The nominees for the 2017 MPSE Golden Reel Awards recently came out, and I’m very happy to report that we got nominated for the fourth episode of Startup! Super happy about this — I really think the entire team did great work on the episode.

It’s also personally rewarding since from a sound design point of view it was by far the busiest and most challenging episode of the season; on top of a slow-mo gun sequence there’s a big exterior shootout as two rival gangs confront each other. That scene in particular was a ton of work since I designed most of the guns from scratch and tried something I never did before: a Heat-style treatment.

Whenever I work on a project that involves guns, there’s a good chance the director will mention the Downtown LA shootout in the movie Heat as a reference. And for good reason: the guns in the sequence sound simply insane and are incredibly scary. What many people don’t know, however, is how that sound was accomplished: all the gun shots are from the original production sound recordings! After spending weeks building the sequence sonically in post and carefully replacing each gun shot, director Michael Mann had the sound team go back to the original production sound recordings. Re-recording mixer Chris Jenkins talks more about it here:

So when director Ben Ketai told me he wanted the sequence to sound like the on in Heat, I got excited; I knew they had used real guns firing blanks on set, so for once there was at least a chance that there would be usable stuff in production roll — no other project I worked on before this had ever used real guns on set. More importantly, it would also match the gritty sonic aesthetic of the show. If ever there was an opportunity to use gnarly, distorted and noisy guns with long tails instead of the short, clean and hyper-detailed modern sonic designs, this was it.

And lo and behold, there was some great stuff in sound roll. The only problem was that there wasn’t nearly enough to cover the entire sequence; most of it was either too far gone in terms of noise and distortion (even when fully embracing the aesthetic) or the guns were way too far away, which just made them sound mushy. So instead, I ended up going with a hybrid approach: for each gun design, I combined a production sound gun recording with one (or multiple) clean designs, either from libraries or again designed from scratch. All of that was then fed through compression, distortion and/or transient shaping to yield the final designs. Fabfilter Pro-MB proved invaluable here to control the tails (which really started getting out of control). The result was a mini library for the sequence that had the grit and acoustic realism of the location while adding some extra punch and clarity (I hope).

The final step was to create more distant versions for some of the guns; towards the end of the shootout, our hero runs off while the gun battle continues in the background.  Most of that was just about flattening out the transients and bringing out the tails more.

Overall, I’m pretty happy with how the sequence turned out, especially given the insane time crunch we were under. You can check out the entire show for free on Crackle here. If you’re interested in the gun sequence, it’s in episode 4 roughly 36 minutes in.

Last summer, I went on a trip to Olympic National Park. One highlight among many was a visit to the One Square Inch of Silence. Located in the Hoh Rain Forest, it is one of the quietest places in the US according to acoustic ecologist and field recordist Gordon Hempton (of Soundtracker fame). It is symbolically marked by a small red stone:

The website explains the project better than I ever could:

One Square Inch of Silence is very possibly the quietest place in the United States. It is an independent research project located in the Hoh Rain Forest of Olympic National Park, which is one of the most pristine, untouched, and ecologically diverse environments in the United States. If nothing is done to preserve and protect this quiet place from human noise intrusions, natural quiet may be non-existent in our world in the next 10 years. Silence is a part of our human nature, which can no longer be heard by most people. Close your eyes and listen for only a few seconds to the world you live in, and you will hear this lack of true quiet, of silence. Refrigerators, air conditioning systems, and airplanes are a few of the things that have become part of the ambient sound and prevent us from listening to the natural sounds of our environment. It is our birthright to listen, quietly and undisturbed, to the natural environment and take whatever meanings we may from it.

The “Silence” at One Square Inch is not a literal one; in fact, I was surprised by how present the environment around me sounded. Instead, Silence refers to how untouched by human noise the natural soundscape is. In my twenty minutes of quiet listening, I couldn’t hear a single manmade sound — no cars, no planes, no hikers. All I heard was wind, wood movements and the distant Hoh River. I went in the middle of the day, but if you’re a morning person you might want to make the trek early in the day when the soundscape is more alive. It’s just off of the Hoh River Trail (which is a pretty easy trail). The map on the One Square Inch website has more details on how to get there.

I couldn’t help making a recording as a memento…

The website also mentions a Jar of Quiet Thoughts where visitors can leave notes. Sadly, I couldn’t find it — it might not be there anymore.

