Producers (and more generally artists of any kind) often keep quiet about how they arrived at a final product, maybe because it removes a certain mystique to their work, or maybe they feel that their unique sound will be threatened when their tools and methods are divulged. I suppose exposing some of your production process does both of those things, but as a fan of knowing how things are created myself, I present a sort-of “commentary track” for Across the Threshold, my latest freebie music album that weaves Japanese sounds with electronic music. During the course of producing this album, I went ahead and bought Ableton, Ozone, NI Komplete and a few other VSTs, along with gathering a collection of freeware, so I can actually talk about the majority of what I’m doing without incriminating myself on warez.
First off, why Japan? I’ve always been attracted to Japanese aesthetics, whether new (moé & iconic abstraction) or old (transience, deepness). My music has always reached for an otherworldly (“deep” I guess) theme filled with soft, frail sounds (transience: wabi-sabi)– contrasted with “violent” drum programming and messy breakdowns (transience: mono no awaré) . At the same time I always liked to personify my music with melancholic, singing synth-leads & melodies and my tendancy to name my tracks after things like characters. I think of a lot of my tracks as theme songs for places and characters. This abstract characterization ties into aesthetics like moé, so as far as theme goes, Japanese aesthetics have always been in my music overtly or subconsciously. And as far as sound goes, I’ve seen several taiko performances (from local groups) in the past. Hijacking Japanese sounds for use in electronic music was something I’ve been wanting to do for a long time. Obviously I’m a big anime/game nut. For this album I’ve been inspired by everything from Ookami to The Tale of Genji 1984 anime film. Haruomi Hosono’s Genji soundtrack in particular is a very haunting mix of traditional Japanese music, synths and digital effects… the soundtrack is what eventually drove me to roll my own for this album. Some songs have a particularly “BGM” feel to them.
I’ve always been attracted to production art that depicts an otherworldly plane seeping into the mundane world, like Kamichu, Mushishi, Natsume Yuujinchou and so on. That, as a vague idea, is a driving theme in Across the Threshold. Every time you put your headphones on, your mundane world intersects the virtual world inside your music player. The tracks in Across illustrate the hidden worlds of the small (Zashiki-warashi), the profound (Daifuku), the virtual (Miku Acid) and the spiritual (Across the Threshold) at the intersection of the past and future (Nakoruru). The album begins with earthly, “live”-like sounds and generally moves towards abstract synth-heavy sounds as it progresses, “crossing the threshold” to an unearthly plane.
A lot of what I do is driven by sound rather than melody.
Melody and harmony (music theory)
has always been my weak point. You’ll hear me wanking the
same
minor keys and
chords over and over across all my albums, but at the same time I’m
shuffling
sounds and styles. Most
of my inspiration comes
from playing with toys–
Miku Acid and
Zashiki-warashi
are among the more obvious of ”check out these cool new
synth-toys!”, as opposed to more aesthetically deliberate
songs like Omamori
and Kuda-gitsune. But even in my more “earthly” tracks I’m
still
driven by toys and sounds. That kind of production style
prevents
me
from having a certain sound or genre. That’s not necessarily
a
good or bad thing, it’s just how I work.
What separates this album from Reiha and Kei is a greater emphasis on
“real world” sounds as opposed to abstract, synth-heavy
sounds.
The idea was to have tracks that provoke certain imagery, scenes and
ideas rather than pure IDM-esque
emotional/aesthetic abstraction. Some songs like Kuda-gitsune
have a
downright soundtrack feel to them, so not only am I wanking “ethnic”
sounds, I’ve been sampling CDs of traditional
Japanese music, grabbing field recordings from The
Freesound Project and elsewhere, sampling anime, as
well as recording my own sounds with a collection of small instruments
I’ve gathered.
Similarly I’ve been trying to edge away from huge build-ups of analog
synths and reverbs for a more focused, stripped-down, tactile
sound.
Occasionally you’ll hear me constructing reverb & pad build-ups
though.
