Sequence loops

[MuTools]

Hello,

In MU.LAB 2.x each sequence part can have its own Start, Loop and End locators which define where the sequence starts playing and where it loops.

This is on a per part basis i.e. the sequence itself does not have any loop info.

Now i'm evaluating changing this concept.

Because i think the sequence loop should be memorized on sequence level rather than on part level.
Because when you draw a new sequence part, and choose a sequence from the sequence pool, then the loop should be ok. Which is not the case now!

So there clearly is a need to memorize the sequence loop on sequence level rather than on part level.

On top of this, i even wonder whether it's actually useful to have the Loop and End locators in a sequence.

Personally, when having fun creating some music, i don't use these sequence locators too often; i even sometimes have a feeling that they're in the way, that they're complicating things.

In Muzys, things were more simple: A sequence simply had a 'Length', and the sequence simply looped over that length. Easy and effective!
And if you wanted a longer loop, then you could simply increase the sequence length (with nice options e.g. repeating the existing notes into the new space) and so you got the longer loop.

I'm not sure how many of you have known Muzys, but for those who don't know it, here is a link to a Muzys doc page which explained the concept of sequences:

http://www.mutools.com/temp7491/quickstart.html

Muzys' options when increasing sequence loop length:



Muzys' options when decreasing sequence loop length:



These options are explained on this Muzys doc page ('Length' function):

http://www.mutools.com/temp7491/sequencemenu.html

(all temporary links/pics)

Coming back to MU.LAB, the only reason why the sequence locators would make sense if when you have a single sequence containing multiple sub-sequences that you want to loop in separate parts.
But i think it would make sense, both musically as practically, to put such sub-sequences in separate sequences.
And if the sequence loop would be memorized on sequence level (and yes there definitely is reason for doing so, cfr above), we would need separate sequences anyway!

So i'm evaluating trashing the Loop-End locators on sequence level and just work with a sequence length which implicitly defines the loop, just like in Muzys.

Only the Start locator would survive. And it would still be on part level.
Because the Start locator is used when splitting parts so that the splitted sequence start playing at the correct position.
This definitely is a great improvement on Muzys because in Muzys a new sequence was created whenever a part was splitted. In MU.LAB this is not necessary, thanks to that Start locator.

Conclusion:

I think that trashing the sequence Loop-End locators, and making each sequence loop over its own sequence length would make the sequence looping easier and more transparant, while gaining back the best of Muzys in this aspect.

What do you think?

[anubis2010]

Hi

Just to get the ball rolling on this one.

I personally like the loop, start and end markers and the fact I can change them quickly with the mouse and hear the result quickly. One problem is that the are hard to see when they overlap each other but you are already aware of this.

I also think it should be at sequence level not part. Therefore when I ghost/clone a sequence or select an existing sequence it already has the same start/loop/end markers already set so it is in fact a mirror of the sequence I was working on.

As for the way Muzy works I will let a qualified user of Muzy better answer that part of your question.

anubis2010

[Eggu]

In fact,
I use the locators for sub-sequencing a lot.
I seldom use the sequence pool,
but I copy a loop to create a new one,
so I do not miss the memorizing on sequence level.

[pljones]

Hmm... I think I was instrumental in wanting to keep the sequences as "pure" as possible in MULAB...

Parts are about how you arrange your composition. Thus I feel that the start/loop/end markers belong with the Part (regardless of whether it's audio or sequence).

However, in the same way that an audio sample can loop internally (e.g. the "tail" playing until fadeout), maybe sequences should have the same facility? That would only need a loop marker, as with audio samples.

I tend to use sequences from the pool but assign them to different instruments and want them to play differently, which is why I prefer to have the markers at a part level. But I can see why the alternative would be useful as well.

[anubis2010]

Hi

Sometimes we make changes for the sake of change.

I have been using loop markers at part level and automatically setting loop markers on the ghost or not etc.

Obviously MULAB is being looked at to move forward towards M3.

Users/Jo need to decide the balance between making MULAB as automated as possible or leave a lot of musical flexibility.

On most of these kind of forum issues I am always in two minds, leave it or change it.

Loop DAWs are very quick but can become a bit boring and not really that useful musically in the long run.

I want MULAB to be a good loop DAW but with plenty of flex as well.

Will the change make it more flexible or more sterile?

anubis2010

[anubis2010]

Hi

I have spent a few more minutes soley on sequences and parts.

I am going to answer my own question.

The reason I like MULAB so much is it seems to work correctly/logically right out of the box, (although you do not actually get a box as such).

IMHO loop/start/end should be at part level, seems to be more room for musical flex.

There may be something to learn from Muzy but that is for another user to comment on.