If you’re interested in finding out more about Gordon Hempton and his efforts to preserve natural soundscapes, here are two amazing podcasts that interviewed him:

Tonebenders – Episode 16: Gordon Hempton
On Being – The Last Quiet Places: Silence and the Presence of Everything

The great thing about sound design is that broken stuff suddenly becomes usable and beautiful in its own right. The other day, the charger of my laptop broke — it just wouldn’t charge anymore and emitted a weird buzzing sound. So naturally I got out my telephone pickup coil and attached it to the charger. Turns out the buzzing was a little swan song:

Rest in peace, little overpriced piece of technology.

For those interested in the tech behind it, a telephone pickup consists of a coil that can “sense” electromagnetic fields: a nearby changing magnetic field will create current in the coil (if it is part of a closed loop) through a process called electromagnetic induction. As such, telephone pickups work differently from regular microphones: while microphones convert acoustic energy (pressure changes in the air) to an electrical signal, coil pickups convert the energy of a magnetic field to an electrical signal. Since any electronic device will create an electromagnetic field proportional to the current flowing through it, you can stick a telephone pickup on any electronic gadget to explore its hidden electromagnetic world. The resulting sounds are a great source for any type of glitchy, techy or digital-sounding sound design. I’ve used coil recordings in the past to simulate the sound of weird static, transmission issues and glitches.

I use the cheapo Radioshack telephone pickup coils that look like this:

You can find them on Amazon. Jez Riley French also sells some modified ones on his website.

There are many reasons to love MS. But here’s another reason why I pretty much always record in MS, even if I never intend to use the recording in stereo: worldizing! Think of MS as giving you two different-sounding mono versions of your recording: one fully on-mic, the other fully off. The two channels of an XY recording, in contrast, tend to sound fairly similar since both microphones are equally off-axis from the source (just once to the left of the mic and once right). You can see the difference here:

In interior spaces, the side channel (pictured above in red) will give you a roomy version of your signal while the mid channel provides a dry (or at least drier) signal. Assuming the space you recorded in is similar to the space you’re cutting for, you can use each channel individually to cut for perspective. For off-screen sounds, for example, I often just cut in the side signal. You can even mix both the M and S signal together to dial in the perceived roominess of your signal. Think of this technique as creating a virtual microphone that can be rotated around: the more mid signal you mix in, the more the virtual microphone rotates from fully off-mic (pointing to the left in the picture above) to on-mic. If you mix M and S equally, you essentially recreate the left channel of the equivalent XY recording (since X = M + S), which points halfway between fully on-mic and off-mic.

Here’s an example from my bathtub recording session:

Mid only:

Side only:

Both mixed together:

For most small to medium-sized rooms the resulting effect is fairly subtle (which I like) and to my ears more realistic-sounding than just using artificial reverb. But that doesn’t mean you can’t combine both techniques. In fact, to make things sound far away in a room or off-screen, I’ve found that the side channel often responds better to artificial reverb than the mid channel; since there is already some roominess baked in, you can get away with a more restraint reverb treatment. In comparison, if you start with a dry mid channel, you usually have to push the reverb more, which often betrays its artificial origin (even when using high quality reverbs). But as usual, all of this is subjective and depends on what effect you’re going for and the style of the project you’re working on. I’ve had good luck with this technique on some projects, but there were others where the results just sounded too raw and gritty. Realism isn’t everything.

There are also some drawbacks to using this technique. The most obvious one is that you can only use it if the room you recorded in matches the room you’re cutting for. Additionally, the signal-to-noise ratio is usually worse on the side channel than on the mid, so you really need a pretty quiet room for this to work.

Just stumbled across some pictures on my phone from a hilarious recording session a couple of years ago. The project required some very specific water movements in a bathtub. Picture the scene: two characters in a bathtub, making out, before one tries to drown the other. All in very foamy water. Not exactly something you can easily find in a library. So, perfect excuse to put on my swimsuit, channel my inner child and splash around!

Here’s the before picture (not pictured: the mayhem after):

The trusted MKH8050 / MKH30 MS combo at work

Since there wasn’t any space for a monitor, the plan was to record everything wild with plenty of variations for each type of sound. Here’s a rough list of everything that was needed:

• General small foam and water movements, from slow and quiet to fast
• Trickles: sparse and dense
• Surfacing and plunging movements
• Underwater hits against tub
• Air bubbles and underwater screams (turns out I can in fact scream like a woman under water)
• Splashes as character thrashes around

As expected, all recording discipline went right out the window within minutes of splashing around; so instead of a set of nicely voice-slated files, categorized by type and intensity of movement, I ended up with about 20 minutes of total recording time more or less randomly spread out over 15 files. Looking forward to rejiggering and tagging this mess before I put it in my library…

Another problem was the foam. Not only is it very quiet to begin with, but it also stops making any sound whatsoever after the first couple of minutes. The only solution? Creating more fresh foam by pouring in more and more body wash. I’m pretty sure by the end of the session there were more chemicals than water in the tub — my skin can attest to that.

