Acoustic Resonant Cavities for Electronic Musical Instruments

If you don't plug in your electric guitar and play a chord, does it sound like an acoustic guitar?  Of course not.  Why does a plucked guitar sound different than a plucked violin of the same pitch?  Why does an upright piano sound different than a grand piano?  In the world of acoustics, we often talk about the physical model of an instrument, or the things that make up an acoustic instrument (size, shape, materials, etc.).  In the electronic world, there are many digital physical modeling synthesizers, where the mode of vibration, materials, shape, size, etc, can be programmed to approximate an acoustic instrument.  That is not what the discussion here is about.  Rather, I will talk about my experiments in the study of acoustic resonant chambers, or cavities, on electronic or recorded sound.

Usually, in electronic music, this task is achieved by filtering and effects.  Here, though, I am going to write about actual, physical, real acoustic bodies and objects to run the sound through.

One of the main challenges to this approach is the transducer.  A transducer is the device that changes the one form of energy into another form of energy - in this case, electronic into acoustic energy.  A speaker or microphone can be considered a transducer, as well as record needles, guitar pickups, piezo elements, etc., etc.

The transfer of energy seen through a resonant cavity can be as simple as a speaker resting on a wooden box, facing toward it, with a microphone placed inside.  Likewise, it is a common studio practice to send a signal into an ampliflier and then mic it back into the system - this is called reamping and when using this technique, the room size and microphone placement have just as much influence on the tone as the type of mic and amp used.

 

Although I use the above techniques from time to time, I mostly make use of a talk box when I run sound into chambers and other spaces, then use various microphones to change the sound back into electronic form.  A talk box is something that is used traditionally by guitarists and inserted into the side of their mouth to make their guitar "talk" - using their vocal tract as an acoustic cavity to shape the sound.

 

Recently, I have been building various small resonance chambers and cavities for this purpose. 

Here are some photos of some cavities that I made out of my mouth using dental alginate and a 2 part silicone rubber.  They are very strong and flexible - one being an "ah" shape and the other being an "eee".

 

I suppose that I could just use my mouth for this purpose, but holding it in a repeatable shape for the duration of a song can be difficult.  This way, I can also stretch and bend these molds in ways that would be impossible to do with my own mouth.

 

Another cavity that I have been using is a simple collapsible chamber that I made with a discarded CDr case, pictured on the right.  I used a silicone rubber margin to make it air tight, but it is still able to move back and forth, changing the dimensions of the cavity.  This sounds like a flanging effect, but smoother, deeper, and more natural, if that makes sense.  I don't have any automation to move it back and forth so it is not really like a flanging sweeping effect - it is more like a fixed flanger that is tunable.  This sounds amazing on white noise. 

So why all the bother?  One thing that I love about actual acoustic instruments is how certain ranges of the instrument change drastically in the tone.  In some cases, such as an oboe or clarinet, the change can be note specific - in other cases, such as a guitar, the change can be due to the area that is plucked or strummed, which string or strings you play, etc.  Using these types of chambers on electronic instruments brings them closer (and sometimes convincingly close) to a natural sounding instrument.  In many ways, I view it as the sonic missing link between the world of electronic and acoustic instruments.

 

Other things that I have tried have a lot to do with sympathetic vibrations, which will have to be a topic for a future post.

Digital Releases GALORE!!!

I have posted 3 new digital releases...  there are 3 different volumes of soundtracks that I have composed for video games, available for purchase through Bandcamp or iTunes.

Be sure to check out these albums and other Travelogue albums.  Links coming soon!

Korg PE-1000 Polyphonic Ensemble

 

 

So the Korg PE1000 is not just a simple string machine.  There are several unique features that make this machine stand out above the other synthesizers out there.  Most of this is due to the fact that every single key has its own oscillator.  There is no divide down circuitry.  There are no top octave dividers.  Most of the other string synths (Korg PE2000, Polymoog, Korg Lambda, Roland RS09, Arp Solina, etc, etc) us these dividers to generate all the notes of the octave from one single oscillator.  In some cases, like the Korg PE2000 or Korg Lambda, there are 3 master oscillators that all the other notes are divided down from so each note can generate the equivalent of a 3 oscillator synth - which sounds very rich with chorus even without the chorus effect on.

 

But back to the PE1000...

