production blog:

( )	jun.12 2008: the aesthetics of mediums and mamoru oshii

( )	jun.06 2008: upcoming anime 1

( )	may.29 2008: the dense language of anime

( )	may.27 2008: anime hightlights 1

From this point up I'm participating in a group blog with friends with a focus on anime. My articles will be linked here.

(+)

may.23 2008:
the old computer music scene and tracker2wav.exe

    The story behind this EP is explained here, so I'll jump straight into the analysis, starting with an overview of the 1990s PC music scene. Back in the days of diskswapping and BBSes, computer music was distributed in a format called module music. A "MOD" file is a self-contained song that houses both note data and sample data. In some sense a mod is like a MIDI file packed with a set of WAV instruments. This allowed MODs to sound the same on every playback system (unlike raw MIDI data), contain original samples and recordings, and still remain small enough for narrowband distribution. Usually these songs were anywhere from a few kilobytes in size to a couple megs, and took much less power to play back than a CD-quality MP3, a crucial difference in the days before the 100MHz+ Pentium.
    Tracker music was also sequenced in a completely different way. In specifics, module music was pattern-sequenced much like the roland TR/TB series. The 1st gen MOD format supported four monophonic "channels" to make a total polyphony of 4. A typical arrangement in a 4-channel module would look something like this:
    1. kick and snare
    2. hats & cymbals
    3. lead instruments
    4. backing instruments, chord pad samples, etc.
    You're probably thinking "4-channel monophonic MIDI? That sucks!", but unlike MIDI, instruments and samples can be swapped in and out of channels on the fly. Four simultaneous sounds was usually enough to make a suitably dense electronic track, and it added an extra layer of creative challenge that made the early computer music scene more exciting. People would get real creative in jamming tons of instruments into every channel. As computers became exponentially faster, the standards were updated, to 8, 16, and eventually unlimited channels. Limitations on sample quality (bits, sampling rate) were updated as well, along with support for more sophisticated panning and volume envelopes and semi-polyphonic channels.
    The necessary limitations of the format were not just for nerds to appreciate when looking at the insides of module files, it encouraged a certain style of music production. For starters, modules lacked effects, peroid. No delays, no reverbs, no choruses, no nothing. But again, unlike MIDI, modules allowed for very precise playback of samples in terms of panning, pitch, offset and so on. This encouraged creative reverse-engineering of sophisticated delay effects, flangers (as heard in Isotoxin), along with mod-specific effects like highly precise sample-cutting and dense panning effects and envelope trickery that is difficult to recreate in a modern MIDI-based DAW. I don't profess to be a ninja master at module-tracking, but the first four tracks of Synthi were composed in a module "tracker" (with post-effects done in old Cool Edit). Cyborg and Graph In in particular exhibit some of those traits.
    Anyway, not only is it fun to see the inner workings of the music you download, it also grants you the ability to tear apart the songs into individual tracks for a proper mix in a modern DAW. I've done this a lot in the past actually, but this is the first time I'm releasing a set for serious.
    Necros: Isotoxin
    Yeah, it's the song from Unreal 1, but it was previously released in the "musicdisk" (old lingo for an album of MOD songs) called Progression. Given the age of it, the level of production is incredible. Those soft pads, those breaks, those synth lines... juicy days. I didn't have to do anything drastic to the track because it sounded great from the start. I had no luck trying to "enhance" the intensity of the song, so instead worked on more of a balanced sound that emphasizes the "richness" and prettiness so to speak.
    Xhale: Attending Earth
