A Dream of Crows: the chords

I am in the process of writing a series of articles about the songs on my sixth Jamendo album, In Unexpected Places.  This is about the tenth track, “A Dream of Crows.”

“A Dream of Crows” started out as two song fragments, older even than my “Rhythm Method” Prototypes.  When I was composing my previous album, Martian Winter, I did a series of prototypes called “The Periodic Table of the Groove.”  There was a lot of good material there, and some of the fragments I meant to come back to later.

The story of how these two parts crystallized into one is a good one, so I’ll write about that next time.  The title “A Dream of Crows” suggests something about its origin.  But today, I want to cover the song structure and chords.

Overall Structure

There are two main parts in this song.  All of the prototypes I created for “The Periodic Table of the Groove” were in odd time signatures — these two parts are actually in different odd time signatures.

• Part A, the “verse”, is in 7/4 swing.  (Technically, I’d call this 33233 7-clave 4:3 swing . . . see my article “Rhythms in Seven.”)
• Part B, the “bridge”, is in 5/4 swing.  (Again, I’d call this by its full name, 2332 5-clave 3:2 swing . . . see my article “Rhythms in Five.”)

Part A (7/4)

The main part of the song is actually a Bluesmorph (as defined in a previous article in this series) — it’s structure can be mapped to the standard blues progression (think minor blues, with some extensions and alterations):

Dbm9	%	%	%
B13#9	%	%	%
Dbm9	%	Em7	%
Eb7#5	%	Dbm9	%

Part B (5/4)

The bridge is probably my favorite part of the song — the part which, to me, suggests dreaming.  It uses the following chord progression:

Abm7	%	BΔ	F#Δ
Abm7	%	EΔ	Ebm7
Abm7	%	BΔ	F#Δ
Abm7	%	EΔ	Ebm7
Db13#11	%	Eb13b9	%

About the Diagrams Below

As I’ve done in previous articles, I’ve given some chord voicing diagrams in the sections below.  However, this time, I had to make a huge simplification:

I want the diagrams to focus on the notes and the voicings.  These songs are written in odd time signatures.  If I included all the correct rhythmic information, they would become much more difficult to understand at a glance.  Therefore, I wrote all the voicings below “as if” they were in 4/4, so I could get away with using clean looking whole notes for each bar.

As usual, the diagrams were prepared with GNU Lilypond 2.12.2

Verse

Below, I’ve taken the chords from Part A and shown the approximate voicing I used for each.

Chord voicings used in the "verse" section of "A Dream of Crows." The rhythms have been simplified for clarity. Click the image above to enlarge.

Probably the most interesting voicing is that mad, mad voicing for B7#9.  I break several “rules” here.  Ordinarily the more traditional music theorists balk at chords with a minor 2nd interval in them . . . not to mention two!  And yet the voicing C## – D# – G# – A represents exactly the color tones of the chord: the #9, the major 3rd, the 13th, and the 7th.

This is a great voicing, but as I said before, voicings of this type won’t work on every keyboard instrument.  A concert grand, yes; a cool electric piano, maybe; a parlor upright, possibly; a Hammond organ set up with a lot of 5th overtones . . . nope.

The Bridge

I used two different types of voicings for the bridge chords, shifting from single handed to two handed.

Single Handed

These are simple enough, but set the stage for what follows.

One handed chord voicings for the bridge of "A Dream of Crows." Rhythms have been simplified for clarity. Click the image above to enlarge.

Two Handed

At the end of the bridge, the song has to “turn around” and go back to the verse with some sort of grand flourish.  I held back on the chords in the first part of the bridge, only to plow through the end with rich, five-note voicings:

Two handed chord voicings for the bridge of "A Dream of Crows." Rhythms have been simplified for clarity. Click the image above to enlarge.

[Note: these voicings should read "D13#11" and "Eb13b9", but I've never been able to get Lilypond to understand that the 13th is particularly important to me -- it always sticks it off on the end, as in "Db7#11add13".  So I simplify the input to remove the clutter.]

These type of voicings are called Upper Structure Triads.  (By the way, as always, I highly recommend Mark Levine’s Jazz Piano Book — it gave me a solid base from which to develop my own chord theories.  Upper Structure Triads were one of the best voicings I learned in that book.)

