[PD] sequencer app for osx

Luke Iannini lukexipd at gmail.com
Sun Aug 10 09:23:10 CEST 2008


Haha, well, perhaps I was overoptimistic about the state of the
sequencer.  There seem to be some bugs in it that I "fixed" already
rearing their heads... I haven't touched it in long enough that I
can't fix them deftly so here it is anyways.  Perhaps you can still
get the idea.

Also, this is a bit like showing you my time machine... it's got a lot
of work inside it so it comes with a nice library of 40 or so
supporting abstractions : ).  The sequencer itself is
[sft.structhausen] and a working model using Phil Stone's
[polywavesynth] is in [STRUCTHAUSEN].

Click the big "OPEN" button to open the piano roll.  Inside
[sft.structhausen] you'll find [pd keycommands] which give a few
shortcuts.  For troubleshooting, one you may need often is "shift-s"
to re-sort the notes and "shift-z" to fix up Z-ordering.  You can
either click the "record" button to record notes (I included a
computer-keyboard to piano-keyboard), or, there's always a single note
in the lower-left of the piano roll you can "clone" by option-dragging
(and then you'll experience the annoying mouse-drag behavior).  The
"prefs" button on the GOP will let you switch on drag-modifiable
velocity displays for each note.  You can delete notes by holding
option and hovering over the lower-left corner of a note until you see
the "X".

Anyway, everything I said is in there in some form or another, and
I'll keep you up to date as I get it to a more usable state.

Best
Luke

On Sat, Aug 9, 2008 at 5:26 PM, Luke Iannini <lukexipd at gmail.com> wrote:
> Yo Damien,
> I've been working on exactly that for the past few months using Pd's
> venerable data structures.  It has quantization, a piano underlay,
> draggable notes, draggable lengths, draggable velocity, zooming X & Y,
> draggable looping, the whole bit (layers coming soon, but you can of
> course run many of them side by side; I'm working on a synchronization
> system since running multiple instances makes it more fun to do
> polyrhythms).  Thanks to some wonderful shortcuts Miller built in to
> DS arrays, you can add and remove notes (and even shift groups of
> notes) really easily.
>
> But, there are some bugs in Pd that prevent it from being everyday
> usable... the two major ones are that the notes are in a "quantized DS
> array" (for performance reasons), which exposes a very painful
> interaction bug where dragging is just close enough to working to be
> horribly frustrating, and editing the notes (to change lengths or
> velocity) only really works for the notes closest to time 0.  And,
> finally, Z-ordering is undefined and gets screwed up every time you
> open the window (but, I've worked around that one for the most part
> with some hacks).
>
> So, I'll wrap it up and post it after I locate all the stuff necessary
> to run it.  It should convince you that such a thing is most
> definitely attainable in Pd, and perhaps Miller will be so excited
> that his language for the future of music is being used to make a
> 30-year old interface, that he will go fix the bugs immediately : )
>
> (I promise though Miller, my actual goal is to build a happy
> environment for both traditional grid-note-editing and graphical type
> stuff like in 4.data.structures/07.sequencer)
>
> Back soon,
> Luke
>
> On Sat, Aug 9, 2008 at 12:43 PM, Damian Stewart <damian at frey.co.nz> wrote:
>> Martin Peach wrote:
>>
>>>> (since i haven't yet found a pure Pd sequencer i'm happy with)
>>> What makes you sad about Pd sequencers? Maybe it's fixable.
>>
>> have you ever used Ableton's MIDI sequencer? it's a piano roll, with blocks
>> for notes. drag start and end points to control noteon/noteoff points.
>> quantize to a grid of any size you want (switchable with a hot-key) or turn
>> off quantization completely to finetune timing (for that little 'humanized'
>> extra, whatever that might mean). super-intuitive scroll and zoom with a
>> single mouse button drag in the top. the main features in this i'm
>> interested in is the quantization grid control, and the ability to super
>> fine-tune note durations.
>>
>> lines down the bottom control velocity. if you have two notes on the piano
>> roll trigger at the same time, it's super-easy to choose which note's
>> velocity you're editing.
>>
>> i don't believe it's do-able in Pd, not without a ridiculous amount of
>> work, and i'm too lazy to do the work myself (and if i wasn't, i'd do it
>> from scratch in a different programming language eg Processing or C++,
>> rather than fighting Tcl/Tk).
>>
>>> Why not run three instances of midifile or even textfile with the same
>>> data?
>>
>> yes, this is what i'm already doing. the problem is generating said
>> midifile/textfiles in the first place.
>>
>>>> - easy access to velocity and note duration
>>>
>>> Do you mean that the thing should output the duration at the same time
>>> as the note-on, or the file should be readable in a text editor?
>>
>> no, i mean that i should be able to easily see and edit the duration and
>> the velocity without having to think or do much work. see above..
>>
>> --
>> damian stewart | +31 6 8178 5197 |  damian at frey.co.nz
>> frey | live art with machines | http://www.frey.co.nz
>>
>> _______________________________________________
>> Pd-list at iem.at mailing list
>> UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
>>
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: structhausen.zip
Type: application/zip
Size: 228085 bytes
Desc: not available
URL: <http://lists.puredata.info/pipermail/pd-list/attachments/20080810/2672ebca/attachment.zip>


More information about the Pd-list mailing list