I like it like it is.

anubis2010

[anubis2010]

Hi

I have read the Muzy docs, but this is no substitue for trying the actual software.

Are there any experienced Muzy users out there that can comment on the pro/cons of the part/sequence system in Muzy and can then compare it to MULAB's current methodology?

anubis2010

[MuTools]

Just a little detail: It's called Muzys, not Muzy.

[anubis2010]

Hi Jo

I am aware of that I was just abbreviating it like it has been in "Muzycian".

anubis2010

[robenestobenz]

I personally quite like them for sorting through different takes, and would miss them if they were taken out. I did find them a bit confusing in the beginning though. Maybe not exactly confusing, but a bit messy, for a few reasons.

First of all, after a split, for example, I'd open up a part and find myself at the beginning of the part, faced with a whole stretch of notes that aren't sounding. I personally think it'd be much better if the sequence editor opened scrolled to the sequence start marker. Having the section of the part that is sounding fitted to the window would be even better.

Secondly, that loop marker. Half the time it obscures the start marker and especially in parts formed by a split, I feel it shouldn't be there by default. I think there is a case for not adding it to a part as default -- the default behaviour of repeating the whole sequence when stretching the part along the timeline is fine, the loop marker is a power user feature IMO, and doesn't need to be enabled by default.

RE. part and sequence level, while I can see advantages for both, I think the section control stuff really belongs in the part level.

[pljones]

Actually, having none of the markers appear until needed would indeed make them much easier to use. If a part and its contained sequence are the same length and the part starts playing the sequence from the beginning, there is no need to display any of them. Right-click the time line to set a new start position and you get a start marker. Right-click it and select "Reset Start" and it's gone. Something like that? (Similarly for end, to have the part play less than of the sequence to fill itself and Loop - very advanced - to say where to loop back to after hitting the end.)

[MuTools]

I've been thinking about this more.

Also based on your feedback here i've decided to keep the current sequence loop system i.e. using a Start, Loop and End locator.

In M3, these locators are displayed in an improved way: They use icons instead of text, and they are not displayed if they are at their default position:



The default positions are:

Start: @ Loop locator
Loop: @ start of sequence
End: @ oo (=infinity => sequence is an unlimited sequence with no practical end)

So for a simple 2-bar sequence loop with Start and Loop at the beginning of the sequence, it looks like this:



This certainly improves things already

But now i've still one thing to decide: Is the sequence loop on part level or on sequence level?

This question is only about the Loop and End locators (which define the loop).
The Start locator will be on part level anyway because it is used to realize splits.

Personally i would like to put the sequence loop on sequence level because:

* It's a sequence loop
* That way sequence functions like "Legato Notes" can take the sequence loop into account
* Whenever choosing/loading a sequence, the loop is clearly defined
* If you want to have different sequence loops for the same sequence, then simply duplicate the sequence; this is musically perfectly relevant imho.

I only see one possible little disadvantage of having the sequence loop on sequence level: Then the Start locator is at part level while the Loop and End locators are on sequence level.

For me i think that's ok.

But what do you think?

[pljones]

I think I'm sold .

Part "Start" is actually the Part-specific offset into it's sequence? The Part's "End" is defined by its length.

The Loop and End markers are actually the sequence Loop Start and Loop End positions?

(Now, if I were triggering the sequence from a MIDI Note, I'd expect it to play from "start->loop start->loop end->(repeat loop)" whilst I'm holding the note and then from "loop end->end" when I release the note -- with "start" and "end" defined outside the sequence, I suppose, similar to a Part.)

[MuTools]

I think I'm sold .

Part "Start" is actually the Part-specific offset into it's sequence?

Yes, indeed.

The Part's "End" is defined by its length.

The Loop and End markers are actually the sequence Loop Start and Loop End positions?

Yep!

(Now, if I were triggering the sequence from a MIDI Note, I'd expect it to play from "start->loop start->loop end->(repeat loop)" whilst I'm holding the note and then from "loop end->end" when I release the note -- with "start" and "end" defined outside the sequence, I suppose, similar to a Part.)

It's not yet possible to trigger sequence parts from a MIDI note, so i leave this topic until later

[MuTools]

Also regarding the sequence loops:

Whe the sequence loop is on sequence level, there will also come a sequence function "Change Sequence Loop"(or "Resize Sequence Loop" or something like that).

This function will do exactly the same as when changing the sequence length in Muzys i.e.

On increasing the loop length: options are: Repeat - Stretch - Increase Only
On decreasing the loop length: options are: Cut - Shrink - Rotate - Decrease Only

For more info on what these options do, see http://www.mutools.com/temp7491/sequencemenu.html