Hello and a suggestion- chord mode

[pdxindy]

Hi all,

I got sidetracked into some other tasks but I'm giving this topic some thought and will post something in the next few days. I did take a few minutes this evening to play with the Chord Trigger feature in Logic, which is a pretty good auto chord tool. The problem I have is that I quickly miss the ability to voice each chord as I choose, so the music sounds boring and repetitive very quickly.

You can make some variations of course and assign them to other keys... and it respects the velocity of the note played. But if someone wants to explore this, I recommend Cthulhu which is more feature rich than the Logic Chord Trigger device.

Cthulhu you can import a midi file that has some chords, it auto assigns them to keys. Then it has a function to create a bunch of variations/inversions of those chords across the rest of the keyboard. It has a number of other useful functions.

[thedommer2000]

Great stuff Roger. Will be experimenting with that mode! The more I read the more I wan't wait to play it.

[PeterAntoniou]

I wonder how it would work to have something like pressure or up/down movement change the chord voicing or inversion? So for example change pressure to vary the density of the voicing, slide up or down to change the inversion?

Also, how about using the low row to switch this?

[Roger_Linn]

I’ve given this topic some thought and spent some time playing with Logic’s Chord Trigger and Xfer’s Cthulhu.

One problem I’m having is that an extensive UI is needed for an auto-chord feature to be useful, but LinnStrument doesn’t have an extensive UI. At the least, users will want to assign each note in the chromatic scale to his or her chosen chord, because tastes are different. And the UI to do that would be kludgey on LinnStrument, and still wouldn’t offer any advantage over computer software.

One option would be to assign a row of buttons to the chord root and another row to chord mode (Ma, Mi, 7th, etc.) with the Root buttons defaulting to diatonic chords (C = C Ma, D = D Mi, E = E Mi, etc.) But for anything more than very simple music, it would take time to develop skills at this, perhaps as much time as learning to play simple chords. And still the chord voicings would be limited.

Another option used in some auto-chord keyboards is to play 2 simultaneous notes in order to select both the root note and chord mode. For example, playing:

C alone sounds C + G (no third)
C + Eb sounds C minor triad
C + E sounds C major triad
C + F sounds C sus4
C + F# sounds C diminished
etc.

But playing 2 notes to hear a triad takes nearly the skill of playing the actual triad. Plus, I don’t think many LinnStrument players want to play simple majors, minors and sevenths. These days it seems to me that even those with limited playing ability are interested in hearing more unusual and interesting chord voicings.

In summary, I can’t think of an auto-chord implementation for LinnStrument that would be truly useful, not a UI kludge, and better than—or even equal to—a computer application.

[John the Savage]

I wonder how it would work to have something like pressure or up/down movement change the chord voicing or inversion? So for example change pressure to vary the density of the voicing, slide up or down to change the inversion?

Also, how about using the low row to switch this?

I thought about this, but in practice it wouldn't be ideal, because regardless of what voicing you were after, you'd still have to hit the pad and trigger the default chord voicing first. So every chord change would constitute a double-tap of sorts: i.e. first landing on the basic chord and then morphing to the desired voicing. There would be no way to simply grab a dominant 7 chord in the third inversion, for instance, without first triggering the basic triad. And it would be a liability at the best of times, as to whether or not you could coax the exact chord you were looking for out of the pad (presumably while playing a solo with your other hand no less).

I did, however, sit down last night and put some serious thought into the idea I proposed earlier about using a system of "modal interchange". I mapped it out on paper, double-checked my music theory math, made a proper diagram, and I'm just waiting to hear what Roger thinks of it. Honestly, I know the terminology may elude some of you who perhaps don't know your theory all that well, but in reality, I think you'd be pleasantly surprised at just how easy it would be to use. It would provide easy navigation of basic chord progressions and voicings, while still allowing more advanced players and composers to create much more sophisticated progressions. It could all be done in real-time, from one 8x8 grid of pads, without the need for labels or complicated fingerings. And my proposed design includes ways to incorporate the arp and make use of the LinnStrument's multidimensional touch.

The modal interchange idea came to mind because it's far more flexible than using the "Circle of Fifths", easier to visualize, and way easier to implement on an 8x8 grid! In fact, the number of possible chord progressions and chord voicings would be virtually limitless with this system; yet you could start playing basic pop progression straight away using just the 7 pads in the bottom row, and later start experimenting with chord substitutions and more complex voicings at your leisure, simply by playing vertically on the pads instead of horizontally. Seriously, it's that easy!

