Hello all,

 Thanks for your feedback. I submitted my application last night. At Mr. 
Santoni's suggestion I made my proposal to rewrite the waveform display. 
I look forward to potentially working with you this summer :).

Thanks again,
RJ Ryan



Adam Davison wrote:
> Hi Russell,
>
> Albert gave a fairly complete response there I guess but I just
> thought I'd chime in and say that given your rather in depth
> experience with Qt and so on perhaps applying to rewrite the waveform
> display would be worthwhile. This is something that definitely needs
> doing, should be reasonably self-contained and educational for
> everyone involved. So out of all of those I'd put my vote in that
> direction.
>
> Adam
>
> On 31/03/2008, Albert Santoni <[EMAIL PROTECTED]> wrote:
>   
>> Hi Russell,
>>
>>
>>  On Sun, 2008-03-30 at 22:15 -0400, Russell Ryan wrote:
>>  > Hi Mixxx-devel,
>>  >
>>  >    My name is RJ Ryan. I'm a junior at MIT, amateur DJ, and user of
>>  > Mixxx. I was glad to see your name on the list of mentoring
>>  > organizations for GSOC. I'd really like to help out this summer and
>>  > beyond, and I feel like GSOC would be a good place to get started. Over
>>  > the past week I've been going over the code, and it's gotten me thinking
>>  > about various projects and features I would love to see built in to
>>  > Mixxx (apart from the ones on the ideas page)
>>  >
>>  > I'd like some feedback as to which of these ideas are ones you would
>>  > consider valuable, and which are ones that aren't too important. (Some
>>  > are big and some are small) I think you'll find as the list goes on they
>>  > get more and more far-fetched. :) That will help me decide which one (or
>>  > ones) I'll focus on in my application.
>>  >
>>  > - One main thing I see that's missing when compared with other dj'ing
>>  > apps is an effects panel (per channel). A simple pluggable effects tool
>>  > would really bring Mixxx up to speed with proprietary apps and also if
>>  > designed well would blaze a trail for later improvements and possibly
>>  > user-contributed effects (I'm thinking a DSL or scripted plugin setup
>>  > here). (the GSOC project I'd propose would be effects units for either
>>  > channel (or the main mix) with some basic effects you find on mid-level
>>  > mixers like flange, phase, filter, wah, tape-echo, 2xecho, etc. Also
>>  > possibly jet/zip like on the CDJ200.)
>>
>>
>> Effects was implemented in the form of LADSPA support last summer by a
>>  GSoC student (Pawel). The GUI part of it wasn't finished, although he
>>  may try to finish it some time in the next few months.
>>
>>
>>  >
>>  > - A loop toolbox -- fine-grained control over looping, loop-in,
>>  > loop-cut, loop-jump, loop-exit, re-loop, etc. It would include visual
>>  > cues in the waveform output. I'd really love to see these features in
>>  > Mixxx.
>>  >
>>
>>
>> Looping is a great feature, but from the under-the-hood point of view,
>>  it's going to be a big challenge. Looping is more complicated than just
>>  seeking back to some loop point and playing again. When Mixxx seeks, the
>>  waveform view incurs a fairly big performance penalty, which can cause
>>  gaps in the audio when you hit the "cue" button. Looping would have the
>>  same problem, so addressing this performance problem (and probably
>>  others) is something that would have to be taken seriously first. It's
>>  universally agreed that the waveform view needs to be rewritten from
>>  scratch anyways, so perhaps that's a better summer project. It would
>>  open up the door for some cool visual integration of the looping as
>>  well. (It's very awkward to modify the waveform as the code exists right
>>  now.)
>>
>>
>>
>>  > - Tempo-bend.. you already have pitch bend, but resampling so that the
>>  > pitch stays the same while tempo increases is something I'd like to see.
>>  >
>>
>>
>> We already have this when you're using "pitch-independent time stretch"
>>  mode.
>>
>>
>>  > - For usability and new user orientation, a setup wizard that helps with
>>  > MIDI configuration and option choices. (this is similar to the one
>>  > suggested under 'MIDI' on the ideas page)
>>
>>
>> The MIDI configuration/mapping project we listed on our Ideas page
>>  already covers this to some extent.
>>
>>
>>  >
>>  > - Mix recording feature -- I'm a bedroom dj for the most part, and I
>>  > like recording my mixes from time to time. The way I've done it with
>>  > Mixxx is routing jack into an audio recorder, but I think it'd be a nice
>>  > touch if you could choose to record from within Mixxx. I also notice
>>  > that you cannot route audio to an external mixer. Since that's also an
>>  > audio routing problem I imagine that sort of fix would fall under this
>>  > project.
>>  >
>>
>>
>> Already in SVN...
>>
>>
>>  > - Support for more than 2 virtual decks. -- This probably isn't key
>>  > whatsoever, but it might be nice if Mixxx supported an arbitrary number