All in all, just another day at the office.

UPDATE: Funnily enough, somebody uploaded that specific scene to youtube. Here’s the final result (Warning: most definitely NSFW):

[youtube https://www.youtube.com/watch?v=ouesPYdFavI&showinfo=0?&theme=light ]

A couple of years ago, we had company from a little mockingbird on our street. Not sure if the little guy got lost, but it decided to camp out in the tree right outside my house and serenade the neighborhood all night (quite loudly so). So naturally, one night I got out my recording gear, tiptoed to the tree, and… scared the little guy off. Luckily, it came back the next day so I set up my gear even more stealthily and further away. This time, I got more lucky: I got a 20 minute recording of the little bird chirping and singing away. Listening back to it, it’s incredible how varied the individual phrases are; over that twenty minute period, there are very few repeating patterns. Really makes you wonder how these birds communicate and what those phrases mean.

Here’s a short snippet of some of the sounds (with some traffic low-end filtered out). Watch your levels:

The sounds are pretty cool in their own right, but just imagine the potential for manipulation! Pitch shifting alone can give you a wide range from weird creatures to alarms and beeps. Here’s a random selection, pitched anywhere from 1 to 4 octaves down and 1 up:

And here’s a weird alarm I made while playing around with the recording:

This is what happens when you naively start feeding one seagull… Apocalypse!

…of microphones. Just got these two babies in the mail:

MKH8040 matched stereo pair!

Two matched Sennheiser MKH8040s! They’re the cardioids in Sennheiser’s 8000 series lineup, which I can’t praise highly enough. I was originally in the market for just one MKH8040 as another, more standard mid microphone option for my MS rig. But the deal on this matched pair was just too good to pass up, especially considering how many additional recording setups I can get out of them: in addition to MS, I can also use them in an ORTF array or even to frankenstein together a Double-MS rig for surround recordings. I’m especially excited about ORTF for ambiences — I was never really too happy with my MKH8050/MKH30 combo for backgrounds. Sadly, ORTF will be an indoors-only affair for now since I don’t have proper windshields for them yet. And given the prohibitive cost of Rycotes, that might take a while…

Whenever I have downtime, I try to chip away at the daunting task of going over my old recordings and tagging them for my library. Every now and then, I stumble across a recording that instantly transports me back in time and space to the actual recording session. For this particular sound, it was by far the scariest session I’ve ever had…

I went through a phase a while ago where I would go out at night and more or less randomly explore LA. One of those adventures led me to a pond in the middle of a park. I’m sure during the day it’s a perfectly nice and family-friendly place, but when I went there at night it had transformed into the stuff of nightmares: the place was pitch-black, surrounded by trees and bushes, with not a single soul around (I hope). The bushes were constantly moving back and forth, swayed by the breeze (I hope). The pond was eerily still save for occasional sudden water splashes when fish came up (I hope). Other than the distant drone of LA freeways, the only sounds I heard were the crackling of the bushes and the little splashes in the pond. But what really made me jump was the sudden screech of some animal that sounded like a pained human scream. Needless to say, I got out of there pretty fast. (I’m fairly certain it actually wasn’t human; I checked before I made my way back to my car.)

On a project a couple of months later, I got a chance to revisit that place. I was working on a scene that required the pitter-patter sound of gentle rain on a pond. Nothing in my library really worked, so I decided to go out and record it. As luck would have it, one night it was raining the perfect amount (and if you live in LA, you know that getting any rain is rare enough to constitute actual luck). Not knowing any other ponds around, I went back to my nightmare pond with my field recording kit. Not quite sure what made me think that was a good idea, especially since I didn’t go out until 1am. At least this time I brought flashlights (yes, multiple). The rain had turned the dirt path into a muddy wet mess, so much so that I slipped and fell multiple times. Towards the end, where it got really steep, I was essentially crawling, holding on to the ground and nearby bushes with my hands so I wouldn’t slip back down. By the time I got to the pond, I was drenched in mud and sweat.