As I finished Reiha EP, I was nearing the end of my interest in the
Confield-era Autechre style– Reiha has shorter, less tedious
tracks than Kei and Kai,
and Across continues that trend by changing course more rapidly and
keeping
some kind of progression going. There are still some “last
gen”-style tracks like Daifuku and Across the Threshold (i.e., build-up
build-up build-up…), but those are in fact last-gen
tracks made
during or just after the completion of Reiha.
I’ve been escalating my use of two-part
song structures (heard previously in songs
like Leviathan and Reiha)
since that structure is an easy way to take a
song in a different direction without having to think too hard about
integration. They’re not unlike a DJ session with only two
songs
and one crossfade. Usually after one transition I have to
clear
the whole set of
instruments to come up with something new. I only made one
chorus-verse-bridge song for the album (Nakoruru).
Most of the other tracks I can think of are either linear oldschool
build-ups,
linear part1+part2 etc. split tracks, or messy adventures (e.g. Pictures).
I’ve also been trying to provide more audible cues in order to create
anticipation for a song’s progression (i.e. the peak of a buildup or
the bottom of a breakdown). Where many of my previous albums
had
songs that were soupy and floating in space in terms of structure, I
try to be more deliberate in dropping elements, killing certain
frequency ranges (and other filtering gimmicks taken out of the DJ
world) and so on, so hopefully Across is easier to follow than my previous albums progression-wise.
Inn
“Tada… where do you think we are?”
There’s not much to say about this one. It’s a mishmash of
sounds and patches I had, thunder crashes, along
with me banging on kitchen bowls, a CPU heat sink and a large ride
cymbal. In making
this album, I scoured P2P networks for whatever traditional Japanese
music I
could find, so you’ll find some gagaku clips in this track. Obviously I was going for a “haunted
house” kind
of soundscape here… though not so much haunted as
otherworldly, saturated with gods and spirits.
I used a certain $500 highly realistic IR reverb effect for
this
one which I will probably never buy, so I try not to use it often.
It’s only useful anyway if you’re aiming to depict believable
spaces… realistic
reverbs tend not to have the same applications as abstract, “musical”
algorithmic
reverbs.
Pictures
of the Floating World
Pictures was a very difficult
track to finish. It went through many drafts
and revisions. There used to be two versions of an additional
bridge and second half that I ultimately tossed out. I
struggled
with
whether this track should remain on the album. The song is a
huge
mess with many disparate parts that have nothing to do with eachother
and a very loose, noisy sound. At the same time, it’s an
explosive intro to the faux-Japanese sound of the album so in terms of
album progression, it serves a purpose. I doubt anyone will
be
replaying the track 5 times in a row, but it makes sense in terms of
the
“journey”.
My new and improved taiko drumset makes its debut here. It’s a big sample set I
meticulously constructed from taiko CDs over the years (an early
version was previously used in Kuruma K).
I was especially happy to find
some taiko group-rimshots and stick sounds since those are practically
the taiko version
of a clapsnare. You’ll hear it in like every
track I use the drumset.
Nowhere in my library do I have a
reasonable shamisen patch, so in tracks 2-4
you’ll hear me improvising
with other asian-esque plucked string sounds. This one in
particular has a lead
instrument assembled from four overdubbed instruments– a String Studio
(Ableton Tension) patch, a Kong Audio GuZheng patch, a live picked
violin,
and a live clean electric guitar.
Also on top is another “bouncing hammer” zithery sound from String
Studio, and a buildup of electric guitars I created when a friend left
his bass and electric at my place. Not knowing how to play
them
didn’t stop me from exploiting them in this track, heh. In
fact
(if this wasn’t obvious) I’m not great at any instrument, so I’ve had
to stretch my ability and willingness to whip up faux live-sounding
performances through meticulous sound editing, midi-editing and
overdubbing. Kuda-gitsune takes the cake though:
Kuda-gitsune
Kuda-gitsune is the last track made for the album, an xxxHolic-derived
name (the short flick and the 2nd TV series are pretty alright btw).