>>  > of songs/audio sources for mixing. This would help the 'live PA' type
>>  > DJ's more than the conventional ones.
>>
>>
>> This one's doable, but again, the waveform view is going to be a
>>  nightmare to work with unless its rewritten. If you want to make a
>>  contribution regardless of GSoC, it would be helpful if we could scrape
>>  together a feature specification on the wiki for the new waveform view.
>>  An example specification can be found here:
>>  http://mixxx.org/wiki/doku.php/configurable_cue_behaviour
>>
>>  We need to take things like extra channels into account when designing
>>  it, and I don't want us to forget things like that when someone ends up
>>  coding it.
>>
>>
>>  >
>>  > - 'hands-free' mode -- While this might be a little off the mark when
>>  > considering this is a mixing app, many people (a couple even asked on
>>  > IRC the past couple days) asked for a feature where they could setup a
>>  > playlist and have Mixxx play songs serially, doing a basic-effort job of
>>  > mixing them together. While not a feature a DJ would use, it could help
>>  > improve Mixxx's visibility / reputation if more people use it.
>>  >
>>
>>
>> "NEXT" mode is all we have right now, and that needs some work too (you
>>  can get a player "stuck", I forget how). In the past, we've forgone
>>  "auto-DJ" support in favour of cooler things. Our primary use case is
>>  the DJ who wants do a set for a few hours in a club, and that usually
>>  involves manning your decks/laptop for that time. Auto-DJ will be more
>>  important for us if we start to target other potential users, like radio
>>  DJs.
>>
>>
>>  > - (oh boy) Support for control of a video system as well as a music
>>  > system. cue/loop video clips and so on to be output on a projector as
>>  > you're mixing. This probably isn't a GSOC task, but it's an interesting
>>  > feature from a performance point of view.
>>  >
>>
>>
>> There's no way this is going to happen in Mixxx, ever. It's an
>>  interesting feature from a design point of view, because not only do you
>>  have to be an audio programming ninja, but you also need to be a video
>>  programming samurai. I can imagine that any sort of video DJing stuff
>>  would involve clever synchronization of videos with the audio (see the
>>  new Serato), and that would take quite a bit of thinking to come up with
>>  a worthy design. Mixxx just isn't built to handle that, and it'd be more
>>  work to add video support to Mixxx than to write a new application from
>>  the ground up with that in mind.
>>
>>
>>  > Some minor tweaks I would probably approach on my own outside of GSOC
>>  > - fullscreen mode has always been slightly quirky on my debian box
>>  > - AFAIK there isn't support for Torq timecoded vinyl or cd.
>>  >
>>
>>
>> Fullscreen's always been quirky on everyone's computers (Windows, Mac,
>>  AND Linux! heh). Since SVG is in the cards as a GSoC project (which
>>  would solve the fullscreen issue), I probably wouldn't hack that in.
>>
>>  Torq vinyls are more or less the same as Ms Pinky vinyls, neither of
>>  which we support. If you'd like to see support for that in Mixxx, I'd
>>  suggest adding support for it to xwax, since we borrow some of xwax's
>>  vinyl control code: http://www.xwax.co.uk
>>
>>
>>
>>  > Now that I've listed off a bunch of lofty goals, here's what I can use
>>  > to help contribute: I have experience with C/C++ (over 10 years),
>>  > multithreaded programming, OpenGL, Qt, and scons. As for real-time
>>  > performance, I write 3d-game engines as a hobby, so I'm familiar with
>>  > that problem space :) . Essentially, looking over your codebase I felt
>>  > right at home. I don't have experience with PortAudio, but I'm a fast
>>  > learner.
>>
>>
>> Right, so although I shot down most of your ideas, you brought up many
>>  important features that people want. If I had to suggest a new project
>>  that's not on the ideas list, I think it might be writing a new waveform
>>  view widget. It's one of these "gateway projects" that would open the
>>  door for many other cool projects. That being said, we're all ears and
>>  I'm willing to discuss any other ideas you have.
>>
>>  Thanks,
>>  Albert
>>
>>  GSoC Mentor
>>  Mixxx Core Developer
>>  Level 32 Ninja
>>
>>
>>
>>  -------------------------------------------------------------------------
>>  Check out the new SourceForge.net Marketplace.
>>  It's the best place to buy or sell services for
>>  just about anything Open Source.
>>  http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace
>>  _______________________________________________
>>  Mixxx-devel mailing list
>>  [email protected]
>>  https://lists.sourceforge.net/lists/listinfo/mixxx-devel
>>
>>     


-------------------------------------------------------------------------
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace
_______________________________________________
Mixxx-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/mixxx-devel

Reply via email to