The basic definition for this type of chord is “A tritone on the left, a triad on the right.”  Here is the analysis of each voicing:

• Db13#11 is voiced as an Eb major triad (Bb – Eb – G) over a B – F tritone.  The B (actually, Cb) and the F are the 7th and 3rd of the chord, the guide tones.  The Eb major triad spells out the color tones — Bb = 13th, Eb = 9th, G = #11th.
• Eb13b9 is voiced as a C major triad (C – E – G) over a Db – G tritone.  Again, the tritone notes are the guide tones for the chord, and the C major triad spells out the color tones — C = 13th, E = b9th, G = restatement of the 3rd.

Step Through This Window: the acoustic guitars

I am in the process of writing a series of articles about the songs on my sixth Jamendo album, In Unexpected Places.  This is about the ninth track, “Step Through This Window.”

Before In Unexpected Places was published on Jamendo, I sent a preview copy of this song out to a few friends of mine.  It’s one of my favorites on the album, and it has a very subtle feature — the complementary acoustic guitars.  I used a certain technique to get that sound which I’ll describe here.

This song owes a lot to the “folk” tradition, at least in concept.  It’s based on one of my earlier “Rhythm Method” prototypes, P35 (hosted on SoundClick, so you can listen to the original and compare).

Overall Structure

Chord wise, there are really only two parts to this song, “A” and “B”.  The rhythm, pacing, and drum line change throughout, and distinguish each of the key development points.

Part A

The A part is simple — only two chords.  (Quite a departure for me!)  Most of the time, the chords alternate between Em and A2 for 16 bars.  (A2 is my way of writing “A suspended 2nd”, meaning: take an A triad, lose the 3rd, replace it with the 2nd.)

Part B

The eight bar “chorus” is relatively simple as well:

C	Em	C	Em
C	Em	D2	D2

Acoustic Guitar Parts

The important thing to realize about the acoustic guitar part is that there are two guitars playing.  One is played as normal; the other is played with a capo at the 5th fret.  I play the same rhythms on both, but choose different chord voicings on each.  This gives a “natural chorus” sound for the resulting chords.

Below, I’ve given diagrams for each of the guitar voicings using GNU Lilypond.  This is the first time I’ve used Lilypond to make fingering diagrams — pretty easy, really!

Part A

On Guitar 1, I play the standard voicing for Em, and an easy voicing for A2.  The trick here is the complementary line on Guitar 2 — the capo at the 5th fret lets me play a “D” shape and get the sound of G, and an “E” shape to get A.

Acoustic guitar parts for "Step Through this Window." Click image to enlarge.

Why does this work?  At one point, I’m playing Em on one guitar, and G on another.  These chords work together because in total, they spell Em7 (E G B D)!  This is just a huge voicing (extending from that low E to the high 7th fret B).

Part B

I use the same trick on the Em chord in the B part.

Acoustic guitar parts for "Step Through this Window." Click image to enlarge.

Commonalities

Another composition principle for these two voicings is revealed by the top note played for each chord.  Notice:

• Guitar 1 always has the 1st string open “E” as its highest voice.  This establishes a focal point for the progression, a folksy “drone.”
• Guitar 2 always has a descending line within its chordal unit.  For example, in the “A” section Em – A2, the highest note is B, then A.  For the “B” section, the highest note descends as C – B – A.

Ubuntu 9.10, Maxima, and wxMaxima

Continuing on the topic I presented in a previous article:

I’ve encountered my first aggravating problem with Ubuntu 9.10.  It took over a week, so that’s heartening.

The Ubuntu 9.10 version of Maxima (5.17.1, built with GCL 2.6.7 [ . .  somehow! see below]) doesn’t work.  It crashes with a bizarre error whenever it is asked to “solve” anything:

mccosar (1) -> ~
$ /usr/bin/maxima

Maxima 5.17.1 http://maxima.sourceforge.net
Using Lisp GNU Common Lisp (GCL) GCL 2.6.7 (aka GCL)
Distributed under the GNU Public License. See the file COPYING.
Dedicated to the memory of William Schelter.
The function bug_report() provides bug reporting information.
(%i1) solve(x^2 = 4,x);
Universal error handler called recursively (:ERROR NIL
    CONDITIONS::CLCS-UNIVERSAL-ERROR-HANDLER
    ""
    "Couldn't protect")
Universal error handler called recursively (:ERROR NIL
    CONDITIONS::CLCS-UNIVERSAL-ERROR-HANDLER
    "" "Couldn't protect")
Maxima encountered a Lisp error:

 Error in CONDITIONS::CLCS-UNIVERSAL-ERROR-HANDLER [or a callee]:
 Caught fatal error [memory may be damaged]

Automatically continuing.
To reenable the Lisp debugger set *debugger-hook* to nil.

