Update: Finding out that this post is being read by those curios, but not having previous knowledge of synthetic sound design, I added links to definitions and resources throughout this post, to inspire further educational research on the subject. An MP3 sample was also added, as I discovered WAV samples had an issue with playback using Firefox. Thanks for the feedback!
I thought I would create a post that somewhat details the sound design process, or at least gives you an idea of how I think and what steps I take during that process. As I’ve mentioned numerous times here on this Sound Design Blog, I love pushing a Synthesizers sound architecture to see if I can actually manifest what I hear in my mind. As you probably guessed, it doesn’t always work as planned.
I consistently run into virtual brick-walls because of the limitations some Synthesizers have. It’s probably why Sampler technology is on top of the sound design pedestal. I mean, if you want chirping crickets or ocean waves wouldn’t it just be easier to record, loop, and map them using a Sampler? Well of course it would. But then it wouldn’t be challenging or fun right?
I had this idea to create an atmospheric pad in Fathom using natural sounds, combined with additive harmonically contrasting tonalities that blended well together, and the final timbre resulting in a cinematic ethereal atmosphere. In my mind, this idea would take about 3 sound layers. I say layers and not oscillators because a layer can have multiple oscillators to get the desired end result. I chose an easier natural sound to build the entire patch around, to gain a better understanding of how to build a more complicated natural sound using Fathom in the future.
This technique has worked well in the past. I once created a realistic Thunder Storm that included distance and intensity using Z-Plane filters morphed through dual function generators in the E-MU Ultraproteus as an end result of starting simple, so I know this is very possible.
Like all true Synthesizers, there are zero Samples used in Fathom. Fathom generates its own waveforms. I must say that creating naturally based sounds using pure synthesis methods will give anyone who succeeds the fundamental understanding of how the very nature of sound works.
So what natural ambient sound would be the “easiest” to start with? Synthetically speaking, there are two …Wind and Water. I chose Water …Specifically, Ocean Surf. I’ve spent countless days running on the beach with children, dogs, and lovers. I know this sound well.
Sound wise, Ocean Surf is considered a variant combination of white, pink, and brown noise (click here for more on the color of noise). This depends on what time, position, and cycle the water is in. Although natural noise can be a complicated structure, I narrowed it down to 3 simple water cycles: Background ambient noise, approaching waves, and beach caressing foam. While the background ambient noise would remain consistent, the approaching waves and beach caressing foam had to cycle seamlessly at slightly parallel intervals for realism.
Fathom happens to have the most detailed noise generator of any Synthesizer I have used, giving me the ability to “surf the perfect wave” 😉 The noise generator uses step, linear, smooth and turbulent algorithms combined with frequency ranges, tracking, grains, detuning, and filters. Not your average synth noise generator to say the least.
I created this first sound layer using 3 noise oscillators, 1 for each surf cycle. The “brownish-white approaching waves”, and “pink beach caressing foam” cycles were both modulated by their frequencies through 2 low frequency oscillators that were timed at parallel intervals, while the “whitish-brown background ambient noise” remained static with slight frequency variation to give it a virtual overall distance effect. And the remaining 2 layers of this patch were centered around this “Ocean Surf” atmosphere.
Fathoms Noise Generator x3 for 3 Ocean Surf water cycles, being timed by modulation
For the second layer of this patch I wanted to create an ethereal pad. I used Spectral Harmonic Additive Synthesis to accomplish this. I needed this pad to have two qualities: A vocal synth choir “Ah” sound, with a smooth “Analog low end” tonality. This required 2 spectral oscillators as a fundamental source, which doubled in octave level pitch variance, resulting in a total of 4 oscillators for a true ethereal sound.
Additive Oscillator 1 Waveform, “Ah” & “Analog low end” x2 using 2 harmonic octaves
Additive Oscillator 2 Waveform, “Ah” & “Analog low end” x2 using 2 harmonic octaves
The third and final layer had to have a contrasted tonality that would sonically harmonize with the second layer pad, and at the same time have ethereal ambiance for a perfect balance between the 2 layers. Once again, I used Spectral Harmonic Additive Synthesis. I decided that the best tone for this harmonic contrast would be a “bellish-chime”.
I also decided to use a modulator so that this chime could change timbral characteristics, using the Mod Wheel to create an atonal temperament. Like layer 2, this layer also needed an octave level pitch variance to create a true contrasting ethereal sound. Layer 3 ended up requiring 2 oscillators to achieve a satisfactory end result with all 3 layers combined.
Additive Oscillator 3 Waveform, the “bellish-chime” x2 using 2 harmonic octaves
With all 3 layers finished, I began finalizing this patch by planning an optimized ambient performance structure. Layer 1, being an atonal noise source, would not perform well over the entire chromatic range combined with the other 2 layers. By the same token, layer 3 only sounds harmonic when the Mod Wheel is positioned as such. But layer 2 sounded perfect as it stood across the entire 88-key chromatic range.
What I did was design layer 1 as a split sound source between C1 and C4 (Middle C), and layer 3 as a split sound source between C4 and C8, while layer 2 was consistent across the entire range. But that’s not all. To optimize this performance planning, I also made layer 1 consistently audible, while layer 2 fades out when you move the Mod Wheel up, and layer 3 is consistently audible, and also changes timbre when you move the Mod Wheel up. Performance optimization complete!
If you’ve done the math thus far, you now realize this patch is now a 9 Oscillator Additive Beast! And it was created in the true form of Additive Synthesis …There were zero component level subtractive filters used on any oscillator. You would also think that 9 Oscillators might be too much for a computers CPU to handle. Well, at the moment I’m using a PC that was made 5 years ago, and this patch performs great so far. The hardest part for me …Sound levels. Mixing this many contrasting sound source variations wasn’t easy. All levels had to be optimized independently at a micro level, to avoid audible Oscillator cross distorted saturation.
If this sounds confusing, all of this will become crystal clear when you listen to the brief audio demo. The demo starts with the Mod Wheel up, then transitions down, then ends back up to finish the demonstration, and was created as a very simple, and slowly played “2 finger” track so you can differentiate the sound sources. Zero external audio processing was done outside of Fathom. I hope this Blog Post was helpful for anyone that might be wondering what my design process was about Here’s what this patch sounds like:
Sample 1: Bitrate 2822kpbs Wave File (Verified working in Chrome – Higher Quality)
Sample 2: Bitrate 320kpbs MP3 File (Verified working in Firefox – Lower Quality)
“Shimmering Shoreline’s” overall patched configuration, which includes internal FX