 

Something that is not obvious from photos of this instrument or many descriptions is that there is a panel below the keyboard that can be removed which reveals a trim pot for the pitch of every note.  You can tune these oscillators so they are slightly out of tune with the scale.  But even without doing this detune, there are plenty of interesting options that Korg piled into this instrument.

 

For one thing, the LFO or vibrato circuit is very interesting.  There are actually 3 separate LFO's (Low Frequency Oscillators) that are controlled by one knob.  These 3 oscillators affect the pitches of different notes across the keyboard.  You can't control which LFO's go where, but Korg arranged this in such a way that when you play a chord or octave, the LFO's affecting each note are likely to be different and therefore the sound is huge and lush. In contrast with most analog synths, when you have a fixed number of oscillators with a single LFO affecting them together, any chord played will rise and fall in the same exact proportion.  With the PE1000, some notes will be rising while others falling or peaking, etc.



Another thing that makes the PE1000 unique is the Pitch Expand control.  This adjusts the pitch across the entire 5 octave keyboard so that if you play an octave, it is slightly higher in pitch than it should be.  If this is adjusted so it slightly affects the sound, then it can bring the keys so slightly out of tune that it sounds interesting when playing chords across the keyboard.  Furthermore, if you increase the value of the Octave Coupler knob, it will  add the volume of the oscillator one octave above to the key that is being depressed.  So if you use a combination of LFO, Pitch Expand, and Octave Coupler, each note plays 2 oscillators (one octave apart), likely having a separate phase of LFO affecting its rise and fall differently, and detuned depending on the amount of Pitch Expand.  Note that the Octave Coupler does not affect the highest octave of this instrument because there aren't higher oscillators to add to the keys.

 

Another thing that is pretty neat, but subtle, is that divider circuits tend to output pulse waves only.  The PE1000 has a saw wave available from every key.  The saw waveform is one of the most harmonically rich waveforms of traditional analog waveforms.  So when you filter the sound, there is a waveform with lots of harmonics to take out via the filter.

 

The filter (or traveler, as they call it) is a series Low Pass/High Pass filter like that found on a Korg 700/700s/800dv.  Each of these have an individual Peak/Resonance control which leads to many tonal possibilities.  Unfortunately, there are no modulation capabilities except for a subtle envelope control on the High Pass filter only, or an expression pedal input again only affecting the High Pass part of this filter.

 

Although this instrument sounds amazing and is packed with numerous features, there are 3 things that I would change (or perhaps modify at some point if I get around to it):

 

1) While the sustain/release time are comprehensively long and it can slowly decay the note long after the key is released, the attack time has much to be desired.  It does not increase slow enough, there is a slight delay before the rise on the "longest" setting, and sometimes a slightly audible thud can be heard before the attack rise.

 

2) There is a chorus waveform which is kind of a PWM (pulse width modulated signal), but unfortunately, the modulation rate of this waveform is fixed by an internal oscillator without any external control of the speed.  I would much like a control over this in order to alter the rate from a slow morphing waveform to a fast, thick, modulated signal.

 

3) Many of the controls are limited and do not affect the presets.  Preset modes like "harpsichord" or "electric piano" only allow a few of the knobs to alter the sound. Only in the "control" preset do ALL the knobs come to life.  I think that this is because most of the presets can be made with the total control of all parameter knobs.  In other words, I don't believe that there are any special circuits or extra filtering that the preset sounds go through in order to make them.  There are a few presets that I wish that they would have at least given the operator control over the peak/resonance of the filter, but because it in part creates the tone of the preset, they don't allow control over these parameters.

 

 

Philosophy of Live Performance

I will be upfront about this and I apologize if this seems like an attack on anyone.  It is not.  It is simply the way I feel about live performance and electronic live performance, in particular.

First off, I admit that I don't care very much for repetition or wild impromptu in music.  I have been to shows where I didn't care much for the music, but thought that the performance and show were amazing.  Vice versa, I have also been disappointed with a performance from an artist who has music I love....

(a side note: I like songs with some sort of structure, planning, melody, theme, motif, but above all; progression.  I think that interesting music progresses during the song.  This could simply be the transition from a verse to a chorus and back to a verse.  It could be a tempo change or key change or nuance in sound.  These changes don't have to be drastic; they can be subtle.  Something that I try and do with my music/song is to never repeat something exactly more than 3 times.  I feel that the brain gets bored (at least mine does) after 3 times of hearing it.  This is a common psychological feeling - often the reason that there are drum fills and breaks every 4 bars in many songs.)