That said, it all depends on what, if anything, Roger wants to do. And so we wait...

Cheers!

[John the Savage]

Wups! It seems the man was posting as I was typing the above message (sheepish grin)...

Cheers!

[John the Savage]

Okay, so, Roger just got back to me with the following reply:

"I think what you came up with is about as good as it gets. But [...] given that adding this feature to LinnStrument offers no advantage over a computer application, and has very little demand from LinnStrument owners..."

Well, you get the gist. Looks like it's learn your chords or bust, kids (wink). Don't say I didn't try. It was a fun academic exercise for me regardless, but ultimately I agree with the man.

Cheers!

[t-IB]

Well I appreciate the effort that has gone into exploring this idea- I'll bow to better judgement/knowledge in terms of a Linnstrument and application and approach, though I will add that I'm coming at this from a computerless setup, so in my case any implementation is better than what is offered by VSTs and such. Appreciate I'm probably in a minority here and that I'm flighty enough to try out an auto-chord implementation once and then revert to standard chord playing - I never did buy that kordbot after all.

Thanks again folks, I think it's great that we can have an open and considered dialogue around potential features- I certainly see the logic in why this idea proabably isn't a goer, although I don't truly understand mr savages modal interchange idea due to my lack of understanding.

[pdxindy]

Well I appreciate the effort that has gone into exploring this idea- I'll bow to better judgement/knowledge in terms of a Linnstrument and application and approach, though I will add that I'm coming at this from a computerless setup, so in my case any implementation is better than what is offered by VSTs and such. Appreciate I'm probably in a minority here and that I'm flighty enough to try out an auto-chord implementation once and then revert to standard chord playing - I never did buy that kordbot after all.

Thanks again folks, I think it's great that we can have an open and considered dialogue around potential features- I certainly see the logic in why this idea proabably isn't a goer, although I don't truly understand mr savages modal interchange idea due to my lack of understanding.

I also appreciate the open discussion which includes the developer!

Since I do use a computer, I have not explored what is available in hardware. Surely there must be some hardware device with such capability and better suited to it?

[John the Savage]

Hello again, my fellow LinnStrumentalists (grin).

After another brief chat with Roger, he has given me permission to post a draft of the Chord Mode I came up with, just to see if the concept garners any real interest from the community. If so, who knows, maybe it becomes a reality, maybe not. Regardless, we both felt as though the conversation thus far has been interesting enough to merit keeping the dialogue going.

It's nearly 5:30 a.m. here, and suffice to say, I'm hittin' the proverbial hay momentarily, so I don't really have time to write anything long-winded about the concept of "modal interchange"; but I will post the PDF diagram I created to illustrate how it might work on the LinnStrument for y'all to study and ponder. And I do ask that you take your time with it and actually read the descriptions below the diagram. —Ahem!

(smirk)

Anyway, if there's anything you don't understand, please feel free to ask questions, and I will do my best to answer them as time allows...

Okay, here ya go. Have a good day all.

Cheers!

*Edit: Oh bloody hell, it seems as though the forum won't allow the upload of a PDF file... WTF? I gotta go to bed. I'll figure this out later. Sorry, folks.

**Edit: Okay, let's try a JPEG...

[t-IB]

Well that looks very interesting to me- I'm only half awake right now so will go back to this later, but first impressions are that looks like a great way of accessing and forming a rather large amount of chords quickly, with the ability to save too.

[John the Savage]

Okay, so...

Hopefully the diagram above is, for the most part, self-explanatory; but just in case there's any confusion, here's a more thorough overview about how it works and how you might use it. I'm writing this with the layman in mind, so forgive me if I'm being too basic.

First of all, if you're new to music theory, don't get overwhelmed by all the chord abbreviations and mode names. They're just there for academic reasons, as a road map of sorts. In practice, you won't be looking at those on the LinnStrument, and don't really need to. If you're a beginner, you'll most likely just stick to the bottom row of pads at first (which represent the major scale), and indeed there is a lot you can do with that alone. "Ionian" mode is just another name for the major scale, and "Aeolian" is just another name for the minor scale. You can all but ignore the other modes if you want to, and still make most pop music happen.

I designed this interface so that all chords can be played with just one finger, while allowing the user to modify (and save) the voicing of any chord on the grid. So you can set up and map out virtually any progression you want, and not have to memorize or perform any complicated fingerings.

