Page 4 of 8

Re: Perfect Layout

Posted: Wed Oct 03, 2018 7:41 pm
by NothanUmber
Still experimenting. All seem to have their strength and weaknesses. At least I come to grips with the "logics" behind all fourth better now (you can still play the same pattern all over the place as long as you shift diagonally). Fourth intervals are still not great with all fourth with fingers on top of each other.
But on the other hand the closer spacing of all fourth is indeed nicer for certain chords.

Am at the point now where I think that the augmented fourth layout at least isn't so vastly superior (in the sense that it would allow to play stuff that isn't possible with all fourth) that it would be worth giving up the instantly-at-home feeling that all fourth offers to guitarists.

So I am starting to lean towards all fourth now. Even though experimenting with stuff is fun I also see the merit in common standards - so pieces can be written for an instrument etc. And for the lack of vastly better alternatives going with what all the others do is just human nature (and at least often not the worst choice) :)

The most logical, best-for-four finger chord/multi-voice-per-hand layout is probably still the linear one (like on the upper, chromatic half of a piano - or in the purest form a Continuum). But even that is not "perfect" in the sense that all imaginable chords are easy to play (anything vastly over an octave is actually impossible with one hand while possible with some of the grid-based layouts).

So probably just several sweet spots in a landscape of possibilities, all more local maxima, probably none "optimal in all regards for grid-instruments" or even "globally perfect for two hand ten finger creatures" - but still fine to have fun with. :)

It can still be interesting to use alternative layouts as they will lead to different sweet-spot patterns and thus different music. Similar to changing the tuning on a guitar to get rid of all the hardwired riffs and venture into new grounds.
That's an advantage of grid layouts compared to linear ones. A line is a line is a line :)

Re: Perfect Layout

Posted: Thu Oct 04, 2018 4:06 pm
by Roger_Linn
Another factor is Y-axis range, where Continuum has the advantage over LinnStrument, albeit at the expense of the advantages of overlapping rows. Interestingly, Madrona Labs’ SoundPlane offers a very nice middle path—the notes are arranged in a grid like LinnStrument but each one is a tall rectangle, providing a more usable Y-axis range:

https://madronalabs.com/soundplane

I suspect that over time, a few note layouts will succeed, just as the instruments of the orchestra succeeded while others became obsolete:

https://artsandculture.google.com/exhib ... 8kJg?hl=en

Re: Perfect Layout

Posted: Sun Oct 07, 2018 6:02 am
by Tj Shredder
Actually I find the Linnstrument y-axis range very usable, even more than Continuum and Seaboard. I just bend my finger. The drawback is, that in most MPE ready presets the y-axis expression is triggered on vibrato too easily and creates big changes in the timbre. You have to be careful to vibrate only the x-axis...; - )
I could imagine to tackle this in the software with a smoothing parameter for the y-expression. It would not react on fast changes or only very little... The smoothing parameter could be controllable by Midi itself...

Re: Perfect Layout

Posted: Sun Oct 07, 2018 7:59 pm
by Roger_Linn
Thanks, TJ. The Y-axis data is sent as fast as it can be and is very dense, so no further message density (for smoothing) is possible in LinnStrument. However, smoothing is generally a function of the sound generator, where the smoothing is not limited by the MIDI message rate. In general, such audio changes are already smoothed in sound generators.

Re: Perfect Layout

Posted: Sun Oct 07, 2018 10:11 pm
by Tj Shredder
The smoothing I think of would produce less midi data, it would simply prevent sudden jumps or too fast big changes. I am not aware of any sound generator which would do that. There would be no reason from the sound generator perspective... Its more like a low pass filter for controller data...

Re: Perfect Layout

Posted: Mon Oct 08, 2018 1:01 am
by Roger_Linn
There are some good reasons why smoothing to occur in the sound generator instead of the MIDI controller:

1) If the controller smooths MIDI data, then if you switch to a different sound in your synth, you must also change LinnStrument's smoothing settings.

2) The smoothing rate should be optimized for what in the synth is being smoothed. For example, volume smoothing should have a different smoothing rate than pitch changes or timbral changes.

