[Written by Tim Walston for Designing Sound]
… And don’t let it get away! One of the things I love most about sound design is the challenge of creating something new. What am I going to make for that on the screen? What the heck does this sound like? How am I ever going to make the sound I hear in my head… come out of these speakers? What do I do if I have no sound in my head yet?
Sometimes we just have to experiment, to explore. This sonic exploration is pure creative impulse, not an exact science. Once you decide on a path to try, you should follow it and see where it leads. Sometimes you achieve your goal, and sometimes you hit a dead end. If so, then you have to try something else. People I admire, cleverer and more experienced than me, have complained: “Why couldn’t I have just [followed the correct approach] first – and saved all the hassle and days of failed experiments!” We all laugh, and then we shake our heads because we’ve all been there. The thing that some people in the non-creative parts of this business don’t understand is that, sometimes, sound design is a journey. It’s an evolution. And sometimes it is possible to never be done. Sometimes it’s only finished when they wrestle it out of your hands and send it off to the mix stage. Times up – it’s now or never!
Getting back to the journey idea now… I’ve found the side trips and detours that sometimes occur to be incredibly fruitful and educational. Personally, I don’t often save mixdowns of “final” sound effects composites. Usually they are too custom tailored to the project, or (hopefully) too unique to be used in another film. What I create and save are the elements I’ve made along the way. These building blocks can be reused or repurposed or reprocessed again later.
The most important lesson for me when making these sonic explorations is that I must follow them to their conclusion, while I’m in the mode, and see what they yield… whether it’s relevant to the current task or not. This is important for two reasons:
First, while experimenting with a new process or approach, I will generally create a lot of interesting material along the way. These “happy accidents” can be an unexpected gold mine of material. Most of the time, I will not EVER get back to that exact frame of mind, with all of those plug-in or hardware settings – even though it’s theoretically possible nowadays. I’ve got to capture the inspiration before it disappears. That way, even the wrong turns are productive – if not for the current project, then for something in the future. Record now – ask questions later!
For example, one day in while working on a show in 2008, I was trying to create a believable “skid” sound for a hovercar (that doesn’t touch the ground, of course, but I digress). I was experimenting with a phaser plug-in, and I discovered by accident that by cranking the input, output, resonance, and mix up all the way, I could generate a sweeping tone that oscillated up and down. I think I needed to feed a little sound into it to seed the feedback – and then it would go for several minutes before fading into nothing. I adjusted the modulation shape and immediately thought: “this sounds like a siren!” I found I could vary the sound by adjusting the Mod depth and rate. I started recording the output immediately. I got stuff like this:
[soundcloud width=”100%” height=”81″ params=”” url=”http://api.soundcloud.com/tracks/9432741″]
If I had the time, I might have edited the source files and tried to worldize it, but I didn’t. So I spent the rest of the afternoon making loops and edits of the material and then creating a patch using a convolution plug-in to try to replicate the sound of the dry tone being projected first through a horn, and then into a reflective urban environment. I looped some steady sounds and made slow Doppler pass bys. I was trying for retro/futuristic distant sirens to add to the backgrounds. I ended up with things like these:
[soundcloud width=”100%” height=”81″ params=”” url=”http://api.soundcloud.com/tracks/9432809″]
The resulting fabricated sirens aren’t completely 100% convincing, but for an animated sci-fi movie I thought they added a nice little flavor. I handed a bunch of them over to the background editor. I don’t know if they’re in the film or not, actually, but they were fun to make and they started out as an accident.
Second, I always try to follow a new sound exploration to its conclusion because you never know what is just around the corner. Sometimes, after trying unsuccessfully to create the sound I hear in my head, when I’m about to give up, I’ll try one more time… and get it. This happened to me on Star Trek, but I’ll discuss that in another article. For now, I’ll illustrate with another sound.
In 2003 I worked on a film that had monks with mystic powers and a bad guy who developed his own evil powers. At some point in the film, there was a battle of “energy” coming from their hands… I got the idea in my head that I wanted the clash of supernatural energies to be like the sound of two light sabers from Star Wars when they are held together during a fight. A crackling energy… I could just hear it in my minds ear!
So I tried a bunch of things. The monk power was supposed to be organic and nature-based so I experimented with thunder. I start making some rippling energy sounds by stacking several thunder effects on separate tracks – each with its own doppler modulation running continuously at different rates, and some with envelope-controlled eq filters. The sum of these tracks went to a bus with another doppler and some compression etc. On the last doppler effect, I moved the “puck” manually to perform the doppler shapes I wanted – one of my all-time favorite tricks. The entire signal chain was inspired by my friend Charles Deenen. It’s similar to his “100 Whooshes in 2 minutes” video, but simpler. I got results like this:
[soundcloud width=”100%” height=”81″ params=”” url=”http://api.soundcloud.com/tracks/9433020″]
Interesting… but I wasn’t getting the sound I wanted. I recorded nearly 5 minutes of stuff like that, recorded into 3 long sound files. I tried adjusting the parameters and plug-ins but I just wasn’t happy. I was about to give up and try something completely different, when I thought I’d throw those sounds into a sampler. I think the effects chain also included the hardware RSP Saturator, which SoundStorm had at the time, a delay and maybe the old extinct Hyperprism Bass plug-in. I performed radical arpeggios on the keyboard – ranging from high to low, with ridiculous wild pitch bends. Finally I was getting somewhere!
I found what types of performances gave me pleasing results. I got very active and nicely overdriven “steady” energy sounds while holding down several spaced out keys. I got more percussive, explosive shots with shorter, rolling key combinations. I recorded performances for a while, then moved the sample start marker further down my source file and recorded more. I think I might have even taken THAT source and loaded it back into the sampler to perform again. In the end I recorded about 20 minutes of varied source that has served me well over the years. Here are some snippets:
[soundcloud width=”100%” height=”81″ params=”” url=”http://api.soundcloud.com/tracks/9448955″]
By processing the results of my original experiments one more time, I finally took the extra step to reach my goal. If I had given up, I’d have missed out. Since I saved the intermediate step’s results (the thunder doppler stuff) I now have the ability to take that material in a completely different direction in the future.
The funny thing is, when I later listened to the light saber crackling energy sound, it was VERY different from my memory! I guess that’s a good thing. What I was chasing was the sound in my head, not a sound effect rip-off.
Sound memory is not always very accurate. When I was a kid, I had an Evel Knieval Stunt Cycle toy, with a wind up gyro launcher thing. My memory tells me it was the coolest sound – kind of like the inertial starter device that Ben Burtt used for the Millenium Falcon. In real life – who knows. Anybody have one I can record?