pattern sequencer; possible with Ruby?

[Nubeat7]

nice work billv, for the drawing and mouse action in the sequencer part i would recommend to use loops, isn`t it boring to repete some code 16 times? you could also change the steps then

you can have a look at my stepseq here
viewtopic.php?f=3&t=1001&start=30

[kortezzzz]

billv,

I'v just dived into your schematic and I understand the logic behind it pretty well, although I don't understand the
Ruby code at all. I saw lot of interesting things in this gold mine, and one of them is the module called "Ticker - Custom+ Clock". this module determines a golabal note length for the full arpeggio.
Now, if it possible, you could seed this module in every note separately instead of making it global, and let the user determine each note's length separately with a new set of "ruby_sliders_x16+clock_fixed" module. the length
menu should start at 1\64 and grow in sequential up to 1\1 (or even 2\1) every time the user pushes the note slider up. in that way, we get almost a full piano roll abilities, but in different approach!!!
and that is a huge step forward to the very first "first class ruby midi piano roll"!

EDIT: on second thought, the "Ticker - Custom+ Clock" has nothing to do with the note length, because note's length
is all about midi note's sustaining
so if we desire to control each note's length separately and pass it as an independent note to the midi event, we should make a independent sequencer for each note... so then the schematic will contain 16 different sequencers, which are connected with separate 16 midi wires to the midi input. each sequencer would be in charge for a single note; you may control it's pitch, velocity and length and all the notes can play simultaneously. do I see it right?

[billv]

Thanks guys..!

like choosing between 3\4-4\4 beats

Yeh, there's a lot of work to do yet. Not really sure yet what happened here with
the beats ...somewhere in the design process I got stuck to 1/4, 1/8 (even beats) ect ect.
Thats one of the reasons why i made a quick stripped down version, its very easy with
seq building to find half way through a design that its no good and you have to start again.
At the moment, it is what it is, cause i can only see one step at a time...so to speak..

Now, if it possible, you could seed this module in every note separately instead of making it global, and let the user determine each note's length separately with a new set of "ruby_sliders_x16+clock_fixed" module. the length

..yeh, I actually did that to test that out. I made a test unit first, 16 sliders with
that send note length for 1/1 to 1/64....it worked fine....but that was just a module sending
the hard values, which is no good, as the values must be controlled by the tempo prim.
So i needed to loop the beat division 16 times...

Code: Select all

@alpha = 60/@tempo
@beta =@alpha/@index
@delta= @beta*4
output 0,@delta

Can't get it to work yet, so i refused to copy it 16 bloody times, and choose to wait until i get the
looping method right. So I just left it there on an array, make it easy for the next guy.

full piano roll abilities, but in different approach!!

Thats the funny thing about seq building....there's seems to be limitless ways to "approach" it,
but we all want 1 consistent end result.

i would recommend to use loops, isn`t it boring to repete some code 16 times?

Boring isn't the word mate ..it does my head in...but I'm getting better at loops...
its just the learning stage I'm at....I'll get it right one day...
Thanks for the link.....not much in my toolbox at the moment, missing a lot of reference material.

[kortezzzz]

its very easy with
seq building to find half way through a design that its no good and you have to start again.
At the moment, it is what it is, cause i can only see one step at a time...so to speak..

Well, we have an excellent start here. I believe we'll find the way to put things together later.

yeh, I actually did that to test that out. I made a test unit first, 16 sliders with
that send note length for 1/1 to 1/64....it worked fine....but that was just a module sending
the hard values, which is no good, as the values must be controlled by the tempo prim.
So i needed to loop the beat division 16 times...

Well on second thought, I think its not needed! because if we desire to manipulate note's length, we should deal with note's sustain, and not to manipulate the whole time line which means, putting a timer between the midi input and the midi to poly that will extend the note's duration to a chosen one. so let say that a given note
is playing for 2 seconds and the the next note starts after 1 second; They both will play together for one second until the first note dies

Thats the funny thing about seq building....there's seems to be limitless ways to "approach" it,
but we all want 1 consistent end result.

