Yesterday I shared some ideas on how to grow into recording complex sound effects. The idea was a three step process:
Today I’ll describe how I put those ideas to work in a session I completed last summer: recording the Honda Indy.
I returned to the Honda Indy in July, hoping to put those three steps to work.
Analyzing the Location
I had recorded the Indy before, so I had a good idea of what to expect. However, I those sessions were brief, and I wanted to record more. I revisited my research, and discovered three substantial challenges:
Weather. It would be fairly hot (24–28 degrees Celsius), however the biggest issue was shade. There wasn’t any. It’s only weather, though, right?
Not really. The sun was strong, and merciless. Being cooked for eight hours a day affects performance and produces weaker recordings.
Noise. Problems sounds included loud PA, helicopters, crowd voices, generators, and HVAC. That wasn’t all. The race was held in downtown Toronto. Ambient noise from distant concerts, highway traffic, and a nearby airport threatened every take.
Restricted access. Many of the best sites were off limits. No recording devices were allowed on site, which meant I had to work stealth.
I used the research to map out the location. I’ve marked the hazards on a Google Map, below. The black line is the race path. The blocks of colour show problem areas. Click on them to see what affected recordings there.
The result? Despite the size of the 2.8 kilometre (1.8 mile) race track, I was left with only slivers where I could capture decent recordings.
That was compounded by the fact that I felt as if I was recording in a cage. The noise and security issues didn’t allow much freedom of movement, or an ability to deal with problem sounds.
Analyzing the Cars
I turned my attention from the race grounds to my target: the races and cars themselves.
I learned that the races were part of The Road to Indy. That’s a path a driver takes from junior races to professional ones. Four of the five race types that weekend were part of this circuit.
I brushed up on racing flags to get an idea on the action I’d see from trackside. I found which races included a parade lap and a rolling or standing start. That would help me learn what to expect, move to capture the differences, and list them in metadata when mastering afterwards.
I learned about the cars of each series. This was the fun part.
How do you record race cars, and races? What makes one car different from another? How could I bring personality out of the recordings? What would an editor need to cut in race car tracks?
To pull off the recordings successfully I needed to capture distinctiveness well, all in a location that was mostly out of my control. How could I do this?
I decided to break the session down into voices. What kind of voices?
Each race has strict specifications for the cars: RPM, top speed, weight, design, and so on. Each of this I knew would contribute to a unique sound.
I thought about cars in general, and race cars in particular. I pulled every aspect of their sounds apart into their smallest kernels.
How did I find these sounds? I asked myself these questions:
- How is this car special?
- How is it performing?
- Where is it performing?
- What do editors need?
In the end I came up with five categories, each with many variations.
- Engines. The car type was the most obvious selection. The RPM, horsepower, and speed ceiling was the easiest way to distinguish the recordings. Each race featured a specific car. So, to collect them I’ll I’d need to capture every race:
- Maneuvers. I also wanted coverage. I wanted to document each performance. That would give editors the tools and flexibility to be able to cut however they needed. So, I needed:
- Perspective. Close as well as distant recordings.
- Performances. Each race evoked different performances from the cars:
- Practice. Unpredictable, experimental, and with character as the drivers tested their cars, and their pit crews.
- Qualifying. Measured, with paced passes.
- Race. Eager, emotional, and tense performances. Dense.
- Density. I would record individual cars. That would be helpful to editors. However, I also wanted to capture the energy of the races: the buzzing, jockeying maneuvers as cars battled for a lead. So, I needed to cover:
- Single passes.
- Constant stream.
- Batches or packs/swells and waves.
- Dense packs.
- Progressions: spread out to tight, and the reverse.
I did the math. Fourteen maneuvers for three race type performances, each with five depths of density and two perspectives, all for five race cars. Assuming success on every attempt, that would be 2,100 takes.
What was I getting myself into?
Thankfully I had three days to get the job done.
So, I was armed with all the information I needed. Here’s what I did:
I had the race schedule on my smartphone. That let me know when each car was racing, and which race type was occurring (practice, qualifying, and race). Cars repeated race types a few times over three days, so if I missed one race type I could try again later.
I uploaded the race map to my smartphone, too. I scouted and found quiet pockets around the track that would capture each maneuver. I found a hillside near a straightaway, an isolated corner near the hairpin, and so on. Often this involved striding out in front of crowds, or into unusual places.
I bought a pricey pit pass. That allowed me unlimited access in the pit lane, and some freedom of movement. Not many people wanted to pay $200 for the pass, which was nice, and I had the pit lane pretty much to myself.
I knew the race duration. Sometimes a qualifying heat was only 45 minutes long. That meant I had to move from straightaway to turn, to hairpin, to swerve, to pit stop, and record them all within 45 minutes. It didn’t leave much room for error.
I timed how long it took me to change locations. This was from three to seven minutes. That took an alarming bite out of race time. But the good news was it told me exactly how much time I could spend recording at one location before moving to the next. I had to keep to this schedule to capture all the takes before the race finished.
My most efficient loop was: pit stop (start line), straightaway (after turn two), hairpin turn (turn three), accelerating out of swerve (turn four), decelerating into swerve (turns eight and nine), pit stop (west end, turn eleven).
I created an Excel sheet checklist, and uploaded that to my iPhone, too. I ticked off which cars and locations I had completed. That gave me an overview of what remained to be done.
I also used the slating app MyPlaces. My recorders were overflowing with hundreds of takes. I paired the take numbers with slates on the app, each with detailed notes, photos, and geolocation. This was incredibly helpful, since slating verbally was impossible in the racket.
I captured over 16 gigabytes of raw material during three sun-bleached days. Here are photos of the shoot, and audio examples.
Analysis gave me scope for the shoot. I wasn’t stumbling blindly into the location. I was forewarned about the differences between the cars. Together they presented thousands of takes I could choose from.
Surely one can get lost among that many takes? It’s possible, I suppose. But the amount of takes was also good news. I had as good a chance of capturing distinctive material as hitting the broad side of a barn with my rycote. There was so much variety that I couldn’t help but capture something interesting. That’s one of the immense benefits of using analysis and articulation when recording. Your options expand.
The result? My perspective changed from being swept under a wave of challenges into a simpler task of scooping cool sounds from an endless sea of audio. Instead, the challenge became: could I record it all?
The sound effects hadn’t changed. They remained challenging subjects. But, with a shift of perspective, some analysis, detailed articulation, and time, tricky sound effects became as simple to capture as the doors and traffic around us.
Try it during your next complex field recording shoot. You’ll find you’ll capture more, better sound confidently and easily.