Whoops!  OK, so something’s wrong.  That’s alright, my recent install of Sage includes Maxima 5.19.1 (compiled with an inbuilt ECL 9.8.4) — it works fine on its own, with no errors.

However, when I tried using it with wxMaxima, it wouldn’t work.

I’ll prove this later, but wxMaxima is really worth taking the time to get working right.

Search and Research

Research time.  I compiled my own version of Maxima 5.19.2, the latest and greatest, using SBCL (1.0.29.11.debian).  It worked fine, and was quite speedy!  But still didn’t work with wxMaxima.

OK.  The Ubuntu / Debian version of Maxima was compiled with GCL . . . I set out to do the same.  Problem: the GCL included with Ubuntu 9.10 [somehow!] wasn’t configured with --enable-ansi.  So in turn I tried to compile a new version of GCL . . . .

What a nightmare!  Gnu Common Lisp hasn’t been released since 2005.  I spent an hour trying to configure, reconfigure, update from CVS, and so on.

In the end, phooey.  GCL seems to be a defunct project (and don’t hand me that “there’s a lot of activity in CVS”, because there’s 0, zero, none, nil in terms of actual releases).

So as I mentioned above — how Debian compiled Maxima with GCL in the first place is a total mystery.  They must have a patch file as big as a hard drive partition.

Rethinking the Problem

Up until this point, it hadn’t occurred to me the problem might be in the Ubuntu version of wxMaxima.  In the modern era of computing, it’s relatively rare for a program to have a severe bug, and a program that uses that program to have an equally severe bug.  With tens of thousands of packages out there, the odds of two striking in the same chain are pretty low.

I guess I’m looking at some pretty compelling evidence it can still happen, however

I compiled wxMaxima 0.8.3a from source, and installed into /usr/local.

It worked FINE with the new version of Maxima (the one compiled with SBCL).

wxMaxima

If you haven’t had time to check out the combination of Maxima and wxMaxima — and if you’re even remotely interested in computer algebra systems — this system is a must-see.

Overall, the system reminds me of the Sage notebook (see previous article).  The pages you construct are interactive with the underlying Maxima system, and produce some nicely formatted mathematics.  The website itself has tutorials, which — get this — are downloadable as wxMaxima pages, making them totally interactive.

That alone is a win!  And the system itself is quite intuitive and easy to use.

Enough talk.  Time for a screenshot.  Below, I’m asking Maxima to evaluate a double integral (one of the problems from Tom Apostol’s Calculus, Volume II).  wxMaxima solves the problem exactly — notice it even outputs the answer as a fraction, not a decimal approximation.  However, it doesn’t do too well with formatting the double integral itself, which should read:

Screenshot

wxMaxima demo screenshot. Click image to enlarge.

Ubuntu 9.10 and the Korg D888

Just a quick note to follow up on my previous article:

For home recording, I use the Korg D888, a reliable 8-track digital recorder.  I do not use it for processing audio, however.  For that, I attach it to my computer via USB.  This lets me offload the audio so I can mix it in Audacity, and even to send the mixed audio back as a new .wav file on the D888.

Well, in previous versions of Linux, the D888 has always shown up as just another USB drive.

Until now!  On Ubuntu 9.10 — it actually shows up as “KORG D888″ on the desktop, and automounts as “/media/KORG D888/” (I do hate the fact there’s a space in the name, but I’ll get over it.)

Here’s a screenshot I took while the D888 was mounted (click on image to enlarge):

Ubuntu 9.10 desktop, showing the Korg D888 icon. Click to enlarge.

[The wallpaper I'm using, by the way, is from the image I used for the cover art on In Unexpected Places.]

Here’s a blowup of the detail in the upper left corner, showing the D888 icon:

Ubuntu 9.10 and Sage 4.2

On October 25th, I installed the 64-bit version of the Ubuntu 9.10 release candidate.  Since then, Ubuntu 9.10 has been officially released.  I thought it was time for an update, which follows up on something I mentioned in my previous article: the Sage mathematics software system.