YEAH!

[billv]

I think it has that behavior now.
I will upload the temporary Duration controller when
I get home from work, so you can experiment with it.
At the time, I was chasing the behavior that tulamide
(forgive wrong spelling) was looking for, where if
the same note was played at different durations, they
wouldnt cancel each other out...it didnt happen.
And the durations of each note are independent from main
timer...I think that part is ok...lots of experimenting to do..

[kortezzzz]

Great billv. Thanks for your hard work.

I still think that it would be much powerful if you could make a separate one note only sequencer that can be
pasted into the schematic as many times as we need and has it's own midi output, but still connected to the master time line in serial connection. In that way, 16 different pasted small sequencers (or even more than 16!!!) can perform there own "one midi note" each time in sequential order and they never cancel each other (no matter how long the duration is) as their midi output is separated. Possible?

[billv]

16 different pasted small sequencers (or even more than 16!!!) can perform there own "one midi note"

Thanks mate, another mad idea ....definatly not ignoring them....just trying to
process them mentally...hands are full ATM..

I still think that it would be much powerful if you could make a separate one note only sequencer that can be
pasted into the schematic as many times as we need and has it's own midi output, but still connected to the master time line in serial connection. In that way, 16 different pasted small sequencers (or even more than 16!!!) can perform there own "one midi note" each time in sequential order and they never cancel each other (no matter how long the duration is) as their midi output is separated. Possible?

...my first thought is what is going to happen when all this hits the voices section
in Midi to Poly....not sure.. ..I'm not good with the inner workings of voice management..
Provided that was all clear, and the Midi to Poly dosn't explode, I'd take another look at it...
Here's that temporary duration controller discussed earlier..

Temp_Duration Control.fsm

(31.97 KiB) Downloaded 794 times

EDIT:
Note it only goes from 1/1---1/64 cause at 1/128 the circuit starts spitting the dummy...

[tulamide]

I still think that it would be much powerful if you could make a separate one note only sequencer that can be
pasted into the schematic as many times as we need and has it's own midi output, but still connected to the master time line in serial connection. In that way, 16 different pasted small sequencers (or even more than 16!!!) can perform there own "one midi note" each time in sequential order and they never cancel each other (no matter how long the duration is) as their midi output is separated. Possible?

...my first thought is what is going to happen when all this hits the voices section
in Midi to Poly....not sure.. ..I'm not good with the inner workings of voice management..
Provided that was all clear, and the Midi to Poly dosn't explode, I'd take another look at it...

I made a quite similar attempt for my fixed freq problem and dare to say that you can have dedicated Midi to Poly modules per midi out (at least with two of those modules I had no problems). But the midi note filtering would be difficult, wouldn't it? Or would it just be a multiplexer switched by a counter per note on?

[kortezzzz]

Provided that was all clear, and the Midi to Poly dosn't explode, I'd take another look at it

Why should it explode? Hard to believe that midi2poly can not overcome few notes at the same time
To be rational, the possibility that the whole all 16 note will hit at the same time in a single arpeggio beat time line
is very low. It can occur in lets say, on drum phrases, when the user wants the kick and the cymbal to hit together.
or in pad phrases, or even a strumming guitar.

Here's that temporary duration controller discussed earlier..

Thanks. how should it be connected?...

you can have dedicated Midi to Poly modules per midi out

I don't see why we need more than 1 midi to poly module. each 1 note sequencer has it own midi out. they all can be sent to the same midi to poly module, don't they?... am i missing something?

[billv]

Why should it explode? Hard to believe that midi2poly can not overcome few notes at the same time

Yeh Im probably just being paranoid....will have to try it one day...

Thanks. how should it be connected?

Don't. I solved my looping problem and can now share the real one.
Just dis-connect the ruby prim with the array on it.. and connect this module to the duration input(timer module)..

Duration Control_fixed.fsm

(32.19 KiB) Downloaded 758 times

midi note filtering would be difficult

Everythings 'difficult'... .