Hello there.

I'm Cláudio Pinheiro and I've been lurking here for some time (maybe to
grab some courage to contribute).
I've been fiddling on and off with LMMS (I use some other DAWs for my
stuff), and I feel it is already a great accomplishment. Of course there's
still a lot to do, but no software is ever complete anyway. Tous mes
compliments!
I'd like to ask some questions, if you don't mind. I beg you pardon in
advance if some of my questions and comments may sound a bit off(ensive),
please take them in a relaxed, good spirit as if coming from a close friend
(that we aren't - yet) and with the best intentions. Pardon me too for my
poor, non-native English grammar skills, and excuse me for mentioning other
DAWs once in a while.

Let's start:


Can I use Qt5 or should I stick to Qt4?
What's up with those Plastik decorations? Are they themeable?
Is there any progress with MIDI IN for OS X/Windows?
Any plans for vectorial widgets and/or HiDPI support?
Any plans for more platforms?

Some things to take in consideration:
More platforms may be good. Caustic supports Win/Lin/OSX/iOS/Droid (the
last two are paid versions), what's good but the software is a tad limited,
something more powerful would be nice. LMMS may fit and even go way beyond.
iOS development now is a lot easier witth XCode 7 (currently beta), as it
allows a developer to test an iOS program in the real hardware without the
need of an Apple Developer license. Android development tools are readily
available. Qt integrates well with both. iOS uses CoreAudio just as OS X,
Android uses tinyalsa, what makes easier to port RtAudio (if it's not
already working). The tools are right here for us to use.
Multiple platforms have different form factors, resolutions and such (duh).
To be shown in a meaningful usable way LMMS must adapt to such multitude of
environments. 4K displays are becoming popular, and multiple-display setups
are common in songwriting/production. Mobile songwriting is common nowadays
too (Garage Band for iOS, FL Studio Mobile, Caustic, Xewton MusicStudio and
so on), but we still have no easy way to be able to accommodate to both use
cases while preserving feature parity and good usability. Mobile DAW apps
are still "toys" for most purposes, lacking firepower and missing
resources. This shouldn't be the case anymore. LMMS may change it for good.
LMMS should plan for HiDPI. A vectorial UI would help this a lot. KDE uses
SVG for it (with cached PNG images). Qt support is planned for 5.6. 5.5 has
partial support.
Qt4 will not receive updates. CopperSpice may be a dead end. Qt5 is where
new development is coming from, so maybe it's time to embrace it
wholeheartedly, not just making LMMS compatible with Qt5, but putting its
"new" features to a good use. I don't know the status of the current LMMS
code in these matters, but I'm somewhat assuming there's a lot of legacy
because of the Plastik decorations. Come on. LMMS, UI-wise, is... "lovely".
Let's spell it: "eff", "you", "gee", "ell", "why". "Lovely", as in "beaten
with the lovely stick". It doesn't diminish its capabilities as a DAW, but
an UI overhaul is long overdue. Plastik is KDE3. I was unable to find a
theme selector. This UI rant is not for the sake of ranting, but because
changing the UI of such a complex project like LMMS is a complex task, but
one that would benefit LMMS and its users a lot. Qt5 offers UI HW
acceleration, theming and animations with Qt Quick (faster than Qt
Widgets). Changes in UI can be tested without recompiling the code(!).
Theming is easier to implement, and an MVC model with LMMS core separated
from the Views would allow better portability to other devices (different
UIs for different form factors).
Another point is feature parity. Currently LMMS fails to act as expected in
non-Linux platforms when dealing with MIDI, and displaying ZynAddSubFx.
Audio on Windows is tricky (the device/backend selector behaves oddly, not
saving settings in some circumstances, preventing LMMS from generating
audio and sticking users in a "restart program" hell). I know these issues
depend on other people, but enforcing feature parity (when reasonable) will
lead to a smoother user experience across platforms.
There are other places of improvement and innovation: Undo/Redo, seamless
editing across platforms (maybe allowing the use of some network storage
solution like Dropbox, Mega to enable automatic project syncing), AVB/TSN,
project spanned between multiple computers on a network, parallel editing,
automatic channel freezing to allow editing more complex sessions on a less
capable device, automatic plugin discovery/mangling/selection (being able
to use the same plugin as an AU on Mac and a VST on Windows), downloadable
content (audio patches, samples, soundfonts, themes) and so on.

I'll try to compile it for OS X and Windows this weekend, and do some
exploration. If the idea of iOS and Android versions flourish I'll be able
to put my hands on it too.


