Whatever This Is: Experimental Arrangement

Here's the structure and the overview that shows all figures.

Somehow the workflow always starts slow. I am dropping stuff and shifting it around until something "clicks" and triggers an idea. Around that something, I begin placing other stuff.  I have a knack for slow melodies that ascend across parts like the oboe here (containers), so I do that. Add some faster melodies (*2) that complement them, also not necessarily 100% inside a container (overlapping).

Then I start inserting other harmony. Duplicating melodies and stacking them until the interval sounds best. Trying to meet different intervals of the chord. Deleting (many) notes to avoid clashes.

And then suddenly, exactly when there is enough critical mass to play with, I experience a kind of Cambrian Explosion. I make aliases and shove them to not so obvious places. This is what happened with the last modulation here. It was a surprise. Beautiful intervals and different melodies emerged. The harmonies this hints at are really shining through. It could start a whole new, much brighter section ...

Lacking the time and expertise though, this will probably remain a pile of ideas without a consistent narrative. If anyone wants to make their hands dirty, feel free to download the file and do whatever comes to your mind (and share it with us).

Good point. This "Edgy Oriental" progression yearns for resolution. It's full of tension. Not a good thing to start with ;-)

Yes, I tend to jump right into the middle immediately. That's a common trap I would think. As you are previewing phrases all the time, you get used to them. That's kind of a "hidden" preparation. A new listener didn't have that.

Well, as expected, this thing doesn't work any more after 24 hours. I mean, not at all. All the things that I thought were beautiful are completely lost now. The melodies don't work, the harmony changes don't move as originally intended. It has fallen apart.

And I am a "prepared" listener that merely moved a few inches closer to an "unprepared" one. 

Ok, I could listen to it a couple times again and they will probably come back, but that's not how music is supposed to work. The goal is that at least 50% come across on first pass.

I refuse to accept that the only way to create something that "works" is to follow a cliché or a well-trodden path. When spontaneous imagination is so prone to leading you astray, wow. I guess I have a lot to learn if I wanted to tame that kind of creativity. In any case, you can't rely at all on the fact that the experiences you have while composing will come out even rudimentarily in the end.

This is such a frustrating experience. It must have a name. How's it called? Composer bias? Creator delusion?

I would like to point out that this is not Synfire's fault. Quite the opposite. Only Synfire made it possible for me to create a sketch like this in just two hours. Just imagine what would have happened if I had spent several days working on it manually, note by note. The frustration would have been devastating.

It's the man in front of the screen who's to blame :-)

Mixing is also important. I find it especially hard to get the articulations and dynamics right, as I hear them in my head. That is also a source of "hidden preparation", i.e. you tend to hear the intentions in your head rather than the actual experience.

Many years ago I did all my music manually. It is true that the risk of veering too far off is lower, but you also come up with a lot of clichés and stereotypes and boring or cheesy stuff. Not everyone is a genius. I rather prefer a bit of creative chaos from which I can pick the things I like best, than being confined to my current manual capabilities.

I think the key is: Kill your darlings. You must not be afraid of deleting stuff that's simply too much or not in the right place. Maybe put it aside in a library.

It's the same with writing. Painting is more difficult to "undo" but I think the urge to start over is not uncommon.

It begins to dawn on me what's wrong with this arrangement. Some important changes are unmotivated (not properly prepared for) and not established enough to have the desired effect. After every suspense, there is a range of possible "surprises" that will "work". If a change is too sudden or outside this range, (or too short) it won't work.

Yes, it is too ambitious on too many levels (chaos). 

Here's the plain progression in open voicing auto-chords. It's full of tension, then lightens up, and gets back to tension, etc. It's rough at times, but I still like it overall. Tanders is right: The tension in the beginning is actually already somewhere in the middle of a drama. It's a very bad opener.

The 2nd MP3 is the full enchilada, pizza with everything, mustard and rusty nails: The auto-chords progression overlaid with the instruments. Beware - it's hefty. Kind of Rorschach test. I listen to it from a distance, sensing the melodies that come up in my head, and will replace them next.