Origins

I have noticed a pattern among musicians.  A lot of us who are into jazz and music theory are also into higher mathematics.  I became interested in math in my early years through astronomy and physics.

One of the challenges of doing math well is checking your computations.  My eternal enemy is the dangling minus sign, placed there in front of a large fraction like it was just a continuation of the bar:

[By the way, it's not a well known fact, but WordPress supports LaTeX markup for its blog posts.  Here's a short FAQ on the subject.]

Computer Algebra Systems

What a calculator does for simple mathematical operations, a computer algebra system does for equations and algebraic operations — even calculus and differential equations.  For Linux, there are some fine — and freely available — systems out there:

• OpenAxiom (a descendant of Axiom, which has been in development since 1971):  I chose OpenAxiom over the original because it compiles cleanly and runs quickly.  [I built OpenAxiom 1.3.0 on Ubuntu 9.10 using Steel Bank Common Lisp (sbcl 1.0.29.11).]
• Maxima (in development since “Macsyma” of 1982): The version released with Ubuntu 9.10 doesn’t work (5.17.1).  However, a working version of Maxima is built as part of the next system . . .
• Sage, the newcomer to the field.  Actually, it’s a mixture of old and new: it uses a modern computer programming language (Python, not Lisp, thankfully) to tie together CAS systems like Maxima with computational systems like GNU R.

About Sage

Now, Sage and I have an interesting history.  I’m already a Python programmer.  However, I’m also a dedicated Linux user, and like to try out different distros and system configurations on my spare disk partitions.  There are three major build tests I apply to any system — can it build Csound, the most current Python 3, and Sage?

Sage became a standard candle because it is almost entirely self-contained — the source package you download has almost all the source code you need (it still needs to link to system libraries, of course).  Compilation takes a long time, and the test suite is exhaustive.  A good way to check out a system, in other words!

I tried building Sage 4.1.2 on Ubuntu 9.10.  It didn’t work.  But then, Sage 4.2 came out.  Success!

Putting Sage to Work

To demonstrate Sage, let me show some typical output.

In the problem of finding the arc length of a curve, there aren’t many easy integrals.  Let’s try a simple example, a modified version of the “arc length of a parabola” problem.  I tested Sage with this indefinite integral:

Terminal Output

Sage can work in an interactive terminal environment, of course, and even as a souped-up numeric version of Python.  Here I’m showing a short terminal session, in which I ask Sage to solve the integral above:

mccosar (0) -> ~
$ sage
----------------------------------------------------------------------
| Sage Version 4.2, Release Date: 2009-10-24                         |
| Type notebook() for the GUI, and license() for information.        |
----------------------------------------------------------------------
sage: integral( sqrt(1 + x^2), x )
1/2*sqrt(x^2 + 1)*x + 1/2*arcsinh(x)

As you can see, it did the math. But in a terminal, it can’t really handle the formatting.

Sage Notebook

The most brilliant feature of Sage, in my opinion, is that the designers completely abdicated on the responsibility of designing a clever GUI for the program.  Instead, they let you turn Sage into a webpage server on your computer.  That way you can use the rendering abilities of your favorite web browser.

Now, you can activate this from the command line, by issuing the command sage -notebook.  However, I’m very cranky, and I hate for any window to pop up on my desktop unless I’m good and ready.  Therefore I put the following two files in ~/bin:

sagenb.sh:

#!/bin/bash

exec sage /home/mccosar/bin/notebook.sage

notebook.sage:

notebook(open_viewer=False)

Now when I issue the command sagenb.sh, the Sage server starts and is ready for me to go to the address http://localhost:8000/.  This is the interactive Sage notebook!

Here’s a screenshot of Sage in action on my original integral problem:

The Sage Notebook in action. Click the image to enlarge.

The difference in the two lines shown is that in the second, I’ve clicked the “Typeset” button at the top — Sage formats the output in nicely rendered mathematical symbols.

Checking Our Work

And now to put the cart before the horse.  Ideally, I like to use Sage to check my calculations.  Yes, that’s right — even with this powerful tool, I like to carry out the calculations by hand.  (Up to a point.  I believe in tables of integrals just like everyone else.)  Here, I’m going to do the reverse: check Sage’s work by hand.

It was very easy to get an answer using Sage.  But was it the right answer?  Can we reproduce the steps that led to this integral?