    From a musician's standpoint, this is the kind of old, experimental track you want to forget you ever made. It has many disconnected parts and many nonsense sounds, but it exhibits Xhale's strong sense of aesthetic and presents a nice, dark vibe. It all comes together in a sort of a funky way, so much of my task was to increase its cohesion. Attending Earth's mix was very loose, so it was a challenge to both enhance (i.e., "bring forward") some aspects, while taming some of the other zanier qualities of the song. For instance, the song opens with an intense, bright effect over the drum pattern. That left me with a decision between "respecting" the original mix and leaving the sonic assault the way it was, or to tame it for a more listenable sound. I opted for the former in that specific case, cuz the goal of the song is to go a little nuts I think, and I didn't want to like, cramp its style.
    Kschzt: gum.flex.tide
    Here's a song I had a lot of fun mixing, and it must be among the more drastic overhauls in this EP. I recognized its "latent quality" of an intensely bright and bassy song, oozing with high-tech style and cyberpunk visuals. With a clear objective, I went about my work: fattening up the pads, balancing and intensifying the drums and so on. I brought the pop-snare to the front and pushed that somewhat lame punk snare to the back, even mixing in a little 808 clapsnare for the last leg of the song to supplant it. I tried to keep all the intense treble sounds in check with eachother, and at some parts I had to tame the wild panning and stereo effects in the song. On the mastering stage I brought the group brightness back up. The result is very intense, but I'm satisfied with the sound because it achieves its aesthetic objective. I still follow Kschzt's work since he always comes up with creative sounds and styles-- practically everything he's made in the past decade is worth checking out.
    Timelord: Impulse
    Somewhat similar to my own music, the objective of this track is to put the listener into a hazy trance. Most of the changes I've made throughout the many sessions of critical listening and revision were to make sure everthing meshed together and nothing jumped out. I had to be very delicate and precise with the levels and effects, because the wrong amount of something would snap me out of the fragile "dream" of the song. Impulse is also another technically amazing song given its age. Like Isotoxin, this is one of the few songs that could almost be printed on a CD in its original form. It's a rare and amazing feat for mid-90s module music to achieve a level of production quality that transcends "computer music".
    Lackluster: Space Age
    While released in 2002, I suspect the song was composed back in the late 90s. I always wanted to mess with one of Lackluster's tracks. He comes from a tradition of minimalistic module production known as the "chiptune" scene. The original song file is only 53 kilobytes, just enough for a few oscillator waveforms ("chips") and some short, blippy sounds. Compare that to Impulse's near-1MB. Anyway, Lackluster avoids the typical NES-like genre sound for a sound of his own well represented in his early albums and occasionally in his latest. The best of his music carries a very thin, sweet yet frail sound. It's kind of hard to put it but there's something rather magical about it. I'll stop myself from making some kind of mystical cheesy metaphor. Space Age is not the strongest track he's ever made in that respect, but it exhibits some of that quality. Loading it up into Ableton Live, I was at a bit of a loss. The song worked because it was thin and simple. What does it mean to "enhance" something like that? I ended up using reverbs to contextualize it in some kind of space to perhaps enhance the thin feeling of the sound, and went about EQing and balancing and whatnot, but like Lackluster's aesthetic, this was all very delicate work. The reverb effect send automation on the hihat for instance went through more revisions than I can count. I added a funky fade out and effect-buildup for the end part that didn't exist in the original. I don't know if I ultimately "enhanced" the song at all. All I know is I ended up with something a little different; more uh, spaced out.

