OK.  I got the impression he wanted to make a step-sequencer like the electribe.  And he would want to program the patterns into lmms as bassline/beat-patterns.  Then, be able to layout patterns as songs.  For DJs, it could queue patterns for creating mixes, etc.  All of this would require more control than midi.  But, this is all a different story if all he wants is a MIDI controller.

-Paul

On Nov 26, 2007 10:07 PM, Vik <vik@japanvik.net> wrote:
Hi,

I'm assuming the sequencer Raffaele will be making is something along
the lines of http://www.ucapps.de/ - creating a hardware midi controller
box.

The question will be, then "How much midi events can LMMS handle?"

The quick answer, as far as I know, (correct me if I'm wrong) will be:
1) Receive and play midi note on/off events - Yes.
2) Receive midi channel change events to dynamically change the
"instrument"(ie track) you play - No.
3) Receive other arbitrary midi events and map them to widget controls - No.

So from a hardware sequencer perspective, you will be able to play the
notes you send to LMMS, but won't be able to switch the instruments
unless you remap the midi channel from within LMMS manually. However,
I'm guessing that this is not really the conventional way a sequencer
operates. At least not for a multitrack sequencer.

You might want to look at how seq24 (software sequencer) does its job.
It can map every sequencer track to a unique midi channel out. You can
then map these to the tracks in LMMS prior to your musical performance.
That way, you can "play" LMMS via seq24, which is what you probably are
looking to do, by replacing seq24 by your hardware sequencer. In theory
it should work, but I haven't tried it, so you might want to play around
with it.

Slightly off topic, but I personally would like to see point 3 of the
above happen. Every automation aware widget should also be able to
receive user configured midi events and record them... maybe I'll try
implementing once the svn trunk code base settles down...

Cheers
Vik


Paul Giblock wrote:
> Raffaele -
>
> I have developed for LMMS for a few months now, not as long as the other
> developers.  I'm sure they will have input as well, but I feel like
> responding to you in the meantime.
>
> My understanding is you want to make a hardware sequencer.  The hardware
> isn't actually a sequencer, but is only a control box that sends
> commands to a host computer? Is that correct?
>
> I'm also guessing you will make the device act as a standard USB HID
> device is that correct?  Doing so will save alot of money on hardware as
> well as help development. Linux has a driver, usbhid, that has an "easy
> to use" userland library, so no kernel hacking required.
>
> About using LMMS as a library. In the current state, this is not
> possible.  I spoke to Toby about this in the past.  We are too closely
> tied to QT to fully realize an independent "lmmslib".  We are doing
> architecture changes currently,  but that still won't solve your endeavor.
>
> Now, it sounds like LMMS has all the features you need, but only lacks
> the input/output you require? For example, a song can actually be a
> "set" or "show", the different patterns are actually different
> Beat+BassLines, etc.. But, you don't have means of fetching or setting
> the BassLine.  Or, there is no way to  create a new Instrument Track,
> etc...
>
> Basically I can think of two different ways to solve this. This is where
> Toby and Javier will probably have something to say.  First way (the
> lazy way):  we can put the usbhid userland driver code straight into
> LMMS.  This is good, especially if you plan on manufacturing and selling
> these control boards  (I might even buy one).   Of course, this bloats
> up the code and there isn't a solution for windows users.
>
> The second method I can think of is LMMS could actually support an API
> that allows remote control.  This API can be accessed through IPC
> (sockets, FIFO files, doesn't matter to me).  The benefit of this is it
> doesn't tie your code straight into ours.  Secondly, it could allow
> developers to make other useful scripts that can control LMMS's
> parameters.  Maybe this could be the foundation of a "General Plugin"
> interface?  There are all sorts of nifty possibilities here.  The
> downside: Someone has to design this API, worst, someone has to
> implement it and we are already very busy.
>
> Anyways, those are my ideas,  let me hear what you think, and do not
> take anything I said as a definitive answer.
>
> -Paul
>
> On Nov 26, 2007 4:38 PM, Raffaele Rossi < rossi.raffaele@gmail.com
> <mailto: rossi.raffaele@gmail.com>> wrote:
>
>     Hi,
>     My name is Raffaele and i'm studying control engieering at the
>     TorVergata UNiversity of Rome.
>     I like music so much that i would like to buy a sequencer, but it's
>     so expensive :s
>     So i would like to make an external sequencer wich one could turn
>     regulator and press button, like a real sequencer.
>     Then signals are sent to lmms via usb (or something other port) and
>     they change current sound parameter
>     (current sound that is also selectable by sequencer self). So one
>     could use lmms' beatbox like if he has ten hands and mice...
>     Electronic cost could be lower then 100EUR (instead of 700EUR for
>     Korg electribe), but to create circuits first i should study how lmms is
>     thought...
>
>     So: is there any guide that explain lmms concepts libraries and
>     sources?
>     Obiously if i could create both electronics and programming then i
>     will share with all community my progresses and also ideas, and
>     discuss on them.
>
>     Let me know,
>     Greetings.
>
>     -------------------------------------------------------------------------
>     This SF.net email is sponsored by: Microsoft
>     Defy all challenges. Microsoft(R) Visual Studio 2005.
>     http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
>     _______________________________________________
>     LMMS-devel mailing list
>     LMMS-devel@lists.sourceforge.net
>     <mailto: LMMS-devel@lists.sourceforge.net>
> ------------------------------------------------------------------------
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2005.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> LMMS-devel mailing list
> LMMS-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/lmms-devel


-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
LMMS-devel mailing list
LMMS-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/lmms-devel