I actually started the song as a variation of the ditty that
closes xxxHolic Kei 13.
When I was developing the tracklist, I had a
problem with how Omamori immediately followed Pictures– two loud
Japsplosions back-to-back. I needed a soft break in between,
so
Kuda-gitsune was created to fit the bill. Aside from the
analog
synth buildup in the middle, the idea was to have a sparse-sounding
track with some electronic bits thrown in.
To aid
in that sparseness, I used Guitar Rig (which is really Distortion, Cabinet
Modelling and Analog Effects-rig
since it’s applicable to things far
beyond electric guitars) to reduce the fidelity of certain patches that
sounded way too rich, like the wind instrument and the backup GuZheng
zither that appears toward the end of the track. In effect
it’s
sort-of lo-fi for the purpose of “realism”, otherwise things start
sounding a little too digital and samply. I also performed
some
simpler fidelity reduction with automated EQs for the lead asian banjo.
The
techno synthline that opens the track is actually an instrument that
came with
Reaktor. When I found it I was like oh shit it’s Arctic
Hospital in a patch! When he said he used mostly
Reaktor to make Citystream he wasn’t kidding.
Thrown into the mix of my usual taiko drumset are some of the taiko
drum samples
included with NI Battery, so the sound of the percussion is a little
different, a little more dry and meatier.
Omamori
Speaking of digital and
samply, Omamori is another track created later as I noticed a
deficit of faux-Japan tracks in the material I had so far. It
opens with a hand drum I bought at a garage sale. It’s so
simple,
you have to tune it by wetting the surface or using a blow
dryer. It’s funny how with enough proximity effect,
comperssion and reverb
you can make it sound like a big
drum (well, maybe not that big). Omamori uses a
bunch of
strum samples so it sounds a little like a rehash of Pictures,
but whatever. It also uses a bunch of arpeggio/glissando
samples
so this track is corny to the max, which makes it kind of
awesome. It has a pretty nice bridge in the middle of the
first segment, using a bunch
of random recordings pulled from anime etc., and many more
clips throughout the rest of the piece (At one point, Mushishi’s Renzu is handed a soda… Listen for it!). Otherwise
there’s not much special to say
about the production– it’s mostly what-you-hear-is-what-it-is. I run a lot of ambient sounds (pads, voices, etc.) under a rhythmic ducking effect. Ableton Live makes it real easy with its compressor’s “sidechain” feature, where you can tell it to “listen” to any track in the project.
I go a bit overboard with the second half but I couldn’t help
myself really. That long string-heavy melancholic bridge used
to
stretch on for the remainder of the song, so I opted to break up the
doldrums with a relatively upbeat part with 808-esque drums and an arpeg Synth1. That
new part lacks finesse, but it was such a late
addition I was like fuck it, Omamori is fixed and it’s time to get this
album out the door already.
Amasawa
Amasawa is among the earlier tracks I’ve made for Across the Threshold,
and marks the beginning of the more electronified majority of the
album. I made a neat idm/techno pattern on my Korg ESX. I
thought
I could develop it into a dance track, but after a while of
struggling I was just like f-ck it. I was watching Dennou Coil at
the time (fun fact: I wrote 95% of that wiki article), so in the spirit
of the series’ music, I filled the song with
arpeggiated glocks, string pads, pizzicatos and an “ahh” voicepad, and
hey… now it’s one of the more interesting tracks in the
album. There was a lot of fine-tuning in the mastering stage
in
terms of multiband compressors and well, everything else, since there’s
a whole lot of high freq.
energy in the track. Tracks where the treble leads the mix
tend to be
very touchy in that dept. so I hope I struck the right balance of
brightness here, though it’ll probably kill the ears of those listening
on “bright” setups. Mind that smiley curve on your EQ.
I also playtested it on a darker stereo setup, and the song
sounded a
bit muffled and dull. I guess the unpredictability of high
trebles is part of why that freq. range is avoided (in favor of ~5khz)
in radio-friendly music.