(+)

may.19 2008:
Anime Central convention

    I just came back from the Anime Central convention and I bought a couple things cuz you know how I like to trick out my nerd cave. One of the problems with that convention in general is the standard of quality in panels is very low. Speakers come in with almost nothing to present, sending the panel into a tailspin of geeky bickering and tangent rants. That's never a fun time, believe me. Me and a bunch of friends brainstormed some panels that were distinctly lacking from the convention... we'll try to apply to host them next year:
    1. Current and Upcoming Anime. It's like the last year of Japanese animation never happened in the isolated world of ACen. A panel that highlights current and upcoming anime would be pretty informative to 'casual' and new fans. I'd be hosting it.
    2. Gay at ACen. For the record: I'm not gay, a friend of mine is. We met another gay dude and they both lamented how there's no place for gays to congregate at the convention. "Yaoi" panels and so on are for the fangirl demographic (of course). What we're talking about organizing is a Honto no Homo assembly, or as he describes it, "no fat chicks".
    3. Black Market. Bring your crazy rare, crazy expensive collectables and trade away. A friend of mine is nuts about figures, kits and gundams, but there's not a lot of rare stuff to be found among the vendors at ACen. Everyone has the same set of Final Fantasy PVC figures and a misc. assortment of erogame PVCs (do those really sell??). I couldn't even fnid a Hatsune Miku. The Black market would create a space for "oh shit!!!" items, the thing that you go to conventions to see IRL.
    4. Some kind of social/hookup thing. Last year's "speed dating" was a logistical disaster (like much of ACen) and was not repeated this year. The panels that had vague promises of socialization did not deliver (Otaku Social Hour, Ouran Host Club). Actually an Ouran Host Club-themed socializing event would be great.
    Also, if I can whip up a decent Japanese-themed live set for dancing, I can apply as a "dj" to offset some of the crappy 'hardcore' DJs at ACen's dance events. That 'hardcore' shit is not made for dancing. Fortunately they always have their very suitable house/D&B/techno regulars to fill in the latter 2/3rds of the Soap Bubble dance. ; Anyway ACen is a gigantic venue to play and it'd be perfect for my music "persona".