Nice! It took me a few listens to get used to the sounds, but it's really more approachable from the start. I always find it fascinating when others take an arrangement and do something different with it.

I'm doing a pop song right now as a new demo track (maybe tutorial). So this experimental stuff needs to be postponed until later.

Don't put too much stuff in it.

It has a few more chords, but a simple and catchy melody.

Minimalism is a challenge. Back in the day when I was a band leader, the A&R manager told us that our songs were becoming too "musical". They were afraid we would actually start making real music and venture into stuff that was no longer mainstream. 

Don't underestimate pop, though. Every now and then I hear a song that really touches me. And when I look at it, it is very simple stuff, done right. Much of Mozart was simple stuff (for the time). Doing simple stuff is hard.

The good thing with simple stuff is that it immediately "works". You may like it or not, but everybody will immediately understand what's going on. Inexplicably, something magical sometimes shines through between the lines of even the simplest music (often it's the vocalist, sometimes a simple but effective chord change). Only after listening to it several times does it disappear (at least for me).

Now I'm getting closer to something more palatable.

I decided for more conventional harmony and stuck to 2-3 motifs that stood out (for me). It is a very different experience, although you will notice many similarities. I think the essence of what I liked with the first attempts could be preserved.

The piece is in Eb Minor and jumps to B Major, Db Major and Gb Major. If you look at the Circle Of 5th, these keys are close neighbors but still somewhat distinct in their sound and chord repertoire. On a notation sheet you would probably not even mention them and rather use accidentals.

As always, I'm too impatient to wait a day or two before posting. Chances are I won't like it any more the next day, so be it. I'm fine with setbacks like that. Let me know if there's something you like, or something that doesn't work for you as an unprepared listener.

Yes, definitely more constructed than composed. Well put. This happens when there are conflicting details and distractions you want to explore all at the same time. It's a puzzle that doesn't match.

This experiment is still a crappy puzzle. I'm struggling to deal with the idea overflow that Synfire throws my way. Too many things flash up in my imagination and I can't get them working together. Sticking with only one or two style elements still eludes me.

Regarding my penchant for hammering marimbas and wood blocks, the Short Ride In a Fast Machine is my favorite of John Adams. He's a true master. The percussion doesn't quite get through in this recording, but I have a CD here where it hammers like mad in the foreground and yes, I like that (also love Muslimgauze, maybe know them).

As far as I remember there are no marimbas in Pale Rider, though?

Happy New Year!

Oh, and the harmony that builds up when the trumpets come at 3:00 still blows me away. Should have analyzed that ... when I think of it.

Ah, OK. I love it (the original, not your clip). I could play the sound all day. It's the essence of the song. Didn't expect it to be so polarizing ;-)

Chromaphone 3 from AAS (Applied Acoustic Systems) is a physical simulation. The entire plug-in is a source of magnificent percussive sounds. Their other simulations are also great, String Studio 2 for example.

Ok, here's now a very smooth version of this (at least that's what it feels like NOW, maybe it'll be totally different tomorrow).

I consequently kept the stuff I like most (the fanfare-like modulation  parts) and remove what didn't work (the constant dissonant staccatos - goodbye oriental chords). In order to prepare for the fanfare parts, I simply re-used the rhythm for some melodies in the between parts. 

The overall rhythm is now very predictable to allow for more harmonic and counterpoint surprises.

Does it work now?

I find it quite interesting how close the extremes, gentle vs. crass, actually are. If you smooth it out a bit, it gets kitschy pretty fast. If you put in some rough elements, it's like flapping mustard on your ice cream. Other genres are more forgiving, that much I can say.

And I can say with confidence that I am not the right person to compile a large library with orchestral building blocks manually (as much as I would love to have and share one). Listening to all kinds of styles and extracting their essential phrases would take considerable time.

Maybe training Factories from MIDI files will do a better job. At least they will deliver more unpredictable and surprising stuff.