It’s also a difficult track
since it’s deliberately overbearing. It’s hard to tell where
the fun ends and irritation begins… I’ve
tried add a few momentary breaks from the brightness and intensity here
and there, like the whole watery breakdown/bridge in the middle.
I don’t know about you, but I certainly like the track.
I find it very vivid so it’s among my favorites in the album.
I
also like the “walking through the forest” scene that
marks the end of its hazy bridge segment. I guess I’ll break the
illusion and say that breathing heavily into a mic and pitching it up
creates the effect of breaths from a little girl.
Zashiki-warashi
Zashiki-warashi is a great example of a toy-oriented production for me.
The title is also another xxxHolic reference. I was cruising
freeware and I came across the (now highly recommended by me) Drumatic
duo of 808-esque drum synths, and Xoxos Surface, which is a cute
physically
modelled percussion toy. Beyond that there’s not much else to
say
about the track. It’s so thin and simple you can pick it
apart by
listening to it. I’m glad to have bought a drum pad
controller
since it sure came in handy here– I’m extra glad it was a Korg
PadKontrol since it’s so absurdly touch sensitive, it might
be the next best thing to a Zendrum. You’ll hear me using a
lot of
Ableton Live’s “grain delay”, which I always like to use as an
alternative reverb. Sometimes I can use it as a reverb sans
the
“distant” and blurry quality that reverbs often have, but in this case
I used it
as a very echoey, chaotic delay. This was also my first
experiment with the Dominion plugin, which is a funky
envelope-driven
dynamics processor. It basically allows you to adjust the
“attack” (the snappiness) and the “sustain” of transient sounds without
having to deal with thresholds, ratios, and dynamic range. I
used
it in this track to reduce the natural snappiness of Drummatic, which
allowed
me to turn up the treble in the mastering stage for a clearer, more
balanced sound. I’ll probably find myself using it a lot in
the
future as a problem-solving tool.
Daifuku
Daifuku is a track I started working
on immediately after Reiha, so it carries that old pad-heavy,
beat-heavy sound and relentless build-up song structure. Here
I
had a lot of fun using Guitar Rig’s distortion amps and effects for the
drum parts. It gives the otherwise samply percussion a dirty
feel.
The odd title is even more oddly
named after Rozen Maiden (as Hina-ichigo’s favorite dessert… it’s just a nonsense working title that stuck), but the
track itself is a big homage to anime
composer Kenji Kawai, in particular his Patlabor: The Movie 2 soundtrack. I found this chord patch that sounds just like
the
one he used in that flick, so I went further and added more of his
trademark sounds like sustained pianos, taiko drums and a female ahhh
voice (which was actually a recorded performance by a friend of
mine, chopped up and reconstituted into layers of overdubs).
Kenji Kawai’s older soundtracks have been an influence in this album,
like his disembodying intersection of synths and faux-archaic sounds in Ghost in the Shell and the Japan-inspired synth-laden melancholic sound of Vampire Princess Miyu (TV) among several others, so I always wanted to do a tribute.
Anyway, songs like these always require a lot of balancing.
There _are_
some lead instruments in this song, but I tried to make everything
blend together like a big soup, as opposed to a more straightforward
lead/backup configuration. I put a lot of work into effects,
dynamics, EQ and so on to make sure nothing pops out and ruins the hazy
dream of the song. At the same time, all the elements have to
add
up to a product worth listening to, so the sheer difficulty of this is
part
of why I’m trying to move away from this soupy aesthetic with my newer
tracks.
Nakoruru
Don’t you love this track?
Here’s a song with a sparse, digital, samply sound. There’s
so
much space in between the drums, you can just taste those long reverb
tails. I mixed in some electronic sounds and a lot of brush
snare
samples with my taiko set. I have no idea how it added up to
sound-texture
magic but it did.