( )	apr.07 2008: the japanese apocalyptic vision: beyond the atomic bomb

(+)

apr.03 2008:
the state of art (or "is music dead" and "why are there no new pure genres?")

    There's a never-ending debate in IDM circles of whether the genre of IDM is dead, due to the stagnation of significant technological advancement in music production, coupled with the recent "home studio revolution"-based onrush of newbie producers, flooding The MySpaces with mediocre "my first" tracks made with a copy of Ableton Live they pirated off the internet just the other day. This is not a development of isolated significance however -- IDM was often considered the final frontier of "new" music. To call it dead says something about the state of music as a whole.
    For a bit of background, IDM ("Intelligent Dance Music" in an arguable wink of self-parody) is a blanket term & genre driven by the explosion of synthesis and sequencing technology. Much of it revolved around subverting the nature of "dance"-oriented devices such as drum machines in order to paint unusually sophisticated or otherwise whacked out beat-y music, hence the name, IDM-- "dance" music for the brain and sometimes the body. A second generation of IDM was created during the advent of softsynths (in particular personal computer-based DSP environments like Max/MSP and Reaktor), which freed artists to make even more unimaginably bizarre material. In its heydays, there was a sense of IDM being at the cutting edge of music especially when compared to the more popular but stagnant electronic genres of techno/trance and hiphop. For clarification, this article refers to IDM, the blanket term for convoluted beat-oriented electronic music and not the archetypical "genre sound" of say, classic IDM or melodic IDM.
    on genres
    IDM was driven by the explosion of electronic music technology in the 80s and 90s. While everyone else was using cutting-edge music gear for their "intended purposes" (e.g. techno, pop, hiphop), IDM was all about exploring and exploiting the new tools of music production in search of new sounds and new styles. If there was a new thing in synthesis, sequencing or control, IDM was there to "push it to the limit" as they say*. The problem is, there hasn't been a significant new thing in music production in a while. The IDM of 2008 bears no significant difference to the IDM of 2000. As a pure genre, IDM is arguably "dead", or at least stagnant.
    However IDM fared better than other pure genres that were of the last to be considered "new" like hiphop and electronic dance, which in their pure forms haven't advanced since 1992 or so. Instead what has been happening in music in the dearth of new "pure" genres is a merging of preexisting genres-- Glitch into hiphop to make glitch-hop (1, 2); Melodic IDM aesthetics and pop/rock to make whatever The Postal Service is, along with the poprock-and-something-else smorgasbord that makes the entire "post-rock" movement. Let's not forget about all the retro hybrid nostalgia-heavy pop music you hear on the radio so often, reiterated and remixed to retardation. However, aside from the standstill of significant music production technology, why are there seemingly no new styles of music left to invent? Is filling in the gaps among preexisting genres the only thing left to do?
    Its one thing to contextualize many of the new genres we've seen in the past century to the advance of recording technology, but it's another to say that everything that can be done with music, fundamentally, has been done in the past. Of course you've heard this argument before -- there are only twelve notes in the scale; how can you come up with a combination that's a radical departure from everything that's been tried in the past? Perhaps it is totally impossible, and the cynical adage of "it's all been done before" is true -- at least in terms of notes and harmony. It's likely that everything that can be done with melody and harmony has already been explored in the days of highly elaborate orchestral music.
   how music has developed
   The thing that recording technology brought to the sonic table was the ability to shape the aesthetic of sound. Previously, the recording medium of choice (sheet music) only dictaed the "note data" so to speak, and only up to a certain point. Everything else was up to the performer, his instrument, and the environment he played it in. Since the arrival of audio recording, popular music saw a gradual but dramatic shift from sophisticated melody/performance-based aesthetics to pure sound aesthetics (2), e.g., a synth line and a techno beat; Some raps and a simplistic "crunky" backing; Rock music which is often simply a set of strummed chords under a lot of fine-tuned distortion; The blanket genres of ambient and "soundscape" with its often heavy reliance on experimental/field recording and effect-tweaking; IDM with its experimental sound and machine-aided sequencing of sophisticated patterns. The frontier had shifted from harmony and melody to audio aesthetics and technology-aided composition. However once the technology of recording and synthesis stopped advancing in "revolutionary" ways, the final frontier of IDM was also stopped in its tracks. Meanwhile, the internet gave exposure to many world genres and most of what's useful in those have been integrated into the core of modern mainstream music. The "extents" of all possible music styles have been discovered and charted. All that is left is to fill in the gaps.
   when the gaps are filled, will music die?
   No, and to understand why, we have to step back and look at the ultimate objective of art & entertainment -- to strike people in some "emotional" way. "Melodies", "sounds", "aesthetics" and all these other techniques are simply a means to that end. A combination of this sound, that melody, and at this pacing will ellicit a certain response... changing one core element even slightly will radically alter the emotional outcome, due to the abstract nature of music in general (unless you're listening to a literal storyteller like Bob Dylan). If you've ever produced a track, you will understand how fragile this all is. Producers put so much effort into getting the right mix and arrangement of sounds just to make their listeners feel something. When you consider the fragility, you'll realize that the possibilities are endless. The "tools" (styles, techniques) have all been charted out in the past, but the emotions you can build with those tools have not yet been exhausted.
   people change
   Additionally, the "emotional effect" that musicians strive for is a very context-sensitive thing. What may sound ridiculously outlandish to one person may in fact be The Jam to someone else on the other side of the globe, all because he has learned to emotionally interpret music in a different way. Similarly, what used to be emotionally affecting centuries ago now makes little sense to the ears of most modern audiences. In fact, what people found emotionally engaging 20 years ago may now ellicit a different response from modern audiences (specifically: teh lolz). What this means is that as people and cultural tastes change, the old combinations of styles and aesthetics lose their effectiveness*. The search for emotionally relevant combinations of aesthetics is a never-ending quest. This is why, even in the face of "the end of pure genres"**, music will never truly be doomed to stagnation. Obviously this is of great importance to me seeing that I'm whacking together one old thing (IDM/electronic) with another (traditional Japanese music) for the next album. Then again, practically everyone is being forced to do something like that these days if they don't want to make some tired old pure genre material.