3) Smoothing the Y-axis value as you suggest could easily result in audio steps if the synth's smoothing rate is too fast to ramp between the sparse steps in Y-axis values. By comparison, if the smoothing is done in the synth, the synth merely ramps at audio rates from the previous Y value to the current Y value, regardless of how far apart they are.

A better way to think of this issue is "How should synth modulation elements be redesigned for expressive control?". For example, envelope generators were designed to overcome the limitations of playing music with on/off switches, but are pretty much useless with expressive controllers. As you have pointed out, performed pressure can't easily produce sharp attacks, nor can it produce consistent long-decay envelopes as on plucked strings. So a better type of envelope generator would be lag generator with the ability to independently quicken or slow the rise and fall of pressure envelopes. For example, slowing the fall time would produce a consistent string-like decay time. Or quickening the rise time (by increasing the amplitude of the rise up the received peak level) would result in a proportionally sharper attack, making it easier to produce sharp plucked-string sounds at lower velocities. But again, this is better implemented as a function of the synth, not the MIDi controller, for the above-stated reasons.

Re: Perfect Layout

Posted: Mon Oct 08, 2018 4:39 am
by Tj Shredder
To certain degree I agree. If done in the Linnstrument it would require a two way communication, as you would have to control that controler data low pass filter. Though I do not expect synth developers to adapt to this too quickly, as the too sensitive y-axis on vibrato is not an issue of all controllers, only the Linnstrument would need it. It could move into the host or a plug-in. In Max its easy to implement, in Bitwig for example it would need a simple Midi fx plug-in. I am sure that will arrive sooner or later...
Roger_Linn wrote: Mon Oct 08, 2018 1:01 am 3) Smoothing the Y-axis value as you suggest could easily result in audio steps if the synth's smoothing rate is too fast to ramp between the sparse steps in Y-axis values. By comparison, if the smoothing is done in the synth, the synth merely ramps at audio rates from the previous Y value to the current Y value, regardless of how far apart they are.
A function which prevents jumps can never result in audio steps, it can prevent them rather... If the jump is fast, Midi would skip values in between, if its smoothed/filtered there would be no steps and the number of Midi events would either be the same or less...

Re: Perfect Layout

Posted: Mon Oct 08, 2018 4:13 pm
by Roger_Linn
I'm a little confused. How are you using Y-axis for vibrato? Are you using it to add LFO? If so, that would seem to be contrary to the fundamental idea of performed expression.

Re: Perfect Layout

Posted: Tue Oct 09, 2018 5:29 am
by Tj Shredder
No, when I do vibrato, I can’t avoid to also move a little bit the y-axis. My finger shakes not always straight on the x-axis. Usually I have a timbral change on the y-axis which will be modulated unintended...

Re: Perfect Layout

Posted: Tue Oct 09, 2018 2:12 pm
by Roger_Linn
Thank you for the clarification, TJ. I have not found that to be a problem, but I generally use Y-axis timbral variations in which any such large change is not objectionable, for example bow position on the Audio Modeling bowed strings or pulse width on a classic subtractive synth sound.

Re: Perfect Layout

Posted: Wed Oct 10, 2018 3:17 am
by John the Savage
When I play synth sounds, I use the Y-axis mostly for mod-wheel duties, with some pretty aggressive modulations, and have not found it to be problematic at all. I keep the Y-axis in "relative" mode with the starting value set to zero, and I curb the modulation range within the synth itself. I find it very playable this way, even when applying vibrato.

That said, I really don't think the Y-axis on the LinnStrument was intended to be used as a mod-wheel. It seems to me that it was primarily meant to help facilitate the finer aspects of "acoustic" expression, as to allow for a more human feel: i.e. the kind of subtleties found in the way one might play a real instrument; like the way a pick never hits the string at exactly the same angle, or how the tilt on a bow is constantly being varied, either deliberately or by accident, resulting in a slightly different timbre with each note attack. It's a technique to be mastered to be sure. At any rate, I don't believe this behaviour needs to be, or should be, smoothed out at all, else we'd be right back where we started with controllers before MPE.

Cheers!