The song is based off a
track I
heard when I was playing Samurai
Shodown 6 (hence the track title), which mixed a
jazz-hiphop
sound with taiko drums and a Japanese choir. I was
immediately
like fuck, I should’ve made this song. So I started on a version, but
with my attention span I knew I’d come up with something a bit
different once I got the basic elements in there.
I hand-made a typical e-piano chord pad
for Nakoruru,
but I tried to find a median between rigid digital samplyness and
organic performance. The patch contains several samples of
the
same chord being “strummed” with timing variations. The
sampler
selects between the variations at random. I still wanted to
keep some of
the cheap hiphop feel of digital resampling, so I turned off the ADSR
release (i.e. the sound decays immediately).
There’s a lot of contrast in Nakoruru.
You have the quantized,
mechanical drumming, mixed with loose sounds like strings, a flabby analog
bassline and a music box. I like the breakdown since it feels
like a literal breakdown of the machinery driving the song.
Then
it snaps you awake with a “ding!” and proceeds with a very pretty
build-up of strings and Ainu (maybe?) vocal clips.
It also has a strong thematic contrast– the world of the old
intersecting the new. With all its old sounds and voice
clips,
it’s arranged in a pop verse/chorus/bridge structure and stylistically
arranged like a hip hop track, with its quantized drum patterns, samply
chord pad, heavy bass line and vocal clips that are, well, not
unlike verses of rap music. It has a bit of a ghostly edge in
how
it conjures the past into a modern structure.
I would’ve liked to use a real violin in the track, but seeing that I
can’t play violin worth shit I
settled with samplers.
Miku Acid
I had
to do it, this track. I used a certain
leekspinning voice
synth that has achieved idol status over the past couple
years. By now
she’s a passing fad, but I’d like to think that I made an
interesting track rather than a disposable voicesynth rendition of
[fill in the
blank].
This song revisits a robogirl theme I’ve
used in previous tracks like Androsynth and Kuruma K’s
second half. Here,
in what is very likely horrifically mangled Japanese, Miku
talks about the emergence of her soul from her “metal body” and
“electronic brian” and that she can “hear the sea of information”
(original indeed).
I go to great lengths to
emphasize the artificiality of Miku’s singing by exploiting some of the
weirder parameters of the synth, along with occasional Auto-tune
effects
(provided by Gsnap). The glitch effects
are provided by the
tremendously awesome, freeware DBlue
Glitch effects sequencer, used
heavily in the second half. The melody itself was created by
pressing buttons in ChordSpace, so like Zashiki-warashi,
this
was another toy-driven track.
I also tried to
keep the song as minimalistic as I could. Much of the song
was
Miku, a limited drum set, a few instances of Synth1 and effects.
Stage
Like Kuda-gitsune, Stage is
another track created when I was
solidifying the tracklist. It’s just something to lead up to
the
next track, Across the Threshold… Without it, the
transition
between Miku Acid and Across is jarring. There’s not
much going on here… some pads, some downloaded sound effects, some
recorded stuff (shuffling paper, strumming on a totally detuned Chapman
stick a friend left at my place). Much of it is reconstituted
parts from other song sketches that didn’t make it into Across.
There’s a lot of low bass and some deliberate
clipping distortion. The track is not that interesting but it
serves the purpose of setting the stage for the next track.
Across the Threshold
For being the title track, it’s ironic that it’s a recycled song from
the Reiha era. It has that good old build-up structure and a
synth-heavy aesthetic with reverb-heavy drums. What prompted
me
to revive this song sketch from the dead was that, despite its Synth1-heavy
dubby instrumentation, it (in my mind) conjures imagery of a serene
snowy scene… at least in the beginning. Eventually the
build-up
reaches a plateau
of spookiness and breaks down, and goes for a spookier finale build-up.
The structure
prompted me to name the track “Across
the Threshold“,
so there’s a bit
of a storytelling element in this track and where it sits in the album.
The song (in my mind) portrays a fissure opening up in the
fabric
of mundane reality to another world.