As seen on the diagram, the 7x7 grid of unlit (white) pads contain all the basic chord (triads) for all 7 modes, while the pads around the perimeter (light blue) contain all the voicing options you need to modify those basic triads to create just about any chord you can imagine.

Let's use the basic and familiar 1-4-5 progression (C-F-G) as an example. As you might expect, this would be performed by simply playing pads 1, 4, and 5 on the bottom row of (white) pads respectively. Easy stuff! Now let's say you want to make your 5 chord into a dominant 7 for a stronger cadence. You would simply select the 5 chord, then press and hold the 4th (light blue) pad in the lower perimeter row to add the 7th and save that chord voicing. The pad for that chord will turn dark blue to indicate that it has been modified, and now whenever you press that pad, it will play a dominant 7 chord instead of the basic triad.

You can also use the voicing options "globally" to modify every chord that you play. If you were to simply select the 7 extension (by tapping the pad instead of holding it), that voicing would now apply to any and all chords that have not otherwise been saved with custom voicings. As an example of how this might be useful, let's take our 1-4-5 progression and move it up to the "Aeolian" mode (row six of the white pads). If you were to play through that very same progression using the pads in this row (where we have yet to save any custom chords), with the 7 extension selected in the lower perimeter, you'd get a slightly schmaltzier cycle of all minor 7 chords instead of just the basic triads.

The "Inversion" options work the same way: i.e. you can either save them separately for each chord, have them apply globally, or a combination of both. Playing over a bass line, for instance, you could use the 2nd inversion globally to avoid clashes between the root notes of your chords and the bass line.

Now let's say you were to play that same progression of minor 7 chords again, only this time you used the 5 chord from the first row (Ionian mode) instead of the 5 chord from the Aeolian mode, you will have effectively performed a modal substitution of sorts, by replacing the minor 7 chord with a dominant 7 chord. This is the basic concept behind "modal interchange", and the possibilities are, for all practical purposes, endless!

Just above the "Inversion" options in the left hand perimeter, you'll see options for sharpening the 3rd and 5th of any chord. Being able to raise the 3rd makes it possible to change any minor chord into a major chord, which means you can essentially "borrow" chords from sympathetic keys: i.e. a common example of this would be to change the D minor chord found in the key of C major into a D major chord instead (à la "Circle of Fifths"). The sharp 5 option is there so you can augment any chord on the grid, which would not otherwise be possible. And finally, there's the option to raise any chord in its entirety by one semitone to facilitate chromatic progressions.

I think that's all I'm gonna say for now. Honestly, there's not much you can't do with this structure, and the workflow is about as straight forward and immediate as it gets; especially considering that there is no graphical display, and the idea is to play every chord with just one finger. Combine this system with the arp functionality, and even I have to admit, it would be a lot of fun (sheepish grin).

Anyway, seriously folks, this is your chance to vote for this feature. If it gains enough support, it may actually become a reality. But as I've stated previously, this was just a fun academic exercise for me (hell, I was the guy trying to talk y'all out of it, remember?), so I'm not going to be campaigning for it. It's up to you. If you think this functionality is of value, then bump this thread and let Roger know. In the meantime, I'm happy to answer any questions you may have about the implementation.

Cheers!

[thedommer2000]

Love what you are doing here. I will have to think about it more as my theory is shady at best. I'd also need to think about from a performance position if its intuitive to play this way.

What I will quickly say is that you are not using many of the lights which leaves a lot of space for an animation section. That could be the strum section or maybe It could be animation types. Hold a button down then click the chord in your chart to do that type of animation on that chord. Since one note plays the whole chord some element of animation would be a helping interesting factor.

THEN use the sequencer in Chord mode and then your in real business

[Roger_Linn]

Thanks John for posting your idea. I’ll visit this conversation occasionally to enjoy the rewards of the journey. If a design emerges that everybody likes and is simple and intuitive enough to work in LinnStrument without any new printed settings, perhaps someone with coding or Max/MSP skills will offer to code it up into an app. Note that everything John has described, including the dynamically-changing pad lights, can be implemented in an external app with the advantage of seeing the text identifiers for each pad on the app’s screen.

If such a design emerges and many like it sufficiently that they are commonly using it, and it requires little or no dependence on an externally screen, it could be a good feature addition. In the meantime, I hope many find the conversation as engaging as I do.

[PeterAntoniou]

The idea is intriguing - if I can find some free time I might try to code it into an app at least as a proof of context.