P.S. To keep this thread on track, I prefer the "guitar" tuning myself. Of course, that's because I'm a guitar player (grin). See, it's my opinion that tuning schemes, especially in this context, are highly subjective, even irrelevant, and that no one tuning scheme is better than the next, any more so than any one instrument is superior to another.

Whether we're talking about what Ana Vidović or Eddie Van Halen can do on the guitar, what Yo-Yo Ma can do on the cello, what Hyung-ki Joo can do on the piano, or what Maceo Parker can do on the sax... They all stand as proof that any well-designed musical interface is as capable as the next in the hands of a proficient player.

So if you're coming from another instrument, I highly recommend that you stick with the tuning scheme that you've already mastered, because learning to play all the same notes, only in a different order, is a waste of time. Otherwise, if you've never played an instrument of any kind before, I would suggest that you latch on to the first layout that speaks to you, and just stick with it. It takes months, even years, of playing an instrument before you can develop a truly informed opinion about the advantages or disadvantages of its design, never mind your own playing technique. By the time you've reached that level of proficiency, most of those perceived stumbling-blocks will have all but disappeared.

My point is, no matter how one chooses to tune it, the LinnStrument is already an extremely efficient playing surface compared to traditional instruments. That is to say, the physicality of its design is definitely not holding you back. Therefore the onus of mastering it falls entirely on the player's dexterity, regardless of where the notes are located (wink).

Re: Perfect Layout

Posted: Wed Oct 10, 2018 8:46 am
by dr_loop
.. I highly recommend that you stick with the tuning scheme that you've already mastered, because learning to play all the same notes, only in a different order, is a waste of time
...I beg to disagree. With the LinnStrument flexibility is king! I like to 'to play all the same notes' in different tunings just to see how it sounds. Very inspiring!

Re: Perfect Layout

Posted: Wed Oct 10, 2018 9:54 am
by John the Savage
dr_loop wrote: Wed Oct 10, 2018 8:46 am
...I highly recommend that you stick with the tuning scheme that you've already mastered, because learning to play all the same notes, only in a different order, is a waste of time.
I beg to disagree. With the LinnStrument flexibility is king! I like to 'to play all the same notes' in different tunings just to see how it sounds. Very inspiring!
Sure, it's novel. But I think what you mean to say is "you like to play all the same notes in different tunings just to see how it feels", because all the same notes are going to sound exactly the same regardless. My point is that such exploration, inspiring though it may be, is arbitrary at best, and poses a waste of time and energy if your objective is to actually master the performance aspect of the instrument, which could otherwise take a lifetime.

The LinnStrument's flexibility in this regard is a benefit to be sure, in the sense that you can customize the layout (and even the fundamental behaviour of the playing surface) to your liking. But if your aim is to be a proficient player, my advice is to pick a setup and stick with it, at least until your technique is fully developed. And if you're already familiar with a particular layout, your time is surely better spent on developing your chops and, you know, picking the right notes and all that (smirk).

I can only presume that proficiency is ultimately what we're talking about here. Hence the title "Perfect Layout". I mean, given the uniformity and open concept of the LinnStrument's design, it's not like any one tuning scheme presents an undeniable advantage over the others. As such, the pursuit of the "perfect layout" quickly becomes a moot point beyond a certain skill level (one's personal preferences notwithstanding).

Otherwise, if we're just talking about changing tuning schemes for the fun of it, this is purely an academic exercise.

Cheers!

Re: Perfect Layout

Posted: Wed Oct 10, 2018 4:13 pm
by Roger_Linn
John makes a good point, but it is also true that sometimes the reward is in the journey.

Re: Perfect Layout

Posted: Wed Oct 10, 2018 5:49 pm
by dr_loop
"you like to play all the same notes in different tunings just to see how it feels", because all the same notes are going to sound exactly the same regardless.
... no, sorry I was'nt clear enough, with 'all the same notes' I mean the *fingering* not the actual musical notes.
So with the same muscle memory pattern you can explore many different musical patterns.
Especially when with the press of a button you can have every tuning you want.
And nobody prevents you from changing the tuning every second if you like :D