Like Daifuku,
this is another soupy reverb-laden mix. To make
things a little more organic, I added some recorded china cymbal
percussion. You can tell I was never quite
pleased with the song by the amount of frantic filtering and
effect-automation that’s in it.
Memories of Earth
Memories of Earth (a reference
to an early BLAME! chapter) was made in the middle of the album’s
production, but was always meant to be an ending track. It’s
not that dissimilar in tone to the enders I’ve had for Kei and Kai and Synthi,
this time with FM8 providing the obligatory long-decay synth
sound. There’s
not much to it… lots of Freesound Project wankery and delay
effects. I actually don’t know what else to say about it,
heh. This track also fits in the storytelling of course, as
the
listener has “crossed the threshold” to some otherworldy plane and
nostalgically remembers Earth, so I tried to aim for nostalgia as an
emotional goal.
Final thoughts and lessons learned…
It seems like when people first discover compression, they go
squash-crazy. I’ve never had much of a tendancy to do this,
but
(in rare exceptions like Daifuku)
I mostly laid off the heavy
compression in Across the Threshold, particularly since I was mostly
aiming for an open, organic sound. Similarly, I used my
mastering chain as more of a final problem-solving stage rather than a
“make pretty” stage. I’ve come to look at a lot of
my
practical
tools (comp, eq, chorus/reverb) as
problem-solvers rather than gimmicks to
turn crap into gold. It seems like an obvious thing to say,
but I
suppose it’s a lesson learned only through experience and
struggle– it’s better to start with good core sounds rather than try
to dress up crappy sounds with effects… also, it’s better to achieve
a good sound in the mix rather than reserve “beautification” to the
mastering stage. With that evolved outlook, I’ve had much
less
trouble with Across’ mixing & mastering than with previous
albums.
I had a
better sense of what I needed to do, both in mixing and composition.
Partially out of necessity (especially for this album), I’ve been
moving away from my old outlook of meticulously hand-making all of my
percussion sets and synth patches in a vain effort to “personalize”
every aspect of the track. In Kei and Kai for
instance, I took some pride in the fact that all the synth patches
were not presets but hand-made from scratch, if I recall correctly.
It also featured several drum sets made out of original
recordings of me tapping and hitting things around the house.
While Across has a lot
of that on the percussion side, much of it was to build a certain sound
that I couldn’t grab offhand in some library. Obviously many
of
the pitched instruments are sampler patches from ethnic libraries and
whatnot, but even some of the synth sounds were pulled out of presets
too. In comparison to my more abstract electronic albums, Across presented me
with a vivid goal to reach for (of a faux-Japan mythological fantasy
world). I didn’t want to be mucking
around in synth patching so much when it had little to do with the big
picture. Instead I reached for many toys and sounds and
tweaked
them to what I needed to convey the scene and the journey.
For mixing and mastering, I snapped and dropped 1.5 Gs on a new
monitoring setup (Event ASP8 + Martin Logan Dynamo), and rearranged my
entire room to get the best possible sound. I wrote a blog
article about it nearly a year ago called Improve Your Monitoring Setup
for Free. I wanted Across to sound substantially better than anything I’ve made in the past.
In terms of a mastering “target”, Across the Threshold continues my
trend in assuming the best from people’s stereos, though I imagine most
people will hear it on desktop speakers and ipod earbuds. The
joke’s on me when I’m mastering it on my full-range monitoring solution
and playtesting it on my high-end Grado headphones and Martin Logan
electrostat home theater setup. That’s
fine
though… ultimately the “target audience” for
my music is primarily myself, so I don’t like to squash the sh-t out of
my sound, give my music an excessively warm curve, duplicate the low
basses to the
upper bass and over-simplify the mix. What’s the point of
compromise if I’m not making a radio-friendly pop album? I
designed Across for people who love sound, because
I love sound too.
A mixing trend I’ve been moving away from is the overuse of reverb,
because of its tendancy to make a mix sound distant. In its place
I’ve been using lots of delay lines if I need to fill space.