I can, I did.  Unfortunately, although WordPress has some nice LaTeX features, it’s not quite up to doing something equivalent to a proof.  I need equation numbering and some nicer formatting.  Therefore, I wrote the following PDF document in standard AMS-LaTeX, using Texmaker and pdflatex:

Sage Verification (PDF, 38 Kb)

If you happened to have blundered onto this page while taking a first year calculus course, this is a fairly good example of using trigonometric substitution to evaluate an integral.

Ten steps.  Wow.  Now you see how valuable a program like Sage can be!

Ubuntu 9.10

Everything I’ve discussed above was performed on my new Ubuntu 9.10 system.  As I said in my previous article, it is operating as an alternative to my main system, Fedora 10 / CCRMA.

I like to test my systems by doing actual work on them.  Today is November the 1st.  A month ago, I released my sixth Jamendo album, In Unexpected Places.  That album was mixed and mastered using Fedora 10 / CCRMA.

It’s a month later, and it’s time to start making more music.

I’m going to try using both systems, and see which works better.  Ubuntu 9.10 has the full advantage of the 64 bit processors.  Fedora 10 has the advantage of reliably configured audio software.  Which will be better?

We shall see.

The Sky in Shadow: bass and chords

I am in the process of writing a series of articles about the songs on my sixth Jamendo album, In Unexpected Places.  This is about the seventh track, “The Sky in Shadow.”

Wow — I’m at track 7.  I still feel a bit guilty about not finishing my previous series of articles for “Martian Winter.”  And yet here I’ve reached the halfway point!  Still, if I’ve learned one thing . . . it’s not to take the future for granted.

In this article, I’m going to focus on the tremolo bass and the chord voicings.  As before, the diagrams that appear in this article were prepared using GNU Lilypond 2.12.2.

Overall Structure

“The Sky in Shadow” is actually based on two of my original “Rhythm Method” prototypes.  You can listen to these early musical sketches on SoundClick (I’ll give a link to the exact tracks below).

The first part of the song is based on P25.  There are two parts, which alternate:

• Part A: 8 bars of Bm7.
• Part B: 4 bars of Dm7.

The “chorus” of the song — that smooth keyboard part — is based on P06.  Here the components do not alternate — they occur as the series C, C, D, then return to the main thread of the song.

• Part C: 4 bars of Dø, then 4 bars of G9#11.
• Part D: 4 bars of Cm9, then 4 bars of Am9.

Tips

Part of the key to tremolo picking on bass is effective palm muting.  Open strings sound great.  Playing far back on the neck — down at the nut, or on open strings — sounds great.  However, the strings have to be muted at least a little bit using the picking hand.  It’s especially important to mute the strings that shouldn’t be sounding.  Under a furious tremolo bombardment, they’re going to start sounding (one way or another) unless they’re muted.

Also, to follow the discussion, it wouldn’t hurt to read my standard chord abbreviations if you’re new to jazz-type harmony.

Part A

This is a straightforward Bm7 riff in the key of A major (making this B dorian).  I’ve shown a clave staff on the diagram below to draw your attention to a subtle feature: even though the bass isn’t explicitly stating the son clave (here written against a 16th note rhythm), it’s there, pushing the music along.  Notice the note changes occur on the “3 side” of the clave.

The "A" bass part for "The Sky in Shadow." Click to enlarge.

There is one addition, however: that final note, which leads in to the next bar.  It’s effectively an eighth note push, and it also helps to propel the music along.

Part B

Here, the Dm7 riff (key C major, so D dorian) occupies only half the bar.  It’s the space that’s important for the feel.  The final note is highly syncopated, almost a “silence push” of an eighth note.  (That is, it’s like the rest in the second half of the bar has bled over past the invisible half-bar line and shoved the notes back by an eighth note).  Again, the note changes line up with the 3-side of the clave.

The "B" bass part for "The Sky in Shadow." Click to enlarge.

Part C

Here the chord voicings and the bass become very important.  The voicing I’ve shown here is a simplified representation of the way the song is played — for clarity, I’ve completely omitted the rhythm.

The chord for the first part of the C section is Dø.  It can be played two ways.  Here I’ve shown it as being in the key of Eb major (and therefore D locrian).  However, it’s also fair to say it could be from the sixth mode of Fμ (F melodic minor, or F jazz minor).

The voicing I’ve shown is, in fact, an inverted F minor triad over a quartal D-G.  That’s the basis for a lot of good chord voicings — a triad or other familiar unit on the right hand, and a simple 4th, tritone, or 5th on the left hand.