Best regards,


Cláudio Pinheiro

2015-06-12 13:06 GMT-03:00 Dave French <dave.fren...@googlemail.com>:

> Sufian
>
> Nice to see your intrested :)
>
> We also have an irc channel on freenode #LMMS, if you have any questions.
>
> Welcome.
>
> On 12 June 2015 at 13:45, Tres Finocchiaro <tres.finocchi...@gmail.com>
> wrote:
>
>> @Sufian,
>>
>> Thanks for chiming in.  We need the help!
>>
>> We use QT for our GUI (and many other items).
>>
>>
>>    - If you're an IDE-type person, Dave French @curlymorphic had a
>>    YouTube video he sent me a while back which outlined how to configure Qt
>>    Creator and CMake to get LMMS to build in an IDE.
>>    - If you're a kate/gedit/vi/emacs type coder, you can get started
>>    right away with the tutorials that @musikBear linked above
>>
>> In regards to OS... are you on Linux, Mac or Windows?
>>
>>    - Windows doesn't have a working build tutorial yet.  I tried for a
>>    few days and was unsuccessful
>>    <https://github.com/tresf/lmms/wiki/Compiling-lmms-(on-Windows)>.
>>    Linux is currently recommended.  Ubuntu is the quickest platform to get
>>    started with.
>>    - Linux building is pretty straight forward on all distros so as long
>>    as you know your way around the package managers to meet dependencies.
>>    Debian and Ubuntu are pretty much a copy and paste of commands to get
>>    started.
>>    - Mac building is pretty straight forward as well.  Homebrew will
>>    probably get you started faster than MacPorts, but our tutorial is still
>>    written around MacPorts, so you'd be fastest looking at this travis
>>    recipe
>>    <https://github.com/LMMS/lmms/blob/master/.travis/osx..install.sh>.
>>
>> In regards to patches, we use Pull Requests exclusively now.
>>
>>    - If you're not familiar with PRs and GitHub, this might be your
>>    biggest battle at first
>>       - They allow us to comment and make tweaks before patches are
>>       merged
>>    - If you're already familiar with GitHub pull requests, you can get
>>    started now, just fork our main branch, master preferably.
>>
>> In regards to what work to do... that is often the second biggest
>> battle...
>>
>>    - Our bug tracker on GitHub is a good place to start.  Just post a
>>    comment before starting if unsure.
>>    - Development is slow at the moment, so there isn't a whole lot of
>>    toes to be stepping on, so to speak.
>>    - GUI vs. Core/DSP
>>       - Depending on your expertise, we have many areas in both GUI and
>>       the core digital signal processing areas which need help. We can point 
>> you
>>       in the right direction if needed.
>>    - Cumulative bug reports -- We have some large bug reports that try
>>    to organize some features and bugs into a centralized area... here they 
>> are:
>>       - Better MIDI support:  https://github.com/LMMS/lmms/issues/1472
>>       - Better Sample Track support:
>>       https://github.com/LMMS/lmms/issues/1471
>>       - Better Jack support:  https://github.com/LMMS/lmms/issues/1467
>>       - New UI / Better Transport:
>>       https://github.com/LMMS/lmms/issues/1911,
>>       https://github.com/LMMS/lmms/issues/1357
>>       - Better Color Handling: https://github.com/LMMS/lmms/issues/1665
>>
>> Hopefully that's enough to get you started.  Welcome. :)
>>
>> -Tres
>>
>>
>>
>> - tres.finocchi...@gmail.com
>>
>> On Fri, Jun 12, 2015 at 6:19 AM, musikbear <mkru...@hotmail.com> wrote:
>>
>>> This link
>>> https://lmms.io/get-involved/
>>> under 'Develop'
>>> And this from the wiki
>>> https://github.com/LMMS/lmms/wiki
>>> about creating your own fork and repository, should give you first
>>> ideas.  I
>>> am not sure everything is fully updated, but someone else will chime in,
>>> if
>>> so :)
>>>
>>>
>>>
>>> --
>>> View this message in context:
>>> http://linux-multimedia-studio-lmms.996328.n3.nabble.com/How-to-get-started-tp11795p11796.html
>>> Sent from the lmms-devel mailing list archive at Nabble.com.
>>>
>>>
>>> ------------------------------------------------------------------------------
>>> _______________________________________________
>>> LMMS-devel mailing list
>>> LMMS-devel@lists.sourceforge.net
>>> https://lists.sourceforge.net/lists/listinfo/lmms-devel
>>>
>>
>>
>>
>> ------------------------------------------------------------------------------
>>
>> _______________________________________________
>> LMMS-devel mailing list
>> LMMS-devel@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/lmms-devel
>>
>>
>
>
> ------------------------------------------------------------------------------
>
> _______________________________________________
> LMMS-devel mailing list
> LMMS-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/lmms-devel
>
>
------------------------------------------------------------------------------
_______________________________________________
LMMS-devel mailing list
LMMS-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/lmms-devel

Reply via email to