as a more personal example, the highly abstract IDM album Confield ellicits the following response from me: oh shit, X-files!! Greys! UFOs! Alien landscapes! Of course to other people it will ellicit a different impression, since these are all aspects of modern American/western culture. Also my mom probably wouldn't be able to connect with it on any level. Better yet, try to imagine if I played it to an audience 100 years ago. It would probably start one of those old-timey concert hall riots of legend.

**until IDM is revived with next-gen BCI-controlled physical modelling synthesizers of course.

(+)	apr.02 2008: improve your monitoring setup for free (and other thoughts on monitoring)
So you shelled out for some Mackie or Event etc. midrange monitors for about $1000/pair and then proceeded to place them at the corners of your computer desk. You should've gone with some lower-end KRK Rockit or M-Audio BX monitors because there are serious monitoring issues with that kind of placement. If you mix music, and especially if you master music then you should seriously reconsider your placement. If you're a home producer/hobbyist on a tight budget with low-end monitors, you can actually get more sound quality out of your monitors by placing them in a more ideal configuration. With bad placement, your frequency response gets all wobbly and your stereo field gets "blurry"; excuse the subjective terms. Trying to compensate for these inaccuracies only makes mixing and mastering more frustrating. I realize a lot of people are forced into cramped setups due to their circumstances (dorm rooms, small apartments) so this article is written under the assumption that you can afford to move your setup around a little. First off, near-field monitoring doesn't mean "2 feet away from you" monitoring -- nearfield just highlights the contrast from traditional (archaic?) far-field monitoring -- i.e., those giant main monitors installed into real studios. Wikipedia has a nice paragraph on this paradigm shift. Ideally I'd say they should be 4 to 6 feet away if you're using 8'' woofer monitors. When your monitors are too close, you run into "axis" issues since your head is going to be closer and more on-axis with either the tweeter or the woofer. If you move your head back a bit you'll hear how the sound character changes and how the woofer & tweeter meshes together better. The 60-degree triangle rule printed in many studio monitor manuals (enter some fake info to download) is a pretty good rule. I see a lot of home studios with monitors placed at extreme perpendicular angles. That is not a fun time for the stereo field. The more the monitors oppose each other, the more wave cancellation occurs resulting in a weirded-out sound. Instead you want to be able to peer deeply into your mix so you can understand precisely how to place tracks and effects. The other part of the equation is reflections and room acoustics, and this is where things get more frustrating if your room is crappy. You either need to move to a softer room, or get lots of acoustic foam, diffusers, bass traps, or even furnature to tame the one you're already in. See the video "How close..." to hear what a bad room will do to your monitoring setup. Your room really is that critical** to the quality of your monitoring setup. Speaking of acoustics, don't place your speakers too close to the back wall or corners of your room, or else your low-mid and bass response will become wobbly. If you have to do this, get a front-ported (or non-ported) monitor, and/or get a separate subwoofer. With all that in mind, you'll have to spend some time experimenting with monitor placement to find an ideal setup. Sweepgen helps to test your bass response here. Also Sound on Sound has a lot of tips on room treatment. Check it out. (I swear I made my blog's graphic before seeing that article! lol) yeah, I know you're going to tell me how so many hit songs from the 80s were mixed on Yamaha NS-10 monitors -- the worst nearfield monitors to ever be so successful -- by the sheer willpower of bandwagon-jumping engineers straining to mix around its blatant flaws. The problem is a lot of songs from the 80s in their original masters sound somewhere between shrill and dull, and this is usually attributed to over-reliance on NS-10s (it's not like pro-level recording technology was lacking back then). In my experience, you certainly can learn to mix around the flaws of your monitoring setup, but the more out of whack your monitoring setup is, the more difficult, frustrating, and time consuming** the process of mixing and mastering becomes. I guarantee you'll be better off monitoring on a balanced, clear setup; something that allows you to dive into your mix in order to make critical and subtle adjustments with relative confidence. After that, feel free to use NS-10s as a peripheral "crappy stereo" reference monitor -- although you can probably emulate a bad setup with some distortion, room reverb and EQ / comb filter plugins in your master bus. Also check this Tweakheadz article for a different, more fundamental discussion of "accuracy" in monitoring. Lastly I wouldn't use me (Renzu) as some kind of golden standard. My mixes are usually experimental and all over the place. I often don't check them on lo-fi systems, nor do I compromise for lame setups. This will change when I (tentatively) put together a minimal/ambient techno EP, cuz I will have to tune and simplify things for noisy, rumbly club-like sound systems.

(+)