The bass and chords for part C of "The Sky in Shadow" (first section). Click to enlarge

Again, above the bass changes line up with the 16th note son clave — with the addition of a final note.  The same idea follows in the complementary piece of the C section, the chord G9#11.  (I tried to get Lilypond to render this as “G9#11″, but it seems to like calling it “G7#11″ instead.)  This chord is in the famous jazz “Lydian Dominant” mode, or the fourth mode of Dμ.

Below, notice the G9#11 voicing might be — to some — a bit bizarre.  There’s no third!  In fact this is an idea I got from listening to Thelonious Monk, and from Mark Levine’s Jazz Piano Book.  The voicing is actually an augmented triad — F+ — over a quartal D-G.  Notice how this fits in with the previous section.

The bass and chords for part C of "The Sky in Shadow" (second section). Click to enlarge.

Another idea to notice is the range of the notes I’ve played so far on the bass.  I went from predominantly the 4th string all the way to this section, which uses mostly the first string.  Playing the full range of the instrument in a song can be an effective way to differentiate the sections.

Part D

The “turnaround” from the chorus is a series of two chords.  The first part of the D section is a particularly tense voicing for Cm9.  (Shown here as C dorian, or the key of Bb major).  The right hand part is an actual “Thelonious Monk” voicing for minor chords.

If you’re a strict piano theorist, you always hear to watch for notes that are a minor second apart.  Well, this voicing luxuriates in the chaos of D-Eb colliding mid-chord.  The trick here is, you have to choose your instrument very carefully.  This voicing is good for electric piano, most large acoustic pianos, and a few synths.  It’s not too great for instruments with a lot of octave and 5th overtones (eg a Hammond organ on a drawbar setting like 888800000).

The bass and chords for part D of "The Sky in Shadow" (first section). Click to enlarge.

Finally, the last chord of the chorus is a variation of the original idea that got us here.  The chord Am9 (A dorian, or key of G major) is voiced as a D major chord on the right hand and an interval of a fifth (C-G, the guide tones for the chord) on the left hand.

The bass and chords for part D of "The Sky in Shadow" (second section). Click to enlarge.

My Own Avalon: the congas

I am in the process of writing a series of articles about the songs on my sixth Jamendo album, In Unexpected Places.  This is about the sixth track, “My Own Avalon.”

The musical feature that most people seem to notice about “My Own Avalon” is that eerie, mystic-sounding synth.  That’s no ordinary synth — that’s the freely available ZynAddSubFX software synthesizer.

I actually wrote this song a long time ago.  I recorded the first version of it on April 11, 2006 — it appeared on my “lost” album, apocrypha.  (If you’d like to hear the original and compare it to the new version, here it is on SoundClick.)

The odd thing is, the melody, the synth, and even the bass all came later.  The first part I wrote was the drum line!

The Three Congas

On December 27, 2007, I wrote an article about the composition of the original song, and described how I developed the underlying rhythm as a chant, then transcribed it to a drum machine.

Flash forward from the original, three years and some change.

I have real congas, three in fact.  They come in three sizes, and they have different names — from largest to smallest, they are tumba, conga, and quinto.  Each of these can be tuned to different resonant notes.  Congas are halfway between pitched percussion like tympani drums and (relatively) pitchless tom toms.

Below, I’ve shown four musical staves.  [This diagram, as usual, was produced using GNU Lilypond 2.12.2]  The first, ensemble, shows the “melody line”, if you will, for the drums.  This can be resolved into three overlapping drum parts — the quinto in C, the conga in G, and the tumba in E:

Arguably the best introductory book I’ve ever read for conga playing is Conga Drumming by Alan Dworsky and Betsy Sansby.  I’ve taken the above conga parts and rendered them below in the “grid drums” style used in the book.  [Key: O = open tone; Δ = "slap"; X = bass tone.]

I’ve been careful to give credit to Lilypond every time I produce a nice sheet music snippet.  Well — the above grid diagram actually was created using OpenOffice Spreadsheet 3.1

Ubuntu 9.10 [64 bit]

Yesterday I upgraded to the new Ubuntu 9.10 Release Candidate (see also the release announcement on DistroWatch).

