already have everything else setup and
running on SF.net for the past 10 years now.



--
Patrick Shirkey
Boost Hardware Ltd

Loading...