dec.04 2007:
old japanese aesthetics meet new

   It's no surprise to see old Japanese aesthetics in animated productions like Mushishi, Mononoke and others that evoke that "classic" atmosphere. It shouldn't also be a terrifying shock to anyone that I occasionally indulge in mo�e-laiden harem anime like Kanon and Clannad. (The animation production quality is hot, there's my excuse.) The interesting thing about those two series in particular is that they highlight the integration of old aesthetics into new aesthetics. More specifically, the integration of (some aspects of) mono no aware, yūgen and wabi sabi into moe But wtf do I mean by that really?
    In those two series, the male protagonist forms a series of relationships with an assortment of archetypical girl characters. Just bear with me here. The twist is these girls are either ghostly/mythological apparitions, scarred by a traumatic past, or cursed by fate to an unavoidable premature death, living their last remaining twilight days on Earth in lonely anguish... or some combination. Watching our protagonist through his sheer curiosity become entangled in the tragic, hopeless fates of these girls is kinda like watching heavily comic book mo�mo�E version of Kawabata's Snow Country, a novel written in the heydays of Japanese literature where the male protagonist indulges in the "tragic beauty" of a cast of onsen geisha (prostitues).
    What makes Kanon& Clannad more interesting as a case study for the merging of old aesthetics and mo�e is that, unlike more obvious examples like Makoto Shinkai's work, they use the old aesthetics to further fetishize the characters. Moe is an aesthetic heavily based in fetishes. Each character archetype found in moe is often nothing but a combination of personality and "character design" points geared towards eliciting an emotional response (on the left side of moefig1.png in this case). However while Snow Country hints at mono no aware (sad transience) as a point of interest for the protagonist, the yūgen, wabi sabi, or mono no aware aspects become one of the many fetishized "charm points" for the girls in Kanon & Clannad.
    As a ridiculous example, one of the girls in Kanon is a mythological fox spirit from an enchanted forest who decides to cut her life short in exchange for taking human form. As the protagonist takes her in, she gradually loses her mental and physical capacity and reverts back to increasingly animalistic behavior before her impending death. What that means specifically is this cute anime chick is crawling around on all fours and mouthing ludicrously cute animal sounds and such, all the while the protagonist is hopelessly crushed by her gradual loss of humanity. It's tragic, it's "yūgen", but it's also moe. Conversely in this case, the aesthetics of "moe" are somewhat subverted to forward the tragedy of transience, so there's a deeper integration going on here.
    SORRY I'LL GET BACK TO WORK NOW. And as a general warning I wouldn't recommend either Kanon or Clannad unless you have a high tolerance for overblown moe anime. Instead you should check out Mononoke and Mushishi (where my namesake is derived).

(+)

nov.28 2007:
evaluating your own music

(+)

nov.27 2007:
thoughts on one-size-fits-all mastering

(+)

nov.26 2007:
on Akihabara Time-Lapse and the inspiration for the new LP

    This is the first time I've maintained a "blog", and it won't get read very often (until I release that new LP, play local shows and start building a fanbase) so I'll just write whatever for now and maybe some crazed fan (lol I know) will read it all later and be like oh my god, this is ~~complete shit~~ a goldmine!
Today's topic is Akihabara Time-Lapse. The mixing, composition and mastering may not be all that great, but I feel it may be one of the most "effective" tracks I've ever written in terms of conjuring a certain mood and imagery. I don�t know about you, but whenever I listen to tracks, they often prompt me to imagine vivid music videos, whether they�re typical singer-centric pop imagery or abstract Gantz Graf-like visualizations. However my own tracks very rarely conjure any kind of imagery for me except when I play the "Japan card" on tracks like Kuruma K (JPG).
Akihabara Time-Lapse is obviously one of those tracks, but like almost everything I write, I went in trying to make one thing (a funky dance track with a rimshot snare) and ended up with something else. As I added the pianos and wacky Japanese "talking clock/calendar" samples, I realized the song was actually turning into something -- a melancholic song conveying the passage of time against the backdrop of otaku culture. I produced the rest of the song with that vague mood in mind. It's very rare that I produce a track with a tangible "message", especially one that conveys it without written lyrics. Actually it's pretty rare that any song does that by juxtaposing a mood with an aesthetic. In case you don't get Akiba Time-Lapse, let me break it down:
    Lonely mood, passage of time, etc that's obvious. There's also a sense of not going anywhere -- stagnation. Like some elements fade in and out from the mix but you never move far beyond your starting point. Usually I let my tracks build or travel, but Time-Lapse doesn't. The drums and pianos and flutes drone on and on. All that is set against the cutsey backdrop of "moe" characters counting up the days, weeks, and months of the year. The voices are triggered mechanically, highlighting their artificiality. They often clash over the order of holidays -- there's not only a passage of time, but also confusion and a general irrelevance of time. All this adds up to a theme song for the lonely otaku -- a social satire created from an intersection of the new Japanese aesthetics of mo�Eand the old aesthetics of transience and the "beauty of imperfection". Despite it's melancholy tone it's a rather pretty track, conjuring the feel of changing seasons and whatnot. However I didn't place all those elements with those specific emotional goals. I just made what felt right and what my mood dictated, and only afterwards did I realize the meaning of what I wrote.
    The more I reflected on that, the more I was like "this is what my next album has to be about" -- exploiting Japanese aesthetics and sounds, old and new, in order to conjure up vivid imagery.