As I wrote in the article Software I Use, my primary system is Fedora 10 / CCRMA.  However, for maximum audio compatibility, I chose to make the Fedora version x86 compatible (running 32 bit on a 64 bit machine).

I have always dual booted.  Occasionally, I’ve even hextuple-booted  So Fedora remains in place, but there is a newcomer:

Ubuntu 9.10 in Glorious 64-bits

I am still in the early stages of testing this newest system, Ubuntu 9.10.  I did not use an install CD — instead, I upgraded my existing 64-bit version of Ubuntu 9.04 using “update-manager -d”.  Here’s my experience so far:

The Trickle Down Effect

Whoo!  The download was slow.  I needed to upgrade and install thousands of packages.  Now, I have a reliable high speed internet connection (Verizon FiOS), which normally lets me pull stuff in at 2 Mb / sec.  I was down to 30 – 100 Kb / sec!

All I can figure is that the mirror was overloaded with enthusiastic Ubuntu fans.  Probably when the big release comes, bandwidth will improve.

The Upgrade Itself

The actual upgrade went smoothly, with not even a single hiccup.  Considering how much stuff was shuffled around, reconfigured, replaced, and so on, that’s really quite a miracle.  I rebooted, and everything cruised along so easily that it would have been possible to miss the fact there’d been an upgrade at all.

What’s Working

I’ve only had about 12 hours to mess with this system, but here are a few highlights and observations:

• The Intel HDA sound card works perfectly.  This has been fixed since Ubuntu 9.04, but it’s nice it didn’t get “unfixed” during the upgrade
• The nvidia driver works.  I turned off desktop effects, though, because I’m cranky, and they’re annoying
• I was pleased to see that some of the programs I use all the time are at the current release version.  GHC is at 6.10.4; Octave is at 3.2.2; Python3 is at 3.1.1+.  Lilypond is at the current stable release, 2.12.2
• I was able to compile fresh versions of Csound 5.11 (double samples) and Phasex 0.12.0.
• I installed the 64-bit version of Opera 10.00 — works fine.
• Hydrogen (0.9.3) on Ubuntu 9.10 is actually behind the times — CCRMA has had 0.9.4 for some time.  The differences aren’t huge.
• Sage is waaaay behind the times (installed as 3.05, when the current version is 4.1.2).  However, I can’t be too much of a smarty pants about this — I tried compiling Sage 4.1.2 myself, and it failed (the included version of R didn’t compile).

I have tried Jack, and it works, but not too well.  On Fedora 10 / CCRMA, I never get xruns, even when I’m doing something audio-intensive that really deserves them.  Evidently I need the realtime kernel, and so far, I haven’t had a chance to try installing and running it.  In the past, that’s led to problems.  Speaking of which:

Boo Boos

The only errors I’ve encountered were warning messages intercepted by apport.  I sent these to Launchpad, using Ubuntu’s automated bug reporting system.  None of the errors visibly affected my system or its performance.

In the End

A very good experience so far.   Will it replace Fedora 10 / CCRMA ?

The advantage Ubuntu 9.10 has is its modernity — the software is across the board at or near the current released versions in every application I checked.

On the other hand, Fedora 10 works extremely well — transparently — with a realtime kernel, and the CCRMA audio applications are perfectly configured.  After all, I recently recorded, mastered, and released an entire album (In Unexpected Places) using this system.

Time will tell.  It’s nice to have a fast 64 bit system as an alternative when number crunching speed becomes important.

If the Future Never Comes: chords and guitar voicings

I am in the process of writing a series of articles about the songs on my sixth Jamendo album, In Unexpected Places.  This is about the fifth track, “If the Future Never Comes.”

I could argue that the most important musical element in “If the Future Never Comes” is that wild bass line — it first made its appearance on the original “Rhythm Method” project prototype, P07 (hosted on SoundClick).  But in this case — unlike most of my music — the bass line was written second.  First came the chords.

Guitar Voicings

Here I’ve shown the guitar voicings I used for the main part (the verse sections) of this song, in both standard notation and tablature.  For clarity, the actual rhythm is not shown.  It may also help to look at my list of standard chord abbreviations if you’re new to jazz.

This illustration was generated using GNU Lilypond 2.12.2, which, as I’ve said, is capable of some very beautiful output.  I’ve crunched it down to a medium resolution PNG here, but if you want to see the original PDF, here it is: ifnc.pdf [83k].

If the Future Never Comes -- guitar voicings (rhythm not shown). Click to Enlarge.