Ableton also has a really nice reverb alternative in the form of
its “Grain Delay” plugin… give it a really large grain size (~1 sec)
and a large “spread” parameter and it gives you a really chaotic delay,
a reverb without the distance and a lot more texture. The “random
pitch” function gives it more of a chorus character. I
practically abuse Grain Delay every day now.
I’ve found a few
nice pieces of freeware… drumatic, dblue glitch and dominion have
already been mentioned, but blockfish has proven to be an interesting compressor, MDA
Stereo invaluable in channels (it’s a versatile “make
awesome” effect), Ambience and Synth1 have always been
in my arsenal, HighLife is kinda cool for
freeware, and Destroy FX Geometer & Buffer
Override are nice
effects. Lazysnake makes a nice electric
piano too.
On the commercial
side of things, I’ve previously used FM8, Massive, Guitar Rig and
Akoustik Piano (all very useful), but having just bought Komplete I’ve
only just begun to explore some of the other NI plugins.
I’ve bought a couple of rather expensive Kong Audio plugins– a GuZheng
zither (similar to a Japanese Koto) and a Nanxiao (bamboo flute) and I
can’t exactly recommend either. Some of their samples have
odd
recording issues (background noise, inconsistant start times etc.) and
the synthedit-based sampling engine & patches are a bit
questionable. The patches aren’t multisampled across the
velocity
range… how lame is that? I instead ripped a bunch of
GuZheng
patches with HighLife and made my own multi velocity zone patch
with it. I ultimately did find both the Kong Audio plugins
useful, but they suffer from some rather poor design choices.
They should give you like 2 awesome, highly versatile patches
rather
than a package of several simplistic ones for each instrument.
String Studio (which I later bought as Ableton Tension) is a neat
plugin, but for physical modelling it’s a bit limited– not so much in
its synthesis, but in its “control”. The excitation models
are
all a bit robotic and there’s not much you can do to make them more
characteristic and human. The developers need to discover
real
instruments and map the many real-life dimensions of expression in
“triggering”
strings. Everything that comes out of String Studio in its
current state sounds like a weird music box or some kind of abstract
“ethnic”
player piano, even after mapping the mod wheel etc. to some of its
parameters. It’s like the physical “motion” of the excitation
engine is too rigid.
Graphics
I’m less able to talk about specifically what software I used here
(haha) but on the
real-world side of things, that’s a friend of mine dressed up in one
of my mom’s for-real family-made kimonos. It took us
a
couple hours of reviewing obi-tying
tutorials and youtube vids to figure out how it goes on, and even then
the back part is kinda half-assed cuz we were in a rush.
After
that we did a photoshoot with her shiny new DSLR. Also, those walkways
of infinity shrine gates actually exist IRL.
The creature she meets in there is a mythological
nine-tailed fox (although in the 3D model I probably made only 5 or 6, heh).
Graphics wise, lighting & rendering are always a big focus for
me
and I always like to pay attention to those elements. Frankly
I’m
not very good with textures and modelling but I’ve always found
lighting & rendering to be much more of a playground. In
terms of
composition, the graphics reflect my music in that they’re very soupy
and scenic rather than having big focal points. The images
with
definite subjects are always subdued into the scene, since I’m much
more interesting in conveying the scene and atmosphere rather than
propping up a polished “lead” element. Did I mention I’m an
INTP?
I gradually became aware of my monitor’s fucked up gamma and ended up
buying a color calibrator (a Spyder 3). My suspicion of
producing
overly dark images was true, as the calibrator toned down (or “up” in
terms of value) the gamma on my monitor, so now I can now trust my
computer monitor like I can my monitor speakers. Yay.
The torii/kyuubi image marked my first (and hopefully last) foray into
photomorphing. I amassed a collection of fox heads, paws and
torsos via google image, and I went to great pains to integrate it with
the rendered body
and the scene lighting. With every new layer of touch ups, I
felt
like I was meticulously reverse-engineering what I take for granted in
a 3D renderer– every aspect of lighting, shading, specular reflection
with fresnel curves, direct & global illumination and so on.