Electronic music has many challenges in live performance - especially now when the computer or modular synth becomes the crutch of musicianship.  I have seen many, many, many shows where the artist is hiding behind equipment or a computer screen and it is not obvious what the artist is actually doing live?  Are they even playing music?   Maybe for the DJ/ dance culture, this is an acceptable format for a "live" show.  I feel that for myself, though, it is not something that I would feel comfortable doing and I don't especially enjoy seeing shows that are like this.  I view live performance to be literally performing something live. 

Most shows that I play require some backing tracks to fill in the rhythms and sounds that are critical to identify a certain song.  Some important challenges come along with this that may or may not be obvious.  Which sounds are on the backing track and which sounds will be played live.  What equipment do you bring to a show to play these live sounds? How much of the performance is props and gimmicks and how much is actually playing a musical instrument?  Vocals aside, how does one put together a set that will capture the audience's attention - even when they are not your friends, but strangers that just happened to catch your set for the first time and have never heard of you before?

Live performance is a delicate balance between playing the music that you create and giving the audience something to watch and hear.  Sometimes the artist sounds so amazing live that just hearing their voice or seeing them play their instrument is captivating.  Sometimes it is more about the visual effects that the band/artist puts on the screen behind them (not that there is anything wrong with this approach, but unless the video/film/slides/photos/images were actually made FOR the band, I feel that this is sort of a cop-out - where the band is sort of taking credit for something that doesn't have anything to do with them [(for this same reason, I rarely play cover songs of other peoples songs)] ).

Live performance is not gimmicky - if you bring a prop or special instrument, it is best to not overuse it and be subtle about it.  Use it so it works well with the song context or space between.  Drawing extra attention to it or calling it a special name on stage is just silly.  No matter what I bring to a show, I tend to follow a rule of not using anything more than two songs during a set.

Live performance should be transparent.  This means that the artist should be doing SOMETHING that the audience can see.  Preferably, this something should change the sound of what they are hearing rather than just being a visual stimulus. This also means that the artist should not be hiding behind equipment.

I should also point out that a live performance should not be endless noodling or soloing around on an instrument while making an expression with your face as if you were constipated.

Come to our show in Cleveland! Sunday 12-28 at the Happy Dog

Five Star Hotel, Juggable Offense, Sodalite, and Travelogue perform at Happy Dog on 12-28!!!  Don't miss this show!  (the other dates on this flyer is the mini tour that Juggable Offense and Five Star Hotel are performing)

Akai S20

 

 

The Akai S20 was the first sampler that I had bought.  I had played with a myriad of samplers that belonged to various friends before this - some were even borrowed and in my studio for a time, but I never had owned a sampler of my own until I bought the S20. 

 

 

The S20 is a decent sounding 12 bit sampler with very limited sample editing capabilities.  There are no filters, envelopes, LFO's, etc, etc.  There is only start time, end time, loop  points, and various MIDI and button assignment functions.  I actually like how basic it is.  It keeps me from spending hours of time over-analyzing the sound...

 

During the late 90's when I used this a LOT, I mostly used it to sample analog sounds around the studio (mostly drum sounds) and then sequence these using my Ensoniq ESQ-1.  (the S20 does have a built in sequencer but it is only 1 track with no obvious quantization)

 

My unit has updated memory so it can store a whopping 5 minutes mono (yes!  count them... FIVE!!!) or 2.5 minutes in stereo.  This then could be stored in memory but was lost once power was off unless you backed this to disk.  The main drawback to this unit is it only uses floppy disks for memory. 

 

Though I have other samplers now ( like my sampling software on my computer or iPad, or like my Emulator 4XT), I still find myself using this sampler from time to time - mostly because of its portability and no-nonsense architecture.  Recently, I used it live at a show with my Arturia Minibrute.  It was MIDI'd to the Minibrute, then the output was ran into the external input of the Minibrute.  The memory of the S20 was loaded with various waveforms at different octaves - sometimes just a simple saw or analog waveform to thicken up the Minibrute, sometimes string sounds or sine waves at different octaves so I can add various overtones to the Minibrute.  This has brought a new purpose to this old piece of equipment. 

Travelogue Performs at Happy Dog on October 4th

Golden Streets of Paradise
High Praise
Travelogue

All performing at Happy Dog on October 4th.

The Happy Dog - 5801 Detroit Ave, Cleveland, Ohio 44102