Key points:

• This is only the structure of the verse sections.  There’s a chorus as well, but it’s a bit more abstract.
• Although there are two standard ii-V sequences in this progression, neither of them end up in the traditional place. Abm7 – Db9 leaps up to Fø; then the minor cadence Fø – Bb7#5 ends up in a major chord, AΔ.
• The first three chords are actually a bluesmorph (see previous article) of the last four bars of the blues.  Normally the progression might go E7 – D7 – A7 – A7.  I’ve changed the chords to Em9 – D9 – AΔ.
• The type of chords I’m using here are a mixture of my drop 2 voicings and another set I’ve not published here yet — which I call “drop 3“.  A good example is the voicing for Abm7.  Take the standard spelling for Abm7: Ab – Cb – Eb – Gb.  Put it in another inversion, for instance Gb – Ab – Cb – Eb.  Now, drop the third note from the top to the bottom: Ab – Gb – Cb – Eb.
• Another great chord voicing is the one I use for AΔ, which is actually AΔ13 (A major 13th): A – G# – C# – F#.  Notice that the top three notes are actually a quartal voicing (based on stacked fourths, not stacked thirds).
• Finally — you have to wait to hear this one — the final chord in the song is the eerie and suspenseful AΔ#5, which can also be thought of as C# / A.  It’s played here as A – G# – C# – E# — only a one note difference from AΔ13.  The perfect chord to suggest an uncertain and mysterious future!

Poison on My Tongue: chords

I am in the process of writing a series of articles about the songs on my sixth Jamendo album, In Unexpected Places.  This is about the fourth track, “Poison on My Tongue.”

Musically, the most important thing about “Poison on My Tongue” is its chord structure.  I call this type of tune a Bluesmorph.

The Basic Blues Structure

Most musicians are familiar with the form of the standard 12-bar blues:

I	I	I	I
IV	IV	I	I
V	V	I	I

The magic of this progression is that this can be played in an almost endless number of variations — not only the variations in the chords, but the underlying rhythms and phrasing as well.  There are traditional blues, jazz blues, minor blues, and so on.

Minor Blues

A specific example: in a minor blues, some reasonable substitutions are as follows.  Here the tonic chord is Em7.

• I: 4 bars of Em7
• IV: 2 bars of Am7
• I: 2 bars of Em7
• V: The V chord is played as (V of V) – V, which would be F#7 – B7, except for the tritone subsitution, which makes it C7 – B7.
• I: 2 bars of Em7

The lead sheet for this progression would look like this, overall:

Em7	Em7	Em7	Em7
Am7	Am7	Em7	Em7
C7	B7	Em7	Em7

Certainly this is a variation I like.  However, variety is the spice of life, so they say, and sometimes shaking things up a bit can work just as well.

Poison on My Tongue

A number of my songs have chord structures that are mappable to the standard blues progression.  I call these songs Bluesmorphs.  The chords can be altered, and the number of chords can expand or contract as well.  For “Poison on My Tongue”, the mapping is

• I: 4 bars of Em7
• “IV”: 4 bars of Dm7 (the IV of IV, if you want to call it that).
• I: 2 bars of Em7
• V: The V chord, B7, played as B13b9.  This lets me take advantage of minor third substitution when I base the chords on the diminished scale (here, B – C – D – D# – F – F# – G# – A) — I can choose from B13b9, D13b9, F13b9, and Ab13b9.  In this variation I start with 4 bars of B13b9, move up to four bars of D13b9, and return for 4 bars of B13b9.  It’s almost a song within a song.
• I: 2 bars of Em7.

All together, it looks like this:  (There’s also a short 16 bar bridge, but this is the predominant chord structure.)

Em7	Em7	Em7	Em7
Dm7	Dm7	Dm7	Dm7
Em7	Em7
B13b9	B13b9	B13b9	B13b9
D13b9	D13b9	D13b9	D13b9
B13b9	B13b9	B13b9	B13b9
Em7	Em7

In this case, the I, “IV”, and V parts are still present — just altered and expanded.  Effectively it’s a “24 bar blues” — but then, there are so many variations out there, any number of them could deserve that title just as well.

This song did not start off in its present form.  I tried an additional variation first, in my original prototype, P22 (hosted on SoundClick; you can read more about the “Rhythm Method” prototypes here).  Can you spot the differences?