Each graphic went through several
drafts but Torii takes the cake.
As I completed
the images, a lingering dissatisfaction grew over the overall
“immersiveness” of the images. Something was missing, and no
amount of fiddling with individual elements, effects, saturation levels
and so on could do it. One day I was looking at the tourou
nagashi graphic I was working on and said to myself this area
needs to be more green… fuck it. I created a new layer on top of my entire
supermassive composite and started airbrushing some green.
And
then I airbrushed some more colors and magic happened. The
problem was I was looking at my composites like they were music mixes.
It
never occured to me to just start painting some colors on top of the
whole damn thing. Unlike audio mixes, graphics can be easily
adjusted with broad strokes, because audio is _only_ additive, where
graphics can blend and occlude. I can’t easily make a portion
of
my music “green” by adding a new green track to the mix– instead I have to
deconstruct the mix and meticulously adjust every part in order to
build a new “green” sound.
I love that tacky photo mashup that makes the album’s back
cover. I laughed the whole time making it, but partially
because
its tackiness reflects some of the album’s overblown faux-Japan
tracks like Pictures and Omamori. It’s also nice to do a
photo/graphics mashup that isn’t meticulously realistic like the other
images. The more discernable elements are
scanned/photographed
items from around the house– dolls, omamoris, and that kimono of
course.
At the last minute, I decided to whack
together a flyby animation using recycled assets from the album cover
and some older projects. The fact that it took, between my
two modern desktop computers (6 cores total), like 48 hours to
render at 720×400 with 1 ray bounce&refraction, low AA and
every scene/rendering/lighting/shader optimization I could think of,
drives home the fact of life that all information technology begins to
feel inadequite just a couple years later. Many of the more
complicated songs in Across were rendered at 44 KHz instead of 88. I thought I could
standardize 88K this time around, but maybe next PC. Some
plugins like Drumatic sound different (better) running at higher
sampling rates.
Anyway you’ll notice lots of rendering artifacts and sloppiness
in the video if you know what to look for, as it was whipped up in a
mad frenzy and published a couple days after the release of the album. I mean damn, it even says “2008″ at the end. Whoops.
Getting it Published
This was a bit of an ordeal but in some
ways it was self-inflicted. The first label I sent it to (an
electronic label) rejected it because it was too chinky too far outside of their sound. That’s fine though… labels
have to control their output because that’s the whole point,
to cull and contextualize the vast (and mostly garbage) output of all the
producers in the world. Sometimes I wish netlabels were even
more discerning quality-wise. Anyway, to some degree Bump
Foot probably made a bit of an exception for this one. I’d
like to say that oh, my album is too unique for genres or whatever, but
in reality I made an album with a heavily ethnic sound, yet
I still wanted to put it out on an electronic label and
not a label that specializes in acoustic or “world” stuff.
For one thing, I don’t know about, don’t listen to and don’t care about
hybrid “world” music, and for another thing I don’t consider Across to be vastly different from my previous all-electronic affairs.
I produced this one just like I did the others, except with
lots of ethnic samples instead of electronic beats and analog synths.
What exacerbates the problem is only the first 12 minutes is
heavily ethnic, and then you don’t hear much of it until track 8, Nakoruru.
It’s not a perfect blend of electronic sounds and ethnic
samples. The two worlds of the electronic and the ethnic are very
distinct, so the album has two sounds. I wonder if I sent it to a
“world”-type label, they’d reject it for having too many heavily
electronic tracks. Regardless, I wanted Across to
be an album for the electronic world, because that’s where I come from
when approaching the pseudo-asian sound.
In my next thing for “Renzu”, I’d like to come up with something more integrated, where the ethnic and electronic cannot be easily separated and categorized. I also want to continue my trends in vivid imagery and easier-to-follow progression. I often say the next thing I’m going to make is a dance album, but well, who knows, cuz I don’t.