Once I had caught my breath I set up my mics. Since I wanted to get the sound of individual drops hitting the pond, I had to get my rig as close as possible to the surface. The only way to do that without risking a nosedive into the water was to set up the stand on the ground of the pond (which, thankfully, is fairly shallow near the edge). Of course, wet soil doesn’t make for the sturdiest surface and the stand was leaning precariously, so I had to hold on to it while remaining as motionless as possible to avoid any handling noise. That turned out to be harder than expected; I constantly had to fight down the urge to turn around and look behind me and to jump at sudden sounds. So there I was: wet, scared, with shivers down my spine and cramps from standing still.

But man, was it worth it! Check out one of the recordings. Unfortunately, even at 2am there still was annoying traffic in the recording, but it responded surprisingly well to drastic EQing: (fellow sound nerds: see below for technical particulars.)

On my way out, I fell once again and slid a good amount downhill, losing one of my flashlights in the process. Surrounded by darkness, I fumbled around my bag to get out my backup flashlight. When I finally turned it on, I saw a gigantic spider a mere foot away from where I had planted my butt. I swear I’m not making this up.

I don’t think I’ve ever been this happy to see my car. (I was considerably less happy the next day when I had to clean it.) And I got another recording! Just some rain on leaves and crickets:

Overall, this was definitely one of my most memorable recording sessions ever. In hindsight, there probably would have been easier ways to get a similar recording, but none as frightening and fun at the same time.

Tech Details

I recorded with my usual MS setup consisting of an MKH8050 and MKH30. It’s interesting listening to the mid and side channels separately; the mid picked up mostly drops on pond (since that’s what it was aimed at), while the side mainly consists of crackly rain on leaves from the bushes around the pond. For rain protection, I used my Rycote Windscreen with the fur on and a couple additional towels wrapped around it. I’ll probably get some Hog Hair next time, but the towels work well enough for now. But the real winner of the session was my field recording backpack. It performed admirably in very adverse (read: wet and dirty) conditions. Couldn’t be happier with it. Of course, I still have to clean it…

Recording around Los Angeles can be a challenge. Not only is it essentially impossible to escape traffic, but the flight patterns of LAX also further complicate things.

After getting thwarted by planes numerous times when I was recording in Angeles Forest, I decided to do some research on flight patterns to see if and how I can best avoid plane noise. It turns out LAX has a realtime flight tracking and noise map online. You can find it here: http://webtrak5.bksv.com/lax4. While it only shows noise information for the immediate area around LAX, it tracks all current flights over an area much bigger than that (including parts of Angeles Forest). Looking at individual flights as they slowly move across the screen is not very useful, but if you have too much time on your hands and stare at it long enough and count planes (like I did in the last couple of days), you’ll start discerning patterns. The western part of Angeles Forest pretty much sucks for recording.

Or you can save yourself the effort and just look at the aircraft traffic flow at LAX (taken from here):

Westerly Traffic to and from LAX (daytime operation)

Pictured here is what LAX calls “Westerly Traffic,” which is the normal LAX daytime flight pattern. (Their definition of “daytime”, by the way, is a bit odd: 6:30am to midnight.) See that bundle of blue lines that comes in from northwest right next to Burbank Airport? Look how nicely it cuts through the west end of Angeles Forest. And as Google tells me, “turboprop” planes are nothing but propeller planes, which is exactly what I hear leaking all over my recordings! Not pictured here are arrivals and departures from any other airports — and while Bob Hope Airport in Burbank is minuscule compared to LAX, I’m sure it contributes to further noise contamination in that area.

Things do look a bit better at night, when LAX is in “Over-Ocean Traffic” mode:

Over-Ocean Traffic to and from LAX (nighttime operation)

At night, LAX switches to “Over-Ocean” traffic flow. Arrivals now occur from the west over the ocean to mitigate the impact of noise on the communities around LAX. So if you don’t mind being alone and afraid in the wilderness at night (as well as potentially being eaten by animals), this is a great time to record.

But no matter what you do, better hope LAX doesn’t switch to “Easterly Operations” (which they do depending on wind conditions), where aircraft both arrive and depart eastbound:

Easterly Traffic to and from LAX

Yikes.

In case you haven’t had enough yet of maps, here’s one showing the impact of aircraft traffic noise on different parts of LA. You can find the original as well as an interactive map on Stephen Murray’s blog here.

Moral of the story: move to Alaska if you want quiet outdoor recordings.